vvEPA
United States                              November 2012
Environmental Protection                       -rmisnnm m/mc/-^
Agency                               EPA/600/R-10/075 C
  Integrated Science Assessment
                    for Lead

          (Third External Review Draft)
            National Center for Environmental Assessment-RTF Division
                 Office of Research and Development
                 U.S. Environmental Protection Agency
                   Research Triangle Park, NC

-------
DISCLAIMER

This document is the third external review draft for review purposes only and does not constitute
U.S. Environmental Protection Agency policy. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
November 2012                            ii                  Draft - Do Not Cite or Quote

-------
CONTENTS
      LEAD PROJECT TEAM	xxii

      AUTHORS, CONTRIBUTORS, AND REVIEWERS	xxv

      CLEAN AIR SCIENTIFIC ADVISORY COMMITTEE  LEAD NAAQS REVIEW PANEL	xxxi

      ACRONYMS AND ABBREVIATIONS	xxxii

      PREAMBLE	xlvi

                        Figure I     Illustration of the key steps in the process of the review of National
                                   Ambient Air Quality Standards.	xlvii
                        Figure II    Illustration of processes for literature search and study selection
                                   used for development of ISAs. 	xlviii
                        Figure III    Characterization of the general process of ISA development.	///
                        Table I     Aspects to aid in judging causality.	Ix
                        Table II     Weight of evidence for causal determination.	 Ixiii
           References for Preamble	 Ixix

      LEGISLATIVE AND HISTORICAL BACKGROUND	Ixxi

           References for Legislative and Historical Background	Ixxvii

      1     EXECUTIVE SUMMARY	1-1

           1.1  Introduction	 1-1
           1.2  Sources, Fate and Transport of Lead (Pb) in the Environment, and the Resulting Human
               Exposure and Dose	 1-2
                        Figure 1-1   Conceptual model of multimedia Pb exposure.	 1-4
           1.3  Integrative Overview of Health and Ecological Effects	 1-5
                        Figure 1-2   Schematic representation  of the relationships between the various
                                   MOAs by which Pb exerts its effects.	 1-6
             1.3.1  Health Effects of Pb	 1-6
                        Table 1-1    Summary of causal determinations for the relationship between
                                   exposure to Pb and health effects.	 1-7
             1.3.2  Ecological Effects of Pb	 1-9
                        Table 1-2   Summary of causal determinations for the relationship between Pb
                                   exposure and effect on plants, invertebrates and vertebrates.	 1-10
           1.4  Policy Relevant Considerations	 1-13
           1.5  Summary	 1-15

      2     INTEGRATIVE SUMMARY	2-1

           2.1  ISA Development and Scope	 2-1
           2.2  Ambient Pb: Source to Concentration	 2-5
             2.2.1  Sources, Fate and Transport of Ambient Pb	2-5
             2.2.2  Monitoring and Concentrations  of Ambient Air Pb	2-6
             2.2.3  Ambient Pb Concentrations in Non-Air Media  and Biota	2-8
                        Table 2-1    Ambient Pb concentrations in non-air media and biota considered for
                                   ecological assessment.	 2-9
           2.3  Exposure to Ambient  Pb	 2-10
           2.4  Toxicokinetics	 2-11
           2.5  Pb Biomarkers	 2-12
           2.6  Health Effects	 2-13
                        Table 2-2   Summary of causal determinations between exposure to Pb and
                                   health outcomes.	 2-14
             2.6.1  Nervous System  Effects	2-16
November 2012                               iii                   Draft - Do Not Cite or Quote

-------
                 2.6.1.1   Children	2-16
                 2.6.1.2   Adults 	2-19
              2.6.2  Cardiovascular Effects	2-21
              2.6.3  Renal Effects	2-24
              2.6.4  Immune System Effects	2-25
              2.6.5  Hematological Effects	2-27
              2.6.6  Reproductive and Developmental Effects	2-28
              2.6.7  Cancer	2-31
           2.7  Ecological Effects of Pb	 2-32
              2.7.1  Summary of Effects on Terrestrial Ecosystems	2-32
              2.7.2  Summary of Effects on Aquatic Ecosystems	2-36
              2.7.3  Determinations of Causality for Effects on Ecosystems  	2-42
                          Table 2-3    Summary of Pb  causal determinations for plants, invertebrates and
                                      vertebrates.	 2-42
                 2.7.3.1   Effects on Development and Reproduction	2-43
                 2.7.3.2   Effects on Growth	2-45
                 2.7.3.3   Effects on Survival  	2-47
                 2.7.3.4   Neurobehavioral Effects	2-49
                 2.7.3.5   Hematological Effects	2-50
                 2.7.3.6   Effects on Physiological Stress	2-51
                 2.7.3.7   Community and Ecosystem Effects 	2-53
           2.8  Integration of Health and Ecological Effects	 2-55
                          Table 2-4    Summary of causal determinations for health and ecological effects. _ 2-56
              2.8.1  Modes of Action Relevant to Downstream Health and Ecological Effects	2-57
                          Table 2-5    MOAs, their related health effects, and information on concentrations
                                      eliciting the MOAs.	 2-58
           2.9  Policy Relevant Considerations	 2-62
              2.9.1  Public Health Significance	2-62
                          Figure 2-1   The effect of a small shift in population mean on the proportion of
                                      individuals in  the population diagnosed with clinical disease (i.e., the
                                      proportion to the right of the "Critical Line").	 2-63
              2.9.2  Air-Pb-to-Blood-Pb Relationships	2-64
                          Table 2-6    Summary of estimated slopes for blood Pb to air Pb relationships
                                      in children.	 2-66
              2.9.3  Concentration-Response Relationships for Human Health Effects	2-67
              2.9.4  Patterns of Pb Exposure and Neurodevelopmental Deficits in Children	2-69
              2.9.5  Reversibility and Persistence of Neurotoxic Effects of Pb	2-71
              2.9.6  Populations Potentially At-Risk for Health Effects	2-72
                          Table 2-7    Summary of evidence for factors that potentially increase the risk of
                                      Pb-related health effects.	 2-73
              2.9.7  Ecological  Effects and Corresponding Pb Concentrations 	2-76
           2.10 Summary	 2-77
                          Table 2-8    Summary of evidence from epidemiologic,  animal toxicological and
                                      ecological studies on the effects associated with exposure to Pb.	 2-78
           References for Chapter 2	 2-85

           AMBIENT LEAD: SOURCE TO CONCENTRATION	3-1

           3.1  Introduction	 3-1
           3.2  Sources of Atmospheric Pb	 3-1
              3.2.1  National Emissions Inventory	 3-2
                          Figure 3-1   Trends in Pb emissions (thousand tons) from stationary and mobile
                                      sources in the U.S., 1970-2008.	 3-3
                          Figure 3-2   Trends in Pb emissions (thousand tons) from stationary and mobile
                                      sources in the U.S., 1990-2008.	 3-4
                          Figure 3-3   Nationwide stationary and mobile source Pb emissions (tons) in the
                                      U.S. by source sector in 2008. 	 3-5
                          Figure 3-4   County-level Pb emissions (tons) in  the U. S. in 2008.	 3-6
                          Figure 3-5   Pb facilities estimated to emit 0.5 tons or more in 2008.	 3-7
              3.2.2  Anthropogenic Sources	 3-7
                          Table 3-1    Pb compounds observed in the environment.	 3-8
November 2012                                 iv                     Draft - Do Not Cite or Quote

-------
                         Figure 3-6   Five-year totals for Pb mining, primary and secondary production,
                                      imports, and exports, 1991-2010.	 3-9
                 3.2.2.1  Pb Emissions from Piston-engine Aircraft Operating on Leaded-Aviation Gasoline and
                         Other Non-road Sources	3-10
                 3.2.2.2  Emissions from Metals  Processing and Mining	3-11
                 3.2.2.3  Fossil Fuel Combustion	3-12
                 3.2.2.4  Waste Incineration 	3-14
                 3.2.2.5  Wood Burning	3-15
                 3.2.2.6  Roadway-Related Sources	3-16
                 3.2.2.7  Deposited Pb	3-18
                         Figure 3-7   Total U. S. Pb additives in on-road gasoline used in on-road vehicles,
                                      1927-1995.                                                      3-21
           3.3
Fate and
3.3.1 Air
3.3.1.1
3.3.1.2
Figure 3-8 Estimated Pb aerosol input
urbanized areas (UAs), fro,
Transport of Pb
Figure 3-9 Fate of atmospheric Pb.

Transport
Deposition
3.3.1.3 Resuspension of Pb from Surface Soil tc
Figure 3-10 Scales of turbulence within
3.3.2 Water
3.3.2.1
3.3.2.2
3.3.2.3
3.3.2.4
3.3.3 Soil
3.3.3.1
3.3.3.2
3.3.3.3
Pb Transport in Water and Sediment
Deposition of Pb within Bodies of Water
Table 3-2 Surface sediment Pb conct
Flux of Pb from Sediments
Pb in Runoff

Deposition of Pb onto Soil from Air
Sequestration of Pb from Water to Soil
Movement of Pb within the Soil Column
's from on-road gasoline into 90 U. S.
m 1950 through 1982.





i Air after Deposition
an urban environment.


and in Sediment
3ntrations for various continental shelves.





3-22
3-23
3-23
3-24
3-24
3-25
3-28
3-30
3-33
3-34
3-35
3-37
3-37
3-39
3-47
3-48
3-49
3-52
Figure 3-11 Schematic model summarizing the estimated flux of Pb within a
                                      typical podzol profile from northern Sweden using data from
                                      Klaminder et al. (2006a).	  3-55
                         Figure 3-12  Eh-pH diagram for Pb in shooting range soils, Jefferson National
                                      Forest, VA.	  3-58
           3.4  Monitoring of Ambient Pb	  3-60
              3.4.1  Measurement Techniques	3-60
                 3.4.1.1  Sample Collection	3-60
                         Figure 3-13  Comparison of particle collection efficiency among different
                                      TSP sampler types.	  3-61
                         Table 3-3    Airborne PM sampling methods potentially applicable for
                                      Pb sampling.	  3-63
                 3.4.1.2  Sample Analysis: Federal Reference and Federal Equivalence Methods	3-68
                 3.4.1.3  Other Analysis Methods for Total Pb 	3-70
                 3.4.1.4  Sequential Extraction 	3-71
                 3.4.1.5  Speciation Techniques	3-72
                 3.4.1.6  Continuous Pb Monitoring	3-75
              3.4.2  Network Design	3-76
                 3.4.2.1  NAAQS Monitoring Network	3-77
                         Figure 3-14  Map of monitoring sites in current Pb NAAQS monitoring network.	  3-79
                         Figure 3-15  Fifteen U.S. locations where a study is currently being performed on
                                      airport Pb emissions.	  3-80
                         Table 3-4    List of 15 airports included in the airport study	  3-81
                 3.4.2.2  Other Pb Monitoring Networks	3-81
                         Figure 3-16  Pb-PM2.s monitoring sites for CSN and IMPROVE networks.	  3-83
                         Figure 3-17  Pb-PMio monitoring sites for NATTS network.	  3-83
           3.5  Ambient Air Pb Concentrations	  3-84
              3.5.1  Spatial  Distribution of Air Pb  	3-84
                 3.5.1.1  Variability across the U.S.	3-84
                         Table 3-5    Summary data for source-oriented Pb monitors across the  U. S.,
                                      2008-2010.                                                       3-86
November 2012                                  v                      Draft - Do Not Cite or Quote

-------
                          Figure 3-18  Highest county-level source-oriented Pb-TSP concentrations
                                      (ug/m3), maximum 3-month average, 2008-2010.	 3-86
                          Table 3-6    Summary data for non-source-oriented Pb monitors across the U.S.,
                                      2008-2010. 	 3-87
                          Figure 3-19  Highest county-level non-source-oriented Pb-TSP concentrations
                                      fijg/m3), maximum 3-month average, 2008-2010.	 3-87
                 3.5.1.2   Intra-urban Variability	3-88
                          Table 3-7    Sample of U.S. near road Pb TSP monitors	 3-90
                          Figure 3-20  Time series of monthly average Pb-TSP concentration at five near-
                                      road monitors.	 3-91
              3.5.2 Temporal Variability	3-93
                 3.5.2.1   Multi-year Trends	3-93
                          Figure 3-21  National trends in Pb concentration (ug/m3), 74 trends sites,
                                      1990-2010. 	 3-94
                 3.5.2.2   Seasonal Variations  	3-94
                          Figure 3-22  Boxplots of average monthly Pb-PMz 5 concentrations measured at
                                      four IMPROVE sites, 2001-2010.	 3-95
              3.5.3 Size Distribution of Pb-Bearing PM	3-96
                 3.5.3.1   Co-located Monitoring Data Analysis	3-97
                          Table 3-8    Summary of comparison data for co-located ambient air Pb
                                      monitors.	 3-98
                 3.5.3.2   Studies of Pb-bearing PM Size Distribution in the Literature	3-98
                          Table 3-9    Summary of studies reporting Pb size distribution  in the peer-
                                      reviewed literature.  	 3-99
                          Figure 3-23  Comparison of urban background and near-road size fractions of Pb-
                                      bearing PM.	 3-103
              3.5.4 Pb Concentrations in a Multipollutant Context	3-105
                          Figure 3-24  Pearson correlations of monitored non-source daily average Pb-TSP
                                      concentration with daily averages of copollutant concentrations,
                                      2008-2010. 	 3-706
                          Figure 3-25  Pearson correlations of monitored Pb-PM2.s concentration with
                                      copollutant concentrations, 2008-2010.  	 3-708
              3.5.5 Background Pb Concentrations	3-109
           3.6  Ambient Pb Concentrations in Non-Air Media and Biota	 3-7 7 7
              3.6.1 Soils	3-112
                          Table 3-10   Soil concentrations  in various cities,  prior to 2005.	 3-773
                          Figure 3-26  Map of median Pb content in soil in New Orleans.	 3-776
                          Figure 3-27  Size distribution of Pb-containing dust collected near busy (HWY 1)
                                      and low traffic (HWY 17) highways. 	 3-7 7 8
              3.6.2 Sediments 	3-119
                          Figure 3-28  WACAP data for Pb concentration in sediment at  eight National
                                      Parks and/or Preserves.	 3-727
                          Figure 3-29  Sediment core data (1992-1994) for the lakes and reservoirs along
                                      the Apalachicola, Chattahoochee, and Flint River  Basin (ACF).	 3-723
                          Figure 3-30  Sediment core data (1975-1995) for the lakes and reservoirs along
                                      the Apalachicola, Chattahoochee, and Flint River  Basin (ACF).	 3-724
              3.6.3 Rain	 3-125
                          Figure 3-31  Trends in Pb concentration in precipitation from various sites in
                                      Norway over the period 1980-2005.	 3-726
              3.6.4 Snowpack	3-127
                          Figure 3-32  Box plots illustrating Pb concentration in snow melt at nine National
                                      Parks and Preserves. 	 3-728
              3.6.5 Natural Waters	3-129
                          Figure 3-33  Boxplots of Pb concentration in surface waters measured at five
                                      National Parks and Preserves.	 3-737
                          Table 3-11   Pb concentrations from stream food-webs; in mining-disturbed areas
                                      of Missouri and the  western  U.S. 	 3-733
              3.6.6 Vegetation 	3-133
                          Figure 3-34  Boxplots of Pb concentration in lichen measured at seven National
                                      Parks and Preserves.                                            3-134
November 2012                                 vi                     Draft - Do Not Cite or Quote

-------
                         Figure 3-35  Trends in regional pollution near a copper (Cu) smelter in Canada
                                     and Pb concentrations at the boundary of heartwood trees within
                                     roughly 75 km of the smelter.	
              3.6.7  Aquatic Bivalves_
              3.6.8  Vertebrate Populations	
                         Figure 3-36 Boxplots of Pb concentration in fish fillet and fish liver, measured at
                                     eight National Parks and/or Preserves.	
                         Figure 3-37  Boxplots of Pb concentration in moose meat and moose liver
                                     measured at Denali National Park and Preserve.	
           3.7  Summary and Conclusions	
              3.7.1  Sources of Atmospheric Pb	
              3.7.2
              3.7.3
              3.7.4
              3.7.5
         Fate and Transport of Pb_
         Ambient Pb Monitoring	
         Ambient Air Pb Concentrations
         Ambient Pb Concentrations in Non-Air Media and Biota	
3.8  Chapter 3 Appendix (Supplemental Material)	
   3.8.1  Variability across the U.S. 	
              Table 3-12   Distribution of 1-month average Pb-TSP concentrations (ug/m )
                          nationwide, source-oriented monitors, 2008-2010.	
                          Distribution of 1-month average Pb-TSP concentrations (ug/m )
                          nationwide, non-source-oriented monitors, 2008-2010.	
                         Table 3-13
                         Table 3-14  Distribution of 3-month moving average Pb-TSP concentrations
                                     (ug/m3) nationwide, source-oriented monitors, 2008-2010. 	
                         Table 3-15  Distribution of 3-month moving average Pb-TSP concentrations
                                     (ug/m3) nationwide, non-source-oriented monitors, 2008-2010.	
                         Table 3-16  Distribution of annual 1-month site maxima TSP Pb concentrations
                                     (ug/m3) nationwide, source-oriented monitors, 2008-2010. 	
                         Table 3-17  Distribution of annual 1-month site maxima TSP Pb concentrations
                                     (ug/m3) nationwide, non-source-oriented monitors, 2008-2010.	
                         Table 3-18  Distribution of annual 3-month site maxima Pb-TSP concentrations
                                     (pg/m3) nationwide, source-oriented monitors, 2008-2010. 	
                         Table 3-19  Distribution of annual 3-month site maxima Pb-TSP concentrations
                         Figure 3-38

                         Figure 3-39
                          (ug/m ) nationwide, non-source-oriented monitors, 2008-2010.	
                          Highest county-level Pb-PMio concentrations (ug/m3), maximum
                          3-month average, 2007-2009. _
              3.8.2
                          Highest county-level Pb-PM2.s concentrations (ug/m ), maximum
                          3-month average, 2007-2009.	
         Intra-urban Variability	
              Figure 3-40  Pb TSP monitor and source locations within Los Angeles County,
                         Figure 3-41

                         Figure 3-42


                         Table 3-20


                         Figure 3-43

                         Figure 3-44

                         Figure 3-45



                         Table 3-21
                          CA (06-037), 2007-2009.	
                          Wind roses for Los Angeles County, CA, from meteorological data at
                          the Los Angeles International Airport, 1961-1990.	
                          Box plots of annual and seasonal 24-h Pb TSP concentrations
                          (ug/m3) from source-oriented and non-source-oriented monitors
                          within Los Angeles County, CA (06-037), 2007-2009.	
                          Comparisons between Pb TSP concentrations from source-oriented
                          and non-source-oriented monitors within Los Angeles County, CA
                          (06-037), 2007-2009.	
                          Pb TSP monitor locations within Hillsborough and Pinellas Counties,
                          FL (12-057 and 12-103), 2007-2009. 	
                          Wind roses for Hillsborough/Pinellas Counties, FL, obtained from
                          meteorological data at Tampa International Airport, 1961-1990.	
                          Box plots of annual and seasonal 24-h Pb TSP concentrations
                          (ug/m3) from source-oriented and non-source-oriented monitors
                          within Hillsborough and Pinellas Counties, FL (12-057 and 12-103),
                          2007-2009. 	
                          Correlations between Pb TSP concentrations from source-oriented
                          and non-source-oriented monitors within Hillsborough and Pinellas
                          Counties, FL (12-057 and 12-103),  2007-2009.	
                         Figure 3-46  Pb TSP Monitor locations within Cook County, IL (17-031),
                                     2007-2009.
3-136
3-136
3-137

3-138

3-139
3-140
3-140
3-140
3-141
3-142
3-143
3-144
3-144

3-144

3-148

3-152

3-156

3-159

3-160

3-161

3-162

3-163

3-164
3-165

3-768

3-769


3-770


3-777

3-774

3-775
3-776


3-777

3-780
November 2012
                                               VII
                                                           Draft - Do Not Cite or Quote

-------
                         Figure 3-47

                         Figure 3-48


                         Table 3-22


                         Figure 3-49

                         Figure 3-50

                         Figure 3-51


                         Table 3-23


                         Figure 3-52

                         Figure 3-53

                         Figure 3-54


                         Table 3-24
                 Wind roses for Cook County, IL, obtained from meteorological data
                 at O'Hare International Airport,  1961-1990. 	
                 Box plots of annual and seasonal 24-h Pb TSP concentrations
                 (ug/m3) from source-oriented and non-source-oriented monitors
                 within Cook County, IL (17-031), 2007-2009..
                 Correlations between Pb TSP concentrations from source-oriented
                 and non-source-oriented monitors within Cook County, IL (17-031),
                 2007-2009. 	
                 Pb TSP Monitor locations within Jefferson County, MO (29-099),
                 2007-2009. 	
                 Wind roses for Jefferson County,  MO, obtained from meteorological
                 data at St. Louis/Lambert International Airport, 1961-1990.
                 Box plots of annual and seasonal 24-h Pb TSP concentrations
                 (ug/m3) from source-oriented and non-source-oriented monitors
                 within Jefferson County, MO (29-099), 2007-2009.
                 Correlations between Pb TSP concentrations from source-oriented
                 and non-source-oriented monitors within Jefferson County, MO
                 (29-099), 2007-2009.	
                 Pb TSP Monitor locations within Cuyahoga County, OH (39-035),
                 2007-2009. 	
                 Wind roses for Cuyahoga County, OH, obtained from meteorological
                 data at Cleveland/Hopkins International Airport,  1961-90.	
                 Box plots of annual and seasonal 24-h Pb TSP concentrations
                 (ug/m3) from source-oriented and non-source-oriented monitors
                 within Cuyahoga County,  OH (39-035), 2007-2009.	
                 Correlations between Pb TSP concentrations from source-oriented
                 and non-source-oriented monitors within Cuyahoga County, OH
                 (39-035), 2007-2009.	
                         Figure 3-55  Pb TSP Monitor locations within Sullivan County, TN (47-163),
                                      2007-2009. 	
                         Figure 3-56

                         Figure 3-57
                 Wind roses for Sullivan County, TN, obtained from meteorological
                 data at Bristol/Th City Airport, 1961-90. 	
                 Box plots of annual and seasonal 24-h Pb TSP concentrations
                 (ug/m3) from source-oriented monitors within Sullivan County, TN
                 (47-163), 2007-2009.	
              3.8.3
     Table 3-25  Correlations between Pb TSP concentrations from source-oriented
                 monitors within Sullivan County, TN (47-163), 2007-2009.	
Seasonal Variation in Pb Concentrations	
     Figure 3-58  Monthly source-oriented Pb-TSP average (ug/m3) over 12 months of
                 the year, 2008-2010.
              3.8.4
     Figure 3-59  Monthly non-source-oriented Pb-TSP average (ug/m ) over 12
                 months of the year, 2008-2010  	
     Figure 3-60  Monthly Pb-PMio average (ug/m ) over 12 months of the year,
                 2007-2009.	
     Figure 3-61  Monthly Pb-PM25 average (ug/m ) over 12 months of the year,
                 2007-2009. 	
Size Distribution of Pb-bearing PM 	
     Table 3-26   Correlations and average of the concentration ratios for co-located
                          Table 3-27
                          Table 3-28
                 monitors, TSP versus PMW, TSP versus PM2 5, and PMW versus
                 PM2.5.	
                 Metadata for studies of Pb-PM size distribution.
              3.8.5
                 Size distribution data for various studies described in Table 3-27. _
Pb Concentration in a Multipollutant Context	
     Figure 3-62  Spearman correlations of monitored non-source Pb-TSP
                 concentration with daily averages of copollutant concentrations,
                 2008-2010.
                         Figure 3-63

                         Figure 3-64

                         Figure 3-65
                 Seasonal correlations of monitored Pb-TSP concentration with
                 copollutant concentrations, 2007-2008.
                 Seasonal correlations of monitored Pb-TSP concentration with
                 copollutant concentrations, 2009.
                 Seasonal correlations of monitored Pb-TSP concentration with
                 copollutant concentrations, 2009.	
3-181


3-182


3-183

3-185

3-186


3-187


3-188

3-191

3-192


3-193


3-194

3-196

3-197


3-198

3-199
3-199

3-200

3-207

3-207

3-202
3-202
3-203
3-206
3-270
3-217
3-277

3-278

3-279

3-220
November 2012
                                               Vlll
                                                  Draft - Do Not Cite or Quote

-------
                         Figure 3-66  Seasonal correlations of monitored Pb-PM2.5 concentration with
                                     copollutant concentrations, 2007-2009. 	 3-221
                         Figure 3-67  Seasonal correlations of monitored Pb-PM2.s concentration with
                                     copollutant concentrations, 2007-2009. 	 3-222
                         Table 3-29   Copollutant exposures for various trace metal studies.	 3-223
           References for Chapter 3	 3-224

           EXPOSURE, TOXICOKINETICS, AND BIOMARKERS	4-1

           4.1  Exposure Assessment	 4-1
              4.1.1  Pathways for Pb Exposure	4-1
                         Figure 4-1   Conceptual model of multimedia Pb exposure.	 4-3
                         Table 4-1    Estimates of Pb measurements for EPA Region 5 from the
                                     NHEXAS study.	 4-4
                 4.1.1.1  Particle Size Distributions for Airborne-Pb, Dust-Pb, and Soil-Pb	4-5
                 4.1.1.2  Estimating Pb Exposure in the Integrated Exposure Uptake Biokinetic (IEUBK) Model 4-7
                         Table 4-2    Predicted concurrent blood Pb levels and source contributions for
                                     children in their seventh year of life.	 4-8
              4.1.2  Environmental Exposure Assessment Methodologies	4-8
              4.1.3  Exposure Studies	4-10
                 4.1.3.1  Airborne Pb Exposure	4-10
                         Table 4-3    Estimates of fixed effects multivariate modeling of Pb levels
                                     measured during the NHEXAS-MD study.	 4-11
                         Table 4-4    Comparison of personal, indoor, and outdoor Pb-PM measurements
                                     from several studies.	 4-14
                 4.1.3.2  Exposure to Pb in Soil and Dust	4-14
                         Table 4-5    Measurements of indoor dust Pb concentration from 2006-2011
                                     studies. 	 4-15
                 4.1.3.3  Dietary Pb Exposure	4-19
                         Figure 4-2   Market basket survey results for Pb concentration in foods.	 4-20
                         Table 4-6    Pb bioaccumulation data for various plants. Bioaccumulation is
                                     expressed as percent of Pb concentration in the plant to the Pb
concentration in the soil.
4.1.3.4 Occupational
4. 1 .3.5 Exposure to Pb from Consumer Products
Table 4-7 Pb content in various consumer products.
4.2 Kinetics
4.2.1 Absorption
4.2.1.1 Inhalation
4.2.1.2 Inqestion
Figure 4-3 Estimated relative bioavailability (RBA, compared to Pb acetate) of
ingested Pb in mineral groups.
4.2.2 Distribution
4.2.2.1 Blood
4-26
4-29
4-29
4-30
4-37
4-32
4-33
4-35
4-39
4-42
4-42
                         Figure 4-4   Plot of blood and plasma Pb concentrations measured in adults and
                                     children.	 4-43
                         Figure 4-5   Relationship between Pb intake and blood Pb concentration in
                                     infants (n = 105, age 13 weeks, formula-fed).	 4-44
                         Figure 4-6   Simulation of quasi-steady state blood and plasma Pb
                                     concentrations in a child (age 4 years) associated with varying Pb
inqestion rates.
4.2.2.2 Bone
4.2.2.3 Soft Tissues
4.2.2.4 Fetus
4.2.2.5 Organic Pb
4.2.3 Elimination
4.3 Pb Biomarkers
4.3.1 Bone-Pb Measurements
4.3.2 Blood-Pb Measurements
4-46
4-47
4-48
4-49
4-50
4-50
4-52
4-55
4-57
                         Figure 4-7   Simulation of temporal relationships between Pb exposure and blood
                                     Pb concentration in children.	 4-60
              4.3.3  Urine-Pb Measurements                                                           4-61
November 2012                                 ix                     Draft - Do Not Cite or Quote

-------
                          Teeth	
                          Hair	
                          S a I i va	
                          Serum 8-ALA and ALAD
           Figure 4-8   Simulation of relationship between urinary Pb excretion and body
                        burden in adults.	
4.3.4  Pb in Other Biomarkers
   4.3.4.1
   4.3.4.2
   4.3.4.3
   4.3.4.4                         	
4.3.5  Relationship between Pb in  Blood and Pb in Bone	
   4.3.5.1  Children	
           Figure 4-9   Simulation of relationship between blood Pb concentration and body
                        burden in children, with an elevated constant Pb intake from age
                        2 to 5 years.	
           Figure 4-10  Simulation of relationship between time-integrated blood Pb
                        concentration and cumulative Pb absorption in children.	
                 4.3.5.2   Adults
                          Figure 4-11
                        Simulation of relationship between blood Pb concentration, bone Pb
                        and body burden in adults with relatively low Pb intake.	
              4.3.6
           Figure 4-12  Simulation of relationship between blood Pb concentration, bone Pb
                        and body burden in adults with relatively high Pb intake.	
      Relationship between Pb in Blood and Pb in Soft Tissues	
           Figure 4-13  Simulation of blood and soft tissue (including brain) Pb in children
                          Figure 4-14
                       and adults who experience a period of increased Pb intake. _
                       Simulation of blood and brain Pb in children and adults who
                       experience a period of increased Pb intake.	
                          Figure 4-15  Relationship between Pb in urine, plasma, blood and bone.
           4.4  Studies of Pb Biomarker Levels	
              4.4.1   Pb in Blood
                          Figure 4-16
                          Table 4-8
                          Figure 4-17

                          Figure 4-18
                          Figure 4-19
                        Temporal trend in blood Pb concentration.	
                        Blood Pb concentrations in the U.S. population.	
                        Box plots of blood Pb levels among U.S. children (1-5 years old at
                        baseline) from the NHANES survey,  1988-2010.	
                        Blood Pb cohort means versus year of exam.	
                       Percent distribution of blood Pb levels by race/ethnicity among U.S.
                       children (1-5 years) from the NHANES survey, 1988-1991 (top) and
                       1999-2004 (bottom).	
           Figure 4-20 Trends in 206Pb/04Pb isotope ratio in blood Pb (a) and trends in
                       blood Pb levels (b) among Australian study populations of children
4-62
4-63
4-63
4-64
4-64
4-65
4-65
4-68
4-70

4-71
4-71

4-74

4-75
4-79

4-80

4-81
4-83
4-84
4-84
4-85
4-86

4-88
4-89
                                                                                                      4-90
4.4.2 Pb in Bone
Table 4-9
Table 4-10
4.4.3 Pb in Urine
Table 4-11
4.4.4 Pb in Teeth
Figure 4-21
4. 5 Empirical Models of Pb
4.5.1 Air Pb-Blood Pb R(
Table 4-12
Figure 4-22
Figure 4-23
Table 4-13
Table 4-14
Table 4-15
during the period 1990-2000.

Epidemiologic studies that provide bone Pb measurements for non-
occupationally exposed populations.
Epidemiologic studies that provide bone Pb measurements for
occupationally exposed populations.

Urine Pb concentrations in the U. S. population.

Comparison of relative temporal changes in tooth enamel Pb
concentration.
Exposure-Blood Pb Relationships
slationships in Children
Summary of estimated slopes for blood Pb to air Pb slope factors
in humans.
Predicted relationship between air Pb and blood Pb based on a meta
analysis of 18 studies.
Blood Pb - air Pb slopes (pg/dL perpg/mj) predicted from
epidemiologic studies.
Environmental Pb levels and blood Pb levels in children in Trail,
British Columbia.
Predicted blood Pb levels and blood-air slopes for Mexico City
children (1987 and 1990 cohorts).
U.S. gasoline Pb consumption and air Pb levels.
4-94
4-95
4-96
4-107
4-112
4-112
4-113
4-114
4-115
4-116
4-118
4-120
4-121
4-122
4-125
4-127
November 2012
                                                         Draft - Do Not Cite or Quote

-------
                         Table 4-16   Air Pb concentrations and blood Pb levels in children in Mumbai,
                                     India.	 4-128
                         Figure 4-24  Predicted relationship between air Pb and blood Pb based on data
                                     from Chicago, ILin children age 0-5 years (1974-1988).	 4-129
                         Figure 4-25  Effect of air-to-blood slope on estimated change in air-related blood
                                     Pb with change in air Pb.	 4-730
              4.5.2  AirPb-Blood Pb Relationships in Occupational Cohorts	4-131
              4.5.3  Environmental Pb-Blood Pb Relationships 	4-132
                         Table 4-17   Linear model relating environmental Pb exposure and blood Pb
                                     concentration in children.	 4-134
                         Figure 4-26  Predicted relationship between soil Pb concentration and blood Pb
                                     concentration in children based on data collected in New Orleans,
                                     Louisiana: 2000-2005.	 4-136
                         Table 4-18   General linear model relating blood Pb concentration in children and
                                     environmental Pb levels—Bunker Hill Superfund Site.	 4-138
           4.6  Biokinetic Models of Pb Exposure-Blood Pb Relationships	 4-138
           4.7  Summary and Conclusions	 4-140
              4.7.1  Exposure	4-140
              4.7.2  Toxicokinetics	4-141
              4.7.3  Pb Biomarkers	4-142
              4.7.4  Air Lead-Blood Lead Relationships	4-144
           References for Chapter 4	 4-145

           INTEGRATED HEALTH EFFECTS OF LEAD EXPOSURE	5-1
5.1
5.2
5.
5.
Introduction
Modes of Action
2.1 Introduction
Figure 5-1 Schematic representation of the relationships between the various
MOAs by which Pb exposure exerts its health effects.
2.2 Altered Ion Status
5.2.2.1 Disruption of Ca2+ Homeostasis
5.2.2.2 Disruption of Ion Transport Mechanisms
5.2.2.3 Displacement of Metal Ions and Perturbed Protein Function
Table 5-1 Enzymes and proteins potentially affected by exposure to Pb and the
5-1
5-2
5-2
5-3
5-4
5-4
5-6
5-10
                                     metal cation cofactors necessary for their proper physiological
activity.
5.2.2.4 Mitochondrial Abnormality
5.2.3 Protein Binding
5.2.3.1 Intranuclear and Cytoplasmic Inclusion Bodies
5.2.3.2 Cytosolic Pb Binding Proteins
5.2.3.3 Erythrocytic Pb Binding Proteins
5.2.3.4 Metallothionein
5.2.4 Oxidative Stress
5.2.4.1 8-ALA Oxidation
5.2.4.2 Membrane and Lipid Peroxidation
5.2.4.3 NAD(P)H Oxidase Activation
5.2.4.4 Antioxidant Enzyme Disruption
5.2.4.5 Nitric Oxide Signaling
5.2.5 Inflammation
5.2.5.1 Cytokine Production
5.2.6 Endocrine Disruption
5.2.6.1 Hypothalamic-Pituitary-Gonadal Axis
5.2.6.2 Hypothalamic-Pituitary-Thyroid Axis
5.2.7 Cell Death and Genotoxicity
5.2.7.1 DMA Damage
5.2.7.2 Mutagenicity
5.2.7.3 Clastogenicity
5.2.7.4 Epigenetic Effects
5.2.7.5 Gene Expression
5.2.7.6 Apoptosis
5.2.8 Summary
5-16
5-17
5-20
5-20
5-21
5-22
5-23
5-25
5-26
5-26
5-28
5-28
5-31
5-32
5-33
5-36
5-36
5-38
5-39
5-39
5-42
5-43
5-48
5-50
5-51
5-51
November 2012                                 xi                     Draft - Do Not Cite or Quote

-------
                          Table 5-2    MOAs, their related health effects, and information on concentrations
                                      eliciting the MOAs.	 5-52
           5.3   Nervous System Effects	 5-55
              5.3.1  Introduction	 5-55
              5.3.2  Cognitive Function	5-59
                  5.3.2.1   Full Scale IQ in Children	5-60
                          Figure 5-2   Associations of blood Pb levels with full-scale IQ (FSIQ) among
                                      children.	 5-62
                          Table 5-3    Additional characteristics and quantitative results for studies
                                      represented in Figure 5-2	 5-63
                  5.3.2.2   Bayley Scales of Infant Development	5-74
                          Table 5-4    Associations of blood Pb level with Bayley MDI in children ages
                                      12 months to 3 years.	 5-75
                  5.3.2.3   Learning and Memory  in Children	5-80
                          Table 5-5    Associations between blood Pb levels and performance on tests of
                                      learning and memory in children.  	 5-81
                          Figure 5-3   Summary of Pb exposure-nervous system concentration-response
                                      information from toxicological studies. 	 5-87
                          Table 5-6    Summary of findings from neurotoxicological concentration-response
                                      array presented in Figure 5-3.	 5-88
                          Table 5-7    Summary of effects of maternal and lifetime Pb exposure on
                                      Fl performance observed by Cory-Slechta and colleagues.	 5-95
                          Figure 5-4   Changes in Fixed Interval performance in (A) female and (B) male
                                      offspring with gestational/lactational Pb exposure plus various
                                      stressors given in adulthood.	 5-97
                          Figure 5-5   Mean basal and final corticosterone levels of female and male
                                      offspring exposed to lifetime Pb.	 5-99
                  5.3.2.4   Executive Function in Children	5-101
                          Table 5-8    Associations between blood or tooth Pb levels and performance of
                                      tests of executive function in children and young adults.	  5-103
                  5.3.2.5   Academic Performance and  Achievement in Children	5-106
                          Table 5-9    Associations between blood or tooth Pb levels and measures of
                                      academic performance and achievement in children and young
                                      adults.	  5-708
                          Figure 5-6   Associations between childhood blood Pb levels and fourth grade
                                      End-of-Grade (EOG) math scores.	  5-113
                          Figure 5-7   Greater reduction in End-of-Grade (EOG) scores with increasing
                                      blood Pb level in lower percentiles of the test score distribution.	  5-114
                  5.3.2.6   Integrated Summary of Cognitive Function in Children 	5-116
                  5.3.2.7   Epidemiologic Studies  of Cognitive Function in Adults	5-121
                          Table 5-10   Associations of blood and bone Pb levels with cognitive function in
                                      adults.	  5-723
                          Figure 5-8   Nonlinear association between patella Pb level and the relative
                                      change over 3.5 years in response latency on the pattern
                                      comparison test in men from the Normative Aging Study.	  5-730
                          Figure 5-9   Nonlinear association of tibia Pb level with annual rate of cognitive
                                      decline, by hemochromatosis genotype in men from the Normative
                                      Aging Study.	  5-737
              5.3.3  Attention-related Behavioral  Problems in Children 	5-139
                  5.3.3.1   Inattention and Impulsivity	5-140
                          Figure 5-10  Associations of blood Pb levels with attention-related behavioral
                                      problems in children.	  5-742
                          Table 5-11   Additional characteristics and quantitative results for studies
                                      presented in Figure 5-10.  	  5-743
                  5.3.3.2   Hyperactivity	5-154
                  5.3.3.3   Ratings of Attention Deficit Hyperactivity Disorder-related Behaviors	5-155
                  5.3.3.4   Attention Deficit Hyperactivity Disorder in Children 	5-157
                  5.3.3.5   Integrated Summary of Attention-related Behavioral Problems	5-159
              5.3.4  Conduct Problems	5-162
                  5.3.4.1   Epidemiologic Studies  of Conduct Problems in  Children	5-162
                          Table 5-12   Associations between blood Pb level and misconduct in children and
                                      young adults.	  5-764
November 2012                                 xii                      Draft - Do Not Cite or Quote

-------
                  5.3.4.2  Toxicological Studies of Aggression	5-171
                  5.3.4.3  Summary of Conduct Problems	5-172
              5.3.5  Internalizing Behaviors in Children	5-174
                  5.3.5.1   Epidemiologic Studies of Internalizing Behaviors in Children	5-174
                  5.3.5.2  Toxicological Studies of Internalizing  Behaviors	5-177
                  5.3.5.3  Summary of Internalizing Behaviors	5-179
              5.3.6  Psychopathological Effects in Adults	5-180
                  5.3.6.1   Epidemiologic Studies of Psychopathological Effects in Adults	5-180
                  5.3.6.2  Toxicological Studies of Mechanisms of Psychopathological Effects	5-182
                          Figure 5-11  Schematic representation of the contribution of Pb exposure to the
                                      development of a phenotype consistent with schizophrenia.	 5-182
                  5.3.6.3  Summary of Psychopathological Effects in Adults	5-183
              5.3.7  Sensory Organ Function 	5-184
                  5.3.7.1   Epidemiologic Studies of Sensory Organ Function in Children 	5-184
                  5.3.7.2  Epidemiologic Studies of Sensory Organ Function in Adults	5-186
                  5.3.7.3  Toxicological Studies of Sensory Organ Function 	5-187
                          Table 5-13   Summary of Pb-related effects observed on the visual system.	 5-790
                          Figure 5-12  Retinal a-wave and b-wave ERG amplitude in adult rats after
                                      prenatal plus early postnatal Pb exposure.	 5-192
                          Figure 5-13  Retinal dopamine metabolism in adult control and gestationally Pb-
                                      exposed (GLE) rats. 	 5-793
                  5.3.7.4  Summary of Sensory Function	5-193
              5.3.8  Motor Function	5-195
              5.3.9  Seizures in Animals	5-198
              5.3.10 Neurodegenerative Diseases	5-199
                  5.3.10.1  Alzheimer's Disease	5-199
                  5.3.10.2  Amyotrophic Lateral Sclerosis	5-201
                  5.3.10.3  Parkinson's Disease	5-204
                  5.3.10.4  Essential Tremor	5-205
                  5.3.10.5  Toxicological Studies of Cell Death Pathways	5-206
              5.3.11 Modes of Action for Pb Nervous System Effects	5-207
                  5.3.11.1  Effects on Brain Physiology and Activity	5-207
                  5.3.11.2  Oxidative Stress	5-210
                  5.3.11.3  Nitrosative Signaling and Nitrosative Stress	5-212
                  5.3.11.4  Synaptic Changes	5-213
                  5.3.11.5  Blood Brain Barrier	5-215
                  5.3.11.6  Cell Adhesion Molecules	5-217
                  5.3.11.7  Effects on Glial Cells	5-217
                  5.3.11.8  Neurotransmitters	5-218
                  5.3.11.9  Neurogenesis	5-221
                          Figure 5-14  Neurogenesis (production of new cells) in the rat hippocampal
                                      dentate gyrus after early postnatal Pb exposure.	 5-223
                  5.3.11.10     Neurite Outgrowth	5-223
                  5.3.11.11     Epigenetics	5-224
                  5.3.11.12     Cholesterol and Lipid Homeostasis	5-225
              5.3.12 Lifestage of Pb Exposure and Neurodevelopmental Deficits	5-225
                          Table 5-14   Associations of cognitive function with blood Pb levels measured at
                                      various lifestages and time periods in prospective studies. 	 5-228
                          Table 5-15   Comparisons of blood Pb-FSIQ associations in groups of children
                                      with different temporal trends in blood Pb levels.	 5-234
                          Figure 5-15  Estimated FSIQ for three patterns of temporal trends in blood Pb
                                      level from ages  2 to 6 years in the Rochester and Cincinnati cohorts. 5-236
              5.3.13 Examination of the Pb Concentration-Response Relationship	5-238
                          Figure 5-16  Comparison of associations between blood Pb level and cognitive
                                      function among various blood Pb strata.	 5-240
                          Table 5-16   Additional characteristics  and quantitative results for studies
                                      presented in Figure 5-16.  	 5-247
              5.3.14 Confounding in Epidemiologic Studies of Nervous System Effects	5-249
              5.3.15 Public Health  Significance of Associations between Pb Biomarkers and Neurodevelopmental
                     Effects  	 5-253
                          Figure 5-17  Hypothetical effect of increasing blood Pb level on the proportion of
                                      the population with IQ <70 and <80 points.	 5-255
November 2012                                 xiii                     Draft - Do Not Cite or Quote

-------
               5.3.16 Summary and Causal Determination	5-257
                  5.3.16.1 Evidence for Cognitive Function in Children	5-257
                  5.3.16.2 Evidence for Attention-related Behavioral Problems in Children	5-263
                  5.3.16.3 Evidence for Conduct Problems in Children and Young Adults	5-267
                  5.3.16.4 Evidence for Internalizing Behaviors in Children 	5-268
                  5.3.16.5 Evidence for Sensory Function in Children	5-271
                  5.3.16.6 Evidence for Motor Function in Children	5-272
                  5.3.16.7 Evidence for Cognitive Function in Adults	5-273
                  5.3.16.8 Evidence for Psychopathological Effects in Adults	5-276
                  5.3.16.9 Evidence for Sensory Function Decrements in Adults	5-278
                  5.3.16.10     Evidence for Neurodegenerative Diseases in Adults	5-279
                          Table 5-17  Summary of Evidence Supporting Nervous System Causal
                                      Determinations.	 5-281
           5.4   Cardiovascular Effects	 5-297
               5.4.1  Introduction	5-291
               5.4.2  Blood Pressure and Hypertension	5-293
                  5.4.2.1   Epidemiology	5-293
                          Figure 5-18  Associations of blood and bone Pb levels with systolic BP, diastolic
                                      BP, and pulse pressure in adults.	 5-295
                          Table 5-18  Additional characteristics and quantitative data for associations of
                                      blood and bone Pb with BP measures for studies presented in
                                      Figure 5-18.	 5-296
                          Figure 5-19  Odds ratios (95% Cl) for associations of blood (closed circles) and
                                      bone (open circles) Pb with hypertension prevalence and incidence3.  5-300
                          Table 5-19  Additional characteristics and quantitative data for results presented
                                      in Figure 5-19 for associations of blood and bone Pb with
                                      hypertension measures.	 5-301
                          Figure 5-20  The relationship between tibia Pb and estimated systolic BP (SBP)
                                      for those with high self-reported stress versus those with low self-
                                      reported stress.	 5-307
                  5.4.2.2   Toxicology	 5-309
                          Figure 5-21  Changes in BP after Pb exposure (represented as blood Pb level) in
                                      unanesthetized adult rats across  studies.	 5-310
                          Table 5-20  Characteristics of studies of blood Pb with BP measures in animals
                                      presented in Figure 5-21. 	 5-311
                  5.4.2.3   Hypertension Modes of Action 	5-313
                  5.4.2.4   Summary of Blood Pressure and Hypertension	5-320
                          Figure 5-22  Meta-analysis of change in systolic BP (SBP), in mmHg with 95% Cl,
                                      associated with a doubling in the  blood Pb concentration.	 5-322
                          Figure 5-23  Meta-analysis of an increase in systolic BP (SBP) and diastolic BP
                                      (DBF) and relative risk of hypertension per 10pg/g increase in bone
                                      Pb levels.	 5-324
               5.4.3  Vascular Effects and Cardiotoxicity	5-325
                  5.4.3.1   Effects on Vascular Cell Types	5-325
                  5.4.3.2   Cholesterol	5-327
                  5.4.3.3   Atherosclerosis	5-328
                  5.4.3.4   Heart Rate Variability	5-329
                  5.4.3.5   Peripheral Artery Disease	5-331
                  5.4.3.6   Ischemic Heart Disease	5-332
                          Table 5-21  Characteristics and quantitative data for associations of blood and
                                      bone Pb with other CVD measures HRV, PAD, and IHD in recent
                                      epidemiologic studies.	 5-334
                  5.4.3.7   Summary of Vascular Effects and Cardiotoxicity	5-336
               5.4.4  Cardiovascular Function and  Blood Pressure in Children	5-338
                  5.4.4.1   Introduction	5-338
                          Table 5-22  Studies of child cardiovascular endpoints and Pb biomarkers.	 5-339
                  5.4.4.2   Cardiovascular Functioning in Children	5-341
                          Figure 5-24  Children's adjusted total peripheral resistance (dyn-s/cm^) responses
                                      to acute stress tasks, as a function of childhood Pb levels.	 5-342
                  5.4.4.3   Blood Pressure in Children	5-343
                  5.4.4.4   Summary of Child Cardiovascular Studies	5-344
               5.4.5  Mortality	5-346
November 2012                                 xiv                     Draft - Do Not Cite or Quote

-------
                          Figure 5-25  Multivariate adjusted relative hazards of all-cause and
                                      cardiovascular mortality per 3.4 pg/dL increase in blood Pb.	 5-348
                          Figure 5-26  Multivariate-adjusted relative hazard (left axis) of mortality
                                      associated with blood Pb levels between 1 pg/dL and 10 pg/dL	 5-349
                          Figure 5-27  Relative risk of all cause mortality for different blood Pb levels
                                      compared with referent level of 1.5 pg/dL (12.5th percentile).	 5-350
                          Figure 5-28  Associations between patella bone Pb level and the log of hazard
                                      ratio (logHR) for all-cause, cardiovascular, and ischemic heart
                                      disease.	 5-352
                          Figure 5-29  Multivariate adjusted relative hazard (left axis) of mortality as a
                                      function of blood Pb levels between 1 pg/dL and 15 pg/dL	 5-354
                  5.4.5.1   Summary of Mortality	5-354
                          Figure 5-30  Hazard ratios for associations of blood Pb or bone Pb with all-cause
                                      mortality and cardiovascular mortality.	 5-356
                          Table 5-23   Additional characteristics and quantitative data for associations of
                                      blood and bone Pb with CVD mortality for studies presented in
                                      Figure 5-30.	 5-357
               5.4.6  Air Pb-PM Studies	 5-359
                  5.4.6.1   Cardiovascular Morbidity	5-359
                  5.4.6.2  Mortality	 5-360
               5.4.7  Summary and Causal Determination	5-361
                  5.4.7.1   Evidence for Hypertension and Increased Blood Pressure 	5-361
                  5.4.7.2  Evidence for Subclinical Atherosclerosis 	5-365
                  5.4.7.3  Evidence for Coronary Heart Disease	5-366
                  5.4.7.4  Evidence for Cerebrovascular Disease	5-368
                          Table 5-24   Summary of evidence supporting cardiovascular causal
                                      determinations.	 5-369
           5.5   Renal Effects	 5-372
               5.5.1  Introduction	5-372
                  5.5.1.1   Kidney Outcome Measures	5-374
               5.5.2  Nephrotoxicity and Renal Pathology	5-375
                  5.5.2.1   Epidemiology in Adults	5-375
                          Figure 5-31  Concentration-response relationships for associations between
                                      blood Pb level or bone Pb level and kidney function outcomes.	 5-379
                          Figure 5-32  Percent change in kidney outcomes across quartiles of blood Pb
                                      level in NHANES. 	 5-387
                          Table 5-25   Additional characteristics and quantitative data for associations of
                                      blood and bone Pb with kidney outcomes for results presented in
                                      Figure 5-31 and Figure 5-32.	 5-382
                          Table 5-26   Prospective patient population studies: kidney function decline. 	 5-387
                          Table 5-27   Clinical randomized chelation trials in chronic kidney disease
                                      patients.	 5-397
                          Figure 5-33  Added variable plot of association between serum creatinine and
                                      blood Pb in 267 Korean Pb workers in the oldest age fertile.	 5-394
                  5.5.2.2  Epidemiology in Children 	5-396
                  5.5.2.3  Associations between  Pb Dose and New Kidney Outcome Measures	5-397
                  5.5.2.4  Reverse Causality	5-398
                  5.5.2.5  Toxicology	 5-400
                          Table 5-28   Animal toxicological studies reporting the effects of Pb exposure
                                      (as blood Pb level) on kidney function.	 5-400
                          Table 5-29   Indicators of renal damage in male rats  exposed to 50 ppm  Pb for 40
                                      and 65 days, starting at parturition.	 5-403
                          Table 5-30   Effects of Pb on the kidney/renal system related to exposure
                                      duration- evidence from animal toxicology studies.	 5-408
               5.5.3  Modes of Action for Pb-lnduced Nephrotoxicity	5-408
                  5.5.3.1   Oxidative Damage	5-408
                  5.5.3.2  Hypertension and Alteration of Renal Vasculature and Reactivity	5-412
                  5.5.3.3  Apoptosis and/or Ischemic Necrosis of Tubules and Glomeruli	5-413
                  5.5.3.4  Renal Gangliosides	5-415
                  5.5.3.5  Altered Uric Acid	5-415
                  5.5.3.6  Role of Metallothionein	5-416
               5.5.4  Effects of Exposure to Pb Mixtures	5-417
November 2012                                 xv                     Draft - Do Not Cite or Quote

-------
                 5.5.4.1  Lead(Pb) and Cadmium(Cd)	5-417
                 5.5.4.2  Lead(Pb), Cadmium(Cd), and Arsenic(As)	5-419
                 5.5.4.3  Lead (Pb) and Zinc (Zn)	5-420
                 5.5.4.4  Lead(Pb) and Mercury(Hg)	5-420
              5.5.5  Summary and Causal Determination	5-422
                 5.5.5.1  Evidence for Reduced Kidney Function	5-422
                         Table 5-31  Summary of evidence supporting renal causal determinations.	 5-427
           5.6  Immune System Effects	 5-428
              5.6.1  Introduction	5-428
                         Figure 5-34  Immunological pathways by which Pb exposure may increase risk of
                                     immune-related diseases.	 5-429
              5.6.2  Cell-Mediated Immunity	5-431
                 5.6.2.1  T Cells	 5-431
                 5.6.2.2  Lymphocyte Activation 	5-434
                 5.6.2.3  Delayed-type Hypersensitivity	5-435
                 5.6.2.4  Macrophages and Monocytes	5-436
                 5.6.2.5  Neutrophils	 5-440
                 5.6.2.6  Dendritic Cells	5-442
                 5.6.2.7  Natural  Killer Cells	5-442
              5.6.3  Humoral Immunity	5-443
                         Table 5-32  Comparison of serum immunoglobulin levels and B cell abundance
among various blood Pb groups.
5.6.4 Inflammation
5.6.5 Immune-based Diseases
5.6.5.1 Host Resistance
5.6.5.2 Asthma and Allergy
5-445
5-453
5-455
5-455
5-457
                         Figure 5-35 Associations of blood Pb levels with asthma and allergy in children. _ 5-458
                         Table 5-33  Additional characteristics and quantitative results for studies
                                     presented in Figure 5-35.  	 5-459
                 5.6.5.3  Other Respiratory Effects	5-465
                 5.6.5.4  Autoimmunity	5-468
                 5.6.5.5  Tumors	 5-468
              5.6.6  Modes of Action for Pb Immune Effects	5-469
                 5.6.6.1  Cytokine Production	5-469
                 5.6.6.2  Decreased Nitric Oxide	5-474
                 5.6.6.3  Increased Reactive Oxygen Species and Prostaglandins	5-476
                 5.6.6.4  Cellular Death (Apoptosis, Necrosis)	5-478
              5.6.7  Immune Effects of Pb within Mixtures	5-479
              5.6.8  Summary and Causal Determination	5-480
                 5.6.8.1  Evidence for an Increase in Atopic and Inflammatory Conditions	5-480
                 5.6.8.2  Evidence for Decreases in Host Resistance	5-485
                 5.6.8.3  Evidence for Autoimmunity	5-487
                         Table 5-34  Summary of evidence  supporting immune causal determinations.	 5-488
           5.7  Hematological Effects	 5-497
              5.7.1  Introduction	5-491
              5.7.2  Red Blood Cell  Function 	5-493
                 5.7.2.1  Pb Uptake, Binding, and Transport into Red Blood Cells	5-493
                 5.7.2.2  Red Blood Cell Survival, Mobility, and Membrane Integrity	5-494
                 5.7.2.3  Red Blood Cell Hematopoiesis	5-502
                 5.7.2.4  Membrane Proteins	5-503
                 5.7.2.5  Red Blood Cell Energy Metabolism Enzymes 	5-504
                 5.7.2.6  Other Enzymes	5-504
                 5.7.2.7  Red Blood Cell Oxidative Stress	5-505
                 5.7.2.8  Summary of Effects on RBC Survival and Function	5-508
              5.7.3  Red Blood Cell  Heme Metabolism	5-509
                         Figure 5-36 Schematic representation of the enzymatic steps involved in the
                                     heme synthetic pathway.	 5-509
                 5.7.3.1  Red Blood Cell 5-Aminolevulinic Acid Dehydratase	5-510
                 5.7.3.2  Other Heme Metabolism Enzymes	5-512
                 5.7.3.3  Hematological Effects	5-512
              5.7.4  Summary and Causal Determination	5-513
                 5.7.4.1  Evidence for Decreased RBC Survival and Function                             5-513
November 2012                                 xvi                    Draft - Do Not Cite or Quote

-------
                 5.7.4.2  Evidence for Altered Heme Synthesis	5-517
                         Table 5-35  Summary of evidence supporting RBC survival and heme synthesis
                                     causal determinations.	 5-519
           5.8  Reproductive and Developmental Effects	 5-523
              5.8.1  Effects on Development	5-523
                 5.8.1.1  Effects on Puberty among Females	5-524
                         Table 5-36  Summary of recent epidemiologic studies of associations between
                                     Pb levels and puberty for females.	 5-525
                         Figure 5-37 Toxicological concentration-response array for reproductive and
                                     developmental effects of Pb.	 5-533
                         Table 5-37  Toxicological concentration-response array summary for
                                     reproductive and developmental effects of Pb presented in
                                     Figure 5-37.	 5-534
                 5.8.1.2  Effects on Puberty among Males	5-536
                         Table 5-38  Summary of recent epidemiologic studies of associations between
                                     Pb levels and puberty for males.	 5-537
                 5.8.1.3  Effects on Postnatal Stature and Body Weight	5-542
                         Table 5-39  Summary of recent epidemiologic studies of associations between
                                     Pb levels and postnatal growth.	 5-543
              5.8.2  Toxicological  Studies of Other Developmental Effects	5-553
                 5.8.2.1  Developmental Effects on Blood and  Liver	5-553
                 5.8.2.2  Developmental Effects on Skin	5-555
                 5.8.2.3  Developmental Effects on the  Retina	5-555
                 5.8.2.4  Developmental Effects on Teeth	5-555
              5.8.3  Effects on Birth  Outcomes	5-556
                 5.8.3.1  Infant Mortality and Embryogenesis	5-556
                 5.8.3.2  Birth Defects	 5-557
                         Table 5-40  Summary of recent epidemiologic studies of associations between
                                     Pb levels and neural tube defects.  	 5-558
                 5.8.3.3  Preterm Birth	5-559
                         Table 5-41  Summary of recent epidemiologic studies of associations between
                                     Pb levels and preterm birth.	 5-560
                 5.8.3.4  Low Birth Weight/Fetal Growth	5-566
                         Table 5-42  Summary of recent epidemiologic studies of associations between
                                     Pb levels and low birth weight and fetal growth.	 5-567
              5.8.4  Effects on Male Reproductive Function	5-580
                 5.8.4.1  Effects on Sperm/Semen Production, Quality, and Function	5-581
                         Table 5-43  Summary of recent epidemiologic studies of associations between
                                     Pb levels and effects on sperm and semen.	 5-582
                 5.8.4.2  Effects on Hormone Levels	5-595
                         Table 5-44  Summary of recent epidemiologic studies of associations between
                                     Pb levels and hormones for males.	 5-596
                 5.8.4.3  Fertility	 5-603
                 5.8.4.4  Effects on Morphology and Histology of Male Sex Organs	5-605
                 5.8.4.5  Summary  of Effects on  Male Reproductive Function	5-605
              5.8.5  Effects on Female Reproductive Function	5-606
                 5.8.5.1  Effects on Female Sex Endocrine System and Estrus Cycle	5-607
                         Table 5-45  Summary of recent epidemiologic studies of associations between
                                     Pb levels and hormones for females.	 5-608
                 5.8.5.2  Effects on Fertility	5-615
                         Table 5-46  Summary of recent epidemiologic studies of associations between
                                     Pb levels and fertility for females.	 5-676
                 5.8.5.3  Ovaries, Embryo Development, Placental function, and Spontaneous Abortions	5-622
                         Table 5-47  Summary of recent epidemiologic studies of associations between
                                     Pb levels and spontaneous abortions. 	 5-623
                 5.8.5.4  Effects on Breast  Milk	5-627
                 5.8.5.5  Summary  of Effects on  Female Reproductive Function	5-628
              5.8.6  Summary and Causal Determination	5-629
                 5.8.6.1  Effects on Development	5-629
                 5.8.6.2  Effects on Birth Outcomes 	5-630
                 5.8.6.3  Effects on Male Reproductive  Function 	5-631
                 5.8.6.4  Effects on Female Reproductive Function	5-632
November 2012                                 xvii                    Draft - Do Not Cite or Quote

-------
                         Table 5-48  Summary of evidence supporting reproductive and developmental
                                     causal determinations.	 5-633
           5.9  Effects on Other Organ Systems	 5-637
              5.9.1  Effects on the Hepatic System	5-637
                 5.9.1.1  Summary of Key Findings of the Effects on the Hepatic System (2006 Pb AQCD)_ 5-637
                 5.9.1.2  Recent Epidemiologic Studies 	5-638
                 5.9.1.3  Recent Toxicological Studies	5-639
                 5.9.1.4  Summary of Effects  on the Hepatic System	5-645
              5.9.2  Effects on the Gastrointestinal System	5-646
                 5.9.2.1  Summary of Key Findings on the Effects on the Gastrointestinal System
                         (2006 Pb AQCD)	 5-646
                 5.9.2.2  Recent Epidemiologic Studies 	5-646
                 5.9.2.3  Recent Toxicological Studies	5-647
                 5.9.2.4  Summary of Gastrointestinal Effects	5-648
              5.9.3  Effects on the Endocrine System	5-648
                 5.9.3.1  Summary of Key Findings of the Effects on the Endocrine System (2006 Pb AQCD)5-648
                 5.9.3.2  Recent Epidemiologic Studies 	5-649
                 5.9.3.3  Recent Toxicological Studies	5-651
                 5.9.3.4  Summary of Endocrine Effects	5-652
              5.9.4  Effects on Bone and Teeth	5-653
                 5.9.4.1  Summary of Key Findings of the Effects on Bone and Teeth (2006 Pb AQCD)	5-653
                 5.9.4.2  Recent Toxicological and Epidemiologic Studies	5-654
                 5.9.4.3  Summary of Effects  on Bone  and Teeth	5-658
              5.9.5  Effects on Ocular Health 	5-659
                 5.9.5.1  Summary of Key Findings of the Effects on Ocular Health (2006 Pb AQCD) 	5-659
                 5.9.5.2  Recent Toxicological and Epidemiologic Studies	5-659
              5.9.6  Effects on the Respiratory System	5-661
           5.10 Cancer	 5-667
              5.10.1 Cancer Incidence and Mortality	5-663
                         Table 5-49  Summary of recent epidemiologic studies of cancer incidence
                                     and overall cancer mortality.	 5-664
                 5.10.1.1 Overall Cancer Mortality	5-676
                 5.10.1.2 Overall Cancer Incidence	5-677
                 5.10.1.3 Lung Cancer	5-677
                 5.10.1.4 Brain Cancer	5-679
                 5.10.1.5 Breast Cancer	5-680
                 5.10.1.6 Other Cancers	5-681
                 5.10.1.7 Animal Models of Carcinogenicity	5-682
              5.10.2 Cancer Biomarkers	5-684
              5.10.3 Modes of Action for Pb-induced Carcinogenicity	5-684
                 5.10.3.1 Neoplastic Transformation Studies, Human Cell Cultures	5-685
                 5.10.3.2 DMA and Cellular Damage	5-686
                 5.10.3.3 Epigenetics	5-695
              5.10.4 Effects of Pb within Mixtures	5-696
              5.10.5 Summary  and Causal Determination	5-697
                         Table 5-50   Summary of evidence supporting cancer and genotoxicity
                                     causal determinations.	 5-707
           References for Chapter 5	 5-702

           POTENTIALLY AT-RISK POPULATIONS	6-1

                         Table 6-1     Classification of evidence for potential at-risk factors.	 6-4
           6.1  Physiological Factors that Influence the Internal Distribution of Pb	 6-4
           6.2  Population Characteristics Potentially Related to Differential Pb Exposure	 6-6
              6.2.1  Age	6-6
                 6.2.1.1  Early Childhood	6-6
                         Table 6-2   Blood Pb levels by age and sex, 2009-2010 NHANES.	 6-8
                         Table 6-3   Percentage of children within six categories/brackets of blood Pb
                                     levels,  1999-2004 NHANES.	 6-9
                 6.2.1.2  Older Adulthood 	6-10
              6.2.2  Sex	  6-11
              6.2.3  Race and  Ethnicity	6-11
November 2012                                xviii                    Draft - Do Not Cite or Quote

-------
                         Figure 6-1   Percent distribution of blood Pb levels by race/ethnicity among U. S.
                                     children (1-5 years).	 6-13
                         Figure 6-2   Soil Pb concentration exposure among the population of three
                                     parishes within greater metropolitan New Orleans.	 6-15
              6.2.4  Socioeconomic Status (SES)	6-16
              6.2.5  Proximity to Pb Sources	6-17
              6.2.6  Residential Factors	6-18
                         Table 6-4   Regression of log-transformed blood Pb level of children 12-60
                                     months old on various factors related to housing condition, from
1999-2004 NHANES dataset.
6.3 Factors Potentially Related to Increased Risk of Pb-lnduced Health Effects
6.3.1 Aqe
6.3.1.1 Childhood
6.3.1.2 Older Adulthood
6.3.2 Sex
6.3.3 Genetics
6.3.3.1 Aminolevulinate Dehydratase
6.3.3.2 Vitamin D Receptor
6.3.3.3 Methylenetetrahydrofolate reductase
6.3.3.4 Hemochromatosis
6.3.3.5 Other Genetic Polymorphisms
6.3.4 Pre-existinq Diseases/Conditions
6.3.4.1 Autism
6.3.4.2 Diabetes
6.3.4.3 Hypertension
6.3.5 Smoking Status
6.3.6 Socioeconomic Status
6.3.7 Race/Ethnicity
6.3.8 Body Mass Index
6.3.9 Alcohol Consumption
6.3. 10 Nutritional Factors
6.3.10.1 Calcium
6.3.10.2 Iron
6.3.10.3 Folate
6.3.10.4 Protein
6.3.11 Stress
6.3.12 Maternal Self-Esteem
6.3.13 Cognitive Reserve
6.3.14 Other Metal Exposure
6.3.14.1 Cadmium
6.3.14.2 Manganese
6.4 Summary
6-19
6-20
6-20
6-21
6-22
6-23
6-27
6-27
6-28
6-29
6-29
6-29
6-30
6-31
6-31
6-32
6-33
6-34
6-34
6-35
6-36
6-36
6-36
6-37
6-37
6-37
6-38
6-38
6-39
6-39
6-39
6-40
6-40
                         Table 6-5   Summary of evidence for factors that potentially increase the risk of
                                     Pb-related health effects.	 6-41
           References for Chapter 6	 6-45

           ECOLOGICAL EFFECTS OF LEAD	7-1

           7.1  Introduction to Ecological Concepts	 7-1
              7.1.1  Ecosystem Scale, Function, and Structure	7-2
              7.1.2  Ecosystem Services	 7-4
              7.1.3  Critical Loads as an  Organizing Principle for Ecological Effects of Atmospheric Deposition  7-6
           7.2  Fate and Transport of Pb in Ecosystems	 7-7
                         Figure 7-1   Fate of atmospheric Pb in ecosystems. 	 7-7
              7.2.1  Fate and Transport	 7-8
              7.2.2  Ecosystem Exposure,  Lag Time and Re-entrainment of Historically Deposited Pb 	7-10
                         Table 7-1   Comparison among several metals: Time to achieve 95% of steady
                                     state metal concentration in soil; example in a temperate system.	 7-12
              7.2.3  Concentrations in Non-Air Media	7-13
                         Table 7-2   Ambient Pb Concentrations in Non-Air Media and Biota Considered
                                     for Ecological Assessment.	 7-14
           7.3  Terrestrial Ecosystem Effects	 7-15
November 2012                                 xix                    Draft - Do Not Cite or Quote

-------
               7.3.1  Introduction to Effects of Pb on Terrestrial Ecosystems 	7-15
               7.3.2  Soil Biogeochemistry and its Influence on Bioavailability 	7-17
                  7.3.2.1  pH, CEC and Salinity 	7-17
                  7.3.2.2  Organic Matter	7-19
                  7.3.2.3  Aging	 7-20
               7.3.3  Bioavailability in Terrestrial Systems	7-22
                          Figure 7-2   Conceptual diagram for evaluating bioavailability processes and
                                       bioaccessibility for metals in soil, sediment, or aquatic systems.	 7-23
                          Figure 7-3   Schematic diagram of the biotic ligand model.	 7-24
                  7.3.3.1  Terrestrial Plants	7-25
                  7.3.3.2  Terrestrial Invertebrates	7-34
                  7.3.3.3  Terrestrial Vertebrates 	7-37
                  7.3.3.4  Food Web	 7-40
               7.3.4  Biological Effects of Pb in Terrestrial Systems	7-42
                  7.3.4.1  Terrestrial Plants and Lichen 	7-43
                  7.3.4.2  Terrestrial Invertebrates	7-48
                  7.3.4.3  Terrestrial Vertebrates 	7-53
               7.3.5  Exposure and Response of Terrestrial Species	7-57
               7.3.6  Terrestrial Community and Ecosystem Effects	7-60
               7.3.7  Critical Loads in Terrestrial Systems	7-66
               7.3.8  Soil Screening Levels	7-68
               7.3.9  Characterization of Sensitivity and Vulnerability	7-69
                  7.3.9.1  Species Sensitivity	7-69
                  7.3.9.2  Nutritional Factors	7-69
                  7.3.9.3  Soil Aging and Site-Specific Bioavailability	7-70
                  7.3.9.4  Ecosystem Vulnerability	7-70
               7.3.10 Ecosystem Services Associated with  Terrestrial Systems	7-72
               7.3.11 Synthesis of New Evidence for Pb Effects in Terrestrial Systems	7-73
               7.3.12 Causal Determinations for Pb in Terrestrial Systems	7-78
                  7.3.12.1 Reproductive and Developmental Effects-Terrestrial Biota	7-78
                  7.3.12.2 Growth Effects-Terrestrial Biota	7-80
                  7.3.12.3 Survival-Terrestrial Biota	7-81
                  7.3.12.4 Neurobehavioral Effects-Terrestrial Biota	7-83
                  7.3.12.5 Hematological Effects-Terrestrial Biota	7-84
                  7.3.12.6 Physiological Stress-Terrestrial  Biota	7-85
                  7.3.12.7 Community and Ecosystem Level Effects-Terrestrial Biota	7-86
            7.4   Aquatic Ecosystem Effects	 7-89
               7.4.1  Introduction to Effects of Pb on Aquatic Ecosystems 	7-89
               7.4.2  Biogeochemistry and Chemical Effects of Pb in Freshwater and Saltwater Systems	7-91
                  7.4.2.1  Other Metals	7-93
                  7.4.2.2  Biofilm	 7-94
                  7.4.2.3  Carbonate	7-95
                  7.4.2.4  Dissolved Organic Matter (DOM)	7-95
                  7.4.2.5  Sulfides	7-97
               7.4.3  Introduction to Bioavailability and Biological Effects of Pb in Freshwater Ecosystems	7-97
               7.4.4  Bioavailability in Freshwater Systems	7-100
                  7.4.4.1  Freshwater Plants and Algae	7-104
                  7.4.4.2  Freshwater Invertebrates	7-109
                  7.4.4.3  Freshwater Vertebrates	7-112
                  7.4.4.4  Food Web	 7-119
               7.4.5  Biological Effects of Pb in Freshwater Systems	7-121
                  7.4.5.1  Freshwater Plants and Algae	7-122
                  7.4.5.2  Freshwater Invertebrates	7-126
                  7.4.5.3  Freshwater Vertebrates	7-130
               7.4.6  Exposure and Response of Freshwater Species	7-143
               7.4.7  Freshwater Community and Ecosystem Effects	7-147
               7.4.8  Critical Loads in Freshwater Systems	7-151
               7.4.9  Characterization of Sensitivity and Vulnerability in Freshwater Systems 	7-152
                  7.4.9.1  Seasonally-Affected Physiological Changes	7-152
                  7.4.9.2  Increased Nutrient Uptake  	7-153
                  7.4.9.3  Temperature and pH	7-153
                  7.4.9.4  Lifestage	7-154
November 2012                                  xx                      Draft - Do Not Cite or Quote

-------
                  7.4.9.5  Species Sensitivity	7-154
                  7.4.9.6  Ecosystem Vulnerability	7-156
               7.4.10 Ecosystem Services Associated with Freshwater Systems	7-156
               7.4.11 Synthesis of New Evidence for Pb Effects in Freshwater Ecosystems	7-157
               7.4.12 Causal Determinations for Pb in Freshwater Systems 	7-161
                  7.4.12.1  Reproductive and Developmental Effects-Freshwater Biota 	7-161
                  7.4.12.2 Growth Effects-Freshwater Biota	7-163
                  7.4.12.3 Survival-Freshwater Biota	7-166
                  7.4.12.4 Neurobehavioral Effects-Freshwater Biota	7-169
                  7.4.12.5 Hematological  Effects-Freshwater Biota	7-170
                  7.4.12.6 Physiological Stress-Freshwater Biota	7-172
                  7.4.12.7 Community and Ecosystem Level Effects-Freshwater Biota	7-173
               7.4.13 Introduction to Bioavailability and Biological Effects of Pb in Saltwater Ecosystems	7-175
               7.4.14 Bioavailability of Pb in Saltwater Systems	7-177
                  7.4.14.1  Saltwater Plants and Algae	7-178
                  7.4.14.2 Saltwater Invertebrates	7-179
                  7.4.14.3 Saltwater Vertebrates	7-182
                  7.4.14.4 Marine Food Web	7-183
               7.4.15 Biological Effects of Pb in Saltwater Systems	7-184
                  7.4.15.1  Saltwater Algae and Plants	7-185
                  7.4.15.2 Saltwater Invertebrates	7-185
                  7.4.15.3 Saltwater Vertebrates	7-188
               7.4.16 Exposure and Response of Saltwater Species	7-189
               7.4.17 Community and Ecosystem  Effects in Saltwater Systems	7-191
               7.4.18 Characterization of Sensitivity and Vulnerability in Saltwater Species 	7-193
                  7.4.18.1  Seasonally Affected Physiological Changes	7-193
                  7.4.18.2 Lifestage	7-194
                  7.4.18.3 Species Sensitivity	7-194
               7.4.19 Ecosystem Services Associated with Saltwater Systems	7-195
               7.4.20 Synthesis of New Evidence for Pb Effects in Saltwater Systems	7-196
               7.4.21 Causal Determinations for Pb in Saltwater Systems	7-199
                  7.4.21.1  Reproductive and Developmental Effects-Saltwater Biota	7-199
                  7.4.21.2 Growth Effects-Saltwater Biota	7-201
                  7.4.21.3 Survival-Saltwater Biota	7-202
                  7.4.21.4 Neurobehavioral Effects-Saltwater Biota 	7-203
                  7.4.21.5 Hematological  Effects-Saltwater Biota	7-204
                  7.4.21.6 Physiological Stress-Saltwater Biota	7-204
                  7.4.21.7 Community and Ecosystem Level Effects-Saltwater Biota	7-205
            7.5   Causal Determinations for Ecological Effects of Pb	 7-206
                          Table 7-3    Summary of Pb causal determinations for plants, invertebrates and
                                      vertebrates.	 7-209
            7.6   Supplemental Material	 7-210
                          Table 7-4    Terrestrial plants, invertebrates and vertebrates; growth,
                                      reproduction and survival.	 7-210
                          Table 7-5    Freshwater plants, invertebrates and vertebrates; growth,
                                      reproduction and survival.	 7-223
                          Table 7-6    Saltwater plants, invertebrates, and vertebrates: growth,
                                      reproduction,  and survival. 	 7-243
            References for Chapter  7	 7-250
November 2012                                 xxi                     Draft - Do Not Cite or Quote

-------
Lead  Project  Team
Executive Direction
              Dr. John Vandenberg (Director)—National Center for Environmental Assessment-RTF
                Division, Office of Research and Development, U.S. Environmental Protection
                Agency, Research Triangle Park, NC
              Ms. Debra Walsh (Deputy Director)—National Center for Environmental Assessment-
                RTF Division, Office of Research and Development, U.S. Environmental Protection
                Agency, Research Triangle Park, NC
              Dr. Mary Ross (Branch Chief)—National Center for Environmental Assessment, Office
                of Research and Development, U.S. Environmental Protection Agency, Research
                Triangle Park, NC
              Dr. Reeder Sams (Acting Deputy Director)—National Center for Environmental
                Assessment-RTP Division, Office of Research and Development, U.S. Environmental
                Protection Agency, Research Triangle Park, NC

              Dr. Doug Johns (Acting Branch Chief)—National Center for Environmental Assessment,
                Office of Research and Development, U.S. Environmental Protection Agency,
                Research Triangle Park, NC


Scientific Staff
              Dr. Ellen Kirrane (Team Leader, ISA for Lead)—National Center for Environmental
                Assessment, Office of Research and Development, U.S. Environmental Protection
                Agency, Research Triangle Park, NC
              Dr. James Brown—National Center for Environmental Assessment, Office of Research
                and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                NC
              Mr. Allen Davis—National Center for Environmental Assessment, Office of Research
                and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                NC
              Dr. Jean-Jacques Dubois—National Center for Environmental Assessment, Office of
                Research and Development, U.S. Environmental Protection Agency, Research
                Triangle Park, NC
              Dr. Tara Greaver—National Center for Environmental Assessment, Office of Research
                and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                NC
              Dr. Erin  Hines—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Dr. Dennis Kotchmar—National Center for Environmental Assessment, Office of
                Research and Development, U.S. Environmental Protection Agency, Research
                Triangle Park, NC
November 2012                            xxii                 Draft - Do Not Cite or Quote

-------
              Dr. Meredith Lassiter—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Stephen McDow—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Qingyu Meng—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Dr. Elizabeth Oesterling Owens—National Center for Environmental Assessment, Office
                 of Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Molini Patel—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr Joseph P. Pinto—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr. Jennifer Richmond-Bryant—National Center for Environmental Assessment, Office
                 of Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Lindsay Stanek—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr. David Svendsgaard—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Lisa Vinikoor-Imler—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC


Technical Support Staff

              Ms. Marieka Boyd—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Mr. Kenneth J. Breito-Senior Environmental Employment Program, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Ellen Lorang—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Mr. J. Sawyer Lucy-Student Services Authority, National Center for Environmental
                 Assessment, Office of Research and Development, U.S. Environmental Protection
                 Agency, Research Triangle Park, NC
November 2012                            xxiii                  Draft - Do Not Cite or Quote

-------
              Ms. Deborah Wales—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Mr. Richard N. Wilson-National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Ms. Barbara Wright—Senior Environmental Employment Program, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
November 2012                            xxiv                  Draft - Do Not Cite or Quote

-------
Authors, Contributors,  and  Reviewers
Authors
             Dr. Ellen Kirrane (Team Leader, ISA for Lead)—National Center for Environmental
                Assessment, Office of Research and Development, U.S. Environmental Protection
                Agency, Research Triangle Park, NC
             Dr. Robyn Blain— Energy, Environment and Transportation, Environmental Science &
                Policy,  ICF International, Lexington, MA
             Dr. James Brown—National Center for Environmental Assessment, Office of Research
                and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                NC
             Dr. Philip Bushnell, National Health and Environmental Effects Research Laboratory,
                Office of Research and Development, U.S. Environmental Protection Agency,
                Research Triangle Park, NC
             Mr. Allen Davis—National Center for Environmental Assessment, Office of Research
                and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                NC
             Dr. Gary Diamond—Syracuse Research Corporation, Akron, NY
             Dr. Rodney Dietert—Cornell University College of Veterinary Medicine, Veterinary
                Medical Center, Ithaca, NY
             Dr. Jean-Jacques Dubois—National Center for Environmental Assessment, Office of
                Research and Development, U.S. Environmental Protection Agency, Research
                Triangle Park, NC
             Dr. Anne Fairbrother—Exponent, Inc., Bellevue, WA
             Dr. Jay Gandy—Department of Environmental and Occupational Health, University of
                Arkansas for Medical Sciences, Little Rock, AR
             Dr. Harvey Gonick—David Geffen School of Medicine, University of California-Los
                Angeles, Los Angeles, CA
             Dr. Margaret Graham—School of Geosciences, University of Edinburgh, Edinburgh,
                Scotland
             Dr. Tara Greaver—National Center for Environmental Assessment, Office of Research
                and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                NC
             Dr. Erin Hines—National Center for Environmental Assessment, Office of Research and
                Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
             Dr. Dennis Kotchmar—National Center for Environmental Assessment, Office of
                Research and Development, U.S. Environmental Protection Agency, Research
                Triangle Park, NC
November 2012                           xxv                 Draft - Do Not Cite or Quote

-------
              Dr. Meredith Lassiter—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Stephen McDow—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Qingyu Meng—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Dr. Bill Mendez—Energy, Environment and Transportation, Environmental Science &
                 Policy, ICF International, Fairfax, VA
              Dr. Howard Mielke—Center for Bioenvironmental Research, Tulane/Xavier Universities,
                 New Orleans, LA
              Ms. Chandrika Moudgal— Energy, Environment and Transportation, Environmental
                 Science & Policy, ICF International, Dublin, CA
              Dr. Elizabeth Oesterling Owens—National Center for Environmental Assessment, Office
                 of Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Katherine Palmquist—Exponent, Inc., Bellevue, WA
              Dr. Molini Patel—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr Joseph P. Pinto—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr. Jennifer Richmond-Bryant—National Center for Environmental Assessment, Office
                 of Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Stephen Rothenberg—National Institute of Public Health, Cuernavaca, Morelos,
                 Mexico
              Dr. Mary Jane Selgrade—Energy, Environment and Transportation, Environmental
                 Science & Policy, ICF International, RTP, NC
              Dr. Lindsay Stanek—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr. David Svendsgaard—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Lisa Vinikoor-Imler—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Virginia Weaver—Johns Hopkins Bloomberg School of Public Health, Baltimore,
                 MD
November 2012                            xxvi                 Draft - Do Not Cite or Quote

-------
              Dr. Marc Weisskopf—Department of Environmental Health and Department of
                 Epidemiology, Harvard School of Public Health, Harvard University, Boston,
                 Massachusetts
              Dr. John Pierce Wise, Sr.—Maine Center for Toxicology and Environmental Health,
                 Department of Applied Medical Sciences, Portland, ME
              Dr. Rosalind Wright—Harvard Medical School and School of Public Health, Harvard
                 University, Boston, MA
              Dr. Robert Wright—Harvard Medical School and School of Public Health, Harvard
                 University, Boston, MA


Contributors
              Mr. Brian Adams—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Dr. Halil Cakir— Office of Air Quality Planning and Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Ye Cao—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Laura Datko—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Mr. Josh Drukenbrod—Office of Air Quality Planning and Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, RTP, NC
              Ms. Meagan Madden—Oak Ridge Institute for Science and Education, National Center
                 for Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Mr. Mark Schmidt—Office of Air Quality Planning and  Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Katie Shumake—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Kaylyn Siporin—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Lauren Turtle—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Adrien Wilkie—Oak Ridge  Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC
November 2012                            xxvii                  Draft - Do Not Cite or Quote

-------
              Ms. Brianna Young—Oak Ridge Institute for Science and Education, National Center for
                 Environmental Assessment, Office of Research and Development,
                 U.S. Environmental Protection Agency, Research Triangle Park, NC


Reviewers

              Dr. David Bellinger - Harvard Medical School and Department of Environmental Health,
                 Harvard School of Public Health, Boston, MA
              Dr. Christal Bowman—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. David Buchwalter—Department of Toxicology, North Carolina State University,
                 Raleigh, NC

              Dr. Barbara Buckley—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Mr. Kevin Cavender—Office of Air Quality Planning and Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Rebecca C. Dzubow—Office of Children's Health Protection, U.S.  Environmental
                 Protection Agency, Washington, DC

              Dr. David DeMarini—National Health and Environmental Effects Research Laboratory,
                 Office of Research and Development, U.S. Environmental Protection Agency,
                 Research Triangle Park, NC

              Dr. Pam Factor-Litvak—Department of Epidemiology, Mailman School of Public Health,
                 New York, NY

              Dr. Gabriel Filippelli—Department of Earth Sciences, Indiana University-Purdue
                 University, Indianapolis, IN
              Dr. Andrew Friedland—Environmental Studies Program, Dartmouth College, Hanover,
                 NH

              Dr. Barbara Glenn—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Washington, DC
              Dr. Jeff Herrick—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr. Marion Hoyer—Office of Transportation and Air Quality, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Ann Arbor, MI

              Dr. Joseph Jacobson - Department of Psychiatry and Behavioral Neurosciences, Wayne
                 State University, Detroit, MI
              Dr. Douglas Johns—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr. Jay S. Kaufman - Department of Epidemiology, Biostatistics and Occupational
                 Health, McGill University, Montreal, Canada
November 2012                            xxviii                 Draft - Do Not Cite or Quote

-------
              Dr. Thomas Luben—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr. Karen Martin—Office of Air Quality Planning and Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Connie Meacham—National Center for Environmental Assessment, Office of
                 Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Marie Lynn Miranda—Environmental Sciences and Policy, Nicholas School of the
                 Environment, Duke University, Durham, NC
              Dr. Deirdre Murphy—Office of Air Quality Planning and Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Dr. Paul Mushak—PB Associates, Durham NC
              Dr. Kris Novak—National Center for Environmental Assessment, Office of Research and
                 Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Mr. David Orlin—Air and Radiation Law Office, Office of General Counsel, U.S.
                 Environmental Protection Agency, Washington, DC
              Dr. Meredith Pedde—Office of Transportation and Air Quality, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Ann Arbor, MI
              Dr. Pradeep Raj an—Office of Air Quality Planning and Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Ms. Joanne Rice—Office of Air Quality Planning and Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Dr. Mary Ross—National Center for Environmental Assessment, Office of Research and
                 Development, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Dr. Joel Schwartz—Department of Environmental Health, Harvard School of Public
                 Health, Boston, MA

              Mr. Jason Sacks—National Center for Environmental Assessment, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Ms. Ginger Tennant—Office of Air Quality Planning and Standards, Office of Air and
                 Radiation, U.S. Environmental Protection Agency, Research Triangle Park, NC
              Dr. Jay Turner—Environmental and Chemical Engineering Department, Washington
                 University, St. Louis, MO
              Dr. John Vandenberg—National Center for Environmental Assessment-RTP Division,
                 Office of Research and Development, U.S. Environmental Protection Agency,
                 Research Triangle Park, NC

              Dr. Robert W. Vanderpool—National Exposure Research Laboratory, Office of Research
                 and Development, U.S. Environmental Protection Agency, Research Triangle Park,
                 NC
              Dr. Nosratola Vaziri—Division of Nephrology and Hypertension, School of Medicine,
                 University of California, Irvine, Irvine, CA
November 2012                            xxix                 Draft - Do Not Cite or Quote

-------
              Ms. Debra Walsh—National Center for Environmental Assessment-RTF Division, Office
                 of Research and Development, U.S. Environmental Protection Agency, Research
                 Triangle Park, NC
              Dr. Nasser Zawia—Department of Biomedical and Pharmaceutical Sciences, University
                 of Rhode Island, Kingston, RI
November 2012                            xxx                  Draft - Do Not Cite or Quote

-------
Clean Air Scientific Advisory Committee

Lead  NAAQS  Review Panel

Chair of the Lead Review Panel
            Dr. Christopher H. Frey*, North Carolina State University, Raleigh, NC

Lead Review Panel Members
            Mr. George A. Allen*, Northeast States for Coordinated Air Use Management
               (NESCAUM), Boston, MA
            Dr. Herbert Allen, University of Delaware, Newark, DE
            Dr. Richard Canfield, Cornell University, Ithaca, NY
            Dr. Deborah Cory-Slechta, University of Rochester, Rochester, NY
            Dr. Cliff Davidson, Syracuse University, Syracuse, NY
            Dr. Philip E. Goodrum, Cardno ENTRIX, Syracuse, NY
            Dr. Sean Hays, Summit Toxicology, Allenspark, CO
            Dr. Philip Hopke, Clarkson University, Potsdam, NY
            Dr. Chris Johnson, Syracuse University,  Syracuse, NY
            Dr. Susan Korrick, Harvard Medical School, Boston, MA
            Dr. Michael Kosnett, University of Colorado School of Medicine, Denver, CO
            Dr. Roman Lanno, Ohio State University, Columbus, OH
            Mr. Richard L. Poirot, Vermont Agency  of Natural Resources, Waterbury, VT
            Dr. Joel Pounds, Pacific Northwest National Laboratory, Richland, WA
            Dr. Michael Rabinowitz, Marine Biological Laboratory, Newport, RI
            Dr. William Stubblefield, Oregon State University, Corvallis, OR
            Dr. Ian von Lindern, TerraGraphics Environmental Engineering, Inc., Moscow, ID
            Dr. Gail Wasserman, Columbia University, New York, NY
            Dr. Michael Weitzman, New York University School of Medicine, New York, NY
            * Members of the statutory Clean Air Scientific Advisory Committee (CASAC)
               appointed by the EPA Administrator

Science Advisory Board Staff
            Mr. Aaron Yeow, Designated Federal Officer, U.S. Environmental Protection Agency,
               Science Advisory Board (1400R), 1200 Pennsylvania Avenue, NW, Washington, D.C.
               20460-0001, Phone: 202-564-2050, Fax: 202-565-2098, (yeow.aaron@epa.gov)


November 2012                         xxxi                Draft - Do Not Cite or Quote

-------
Acronyms and Abbreviations
a       alpha
aT      the extent of DNA denaturation per cell
A       Angstrom (10"10 meter)
AA     African American; arachidonic acid, atomic
        absorption
AALM  All Ages Lead Model
AAS    atomic absorption (spectrophotometry,
        spectrometry, spectroscopy)
Ab      amyloid-beta peptide
ABL    atmospheric boundary layer
ACE    angiotensin converting enzyme
ACF    Apalachicola, Chattahoochee, and Flint River
        Basin
ACh    acetylcholine
ACP    acid phosphatase
ACR    acute to chronic ratio
Acyl-CoA
        acyl-coenzyme A
AD     axial diffusivity
ADHD  attention deficit hyperactivity disorder
ADP    adenosine diphosphate
AE     anion exchanger
AEC    adenylate energy charge
AERMOD
        atmospheric dispersion model
AF      absorbed fraction; absorption fraction
aff'd    affirmed
A/G     albumin/globulin
Ag      silver
AGL    above-ground level
A-horizon:
        Topsoil horizon (surface soil)
AKI     acute kidney injury
Al      aluminum
ALA    aminolevulinic acid
ALAD  aminolevulinic acid dehydratase;
ALAD 1-1:
        aminolevulinate delta-dehydratase 1-1
ALAD-2
        aminolevulinate delta-dehydratase-2
ALD    aldehyde dehydrogenase
ALM    Adult Lead Methodology
ALP    alkaline phosphatase
ALS    Amyotrophic Lateral Sclerosis (Lou Gehrig's
        disease)
ALT    alanine aminotransferase
AM     Alveolar macrophages
AMF    arbuscular mycorrhizal fungi
AMP    adenosine monophosphate
ANC    acid neutralizing capacity; absolute neutrophil
        counts
ANF    atrial natriuretic factor
Angll    renal angiotensin II
ANOVA analysis of variance
ANPR   advance notice of proposed rulemaking
AOP    adverse outcome pathway
AP-1    activator protein-1
Apal    polymorphism of the VDR in humans
APC    antigen-presenting cell
APOE   Apolipoprotein E
APRT   adenine phosphoribosyltransferase
AQCD   Air Quality Criteria Document
AQS    (U.S. EPA) Air Quality System (database)
Ar      argon
As      arsenic
AST    aspartate aminotransferase
ASV    anode stripping voltammetry
ATLD   ataxia-telangiectasia-like  disorder
ATOFMS:
        aerosol time-of-flight mass spectrometry
ATP     adenosine-triphosphate
ATPase  adenosine triphosphatase; adenosine
        triphosphate synthase
ATS     American Thoracic Society
ATSDR  Agency for Toxic Substances and Disease
        Research
November 2012
                                            xxxu
             Draft - Do Not Cite or Quote

-------
Au
gold
avg     average
AVS    acid-volatile sulfides
a-wave  initial negative deflection in the
        electroretinogram
AWQC  Ambient Water Quality Criteria
P

3P-HSD
Beta; Beta coefficient; regression coefficient;
standardized coefficient
        3-beta-hydroxysteroid dehydrogenase
17P-HSD:
        17-beta-hydroxysteroid dehydrogenase
Ba      barium
BAF    bioaccumulation factors
BAL    2,3-dimercaptopropanol
BASC  Behavior Assessment System for Children
BASC-PRS:
        Behavior Assessment System for Children-
        Parent Ratings Scale
BASC-TRS:
        Behavior Assessment System for Children-
        Teacher Rating Scale
BC     black carbon, soot
BCB    blood cerebrospinal fluid barrier
B-cell   Bone marrow-derived lymphocytes, B
        lymphocyte
BCF    bioconcentration factors
Bcl-x   member of the B-cell lymphoma-2 protein
        family
Bcl-xl  B-cell lymphoma-extra large
B-horizon:
        subsoil horizon
bio     biological
Bi2S3   bismuth (III) sulfide
BK     biokinetics
BLM   biotic ligand model
BMD   benchmark dose; bone mineral density
BMDL  benchmark dose limit
BMI    body mass index
BMP   bone morphogenetic protein
BMS   Baltimore Memory Study
BMW  battery manufacturing workers
BP     blood pressure
Br      bromine
BR     bronchial responsiveness
BrdU   bromo-2'-deoxyuridine
8-Br-GMPc:
        8-bromo-cyclic guanosine monophosphate
Bs-horizon;
        subsoil horizon with accumulation of
        sesquioxides
BSI     Brief Symptom Inventory
BSID-II Bay ley Scale for Infant Development-II
BsmI   polymorphism of the VDR in humans
Bt20    Birth-to-age Twenty (cohort)
BUN   blood urea nitrogen
bw     body weight
b-wave  initial positive deflection in the
        electroretinogram
C       carbon; Celsius; soil or dry sediment Pb
        concentration; Caucasian; Cysteine
Ca     calcium
Ca2+    calcium ion
CAA   Clean Air Act
CaBP   calcium binding protein
CaCl2   calcium chloride
CaCO3  calcium carbonate; calcite
CaEDTA
        calcium ethylenediaminetetraacetic acid
CaMKII
        calmodulin-dependent protein kinase II
cAMP  cyclic adenosine monophosphate
CASAC Clean Air Scientific Advisory Committee
CASM  Comprehensive Aquatic Systems Model
CaSO4  calcium sulfate
CaS04-2H20:
        gypsum
CAT    catalase
CBLI   cumulative blood Pb index
CBSA  core based statistical area
CCSEM
        computer-controlled scanning electron
        microscopy
CD     cluster of differentiation
Cd     cadmium
Cd(II)   cadmium (II)
Cd2+    cadmium ion
November 2012
                                               XXXlll
                                                                Draft - Do Not Cite or Quote

-------
CD3+   T lymphocyte
CD4+   T helper cell
CDC    Centers for Disease Control
CEA    carcinoembryonic antigen
CEC    cation exchange capacity
cent    central
cert.    certiorari
cf      correction factor; latin abbreviation for
        conferre (used as "compared with)
CFL    constant flux layer
CFR    Code of Federal Regulations
cGMP  cyclic guanosine monophosphate
C-H    carbon-hydrogen (bond)
CHAD  Consolidated Human Activity Database
ChAT   chlorine acetyltransferase
CHD    coronary heart disease
CHL    Chinese hamster lung
CHO    Chinese hamster ovary cell line
C-horizon:
        Soil horizon underneath A- and B-horizons,
        may contain lumps or shelves of rock and
        parent material
CHV79 Chinese hamster lung cell line
CI      confidence interval
Cir.     circuit
CKD    chronic kidney disease
CKD-EPI:
        Chronic Kidney Disease Epidemiology
        Collaboration
CL     confidence limit
Cl      chlorine
Cl"     chlorine ion
C12     molecular chlorine
CLACE 5:
        Fifth Cloud and Aerosol Characterization
        Experiment in the Free Troposphere campaign
CLS    Cincinnati Lead Study
CO     carbon monoxide
CO2    carbon dioxide
CO32"   carbonate ion
Co     cobalt
CoA    coenzyme A
COD    coefficient of difference
Coeff   coefficient
COMP aT:
        The percentage of sperm with increased
        sensitivity to DNA denaturation
Con    control
Cone.   concentration
Cong.   congress
Corr    correlation
COX    cyclooxygenase; cytochrome oxidase subunits
COX-2  cyclooxygenase-2
cPLA2  cytosolic phospholipidase A2
CPRI    coarse particle rotary impactor
CPRS-R
        Conners' Parent Rating Scale-Revised
Cr      chromium; creatine
C-R    concentration-response (relationship)
Cr III    chromium III
CRAC  Ca2+ release activated calcium
CRACI calcium release activated calcium influx
CREB  cyclic adenosinemonophosphate (cAMP)
        response element-binding
CRP    C-reactive protein
CSF    colony-stimulating factor
CSN    Chemical Speciation Network
CT     zinc-adequate control
Cu     copper
Cu(II)  copper (II)
CV     coefficient of variation
CVD    cardiovascular disease
CYP    cytochrome
CYPlAl,CyplAl:
        cytochrome P450 family 1 member Al
CYP 1A2, CyplA2:
        cytochrome P450 family 1 member A2
CYPP450:
        cytochrome P450
A       delta, difference, change
A5-3P-HSD :
        delta-5-3-beta-hydroxysteroiddehydrogenase
5-ALA
        5-aminolevulinic acid; delta-aminolevulinic
        acid
November 2012
                                               xxxiv
              Draft - Do Not Cite or Quote

-------
5-ALAD:

D2,D3
D50
d
db,dB
DbH
DBF
DENA

dep
dev.
DEX
DG
2-dG
DHAA
diff
DIT
DMPS
DMSA
DMSO
DNA
DoAD
DOC
DOM
DP-109
DP-460
DR
DRD4
DRD4.7
DRUM
D-serine

DSM-IV

DTK
DTPA

E
E2
e
delta-aminolevulinic acid dehydratase
dopamine receptors
size at 50% efficiency
day(s); depth
decibel
dopamine beta-hydroxylase
diastolic blood pressure
Denali National Park and
Preserve, Alaska
dependent
deviation
exogenous dexamethasone
degenerate gyrus
2-deoxyguanosine
dehydroascorbate
differentiation
developmental immunotoxicity
2,3-dimercaptopropane-l-sulfonic acid
dimercaptosuccinic acid
dimethyl sulfoxide
deoxyribonucleic acid
developmental origins of adult disease
dissolved organic carbon
dissolved organic matter
metal chelator
metal chelator
diet-restricted
dopamine 4 receptor
dopamine 4 receptor repeat alleles
Davis Rotating Unit for Monitoring

neuronal signal

Diagnostic Statistical Manual-IV
delayed-type hypersensitivity
diethylene triamine pentaacetic acid;
technetium-diethylenetriamine-pentaacetic acid
east; expression for exposure
estradiol
exponential function
EC      elemental carbon, endothelial cell
ECjo     effect concentration for 10% of test population
EC2o     effect concentration for 20% of test population
EC50     effect concentration for 50% of test population
ECG     electrocardiography; electrocardiogram
ECOD   7-ethoxycoumarin-o-deethylase
Eco-SSLs:
         ecological soil screening levels
EDjo     effect dose for 10% of population
EDTA   ethylenediaminetetraacetic acid
EPS     electrical field stimulus
EGF     epidermal growth factor
EGFR   epidermal growth factor receptor
eGFR    estimated glomerular filtration rate
Eh      electrochemical potential
E-horizon:
         Eluviated horizon; soil horizon which is
         eluviated or leached of mineral and/or organic
         content
EI-MS   electron impact ionization mass spectrometry
ELPI     electrical low-pressure impactor
eNOS    endothelial nitric oxide synthase
EOG     end-of-grade
EPA     U.S. Environmental Protection Agency
EPT     ephemeroptera, plecoptera, trichoptera
ER      endoplasmic reticulum
Erg-1     ether-a-go-go related gene
ERG     electroretinogram
ERK     extracellular signal regulated kinase
ERK1/2
         extracellular signal-regulated kinases 1 and 2
EROD   7-ethoxyresorufin-o-deethylase
ESCA   electron spectroscopy for chemical analysis
ESI-MS
         electrospray ionization mass spectrometry
ESRD   end stage renal disease
ET      endothelin
ET-1     vasoconstrictor endothelin-1
ETA-type receptors:
         endothelin type A receptors
ETS     environmental tobacco smoke
EU      European Union
November 2012
                                                 XXXV
                                                                  Draft - Do Not Cite or Quote

-------
EURO  European emission standard
eV     electronvolts
EXAFS
        X-ray absorption fine structure spectroscopy
F0      filial "zero" generation (parental stock)
Fj      first filial generation (offspring of F0)
F2      second filial generation (offspring of FI)
FAA    Federal Aviation Agency
FAI     free androgen index
FAS    apoptosis stimulating fragment
Fas-L   apoptosis stimulating fragment ligand
Fe      iron
Fe(III)  iron III
FEM    Federal equivalence method
FEV1   forced expiratory volume in 1 second
FI      fixed interval
FI-Ext  fixed interval with extinction
Fl      fluoride
Fokl    polymorphism of the VDR in humans
FR     Federal Register (Notice)
FrA    fractional anisotropy
FR-FI   fixed ratio-fixed interval
FRM    Federal reference method
FSH    follicle-stimulating hormone
FSIQ    full scale intelligence quotient (IQ)
FT3    free triiodothyronine
FT4    free thyroxine
G      pregnancy; guanine
G2     gap 2 Phase
g, kg, mg, ug, ng,  pg:
        gram(s),  kilogram(s), milligram(s),
        microgram(s),, nanogram(s), picogram(s)
G93A   mouse model
GAAR  Gates of the Arctic National Park and Preserve,
        Alaska
GABA  y-aminobutyric acid; gamma aminobutyric acid
GABAergic:
        inhibitory neurons that relase the
        neurotransmitter GABA
GAD    generalized anxiety disorder
GC     gas chromatography
G-CSF  granulocyte colony-stimulating factor
GD      gestational day
GEE     generalized estimating equations
GFAAS  graphite furnace atomic absorption
         spectrometry
GFAP    glial fibrillary acidic protein
GFR     glomerular filtration rate
GGT     gamma-glutamyl transpeptidase
GH      growth hormone
GHRH   growth-hormone releasing hormone
GI       gastrointestinal
GIS      Geographic Information System
G+L     pregnancy plus lactation
GLAC   Glacier National Park, Montana
GLE     gestationally-lead exposed
GM      geometric mean
GMR    geometric mean blood Pb ratio
GnRH    gonadotropin-releasing hormone
G6PD    glucose-6-phosphate dehydrogenase
GPEI    glutathione transferase P (G ST-P) enhancer I
GPT     glutamate pyruvate transaminase
GPx     glutathione peroxidase
GPX1    gene encoding for glutathione peroxidase 1
GR      glutathione reductase
GRP78   glucose-regulated protein 78
GRP94   glucose-regulated protein 94
Grp      glucose-regulated protein
GSD     geometric standard deviation
GSH     glutathione
GSSG    glutathione disulfide
GST     glutathione S-transferase
GSTM1  glutathione S-transferase Mu 1
GST-P   glutathione transferase P
GTP     guanosine-5'-triphosphate; guanine
         triphosphate
H       hydrogen
H+       hydrogen ion
h        hour(s)
ha       hectare
HAD    hydroxy-alkenals
HAP     hazardous air pollutant
November 2012
                                                xxxvi
               Draft - Do Not Cite or Quote

-------
Hb     hemoglobin
HC5    acute toxicity hazardous concentration for 5%
        of species
HC10    acute toxicity hazardous concentration for 10%
        of species
HC1    hydrochloric acid
HCO3"  bicarbonate; hydrogen carbonate
Hct     hematocrit
HDL    high-density lipoprotein
HERO  Heatlh and Environmental Research online
        (database)
HEW   U.S. Department of Health, Education, and
        Welfare
HF     hydrogen fluoride
HFE    hemochromatosis gene
HFE C282Y:
        hemochromatosis gene with C282 Y mutation
HFE H63D :
        hemochromatosis gene with H63D mutation
Hg     mercury
HgCl2   mercury(II) chloride
5-HIAA
        5-hydroxyindoleacetic acid
HIV    human immunodeficiency virus
HLA-DRB:
        human leukocyte antigen genes
HMEC  human dermal microvascular endothelial cells
HMGR  3-hydroxy-3-methylglutaryl-CoA reductase
HMOX-1:  heme oxygenase-1
HNO3   nitric acid
HO-1   heme oxygenase; heme oxidase-1
H2O    water
H2O2    hydrogen peroxide
HOME  Home Observation for Measurement of the
        Environment
HPA    hypothalamic-pituitary-adrenal
HPb,h-Pb:
        high Pb
HPG    hypothalamic-pituitary-gonadal
HPLC   high-performance liquid chromatography
HPRT   hypoxanthine-guanine
        phosphoribosyltransferase
HPT    hyperparathyroidism; hypothalamic-pituitary-
        thyroid
HR     heart rate; hazard ratio
HRV    heart rate variability
hsp     heat shock proteins
5HT    serotonin
5-HT    5-hydroxytryptamine
5-HT2B
        5-hydroxytryptamine receptor 2B
hTERT  telomerase reverse transcriptase
HVA    homovanillic acid
I       interstate
IARC   International Agency for Research on Cancer
IC50    half maximal inhibitory concentration
ICAP   inductively coupled argon plasma
ICP-AES
        Inductively coupled plasma atomic emission
        spectroscopy
ICPMS, ICP-MS:
        Inductively coupled plasma mass spectrometry
ICR    imprinting control region
ICRP    International Commission on Radiological
        Protection
ID      identification
IDA    iron-deficiency anemia
IDE    insulin-degrading enzyme
IEPA    Illinois Environmental Protection Agency
IEUBK  Integrated Exposure Uptake Biokinetic
IFN-y   interferon-gamma
Ig      immunoglobulin
IgA    immunoglobulin A
IgE     immunoglobulin E
IGF-1   insulin-like growth factor 1
IgG    immunoglobulin G
IgM    immunoglobulin M
IHD    ischemic heart disease
IL      interleukin
IL-1 p   interleukin-1 Beta
IL-2    interleukin-2
IL-4    interleukin-4
IL-5    interleukin-5
IL-6    interleukin-6
IL-8    interleukin-8
November 2012
                                               xxxvii
              Draft - Do Not Cite or Quote

-------
IL-10    interleukin-10
IL-12    interleukin-12
IMPROVE:
         Interagency Monitoring of Protected Visual
         Environment
IMT     intimal medial thickening
INL     inner neuroblastic layers of the retina
iNOS    inducible nitric oxide synthase
IOM     Institute of Medicine (provides health
         information to the NAS [National Academy of
         Sciences])
i.p.      intraperitoneal (route)
IQ       intelligence quotient
IQR     interquartile range
IRE1     inositol-requiring enzyme-1
IRP     integrated review plan
ISA     Integrated Science Assessment
ISC-PRIME
         Industrial Source Complex-Plume Rise Model
         Enhancements
ISF      intake slope factor
ISL      inertial sublayer
ISO     International Standards Organization
i.v.      intravenous
IVBA    in vitro bioaccessibility
IVF     in vitro fertilization
JNK     jun N-terminal kinase
K       Kelvin (temperature); potassium;
         resuspsension factor
K+       potassium ion
K0.5     concentration of free metal giving half
         maximal metal-dependent release
KART   Karters of American Racing Triad
IQ       dissociation constant
Kd      partition coefficient; ratio of the metal
         concentration in soil to that in soil solution
kDa, kD
         kiloDalton
KEDI-WISC:
         Korean Educational Development Institute -
         Wechsler Intelligence Scale for Children
6-keto-PGFla:
         6-keto-prostaglandin Fla (vasodilatory
         prostaglandin)
keV     kiloelectron volt
Ki-67    antigen, cell cycle and tumor growth marker
Kim-1    kidney injury molecule-1
Kinder-KITAP :
         Kinder-Testbatterie zur
         Aufmerksamkeitsprufung fur Kinder
K-ras    specific proto-oncogene
K-XRF  K-x-ray fluorescence method of scanning for
         bone Pb
A       lambda; resuspension rate
L        length
L, dL, mL:
         Liter(s) [1000 mL/L], deciliter(s) [100 mL/dL],
         milliliter(s) [1 mL/mL]
LA-ICP-MS:
         laser ablation inductively coupled plasma mass
         spectrometry
LC50     lethal concentration (at which 50% of exposed
         organisms die)
LD50     lethal dose (at which 50% of exposed
         organisms die)
LDH     lactate dehydrogenase
LDL     low-density lipoproteins
LFH-horizons:
         organic soil horizons located above well-
         drained surface soil
LF/HF   low frequency to high frequency ratio
LH      luteinizing hormone
LHRH   luteinizing hormone releasing hormone
LINE    long interspersed nuclear element
LINE-1  long interspersed nucleotide elements-1
LLNA   local lymph node  assay
In       natural logarithm
L-NAME:
         L-NG-nitroarginine methyl ester
L-NOARG:
         L-nitroarginine
LOD     limit of detection
LOEC    lowest-observed-effect concentration
log      logarithm
LPb     low Pb
LPS     lipopolysaccharide
LSO     lateral superior olive
LTP     long-term potentiation
M       metal
November 2012
                                                xxxvm
               Draft - Do Not Cite or Quote

-------
M, mM, |jM, nM, pM:
        Molar, millimolar (10"3 M), micromolar
        (10~6 M), nanomolar (10~9 M), picomolar
        (1(T12M)
m, km, cm, mm, um, nm:
        meter(s), kilometer(s), centimeter(s),
        millimeter(s), micrometer(s) [micron(s)],
        nanometer(s)
MAP   mean arterial pressure
MAPK  mitogen-activated protein kinase(s), MAP
        kinase
MATC  maximum acceptable toxicant concentration
max    maximum, maxima
MBP   myelin basic protein
MCH   mean corpuscular hemoglobin
MCHC  mean corpuscular hemoglobin concentration
MchDMSA:
        mono-cyclohexyl dimercaptosuccinic acid
MCL   maximum containment level
MCP-1  monocyte chemotactic protein-1
MCV   mean corpuscular volume
MD    mean diffusivity
MDA   malondialdehyde
MOD   major depressive disorder
MDI    Mental Development Index
MDL   method detection limit
MDRD  Modification of Diet in Kidney Disease
Med, med:
        median
MEK1  dual specificity mitogen-activated protein
        kinase 1
MEK2  dual specificity mitogen-activated protein
        kinase 2
MENTOR
        Modeling Environment for Total Risk
        (framework)
Mg     magnesium
Mg2+   magnesium ion
MHC   maj or histocompatibility complex
MI      myocardial infarction, "heart attack";
        myocardial ischemia
ml      myoinositol
min    minimum; minima; minute(s)
MKK1/2
        MAPK kinase 1 and 2
ML     mixed layer
MMAD mass median aerodynamic diameter
     MMDD mental retardation or developmental
                        disabilities
MMF   mycophenolate mofetil
mmHg  millimeters of mercury
mmol, umol, nmol:
        millimole(s), micromole(s), nanomole(s)
MN     micronuclei formation; mononuclear
Mn     manganese
MNE   micronucleated erythrocytes per thousand
MnO2   manganese dioxide
Mo     molybdenum
mo     month(s)
MOA(s) mode(s) of action
MORA  Mount Ranier National Park, Washington State
MOUDI
        multi-orifice uniform deposit impactor
MPb, m-Pb: moderate Pb
MPO   myeloperoxidase
MRI    magnetic resonance imaging
mRNA  messenger ribonucleic acid
MRS   magnetic resonance spectroscopy
MS     maternal stress
MSC   mesenchymal cell
MSWI  municipal solid waste incineration
Mt     metallothionein
MTHFR
        methylenetetrahydrofolatereductase
MTP   mitochondrial transmembrane pore
MW    molecular weight
MZ     marginal zinc
N      nitrogen; normal; north; number; population
n       number of observations
Na     sodium
Na+     sodium ion
NAAQS
        National Ambient Air Quality Standards
NAC   N-acetyl cysteine; nucleus accumbens
Na2CaEDTA:
        calcium disodium ethylenediaminetetraacetic
        acid
November 2012
                                               XXXIX
              Draft - Do Not Cite or Quote

-------
NaCl   sodium chloride
NAD   nicotinamide adenine dinucleotide
NADH  nicotinamide adenine dinucleotide
        dehydrogenase
NADP  nicotinamide adenine dinucleotide phosphate
NADPH, NAD(P)H:
        reduced nicotinamide adenine dinucleotide
        phosphate
NAEC  no-adverse-effect concentration
NAG   N-acetyl-p-D-glucosaminidase;
        N-acetylglucosamine
NaHC03
        sodium bicarbonate; sodium hydrogen
        carbonate
NANC  non-adrenergic non-cholinergic
NAS   U. S. Department of Veteran's Affair's
        Normative Aging Study; National Academy of
        Sciences
NASCAR:
        National Association for Stock Car Automobile
        Racing
NATTS National Air Toxics Trends Station
NAWQA
        National Water Quality Assessment
NCAM neural cell adhesion molecule
NCEA  National Center for Environmental Assessment
NCore  National Core multi-pollutant monitoring
        network
N.D.    not detected
NDMAR
        N-nitrosodimethylamine receptor
NE     norepinephrine
NEC AT New England Children's Amalgam Trial
NEI    National Emissions Inventory
NFI    non-fixed interval
NF-KB  nuclear factor kappa B
NGAL  neutrophil gelatinase-associated lipocalin
NGF   nerve growth factor
NH     non-hispanic
NHANES:
        National Health and Nutrition Examination
        Survey
NH4C1  ammonium chloride
NHEJ   non-homologous end joining
         NHEXAS: National Human Exposure Assessment
                  Survey
         NH4OAc
                  ammonium acetate
         7-NI     7-nitroinidazole
         Ni       nickel
         NICA    non-ideal competitive absorption
         NIOSH  National Institute for Occupational Safety and
                  Health
         NIST    National Institute of Standards and Technology
         NK      natural killer
         NKF-K/DOQI:
                  National Kidney Foundation - Kidney Disease
                  Outcomes Quality Initiative
         NMDA  N-methyl-D-aspartate
         NMR    nuclear magnetic resonance
         nNOS    neuronal nitric oxide synthase (NOS)
         NO      nitric oxide; nitrogen monoxide
         NO2     nitrogen dioxide
         No.      number
         NOAA  National Oceanic and Atmospheric
                  Administration
         NOAEL  no observed adverse effect level
         NO AT   Noatak National Preserve, Alaska
         NOCA  North Cascades National Park, Washington
                  State
         NOEC   no-observed-effect concentration
         NOEL   no-observed-effect level
         NOS     nitric oxide synthase; nitric oxide systems
         NOX     nitrogen oxides, oxides of nitrogen
                  (NO + NO2)
         NP      nanoparticle
         NPSH    nonprotein sulfhydryl
         NQO1    NAD(P)H-quinone oxidoreductase (genotype)
         NRC     National Research Council
         NRCS    Natural Resources Conservation Service
         Nrf2     nuclear factor erythroid 2-related factor 2
         NS      not specified
         NTP     National Toxicology Program
         NTPDase:
                  nucleoside triphosphate diphosphohydrolase
         NW     northwest
         NYC     New York City
November 2012
xl
Draft - Do Not Cite or Quote

-------
NZ     New Zealand
O2      molecular oxygen
O2~     superoxide
O3      ozone
9-O-Ac-GD3:
        9-O-acetylated-GD3
OAQPS
        U.S. EPA Office of Air Quality Planning and
        Standards, in OAR
OAR   U. S. EPA Office of Air and Radiation
OBS    observations
OC     organic carbon
OEPA  Ohio Environmental Protection Agency
OH"    hydroxide ion
1,25-(OH)2D3:
        1,25-dihydroxy vitamin D
O-horizon:
        horizon forest floor, organic soil horizon
        (above surface soil)
OLC    osteoblast-like cells
OLYM  Olympic National Park, Washington State
OM    organic matter
ONL    outer neuroblastic layers of the retina
ONOO" peroxynitrate ion
OR     odds ratio
ORD   U.S. EPA Office of Research and Development
OS     offspring stress
OSHA  Occupational Safety and Health Administration
OVA    ovalbumin
8-oxo-dG:
        8-hy droxy-2' -deoxy guanosine
P       percentile; phosphorus
PO      parent generation
P450    cytochrome P450
p       probability value; number of paired hourly
        observations; statistical significance
PA     policy assessment
PAD    peripheral arterial disease
PAH(s)  polycyclic aromatic hydrocarbon(s)
Pb      lead
203Pb    lead-203 radionuclide
204Pb    stable isotope  of lead-204
206Pb    stable isotope  of lead-206
207Pb    stable isotope  of lead-207
208Pb    stable isotope  of lead-208
210Pb    stable isotope  of lead-210
Pb++    divalent Pb ion
Pb°     elemental lead
Pb(II)   lead (II)
          Pb2+     lead ion
          Pb(Ac)2 lead acetate
          PbB     blood lead concentration
          PbBrCl  lead bromochloride
          Pb(C2H302)2:
                  lead (II) acetate
          PbCl+   lead chloride
          PbCl2   lead chloride
          PbCl3   lead (III) chloride; lead trichloride
          PbCl4   lead (IV) chloride; lead tetrachloride
          PbCO3  cerussite; lead carbonate
          Pb(C03)2
                  lead (IV) carbonate
          Pb(CO3)2(OH)2:
                  hydrocerussite
          PbCrO4  lead (II) chromate
          PbD     floor dust lead
          PbFe6(S04)4(OH)12:
                  plumbjarosite
          PEG    porphobilinogen
          Pb(N03)2:
                  lead(II) nitrate
          Pb-NS  lead-no stress
          PbO     lead oxide; litharge; massicot
          PbO2    lead dioxide
          Pb(IV)O2:
                  lead dioxide
          Pb3O4   minimum or "red Pb"
          Pb(OH)2
                  lead hydroxide
          Pb5(P04)3Cl:
                  pyromorphite
          Pb5(PO4)3OH:
                  hydroxypyromorphite
          PbS     galena; lead sulfide; soil lead concentration
          PbSe    lead selenide
          PbSO4  anglesite; lead sulfate
          Pb4SO4(CO3)2(OH)3:
                  macphersonite
          PbxS    lead by stress
          Pb5(V04)3Cl:
                  vanadinite
          PC 12    pheochromocytoma 12 (adrenal / neuronal cell
                  line)
          PCA    principal component analysis
          PCE     polychromatic erythrocyte
          PCR    polymerase chain reaction
          Pet      percent
          PCV    packed cell volume
          PD      Parkinson's disease
          PDI     Psychomotor Development Index
November 2012
xli
Draft - Do Not Cite or Quote

-------
PEC    probable effect concentration
PEL    permissible exposure limit
PER    partial exfiltration reactor
PG     prostaglandin
PGE2, PGE2:
        prostaglandin E2
PGF2   prostaglandin F2
pH     relative acidity; Log of the reciprocal of the
        hydrogen ion concentration
PHA    polyhydroxyalkanoates
PHE    phenylalanine
PIH    pregnancy  induced hypertension
PIQ    performance intelligence quotient (IQ)
PIR    poverty-income ratio
PIXE   particle induced X-Ray emission; proton-
        induced x-ray emission
PKC    protein kinase C
PLP    proteolipid protein
PM     particulate  matter
PMX    Particulate  matter of a specific size range not
        defined for regulatory use. Usually X refers to
        the 50% cut point, the aerodynamic diameter at
        which the sampler collects 50% of the particles
        and rejects  50% of the particles. The collection
        efficiency,  given by a penetration curve,
        increases for particles with smaller diameters
        and decreases for particles with larger
        diameters. The definition of PMX is sometimes
        abbreviated as "particles with a nominal
        aerodynamic  diameter less than or equal to
        X um" although X is usually a 50% cut point.
PM10   In general terms, particulate matter with a
        nominal aerodynamic diameter less than or
        equal to 10 um; a measurement of thoracic
        particles (i.e., that subset of inhalable particles
        thought small enough to penetrate beyond the
        larynx into the thoracic region of the
        respiratory tract) in regulatory terms, particles
        with an upper 50% cut-point of 10± 0.5 um
        aerodynamic  diameter (the 50% cut point
        diameter is the diameter at which the sampler
        collects 50%  of the particles and rejects 50% of
        the particles)  and a penetration curve as
        measured by  a reference method based on
        Appendix J of 40 CFR Part 50 and designated
        in accordance with 40 CFR Part 53 or by an
        equivalent  method designated in accordance
        with 40 CFR  Part 53.
          PM2.5   In general terms, particulate matter with a
                  nominal aerodynamic diameter less than or
                  equal to 2.5 um; a measurement of fine
                  particles in regulatory terms, particles with an
                  upper 50% cut-point of 2.5 um aerodynamic
                  diameter (the 50% cut point diameter is the
                  diameter at which the sampler collects 50% of
                  the particles and rejects 50% of the particles)
                  and a penetration curve as measured by a
                  reference method based on Appendix L of 40
                  CFR Part 50 and designated in accordance with
                  40 CFR Part 53, by an equivalent method
                  designated in accordance with 40 CFR Part 53,
                  or by an approved regional method designated
                  in accordance with Appendix C of 40 CFR Part
                  58.
          PMio_2 5 In general terms, particulate matter with a
                  nominal aerodynamic diameter less than or
                  equal to 10 um and greater than a nominal
                  2.5 um; a measurement of thoracic coarse
                  particulate matter or the coarse fraction of
                  PMjo in regulatory terms, particles with an
                  upper 50% cut-point of 10 um aerodynamic
                  diameter and a lower 50% cut-point of 2.5 um
                  aerodynamic diameter (the 50% cut point
                  diameter is the diameter at which the sampler
                  collects 50% of the particles and rejects 50% of
                  the particles) as measured by a reference
                  method based on Appendix O of 40 CFR Part
                  50 and designated in accordance with 40 CFR
                  Part 53  or by an equivalent method designated
                  in accordance with 40 CFR Part 53.
          PM10c  The PM10.2.5 concentration of PM10.2.5
                  measured by the 40 CFR Part 50 Appendix O
                  reference method which consists of currently
                  operated, co-located low-volume  (16.7 Lpm)
                  PM10 and PM2 5 reference method samplers.
          p38MAPK:
                  p38 mitogen-activated protein kinase(s)
          PMN   polymorphonuclear leukocyte
          P5N    pyrimidine 5'-nucleotidase
          PND    post natal  day
          POC    particulate organic carbon
          PP      polypropylene; pulse pressure
          ppb     parts per billion
          ppm    parts per million
          PRP    post-reinforcement pause
          PS      dam stress; prenatal stress; phosphatidylserine
          PSA    prostate specific antigen
          PSA-NCAM:
                  polysialylated-neural cell adhesion molecule
          PT      proximal tubule
          PTFE   polytetrafluoroethylene
          PTH    parathyroid hormone
          PTHrP  parathyroid hormone-related protein
          PUFA  polyunsaturated fatty acid
November 2012
xlii
Draft - Do Not Cite or Quote

-------
PVC    polyvinyl chloride
PVD    peripheral vascular disease
Q      quantile; quartile; quintile
QRS    QRS complex in EGG
QT     QT interval in EGG
QTc    corrected QT Interval
p       rho; bulk density; correlation
pS      Pearson's r correlation coefficient
R      net drainage loss out of soil depth of concern;
        Spearman correlation coefficient; upward
        resuspension flux; correlation
r       Pearson correlation coefficient
R2      multiple regression correlation coefficient
r       correlation coefficient
RAAS  renin-angiotensin-aldosterone system
RAC2  gene encoding for Rac2
RBA    relative bioavailability
RBC    red blood cell
RBP    retinol binding protein
RD     radial diffusivity
REA    Risk/Exposure Assessment
Ref    reference (group)
RI-RI   concurrent random interval
RL     repeated learning
220Rn   radon isotope
222Rn   stable isotope of radon-222
RNA    ribonucleic acid
ROI    reactive oxygen intermediate/superoxide anion;
        regions of interest
ROMO Rocky Mountain National Park, Colorado
ROS    reactive oxygen species
RR     relative risk; risk ratio
RSL    roughness sublayer (transition layer, wake
        layer, interfacial layer)
rtPCR  reverse transcription polymerase chain reaction
a       sigma, standard deviation
S       south; sulfur; synthesis phase
SAB    U. S. EPA Science Advisory Board
SATs    Standard Assessment Tests
SBP    systolic blood pressure
SCE    sister chromatid exchange
Sena    a-synuclein
SD     standard deviation
SDN    sexually dimorphic nucleus
SE     standard error
Se      selenium
sec     second(s)
SEKI   Sequoia and Kings Canyon  National Park,
        California
                   scanning electron microscopy; simultaneously
                   extracted metal; standard error of the mean
                   socioeconomic status
                   Session
                   stacked filter unit(s)
                   small for gestational age
                   soluble guanylate cyclase
SEM

SES
Sess.
SFU
SGA
sGC
sGC-pl
        soluble guanylate cyclase-beta 1
SGOT  serum glutamic oxaloacetic transaminase
SGPT   serum glutamic pyruvic transaminase
SHBG  sex hormone binding globulin
SHEDS Stochastic Human Exposure and Dose (model)
SHM    Stockholm humic model
siRNA  small interfering RNA
SJW    silver jewelry workers
SLAMS
        State and Local Air Monitoring Stations
SMC    smooth muscle cells
SNAP-25:
        synaptosomal-associated protein 25
SNARE soluble NSF attachment receptor
SNP    single-nucleotide polymorphism;  sodium
           SNS
           SO
           S02
           So
           soc
           SOD
           SOD1
           SOF
           SOM
           SP
           SP1, Spl
        nitroprusside
        sympathetic nervous system
        stratum oriens
        sulfur dioxide
        south
        superior olivary complex
        superoxide dismutase
        superoxide dismutase-1
        study of osteoporotic fractures
        self-organizing map; soil organic matter
        spray painters
                   specificity protein 1
           SPM    suspended particulate matter
           SPT     skin prick test
           SREBP-2:
                   sterol regulatory element binding protein-2
           S. Rep.  Senate Report
           SRIXE  synchrotron radiation induced X-ray emission
           StAR    steroidogenic acute regulatory protein
           STAT    signal transducer and activator of transcription
           STATS  signal transducer and activator of
                   transcription 3
November 2012
xliii
               Draft - Do Not Cite or Quote

-------
STATS  signal transducer and activator of
        transcription 5
STD.    Standard
ST Interval:
        measured from the J point to the end of the T
        wave in an EGG
STN    Speciation Trends Network
Syb     synaptobrevin
Syn     synaptophysin
Syt     synaptotagmin
SZn     supplemental zinc
T, t     time
T3, T3   triiodothyronine
T4, T4   thyroxine
tj/2      half-life (-lives); time required to reduce the
TEARS
initial concentration by 50%

thioBarbituric acid reactive substances;
thiobarbituric acid-reactive species
T cell, T-cell:
        T lymphocyte
TE     trace elements
TEC    threshold effect concentrations
TEOM  tapered element oscillating microbalance, type
        of PM sampler
TF     ratio of the metal concentration in plant to that
        in soil; transferrin
TFIIIA
        transcription factor IIIA
Tg     transgenic
TGF    transforming growth factor
TGF-P  p transforming growth factor
TGFP1,TGF-P1:
        pi transforming growth factor
TH     tyrosine hydroxylase
THl,Thl:
        T-derived lymphocyte helper 1
TH2, Th2:
        T-derived lymphocyte helper 2
Th     T-helper lymphocyte
TIMP-1 tissue inhibitor of metalloproteinases-1
TIMS   thermal ionization mass spectrometry
TLC    Treatment of Lead-exposed Children (study)
T/LH   testosterone/luteinizing hormone - measure of
        Leydig cell function
TNF    tumor necrosis factor (e.g., TNF-a)
TNP-Ficoll:
        trinitrophenyl-Ficoll
TNP-OVA:
        trinitrophenyl-ovalbumin
TPR    total peripheral vascular resistance
TS      transferrin saturation
TSH    thyroid stimulating hormone; total sulfhydryl
TSP    total suspended particles
TSS    total suspended solids
TXB2   thromboxane
U      uranium
UA     urbanized area
UBL    urban boundary layer
UCL    urban canopy layer
HDDS  urban dynamometer driving schedule
UDPGT
        uridine diphosphate (UDP)-
        glucuronosyltransferase(s)
UIUC   University of Illinois at Urbana Champaign
U.K.    United Kingdom
U.S.    United States of America
USC    U.S. Code
U.S. EPA:
        U.S. Environmental Protection Agency
USF    uptake slope factor
USGS   U.S. Geological Survey
USL    urban surface layer
UUDS  urban dynamic driving schedule
UV     ultraviolet radiation
V      vanadium
V79    Chinese hamster lung cell line
VA     Veterans Administration
VAChAT
        vesicular acetylcholine transporter
VAMP-2
        vesicle-associated membrane protein-2
VA-NAS
        Veterans Administration Normative Aging
        Study
VDAC  voltage-dependent anion channel
VDR    vitamin D receptor
VGAT  vesicular gamma aminobutyric acid (GABA)
        transporter
VGCC  voltage gated calcium channel(s)
VGLUT1:
        vesicular glutamate transporter 1
VIQ    verbal intelligence quotient (IQ)
VLPb   very low Pb
November 2012
                                        xliv
              Draft - Do Not Cite or Quote

-------
VMAT2
        vesicular monoamine transporter-2
VO43"   vanadate ion
VOC(s) volatile organic compound(s)
vs.,v.   versus
VSCC  very sharp cut cyclone
VSMC  vascular smooth muscle cells
WACAP
        Western Airborne Contaminants Assessment
        Project
WBC   white blood cell
WCST  Wisconsin Card Sorting Test
WHAM Windermere humic aqueous model
WHO   World Health Organization
WIAT   Wechsler Individual Achievement Test
WINS   well impactor ninety six
WISC   Wechsler Intelligence Scale for Children
WISC-R
        Wechsler Intelligence Scale for Children-
        Revised
wk     week(s)
WML   white matter lesions
WPPSI-III:
        Wechsler Preschool and Primary Scales of
        Intelligence-Ill
          WPPSI-R:
                  Wechsler Preschool and Primary Scale of
                  Intelligence-Revised
          WRAT  Wide Range Achievement Test
          W/S    winter/summer
          WT     wild type
          wt.      weight
          XAFS   X-ray absorption fine structure
          XANES X-ray absorption near edge structure
          XDH   xanthine dehydrogenase
          Xjj      observed hourly concentrations for time period
                  i at site j
          Xik      observed hourly concentrations for time period
                  i at site k
XPS
XRF
yr
Zn
Zn2+
ZPP
                  X-ray photoelectron spectroscopy
                  X-ray fluorescence
                  year(s)
                  zinc
                  zinc ion
                  zirconium-potassium perchlorate; zinc
                  protoporphyrin
          Z-score standard score
November 2012
xlv
               Draft - Do Not Cite or Quote

-------
      Preamble
            Process of ISA Development
 1                  This preamble outlines the general process for developing an Integrated Science
 2                  Assessment (ISA) including the framework for evaluating weight of evidence and
 3                  drawing scientific conclusions and causal judgments. The ISA provides a concise review,
 4                  synthesis, and evaluation of the most policy-relevant science to serve as a scientific
 5                  foundation for the review of the National Ambient Air Quality Standards (NAAQS). The
 6                  general process for NAAQS reviews is described at http://www.epa.gov/ttn/naaqs/review.html.
 7                  Figure I depicts the general NAAQS review process and information for individual
 8                  NAAQS reviews is available at www.epa.gov/ttn/naaqs. This preamble is a general discussion
 9                  of the basic steps and criteria used in developing an ISA; for each ISA, specific details
10                  and considerations are included  in the introductory section for that assessment.

11                  The fundamental process for developing an ISA includes:

12                      •  literature searches;
13                      •  study selection;
14                      •  evaluation and integration of the evidence; and
15                      •  development of scientific conclusions and causal judgments.

16                  An initial step in this process is publication of a call for information in the Federal
17                  Register that invites the public to provide information relevant to the assessment, such as
18                  new or recent publications on health or welfare1 effects of the pollutant, or from
19                  atmospheric and exposure sciences fields. The U.S. Environmental Protection Agency
20                  (EPA) maintains an ongoing literature search process for identification of relevant
21                  scientific studies published since the last review of the NAAQS. Search strategies are
22                  designed for pollutants and scientific disciplines  and iteratively modified to optimize
23                  identification of pertinent publications. Papers are identified for inclusion in several
24                  additional ways: specialized searches on specific topics; independent review of tables of
25                  contents for journals in which relevant papers may be published; independent
26                  identification of relevant literature by expert scientists; review of citations in previous
27                  assessments and identification by the public and the Clean Air Scientific Advisory
28                  Committee (CASAC) during the external review process. This literature search and study
29                  selection process is depicted in Figure II. Publications considered for inclusion in the ISA
      1 Welfare effects as defined in Clean Air Act (CAA) Section 302(h) [42 U.S.C. 7602(h)] include, but are not limited
      to, "effects on soils, water, crops, vegetation, man-made materials, animals, wildlife, weather, visibility and climate,
      damage to and deterioration of property, and hazards to transportation, as well as effects on economic values and on
      personal comfort and well-being."
      November 2012                             xlvi                   Draft - Do Not Cite or Quote

-------
 1
 2

 3
 4
 5
 6
 7
 8
 9
10
11
12
13
are added to the Health and Environmental Research Online (HERO) database developed
by EPA (http://hero.epa.gov/); the references in the ISA include a hyperlink to the database.

Studies that have undergone scientific peer review and have been published or accepted
for publication and reports that have undergone review are considered for inclusion in the
ISA. Analyses conducted by EPA using publicly available data are also considered for
inclusion in the ISA. All relevant epidemiologic, controlled human exposure,
toxicological, and ecological and welfare effects studies published since the last review
are considered, including those related to exposure-response relationships, mode(s) of
action (MOA), and potentially at-risk populations and lifestages. Studies on atmospheric
chemistry, environmental fate and transport, dosimetry, toxicokinetics and exposure are
also considered for inclusion in the document, as well as analyses of air quality and
emissions data. References that were considered for inclusion in a specific ISA can be
found using the HERO website (http://hero.epa.gov).
                       EPA
                     proposed
                    decisions on
                    v standards .'
                                 Integrated Review Plan (IRP) timeline and key
                                  policy-relevant issues and scientific questions
                                        Integrated Science Assessment (ISA): evaluation and
                                             synthesis of most policy-relevant sludies
                                               Risk/Exposure Assessment (REA):
                                           quantitative assessment, as warranted, focused
                                           on key results, observations, and uncertainties
                                                 Policy Assessment (PA): staff analysis of
                                                  policy options based on integration and
                                               interpretation of information in Ihe ISA and REA
      Figure I
Agency decision
making and draft
final notice


Interagency
review
                                                                 Clean Air Scientific
                                                                 Advisory Committee
                                                                  (CASAC) review
                                                                  Public comment
    Illustration of the key steps in the process of the review of
    National Ambient Air Quality Standards.
      November 2012
                             xlvii
Draft - Do Not Cite or Quote

-------
                        Title ;^Yes-e-Yes
         Literature
            Search
         Strategies
            Recommendations
            during Peer Review
                              Citations from
                            Past Assessments
                                 Criteria for study evaluation
                                 include:
                                 •Are the study populations, subjects,
                                 or animal  models adequately
                                 selected,  and are they sufficiently
                                 well defined to allow for meaningful
                                 comparisons between study or
                                 exposure groups?
                                 •Are the statistical analyses
                                 appropriate, properly performed, and
                                 properly interpreted? Are likely
                                 covariates adequately controlled or
                                 taken into account in the study
                                 design and statistical analysis?
                                 •Are the air quality data, exposure,
                                 or dose metrics of adequate quality
                                 and sufficiently representative of
                                 information regarding ambient
                                 conditions?
                                 •Are the health, ecological or welfare
                                 effect measurements meaningful,
                                 valid and  reliable?
                                 •Do the analytical methods provide
                                 adequate sensitivity and precision to
                                 support conclusions?
      Figure II          Illustration of processes for literature search and study selection
                          used for development of ISAs.

 1                    Each ISA builds upon the conclusions of previous assessments for the pollutant under
 2                    review. EPA focuses on peer reviewed literature published following the completion of
 3                    the previous review and on any new interpretations of previous literature, integrating the
 4                    results of recent scientific studies with previous findings. Important earlier studies may
 5                    be discussed in detail to reinforce key concepts and conclusions or for reinterpretation  in
 6                    light of newer data. Earlier studies also are the primary focus in some areas of the
 7                    document where research efforts have subsided,  or if these earlier studies remain the
 8                    definitive works available in the literature.

 9                    Selection of studies for inclusion in the ISA is based on the general scientific quality of
10                    the study, and consideration of the extent to which the study is informative and policy-
11                    relevant. Policy-relevant and informative studies include those that provide a basis for  or
12                    describe the relationship between the criteria pollutant and effects, including studies that
13                    offer innovation in method or design  and studies that reduce uncertainty on critical issues,
14                    such as analyses of confounding or effect modification by copollutants or other variables,
15                    analyses of concentration-response or dose-response relationships, or analyses related to
16                    time between exposure and response. Emphasis is placed on studies that examine effects
17                    associated with pollutant concentrations relevant to current population and ecosystem
18                    exposures, and particularly those pertaining to concentrations currently found in ambient
19                    air. Other studies are included if they contain unique data, such as a previously
      November 2012
xlviii
Draft - Do Not Cite or Quote

-------
 1                   unreported effect or MOA for an observed effect, or examine multiple concentrations to
 2                   elucidate exposure-response relationships. In general, in assessing the scientific quality
 3                   and relevance of health and welfare effects studies, the following considerations have
 4                   been taken into account when selecting studies for inclusion in the ISA.

 5                       •  Are the study populations, subjects, or animal models adequately selected, and
 6                         are they sufficiently well defined to allow for meaningful  comparisons
 7                         between study or exposure groups?
 8                       •  Are the statistical analyses appropriate, properly performed, and properly
 9                         interpreted? Are likely covariates adequately controlled or taken into account
10                         in the study design and statistical analysis?
11                       •  Are the air quality data, exposure, or dose metrics of adequate quality and
12                         sufficiently representative of information regarding ambient conditions?
13                       •  Are the health, ecological or welfare effect measurements meaningful, valid
14                         and reliable?
15                       •  Do the analytical methods provide adequate sensitivity and precision to
16                         support conclusions?

17                   Additional considerations are specific to particular disciplines. In selecting epidemiologic
18                   studies, EPA considers whether a given study: (1) presents information on associations
19                   with short- or long-term pollutant exposures at or near conditions relevant to ambient
20                   exposures; (2) addresses potential confounding by other pollutants; (3) assesses potential
21                   effect modifiers; (4) evaluates health endpoints and populations not previously
22                   extensively researched; and (5) evaluates important methodological issues related to
23                   interpretation of the health evidence (e.g., lag or time period between exposure and
24                   effects, model specifications, thresholds, mortality displacement).

25                   Considerations for the selection of research evaluating controlled human exposure or
26                   animal toxicological studies include a focus on studies conducted using relevant pollutant
27                   exposures. For both types of studies, relevant pollutant exposures are considered to be
28                   those generally within one or two orders of magnitude of ambient concentrations. Studies
29                   in which higher doses were used may also be  considered if they provide information
30                   relevant to understanding MOA or mechanisms, as noted below.

31                   Evaluation of controlled human exposure studies focuses on those that approximated
32                   expected human exposure conditions in terms of concentration and duration. Studies
33                   should include control exposures to filtered air, as appropriate. In the selection of
34                   controlled human exposure studies, emphasis is placed on studies that:  (1) investigate
35                   potentially at-risk populations and lifestages such as people with asthma or
36                   cardiovascular diseases, children or older adults; (2) address issues such as concentration-
      November 2012                              xlix                   Draft - Do Not Cite or Quote

-------
 1                   response or time-course of responses; and (3) have sufficient statistical power to assess
 2                   findings.

 3                   Review of the animal toxicological evidence focuses on studies that approximate
 4                   expected human dose conditions, which vary depending on the dosimetry, toxicokinetics
 5                   and biological sensitivity of the particular laboratory animal species or strains studied.
 6                   Emphasis is placed on studies that: (1) investigate animal models of disease that can
 7                   provide information on populations potentially at increased risk of effects; (2) address
 8                   issues such as concentration-response or time-course of responses; and (3) have sufficient
 9                   statistical power to assess findings. Due to resource constraints on exposure duration and
10                   numbers of animals tested, animal studies typically utilize high-concentration exposures
11                   to acquire data relating to mechanisms and assure a measurable response. Emphasis is
12                   placed on studies using doses or concentrations generally within 1-2 orders of magnitude
13                   of current levels. Studies with higher concentration exposures or doses are considered to
14                   the extent that they provide useful information to inform understanding of interspecies
15                   differences between healthy and at-risk human populations. Results from in vitro studies
16                   may also be included if they provide mechanistic insight or further support for results
17                   demonstrated in vivo.

18                   These criteria provide benchmarks for evaluating various studies and for focusing on the
19                   policy-relevant  studies in assessing the body of health, ecological and welfare effects
20                   evidence. As stated initially, the intent of the ISA is to provide a concise review,
21                   synthesis, and evaluation of the most policy-relevant science to serve as a scientific
22                   foundation for the review of the NAAQS, not extensive summaries of all health,
23                   ecological and welfare effects studies for a pollutant. Of most relevance for inclusion of
24                   studies is whether they provide useful qualitative or quantitative information on
25                   exposure-effect or exposure-response relationships for effects associated with pollutant
26                   exposures at doses or concentrations relevant to ambient conditions that can inform
27                   decisions on whether to retain or revise the standards.

28                   The general process for ISA development is illustrated in Figure III. In developing an
29                   ISA, EPA reviews and summarizes the evidence from: studies of atmospheric sciences;
30                   human exposure, animal toxicological, controlled human exposure and epidemiologic
31                   studies; and studies of ecological and welfare effects. In the process of developing the
32                   first draft ISA, EPA may convene a peer input meeting in which the scientific content of
33                   preliminary draft materials is reviewed to ensure that the ISA is up to date and is focused
34                   on the most policy-relevant findings, and to assist EPA with integration of evidence
35                   within and across disciplines. EPA integrates the evidence from across scientific
36                   disciplines or study types and characterizes the weight of evidence for relationships
37                   between the pollutant and various outcomes. The integration of evidence on health, and
      November 2012                               1                    Draft - Do Not Cite or Quote

-------
 1                   ecological or welfare effects, involves collaboration between scientists from various
 2                   disciplines. As an example, an evaluation of health effects evidence would include the
 3                   integration of the results from epidemiologic, controlled human exposure, and
 4                   toxicological studies, and application of the causal framework (described below) to draw
 5                   conclusions. Integration of results on health or ecological effects that are logically or
 6                   mechanistically connected (e.g., a spectrum of effects on the respiratory system) informs
 7                   judgments of causality.  Using the causal framework described in the following section,
 8                   EPA scientists consider aspects such as strength, consistency, coherence, and biological
 9                   plausibility of the evidence, and develop causality determinations on the nature of the
10                   relationships. Causality determinations often entail an iterative process of review and
11                   evaluation of the  evidence. Two drafts of the ISA are typically released for review by the
12                   CASAC and the public, and comments received on the characterization of the science as
13                   well as the implementation of the causal framework are carefully considered in revising
14                   and completing the final ISA.
      November 2012                               li                    Draft - Do Not Cite or Quote

-------
       Integrated Science Assessment  Development  Process
                              Evergreen Literature Search and
                                     Study Selection
                                      (See Figure II)
                               Characterization of Evidence
               Develop initial sections or "building blocks" of scientific evidence for
               assessment: review and summarize new study results, by
               outcome/effect category and discipline, forexample, toxicological
               studies of lung function. Summarize findings and conclusions from
               previous assessment. As appropriate, develop initial conclusions about
               the available evidence.
                                        Peer Input
                           Review of initial draft materials for scientific
                           quality and to facilitate integration.
                      Evaluation, Synthesis, and Integration of Evidence
               Integrate evidence from scientific disciplines orstudy types - for
               example, toxicological, controlled human exposure and epidemiologic
               study findings for particular health outcome. Evaluate evidence for
               related groups of endpoints oroutcomes to draw conclusions regarding
               health or welfare effect categories.
                   Development of Conclusions and Causal Determinations
               Evaluate weight of evidence and develop judgments regarding causality
               forhealth orwelfare effect categories, integrating health orwelfare
               effects evidence with information on mode of action and exposure
               assessment. Develop conclusions regarding concentration-ordose-
               response relationships, potentially at-risk populations orecosystems.
        Draft Integrated Science Assessment
                 re leased for review
                             Clean Air Scientific Advisory Committee
                             review in public meeting; anticipated review
                                       of two drafts of ISA
        Final Integrated Science Assessment
Figure
Characterization of the general process of ISA development.
November 2012
                       lii
Draft - Do Not Cite or Quote

-------
           EPA Framework for Causal Determination
 1                  EPA has developed a consistent and transparent basis for integration of scientific
 2                  evidence and evaluation of the causal nature of air pollution-related health or welfare
 3                  effects for use in developing ISAs. The framework described below establishes uniform
 4                  language concerning causality and brings more specificity to the findings. This
 5                  standardized language was drawn from sources across the federal government and wider
 6                  scientific community, especially the National Academy of Sciences (NAS) Institute of
 7                  Medicine (IOM) document, Improving the Presumptive Disability Decision-Making
 8                  Process for Veterans (2008). a comprehensive report on evaluating causality. This
 9                  framework:

10                      • describes the kinds of scientific evidence used in establishing a general causal
11                        relationship between exposure and health effects;
12                      • characterizes the process for integration and evaluation of evidence necessary
13                        to reach a conclusion  about the existence of a causal relationship;
14                      • identifies issues and approaches related to uncertainty; and
15                      • provides a framework for classifying and characterizing the weight of
16                        evidence in support of a general causal relationship.

17                  Approaches to assessing the separate and combined lines of evidence
18                  (e.g., epidemiologic, controlled human exposure, and animal toxicological studies) have
19                  been formulated by a number  of regulatory and science agencies, including the IOM of
20                  the NAS (2008). International Agency for Research on Cancer (IARC) (2006b).
21                  U.S. EPA (2005c). and Centers for Disease Control and Prevention  (CDC) (2004). Causal
22                  inference criteria have also been described for ecological effects evidence (U.S. EPA.
23                  1998; Fox. 1991). These formalized approaches offer guidance for assessing causality.
24                  The frameworks are similar in nature, although adapted to different  purposes, and have
25                  proven effective in providing a uniform structure and language for causal determinations.
                    Evaluating Evidence for Inferring Causation

26                  The 1964 Surgeon General's report defined "cause" as a "significant, effectual
27                  relationship between an agent and an associated disorder or disease in the host" (HEW.
28                  1964). More generally, a cause is defined as an agent that brings about an effect or a
29                  result. An association is the statistical relationship among variables; alone, however, it is
30                  insufficient proof of a causal relationship between an exposure and a health outcome.
31                  Unlike an association, a causal claim supports the creation of counterfactual claims, that
      November 2012                             liii                  Draft - Do Not Cite or Quote

-------
 1                   is, a claim about what the world would have been like under different or changed
 2                   circumstances (Samet and Bodurow. 2008).

 3                   Many of the health and environmental outcomes reported in these studies have complex
 4                   etiologies. Diseases such as asthma, coronary heart disease (CHD) or cancer are typically
 5                   initiated by multiple agents. Outcomes depend on a variety of factors, such as age,
 6                   genetic susceptibility, nutritional status, immune competence, and social factors (Samet
 7                   and Bodurow. 2008; Gee and Payne-Sturges. 2004). Effects on ecosystems are often also
 8                   multifactorial with a complex web of causation. Further, exposure to a combination of
 9                   agents could cause synergistic or antagonistic effects. Thus, the observed risk may
10                   represent the net effect of many actions and counteractions.

11                   Scientific findings incorporate uncertainty. "Uncertainty" can be  defined as having
12                   limited knowledge to exactly describe an existing state or future outcome, e.g., the lack of
13                   knowledge about the correct value for a specific measure or estimate. Uncertainty
14                   analysis may be qualitative or quantitative in nature. In many cases, the  analysis is
15                   qualitative, and can include professional judgment or inferences based on analogy with
16                   similar situations. Quantitative  uncertainty analysis may include use of simple measures
17                   (e.g., ranges) and analytical techniques. Quantitative uncertainty analysis might progress
18                   to more complex measures and techniques, if needed for decision support. Various
19                   approaches to evaluating uncertainty include classical statistical methods, sensitivity
20                   analysis, or probabilistic uncertainty analysis, in order of increasing complexity and data
21                   requirements. However, data may not be available for all aspects  of an assessment and
22                   those data that are available may be of questionable or unknown quality. Ultimately, the
23                   assessment is based on a number of assumptions with varying degrees of uncertainty. The
24                   ISA generally evaluates uncertainties qualitatively in assessing the evidence from across
25                   studies; in some situations quantitative analysis approaches, such as meta-regression, may
26                   be used.

27                   Publication bias is a source of uncertainty regarding the magnitude of health risk
28                   estimates. It is well understood that studies reporting non-null findings are more likely to
29                   be published than reports of null findings. Publication bias can  result in  overestimation of
30                   effect estimate sizes (loannidis. 2008).  For example, effect estimates from single-city
31                   epidemiologic studies have been found to be generally larger than those from multicity
32                   studies which is an indication of publication bias in that null or negative single-city
33                   results may be reported in a multicity analyses but might not be published independently
34                   (Bell et al.. 2005).
      November 2012                              liv                   Draft - Do Not Cite or Quote

-------
                     Consideration of Evidence from Scientific  Disciplines

 1                   Moving from association to causation involves the elimination of alternative explanations
 2                   for the association. The ISA focuses on evaluation of the findings from the body of
 3                   evidence, drawing upon the results of all studies determined to meet the criteria described
 4                   previously. Causality determinations are based on the evaluation, integration, and
 5                   synthesis of evidence from across scientific disciplines. The relative importance of
 6                   different types of evidence varies by pollutant or assessment, as does the availability of
 7                   different types of evidence for causality determination. Three general types of studies
 8                   inform consideration of human health effects: controlled human exposure, epidemiologic
 9                   and toxicological studies. Evidence on ecological or welfare effects may be drawn from a
10                   variety of experimental approaches (e.g., greenhouse, laboratory, field) and numerous
11                   disciplines (e.g., community ecology, biogeochemistry and paleontological/historical
12                   reconstructions).

13                   Direct evidence of a relationship between pollutant exposures and human health effects
14                   comes from controlled human exposure studies. Such studies experimentally evaluate the
15                   health effects of administered exposures in human volunteers under highly controlled
16                   laboratory conditions. Also referred to as human clinical studies, these experiments allow
17                   investigators to expose subjects to known concentrations of air pollutants under carefully
18                   regulated environmental conditions and activity levels. These studies provide important
19                   information on the biological plausibility of associations observed in epidemiologic
20                   studies. In some instances, controlled human exposure studies can also be used to
21                   characterize concentration-response relationships at pollutant concentrations relevant to
22                   ambient conditions. Controlled human exposures are typically conducted using a
23                   randomized crossover design, with subjects exposed both to the pollutant and a clean air
24                   control. In this way,  subjects serve as their own experimental controls, effectively
25                   limiting the variance associated with many potential confounders. However, controlled
26                   human exposure studies are limited by a number of factors, including small sample size
27                   and short exposure time. For example, exposure patterns relevant to understanding real-
28                   world exposures,  especially long-term exposures, are generally not practical to replicate
29                   in a laboratory setting. In addition, although subjects do serve as their own controls,
30                   personal exposure to pollutants in the hours and days preceding the controlled exposures
31                   may vary significantly between and within individuals. Finally,  controlled human
32                   exposure  studies require investigators to adhere to stringent health criteria for subjects
33                   included in the  study, and therefore the results often cannot be generalized to an entire
34                   population. Although some controlled human exposure studies have included health-
35                   compromised individuals such as those with respiratory or cardiovascular disease, these
36                   individuals must also be relatively healthy and may not represent the most sensitive
37                   individuals in the population. Thus, observed effects in these studies may underestimate

      November 2012                              Iv                   Draft - Do Not Cite or Quote

-------
 1                   the response in certain populations. In addition, the study design is limited to exposures
 2                   and endpoints that are not expected to result in severe health outcomes.

 3                   Epidemiologic studies provide important information on the associations between health
 4                   effects and exposure of human populations to ambient air pollution. In epidemiologic or
 5                   observational studies of humans, the investigator generally does not control exposures or
 6                   intervene with the study population. Broadly, observational studies can describe
 7                   associations between exposures  and effects. These studies fall into several categories:
 8                   e.g., cross-sectional, prospective cohort, panel, and time-series studies.  "Natural
 9                   experiments" offer the opportunity to investigate changes in health related to a change in
10                   exposure, such as closure of a pollution source.

11                   In evaluating epidemiologic studies, consideration of many study design factors and
12                   issues must be taken into account to properly inform their interpretation. One key
13                   consideration is evaluation of the potential contribution of the pollutant to a health
14                   outcome when it is a component of a complex air pollutant mixture. Reported effect
15                   estimates in epidemiologic studies may reflect (1) independent effects on health
16                   outcomes; (2) effects of the pollutant acting as an indicator of a copollutant or a complex
17                   ambient air pollution mixture; and (3) effects resulting from interactions between that
18                   pollutant and copollutants.

19                   In the evaluation of epidemiologic evidence, one important consideration is potential
20                   confounding. Confounding is "... a confusion of effects. Specifically, the apparent effect
21                   of the exposure of interest is distorted because the effect of an extraneous factor is
22                   mistaken for or mixed with the actual exposure effect (which may be  null)" (Rothman
23                   and Greenland. 1998). One approach to remove spurious  associations due to possible
24                   confounders is to control for characteristics that may differ between exposed and
25                   unexposed persons; this is frequently termed "adjustment." Scientific judgment is needed
26                   to evaluate likely sources and extent of confounding, together with consideration of how
27                   well the existing constellation of study designs, results, and analyses  address the potential
28                   for erroneous inferences. A  confounder is associated with both the exposure and the
29                   effect; for example, confounding can occur between correlated pollutants that are
30                   associated with the same effect.

31                   Several statistical methods are available to detect and control for potential confounders;
32                   however, none of these methods are being completely satisfactory. Multivariable
33                   regression models constitute one tool for estimating the association between exposure
34                   and outcome after adjusting for  characteristics of participants that might confound the
35                   results. The use of multipollutant regression models has been the prevailing approach for
36                   controlling potential confounding by copollutants in air pollution health effects studies.
37                   Finding the likely causal pollutant from multipollutant regression models is made


      November 2012                              Ivi                   Draft - Do Not  Cite or Quote

-------
 1                   difficult by the possibility that one or more air pollutants may be acting as a surrogate for
 2                   an unmeasured or poorly measured pollutant or for a particular mixture of pollutants. In
 3                   addition, pollutants may independently exert effects on the same system; for example,
 4                   several pollutants may be associated with respiratory effect through either the same or
 5                   different modes of action. The number and degree of diversity of covariates, as well as
 6                   their relevance to the potential confounders, remain matters of scientific judgment.
 7                   Despite these limitations, the use of multipollutant models is still the prevailing approach
 8                   employed in most air pollution epidemiologic studies and provides some insight into the
 9                   potential for confounding or interaction among pollutants.

10                   Confidence that unmeasured confounders are not producing the findings is increased
11                   when multiple studies are conducted in various settings using different subjects or
12                   exposures, each of which might eliminate another source of confounding from
13                   consideration. For example, multicity studies can provide insight on potential
14                   confounding through the use of a consistent method to analyze data from across locations
15                   with different levels of copollutants and other covariates. Intervention studies, because of
16                   their quasi-experimental nature, can be particularly useful in characterizing causation.

17                   Another important consideration in the evaluation of epidemiologic evidence is effect
18                   modification, which occurs when the effect differs between subgroups or strata; for
19                   example, effect estimates that vary by age group or potential risk factor. As stated by
20                   Rothman and Greenland (1998):

21                          "Effect-measure modification differs from confounding in several ways. The
22                          main difference is that, whereas confounding is a bias that the investigator hopes
23                          to prevent or remove from the effect estimate, effect-measure modification is a
24                          property of the effect under study ... In epidemiologic analysis one tries to
25                          eliminate confounding but one tries to detect and estimate effect-measure
26                          modification."

27                   When a risk factor is a confounder, it is the true cause of the association observed
28                   between the exposure and the outcome; when a risk factor is an effect modifier, it
29                   changes the magnitude of the association between the exposure  and the outcome in
30                   stratified analyses. For example, the presence of a preexisting disease or indicator of low
31                   socioeconomic status may act as effect modifiers if they are associated with increased
32                   risk of effects related to air pollution exposure. It is often possible to stratify the
33                   relationship between health outcome and exposure by one or more of these potential
34                   effect modifiers. For variables that modify the association, effect estimates in each
35                   stratum will be different from one another and different from the overall estimate,
36                   indicating a different exposure-response relationship may exist in populations represented
37                   by the se variable s.
      November 2012                              Ivii                   Draft - Do Not Cite or Quote

-------
 1                   Exposure measurement error, which refers to the uncertainty associated with the exposure
 2                   metrics used to represent exposure of an individual or population, can be an important
 3                   contributor to uncertainty in air pollution epidemiologic study results. Exposure error can
 4                   influence observed epidemiologic associations between ambient pollutant concentrations
 5                   and health outcomes by biasing effect estimates toward or away from the null and
 6                   widening confidence intervals around those estimates (Zeger et al.. 2000). There are
 7                   several components that contribute to exposure measurement error in air pollution
 8                   epidemiologic studies, including the difference between true and measured ambient
 9                   concentrations, the difference between average personal exposure to ambient pollutants
10                   and ambient concentrations at central monitoring sites, and the use of average population
11                   exposure rather than individual exposure estimates. Factors that could influence exposure
12                   estimates include nonambient sources of exposure, topography of the natural and built
13                   environment, meteorology, measurement errors, time-location-activity patterns, and the
14                   extent to which ambient pollutants penetrate indoor environments. The importance of
15                   exposure error varies with study design and is dependent on the spatial and temporal
16                   aspects of the design.

17                   The third main type of health effects  evidence, animal toxicological studies, provides
18                   information on the pollutant's  biological action under controlled and monitored exposure
19                   circumstances. Taking into account physiological differences of the experimental species
20                   from humans, these studies inform characterization of health effects of concern,
21                   exposure-response relationships and MOAs. Further, animal models can inform
22                   determinations of at-risk populations. These studies evaluate the effects of exposures to a
23                   variety of pollutants in a highly controlled laboratory setting and allow exploration of
24                   toxicological pathways or mechanisms by which a pollutant may cause effects.
25                   Understanding the biological mechanisms underlying various health outcomes can prove
26                   crucial in establishing or negating causality. In the absence of human studies data,
27                   extensive, well-conducted animal toxicological studies can support determinations of
28                   causality, if the evidence base  indicates that similar responses are expected in humans
29                   under ambient exposure conditions.

30                   Interpretations of animal toxicological studies are affected by limitations associated with
31                   extrapolation between animal and human responses. The differences between humans
32                   and other species have to be taken into consideration, including metabolism, hormonal
33                   regulation, breathing pattern, and differences in lung structure and anatomy. Also, in spite
34                   of a high degree of homology and the existence of a high percentage of orthologous
35                   genes across humans and rodents (particularly mice), extrapolation of molecular
36                   alterations at the gene level is complicated by species-specific differences in
37                   transcriptional regulation. Given these differences, there are uncertainties associated with
38                   quantitative extrapolations of observed pollutant-induced pathophysiological alterations
      November 2012                              Iviii                   Draft - Do Not Cite or Quote

-------
 1                  between laboratory animals and humans, as those alterations are under the control of
 2                  widely varying biochemical, endocrine, and neuronal factors.

 3                  For ecological effects assessment, both laboratory and field studies (including field
 4                  experiments and observational studies) can provide useful data for causality
 5                  determination. Because conditions can be controlled in laboratory studies, responses may
 6                  be less variable and smaller differences may be easier to detect. However, the control
 7                  conditions may limit the range of responses (e.g., animals may not be able to seek
 8                  alternative food sources) or incompletely reflect pollutant bioavailability, so they may not
 9                  reflect responses that would occur in the natural environment. In addition, larger-scale
10                  processes are difficult to reproduce in the laboratory.

11                  Field observational studies measure biological changes in uncontrolled situations, and
12                  describe an association between a disturbance and an ecological effect. Field data can
13                  provide important information for assessments of multiple stressors or where site-specific
14                  factors significantly influence exposure. They are also often useful for analyses of larger
15                  geographic scales and higher levels of biological organization. However, because
16                  conditions are not controlled, variability is expected to be higher and differences harder
17                  to detect. Field surveys are most useful for linking stressors with effects when stressor
18                  and effect levels are measured concurrently. The presence of confounding factors can
19                  make it difficult to attribute observed effects to specific stressors.

20                  Some studies are considered "intermediate" and are categorized as being between
21                  laboratory and field are studies. Some use environmental media collected from the field
22                  to examine the responses in the laboratory.  Others are experiments that are performed in
23                  the natural environment while controlling for some, but not all, of the environmental
24                  conditions (i.e., mesocosm studies). This type of study in manipulated natural
25                  environments can be considered a hybrid between a field experiment and laboratory study
26                  since some aspects are performed under controlled conditions but others are not. They
27                  make it possible to observe community and/or ecosystem dynamics, and provide  strong
28                  evidence for causality when combined with findings of studies that have been made
29                  under more controlled conditions.
                    Application of Framework for Causal Determination

30                  In its evaluation of the scientific evidence on health or welfare effects of criteria
31                  pollutants, EPA determines the weight of evidence in support of causation and
32                  characterizes the strength of any resulting causal classification. EPA also evaluates the
33                  quantitative evidence and draws scientific conclusions, to the extent possible, regarding
      November 2012                              lix                   Draft - Do Not Cite or Quote

-------
                the concentration-response relationships and the loads to ecosystems, exposures, doses or
                concentrations, exposure duration, and pattern of exposures at which effects are observed.
Table I
                Aspects to aid in judging causality.
Aspect
                          An inference of causality is strengthened when a pattern of elevated risks is observed
                          across several independent studies. The reproducibility of findings constitutes one of the
                          strongest arguments for causality. If there are discordant results among investigations,
                          possible reasons such as differences in exposure, confounding factors, and the power of
                          the study are considered.
Consistency of the
observed association
                          An inference of causality from one line of evidence (e.g., epidemiologic, clinical, or
                          animal studies) may be strengthened by other lines of evidence that support a
                          cause-and-effect interpretation of the association. Evidence on ecological or welfare
                          effects may be drawn  from a variety of experimental approaches (e.g., greenhouse,
Coherence                laboratory, and field) and subdisciplines of ecology (e.g., community ecology,
                          biogeochemistry, and  paleontological/historical reconstructions). The coherence of
                          evidence from various fields greatly adds to the strength of an inference of causality. In
                          addition, there may be coherence in demonstrating effects across multiple study designs
                          or related health endpoints within one scientific line of evidence.

                          An inference of causality tends to be strengthened by consistency with data from
                          experimental studies or other sources demonstrating plausible biological mechanisms. A
                          proposed mechanistic linking between an effect and exposure to the agent is an
                          important source of support for causality, especially when data establishing the existence
                          and functioning of those mechanistic links are available.
Biological plausibility.
Biological gradient
(exposure-response
relationship)
                          A well-characterized exposure-response relationship (e.g., increasing effects associated
                          with greater exposure) strongly suggests cause and effect, especially when such
                          relationships are also observed for duration of exposure (e.g., increasing effects
                          observed following longer exposure times).
Strength of the observed
association
Experimental evidence
Temporal relationship of
the observed association
                          The finding of large, precise risks increases confidence that the association is not likely
                          due to chance, bias, or other factors. However, it is noted that a small magnitude in an
                          effect estimate may represent a substantial effect in a population.

                          Strong evidence for causality can be provided through "natural experiments" when a
                          change in exposure is found to result in a change in occurrence or frequency of health or
                          welfare effects.

                          Evidence of a temporal sequence between the introduction of an agent, and appearance
                          of the effect,  constitutes another argument in  favor of causality.
                          Evidence linking a specific outcome to an exposure can provide a strong argument for
                          causation. However, it must be recognized that rarely, if ever, does exposure to a
                          pollutant invariably predict the occurrence of an outcome, and that a given outcome may
                          have multiple causes.
Specificity of the
observed association
                          Structure activity relationships and information on the agent's structural analogs can
                          provide insight into whether an association is causal. Similarly, information on mode of
                          action for a chemical, as one of many structural analogs, can inform decisions regarding
                          likely causality.
Analogy
November 2012
                                                Ix
Draft - Do Not Cite or Quote

-------
 1                   To aid judgment, various "aspects"1 of causality have been discussed by many
 2                   philosophers and scientists. The 1964 Surgeon General's report on tobacco smoking
 3                   discussed criteria for the evaluation of epidemiologic studies, focusing on consistency,
 4                   strength, specificity, temporal relationship, and coherence (HEW. 1964). Sir Austin
 5                   Bradford Hill (Hill, 1965) articulated aspects of causality in epidemiology and public
 6                   health that have been widely used (Samet and Bodurow. 2008: IARC. 2006b: U.S. EPA.
 7                   2005c: CDC. 2004). These aspects  (Hill 1965) have been modified (Table I) for use in
 8                   causal determinations specific to health and welfare effects for pollutant exposures (U.S.
 9                   EPA. 2009a).2 Although these aspects provide a framework for assessing the evidence,
10                   they do not lend themselves to being considered in terms of simple formulas or fixed
11                   rules of evidence leading to conclusions about causality (Hill. 1965). For example, one
12                   cannot simply  count the number of studies reporting statistically significant results or
13                   statistically nonsignificant results and reach credible conclusions about the relative
14                   weight of the evidence and the likelihood of causality. Rather, these aspects are taken into
15                   account with the goal of producing an objective appraisal of the evidence, informed by
16                   peer and public comment and advice, which includes weighing alternative views on
17                   controversial issues. In addition, it is important to note that the aspects in Table I cannot
18                   be used as a strict checklist, but rather to determine the weight of the evidence for
19                   inferring causality. In particular, not meeting one or more of the principles does not
20                   automatically preclude a determination of causality [see discussion in (CDC. 2004)1.
                     Determination of Causality

21                   In the ISA, EPA assesses the body of relevant literature, building upon evidence available
22                   during previous NAAQS reviews, to draw conclusions on the causal relationships
23                   between relevant pollutant exposures and health or environmental effects. IS As use a
24                   five-level hierarchy that classifies the weight of evidence for causation3. In developing
25                   this hierarchy, EPA has drawn on the work of previous evaluations, most prominently the
26                   lOM's Improving the Presumptive Disability Decision-Making Process for Veterans
27                   (Samet and Bodurow. 2008). EPA's Guidelines for Carcinogen Risk Assessment (U.S.
28                   EPA. 2005c). and the U.S. Surgeon General's smoking report (CDC. 2004). This weight
      1 The "aspects" described by Sir Austin Bradford Hill (Hill. 1965) have become, in the subsequent literature, more
      commonly described as "criteria." The original term "aspects" is used here to avoid confusion with "criteria" as it is
      used, with different meaning, in the Clean Air Act.
      2 The Hill aspects were developed for interpretation of epidemiologic results. They have been modified here for use
      with a broader array of data, i.e., epidemiologic, controlled human exposure, ecological, and animal toxicological
      studies, as well as in vitro data, and to be more consistent with EPA's Guidelines for Carcinogen Risk Assessment.
      3 The Center for Disease Control (CDC) and IOM frameworks use a four-category hierarchy for the strength of the
      evidence. A five-level hierarchy is used here to be consistent with the EPA Guidelines for Carcinogen Risk
      Assessment and to provide a more nuanced set of categories.
      November 2012                               Ixi                   Draft - Do Not Cite or Quote

-------
 1                   of evidence evaluation is based on various lines of evidence from across the health and
 2                   environmental effects disciplines. These separate judgments are integrated into a
 3                   qualitative statement about the overall weight of the evidence and causality. The five
 4                   descriptors for causal determination are described in Table II.

 5                   Determination of causality involves the evaluation of evidence for different types of
 6                   health, ecological or welfare effects associated with short- and long-term exposure
 7                   periods. In making determinations of causality, evidence is  evaluated for major outcome
 8                   categories or groups of related endpoints (e.g., respiratory effects, vegetation growth),
 9                   integrating evidence from across disciplines, and evaluating the coherence of evidence
10                   across a spectrum of related endpoints to draw conclusions  regarding causality. In
11                   discussing the causal determination, EPA characterizes the evidence on which the
12                   judgment is based, including strength of evidence for individual endpoints within the
13                   outcome category or group of related endpoints.

14                   In drawing judgments regarding causality for the criteria air pollutants, the ISA focuses
15                   on evidence  of effects in the range of relevant pollutant exposures or doses, and not on
16                   determination of causality at any dose. Emphasis is placed on evidence of effects at doses
17                   (e.g., blood Pb concentration) or exposures (e.g., air concentrations) that are relevant to,
18                   or somewhat above, those currently experienced by the population. The extent to which
19                   studies of higher concentrations are considered varies by pollutant and major outcome
20                   category, but generally includes those with doses or exposures in the range of one to two
21                   orders of magnitude above current or ambient conditions. Studies that use higher doses or
22                   exposures may also be considered to the extent that they provide useful information to
23                   inform understanding of mode of action, interspecies differences, or factors that may
24                   increase risk of effects for a population. Thus, a causality determination is based on
25                   weight of evidence evaluation for health, ecological or welfare effects, focusing on the
26                   evidence from exposures or doses generally ranging from current levels to one or two
27                   orders of magnitude above current levels.

28                   In addition, EPA evaluates evidence relevant to understand the quantitative relationships
29                   between pollutant exposures and health, ecological or welfare effects. This includes
30                   evaluation of the form of concentration-response or dose-response relationships and, to
31                   the extent possible, drawing conclusions on the levels at which effects are observed. The
32                   ISA also draws scientific conclusions regarding important exposure conditions for effects
33                   and populations that may be at greater risk for effects, as described in the  following
34                   section.
      November 2012                              Ixii                   Draft - Do Not Cite or Quote

-------
Table II        Weight of evidence for causal determination.
                 Health Effects
                                                     Ecological and Welfare Effects
Causal
relationship
Evidence is sufficient to conclude that there is a causal
relationship with relevant pollutant exposures
(i.e., doses or exposures generally within one to two
orders of magnitude of current levels). That is, the
pollutant has been shown to result in health effects in
studies in which chance, bias, and confounding could
be ruled out with reasonable confidence. For example:
a) controlled human exposure studies that demonstrate
consistent effects; or b) observational studies that
cannot be explained by plausible alternatives or are
supported by other lines of evidence (e.g.,  animal
studies or mode of action information). Evidence
includes multiple high-quality studies
Evidence is sufficient to conclude that there is a causal
relationship with relevant pollutant exposures
i.e., doses or exposures generally within one to two
orders of magnitude of current levels). That is, the
pollutant has been shown to result in effects in studies
in which chance, bias, and confounding could be  ruled
out with reasonable confidence. Controlled exposure
studies (laboratory or small- to medium-scale field
studies) provide the strongest evidence for causality,
but the scope of inference may be limited. Generally,
determination is based on multiple studies conducted
by multiple research groups, and evidence that is
considered sufficient to  infer a  causal relationship is
usually obtained from the joint  consideration of many
lines of evidence that reinforce each other.
                 Evidence is sufficient to conclude that a causal
                 relationship is likely to exist with relevant pollutant
                 exposures, but important uncertainties remain. That is,
                 the pollutant has been shown to result in health effects
                 in studies in which chance and bias can be ruled out
                 with reasonable confidence but potential issues
Likely to be a    remain. For example: a) observational studies show an
causal           association, but copollutant exposures are difficult to
relationship      address and/or other lines of evidence (controlled
                 human exposure, animal, or mode of action
                 information) are limited or inconsistent; or b) animal
                 toxicological evidence from multiple studies from
                 different laboratories that demonstrate effects, but
                 limited or no human data are available. Evidence
                 generally includes multiple high-quality studies.
                                                     Evidence is sufficient to conclude that there is a likely
                                                     causal association with relevant pollutant exposures.
                                                     That is, an association has been observed between the
                                                     pollutant and the outcome in studies in which chance,
                                                     bias, and confounding are minimized, but uncertainties
                                                     remain. For example, field studies show a relationship,
                                                     but suspected interacting factors cannot be controlled,
                                                     and other lines of evidence are limited or inconsistent.
                                                     Generally, determination is based on multiple studies in
                                                     multiple research groups.
                 Evidence is suggestive of a causal relationship with
                 relevant pollutant exposures, but is limited. For
                 example, (a) at least one high-quality epidemiologic
Suggestive of   study shows an association with a given health
a causal         outcome but the  results of other studies are
relationship      inconsistent; or (b) a well-conducted toxicological
                 study, such as those conducted  in the National
                 Toxicology  Program (NTP),  shows effects in animal
                 species,
                                                     Evidence is suggestive of a causal relationship with
                                                     relevant pollutant exposures, but chance, bias and
                                                     confounding cannot be ruled out. For example, at least
                                                     one high-quality study shows an effect, but the results
                                                     of other studies are inconsistent.
Inadequate to
infer a causal
relationship
Evidence is inadequate to determine that a causal
relationship exists with relevant pollutant exposures.
The available studies are of insufficient quantity,
quality, consistency, or statistical power to permit a
conclusion regarding the presence or absence of an
effect.
The available studies are of insufficient quality,
consistency, or statistical power to permit a conclusion
regarding the presence or absence of an effect.
Not likely to
be  a causal
relationship
Evidence is suggestive of no causal relationship with
relevant pollutant exposures. Several adequate
studies, covering the full range of levels of exposure
that human beings are known to encounter and
considering at-risk populations, are mutually consistent
in not showing an effect at any level of exposure.
Several adequate studies, examining relationships with
relevant exposures, are consistent in failing to show an
effect at any level of exposure.
November 2012
                                    Ixiii
          Draft - Do Not Cite or Quote

-------
            Quantitative Relationships:  Effects on Human Populations
 1                   Once a determination is made regarding the causal relationship between the pollutant and
 2                   outcome category, important questions regarding quantitative relationships include:

 3                      •  What is the concentration-response, exposure-response, or dose-response
 4                         relationship in the human population?
 5                      •  What is the interrelationship between incidence and severity of effect?
 6                      •  What exposure conditions (dose or exposure, duration and pattern) are
 7                         important?
 8                      •  What populations and lifestages appear to be differentially affected (i.e., more
 9                         at risk of experiencing effects)?

10                   In order to address these questions, the entirety of quantitative evidence is evaluated to
11                   characterize pollutant concentrations and exposure durations at which effects were
12                   observed for exposed populations, including populations and lifestages potentially at
13                   increased risk. To accomplish this, evidence is considered from multiple and diverse
14                   types of studies, and a study or set of studies that best approximates the concentration-
15                   response relationships between health outcomes and the pollutant may be identified.
16                   Controlled human exposure studies provide the most direct and quantifiable exposure-
17                   response data on the human health effects of pollutant exposures. To the extent available,
18                   the ISA evaluates results  from across epidemiologic studies that characterize the form of
19                   relationships between the pollutant and health outcomes and draws conclusions on the
20                   shape of these relationships. Animal data may also inform evaluation of
21                   concentration-response relationships, particularly relative to MOAs and characteristics of
22                   at-risk populations.

23                   An important consideration in characterizing the public health impacts associated with
24                   exposure to a pollutant is whether the concentration-response relationship is linear across
25                   the range of concentrations or if nonlinear relationships exist along any part of this range.
26                   The shape of the concentration-response curve at and below the level of the current
27                   standards is of particular  interest. Various sources of variability and uncertainty, such as
28                   low data density in the  lower concentration range, possible influence of exposure
29                   measurement error, and variability between individuals in susceptibility to air pollution
30                   health effects, tend to smooth and "linearize" the concentration-response function, and
31                   thus can obscure the existence of a threshold or nonlinear relationship. Since individual
32                   thresholds vary from person to person due to individual differences such as genetic level
33                   susceptibility or preexisting disease conditions (and even can vary from one time to
34                   another for a given person), it can be difficult to demonstrate that a threshold  exists in a
35                   population study. These sources of variability and uncertainty may explain why the
      November 2012                             Ixiv                  Draft - Do Not Cite or Quote

-------
 1                  available human data at ambient concentrations for some environmental pollutants
 2                  (e-g-, particulate matter [PM], O3, lead [Pb], environmental tobacco smoke [ETS],
 3                  radiation) do not exhibit thresholds for cancer or noncancer health effects, even though
 4                  likely mechanisms include nonlinear processes for some key events. These attributes of
 5                  human population dose-response relationships have been extensively discussed in the
 6                  broader epidemiologic literature (Rothman and Greenland. 1998).

 7                  Finally, identification of the population groups or lifestages that may be at greater risk of
 8                  health effects from air pollutant exposures contributes to an understanding of the public
 9                  health impact of pollutant exposures. In the ISA, the term "at-risk population" is used to
10                  encompass populations or lifestages that have a greater likelihood of experiencing health
11                  effects related to exposure to an air pollutant due to a variety of factors; other terms used
12                  in the literature include susceptible, vulnerable, and sensitive. These factors may be
13                  intrinsic, such as genetic or developmental factors, race, gender, lifestage, or the presence
14                  of preexisting diseases, or they may be extrinsic, such as socioeconomic status (SES),
15                  activity pattern and exercise level, reduced access to health care, low educational
16                  attainment, or increased pollutant exposures (e.g., near roadways). Epidemiologic studies
17                  can help identify populations potentially at increased risk of effects by evaluating health
18                  responses in the study population. Examples include testing for interactions or effect
19                  modification by factors such as gender, age group, or health status. Experimental studies
20                  using animal models of susceptibility or disease can also inform the extent to which
21                  health risks are likely greater in specific population groups.

            Quantitative Relationships: Effects on Ecosystems or Public
            Welfare
22                  Key questions for understanding the quantitative relationships between exposure (or
23                  concentration or deposition) to a pollutant and risk to ecosystems or the public welfare
24                  include:

25                      •  What elements of the ecosystem (e.g., types, regions, taxonomic groups,
26                         populations, functions, etc.)  appear to be affected, or are more sensitive to
27                         effects? Are there differences between locations or materials in welfare effects
28                         responses, such as impaired visibility or materials damage?
29                      •  Under what exposure conditions (amount deposited or concentration, duration
30                         and pattern) are effects seen?
31                      •  What is the shape of the concentration-response  or exposure-response
32                         relationship?

33                  Evaluations of causality generally consider the probability of quantitative changes in
34                  ecological and welfare effects in response to exposure. A challenge to the quantification

      November 2012                             Ixv                  Draft - Do Not Cite or Quote

-------
 1                   of exposure-response relationships for ecological effects is the great regional and local
 2                   spatial variability, as well as temporal variability, in ecosystems. Thus, exposure-
 3                   response relationships are often determined for a specific ecological system and scale,
 4                   rather than at the national or even regional scale. Quantitative relationships therefore are
 5                   estimated site by site and may differ greatly between ecosystems.

            Concepts  in Evaluating Adversity of Health Effects
 6                   In evaluating health evidence, a number of factors can be considered in delineating
 7                   between adverse and nonadverse health effects resulting from exposure to air pollution.
 8                   Some health outcomes, such as hospitalization for respiratory or cardiovascular diseases,
 9                   are clearly considered adverse. It is more difficult to determine the extent of change that
10                   constitutes adversity in more subtle health measures. These include a wide variety of
11                   responses, such as alterations in markers of inflammation or oxidative stress, changes in
12                   pulmonary function or heart rate variability, or alterations in neurocognitive function
13                   measures. The challenge is determining the magnitude of change in these measures when
14                   there is no clear point at which a change becomes adverse. The extent to  which a change
15                   in health measure constitutes an adverse health effect may vary between  populations.
16                   Some changes that may not be considered adverse in healthy individuals  would be
17                   potentially adverse in more at-risk individuals.

18                   The extent to which changes in lung function are adverse has been discussed by the
19                   American Thoracic Society (ATS) in an official  statement titled What Constitutes an
20                   Adverse Health Effect of Air Pollution? (ATS. 2000). An air pollution-induced shift in
21                   the population distribution of a given risk factor  for a health outcome was viewed as
22                   adverse, even though it may not increase the risk of any one individual to an unacceptable
23                   level. For example, a population of asthmatics could have a distribution of lung function
24                   such that no identifiable individual has a level associated with significant impairment.
25                   Exposure to air pollution could shift the distribution such that no identifiable individual
26                   experiences clinically relevant effects. This shift toward decreased lung function,
27                   however, would be considered adverse because individuals within the population would
28                   have diminished reserve function and therefore would be at increased risk to further
29                   environmental insult. The committee also observed that elevations of biomarkers, such as
30                   cell number and types, cytokines and reactive oxygen species, may signal risk for ongoing
31                   injury and clinical effects or may simply indicate transient responses that can provide
32                   insights into mechanisms of injury, thus illustrating the lack of clear boundaries that
33                   separate adverse from nonadverse effects.

34                   The more subtle health outcomes may be connected mechanistically to health events that
35                   are clearly adverse. For example, air pollution may affect markers of transient  myocardial
      November 2012                             Ixvi                  Draft - Do Not Cite or Quote

-------
 1                  ischemia such as ST-segment abnormalities or onset of exertional angina. These effects
 2                  may not be apparent to the individual, yet may still increase the risk of a number of
 3                  cardiac events, including myocardial infarction and sudden death. Thus, small changes in
 4                  physiological measures may not appear to be clearly adverse when considered alone, but
 5                  may be a part of a coherent and biologically plausible chain of related health outcomes
 6                  that range up to responses that are very clearly adverse, such as hospitalization or
 7                  mortality.

            Concepts in  Evaluating  Adversity of Ecological Effects
 8                  Adversity of ecological effects can be understood in terms ranging in biological level of
 9                  organization; from the cellular level to the individual organism and to the population,
10                  community, and ecosystem levels. In the context of ecology, a population is a group of
11                  individuals of the same  species, and a community is an assemblage of populations of
12                  different species interacting with one another that inhabit an area. An ecosystem is the
13                  interactive system formed from all living organisms and their abiotic (physical and
14                  chemical) environment within a given area (IPCC. 2007). The boundaries of what could
15                  be called an ecosystem are somewhat arbitrary, depending on the focus of interest or
16                  study. Thus, the extent of an ecosystem may range from very small spatial scales to,
17                  ultimately, the entire Earth (IPCC. 2007).

18                  Effects on an individual organism are generally not considered to be adverse to public
19                  welfare. However if effects occur to enough individuals within a population, then
20                  communities and ecosystems may be disrupted. Changes to populations, communities,
21                  and ecosystems can in turn result in an alteration of ecosystem processes. Ecosystem
22                  processes are defined as the metabolic functions of ecosystems including energy flow,
23                  elemental cycling, and the production, consumption and decomposition  of organic matter
24                  (U.S. EPA. 2002a). Growth, reproduction, and mortality are species-level endpoints that
25                  can be clearly linked to community and ecosystem effects and are considered to be
26                  adverse when negatively affected. Other endpoints such as changes in behavior and
27                  physiological stress can decrease ecological fitness of an organism, but are harder to link
28                  unequivocally to effects at the population, community, and ecosystem level. The degree
29                  to which pollutant exposure is considered adverse may also depend on the location and its
30                  intended use (i.e., city park, commercial, cropland). Support for consideration of
31                  adversity beyond the species level by making explicit the linkages between stress-related
32                  effects at the species and effects at the ecosystem level is found in A Framework for
33                  Assessing and Reporting on Ecological Condition: an SAB report (U.S.  EPA. 2002a).
34                  Additionally, the National Acid Precipitation Assessment Program (NAPAP.  1991) uses
35                  the following working definition of "adverse ecological  effects" in the preparation of
36                  reports to Congress mandated by the Clean Air Act: "any injury (i.e., loss of chemical or
      November 2012                             Ixvii                  Draft - Do Not Cite or Quote

-------
 1                   physical quality or viability) to any ecological or ecosystem component, up to and
 2                   including at the regional level, over both long and short terms."

 3                   On a broader scale, ecosystem services may provide indicators for ecological impacts.
 4                   Ecosystem services are the benefits that people obtain from ecosystems (UNEP, 2003).
 5                   According to the Millennium Ecosystem Assessment, ecosystem services include:
 6                   "provisioning services such as food and water; regulating services such as regulation of
 7                   floods, drought, land degradation, and disease; supporting services such as soil formation
 8                   and nutrient cycling; and cultural services such as recreational, spiritual, religious and
 9                   other nonmaterial benefits." For example, a more subtle ecological effect of pollution
10                   exposure may result in a clearly  adverse impact on ecosystem services if it results in a
11                   population decline in a species that is recreationally or culturally important.
      November 2012                             Ixviii                  Draft - Do Not Cite or Quote

-------
References for Preamble
    ATS (American Thoracic Society). (2000). What constitutes an adverse health effect of air pollution? This
       official statement of the American Thoracic Society was adopted by the ATS Board of Directors, July
       1999. Am J Respir Crit Care Med 161: 665-673.

    Bell. ML; Dominici. F; Samet. JM. (2005). A meta-analysis of time-series studies of ozone and mortality
       with comparison to the national morbidity, mortality, and air pollution study. Epidemiology 16:
       436-445. http://dx.doi.org/10.1097/01.ede.0000165817.40152.85

    CDC (Centers for Disease Control and Prevention). (2004). The health consequences of smoking: A
       report of the Surgeon General. Washington, DC: U.S. Department of Health and Human Services.
       http ://www. surgeongeneral.gov/librarv/smokingconsequences/

    Fox. GA. (1991). Practical causal inference for ecoepidemiologists. JToxicol Environ Health A 33:
       359-373. http://dx.doi.org/10.1080/15287399109531535

    Gee. GC: Pavne-Sturges. DC. (2004). Environmental health disparities: A framework integrating
       psychosocial and environmental concepts. Environ Health Perspect 112: 1645-1653.
       http://dx.doi.org/10.1289/ehp.7074

    HEW (U.S. Department of Health, Education and Welfare). (1964). Smoking and health: Report of the
       advisory committee to the surgeon general of the public health service. Washington, DC: U.S.
       Department of Health, Education, and Welfare.
       http://profiles.nlm.nih.gov/ps/retrieve/ResourceMetadata/NNBBMQ

    Hill. AB. (1965). The environment and disease: Association or causation? Proc R Soc Med 58: 295-300.

    IARC (International Agency for Research on Cancer). (2006b). Preamble to the IARC monographs. Lyon,
       France. http://monographs.iarc.fr/ENG/Preamble/

    loannidis. JPA. (2008). Why most discovered true associations are inflated [Review]. Epidemiology 19:
       640-648. http://dx.doi.org/10.1097/EDE.Ob013e31818131e7

    IPCC (Intergovernmental Panel on Climate Change). (2007). Climate change 2007: Impacts, adaptation
       and vulnerability. Cambridge, U.K.: Cambridge University Press.

    NAPAP (National Acid Precipitation Assessment Program). (1991). The experience and legacy of
       NAPAP: Report of the Oversight Review Board of the National Acid Precipitation Assessment
       Program. Washington, D.C.

    Rothman. KJ; Greenland.  S. (1998).  Modern epidemiology (2nd ed.). Philadelphia, PA: Lippincott,
       Williams, & Wilkins.

    Samet. JM; Bodurow. CC. (2008). Improving the presumptive disability decision-making process for
       veterans. In JM Samet; CC Bodurow (Eds.). Washington, DC: National Academies Press.
       http://www.nap.edu/openbook.php7record id=l 1908

    U.S. EPA (U.S.  Environmental Protection Agency). (1998). Guidelines for ecological risk assessment
       [EPA Report]. (EPA/630/R-95/002F). Washington, D.C.
       http://www.epa.gov/raf/publications/guidelines-ecological-risk-assessment.htm

    U.S. EPA (U.S.  Environmental Protection Agency). (2002a). A framework for assessing and reporting on
       ecological condition: An SAB report [EPA Report]. (EPA-SAB-EPEC-02-009). Washington, D.C.
       http ://www.ntis. gov/search/product. aspx?ABBR=PB2004100741

    U.S. EPA (U.S.  Environmental Protection Agency). (2005c). Guidelines for carcinogen risk assessment
       [EPA Report]. (EPA/63O/P-03/00IF). Washington, D.C. http://www.epa.gov/cancerguidelines/
November 2012                              Ixix                   Draft - Do Not Cite or Quote

-------
    U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for paniculate
       matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
       http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546

    UNEP (United Nations Environment Programme). (2003). Ecosystems and human well-being: A
       framework for assessment. Washington, DC: Island Press.

    Zeger. SL; Thomas. D; Dominici. F; Samet JM; Schwartz. J; Dockerv. D; Cohen. A. (2000). Exposure
       measurement error in time-series studies of air pollution: Concepts and consequences. Environ Health
       Perspect 108: 419-426.
November 2012                              Ixx                   Draft - Do Not Cite or Quote

-------
      Legislative  and  Historical  Background


           Legislative Requirements for the NAAQS Review
 1                 Two sections of the Clean Air Act (CAA) govern the establishment and revision of the
 2                 NAAQS. Section 108 (42:U.S.C.:7408) directs the Administrator to identify and list
 3                 certain air pollutants and then to issue air quality criteria for those pollutants. The
 4                 Administrator is to list those air pollutants that in her "... judgment, cause or contribute
 5                 to air pollution which may reasonably be anticipated to endanger public health or
 6                 welfare; ..." and, "... the presence of which in the ambient air results from numerous or
 7                 diverse mobile or stationary sources;" and, "... for which ... [the Administrator] plans to
 8                 issue air quality criteria	" Air quality criteria are intended to "accurately reflect the
 9                 latest scientific knowledge useful in indicating the kind and extent of all identifiable
10                 effects on public health or welfare which may be expected from the presence of [a]
11                 pollutant in the ambient air ..." (42:U.S.C.:7408([b]). Section 109 (42:U.S.C.:7409)
12                 directs the Administrator to propose and promulgate "primary" and "secondary" NAAQS
13                 for pollutants for which air quality criteria are issued.  Section 109(b)(l) defines a primary
14                 standard as  one ".. .the attainment and maintenance of which in the judgment of the
15                 Administrator, based on such criteria and allowing an adequate margin of safety, are
16                 requisite to  protect the public health." The legislative history of Section 109 indicates that
17                 a primary standard is to be set at "... the maximum permissible ambient air level ...
18                 which will protect the health of any [sensitive] group of the population," and that for this
19                 purpose "... reference should be made to a representative sample of persons comprising
20                 the sensitive group rather than to a single person in such a group..." (S. Rep. No.
21                 91:1196,  91st Cong., 2d Sess. 10 [1970]). A secondary standard, as defined in Section
22                 109(b)(2), must "... specify a level of air quality the attainment and maintenance of
23                 which, in the judgment of the Administrator, based on such criteria, is requisite to protect
24                 the public welfare from any known or anticipated adverse effects associated with the
25                 presence  of [the] pollutant in the ambient air." Welfare effects (as defined in Section
26                 302(h); 42:U.S.C.:7602[h]) include, but are not limited to, "... effects on soils, water,
27                 crops, vegetation, man-made materials, animals, wildlife, weather, visibility and climate,
28                 damage to and deterioration of property, and hazards to transportation, as well as effects
29                 on economic values and on personal comfort and well-being."

30                 The requirement that primary standards provide an adequate margin of safety was
31                 intended to  address uncertainties associated with inconclusive scientific and technical
32                 information available at the time of standard setting. It was also intended to provide a
33                 reasonable degree of protection against hazards that research has not yet identified (Lead
34                 Industries Association v. EPA, 647:F.2d: 1130-1154 [D.C.Cir 1980]; American Petroleum

     November 2012                            Ixxi                  Draft - Do Not Cite or Quote

-------
 1                   Institute v. Costle, 665:F.2d: 1176-1186 [D.C.Cir. 1981]; American Farm Bureau
 2                   Federation v. EPA, 559:F.3d:512-533 [D.C. Cir. 2009]; Association of Battery Recyclers
 3                   v. EPA, 604:F.3d:613, 617-618 [D.C. Cir. 2010]). Both kinds of uncertainties are
 4                   components  of the risk associated with pollution at levels below those at which human
 5                   health effects can be said to occur with reasonable scientific certainty. Thus, in selecting
 6                   primary standards that provide an adequate margin of safety, the Administrator is seeking
 7                   not only to prevent pollution levels that have been demonstrated to be harmful but also to
 8                   prevent lower pollutant levels that may pose an unacceptable risk of harm, even if the risk
 9                   is not precisely identified as to nature or degree. The CAA does not require the
10                   Administrator to establish a primary NAAQS at a zero-risk level or at background
11                   concentration levels (LeadIndustries v. EPA, [647:F.2d:at 1156 n.51]), but rather at a
12                   level that reduces risk sufficiently so as to protect public health with an adequate margin
13                   of safety.

14                   In addressing the requirement for an adequate margin of safety, the EPA considers such
15                   factors as the nature and severity of the health effects involved, the size of sensitive
16                   population(s) at risk, and the kind and degree of the uncertainties that must be addressed.
17                   The selection of any particular approach to providing an adequate margin of safety is a
18                   policy choice left specifically to the Administrator's judgment (Lead Industries
19                   Association v. EPA,  [647:F.2d: 1161-1162]; Whitman v. American Trucking Associations,
20                   [531:U.S.:457-495(2001)]).

21                   In setting standards that are "requisite" to protect public health and welfare as provided in
22                   Section 109(b), EPA's task  is to establish standards that are neither more nor less
23                   stringent than necessary for these purposes. In so doing, EPA may not consider the costs
24                   of implementing the standards (see generally, Whitman v. American Trucking
25                   Associations, [531:U.S.:457, 465-472, 475-476 (2001)]). Likewise, "... Attainability and
26                   technological feasibility are not relevant considerations in the promulgation of national
27                   ambient air quality standards." (American Petroleum Institute v. Costle,
28                   [665:F.2d:1185]).
      November 2012                              Ixxii                  Draft - Do Not Cite or Quote

-------
 1                  Section 109(d)(l) requires that "not later than December 31, 1980, and at 5-year intervals
 2                  thereafter, the Administrator shall complete a thorough review of the criteria published
 3                  under Section 108 and the national ambient air quality standards ... and shall make such
 4                  revisions in such criteria and standards and promulgate such new standards as may be
 5                  appropriate ... ." Section 109(d)(2) requires that an independent scientific review
 6                  committee "shall complete a review of the criteria ... and the national primary and
 7                  secondary ambient air quality  standards ... and shall recommend to the Administrator any
 8                  new ...  standards and revisions of existing criteria and standards as may be appropriate
 9                  ...  ." Since the early 1980's, this independent review function has been performed by the
10                  Clean Air Scientific Advisory Committee (CASAC).

           History of the NAAQS for Pb
11                  Unlike pollutants such as PM and carbon monoxide (CO), air quality criteria had not
12                  been issued for Pb as of the enactment of the Clean Air Act of 1970, which first set forth
13                  the requirement to set national ambient air quality standards for criteria pollutants. EPA
14                  did not intend to issue air quality criteria for lead, and accordingly had not listed lead
15                  under Section 108. EPA had determined to control lead air pollution through regulations
16                  to phase-out use of lead additives in gasoline and EPA viewed those controls, and
17                  possibly additional federal controls, as the best approach to controlling lead emissions
18                  (See 41 FR 14921 (April 8, 1976). However, the decision not to list lead under Section
19                  108 was challenged by environmental and public health groups and the U.S. District
20                  Court for the Southern District of New York concluded that EPA was required to list lead
21                  under Section 108. (Natural Resources Defense Council v.  EPA, 411  F. Supp. 864
22                  [S.D. N.Y. 1976], aff d, 545 F.2d 320 [2d Cir. 1978]).

23                  Accordingly, on April 8, 1976, EPA published a notice that Pb had been listed under
24                  Section 108 as a criteria pollutant (41 FR 14921) and on October 5, 1978, EPA
25                  promulgated primary and secondary NAAQS for Pb under Section 109 of the Act
26                  (43 FR 46246). Both primary and secondary standards were set at a level of
27                  1.5 micrograms per cubic meter (ug/m3), measured as Pb in total suspended particles (Pb-
28                  TSP), not to be exceeded by the  maximum arithmetic mean concentration averaged over
29                  a calendar quarter. These standards were based on the 1977 Pb Air Quality Criteria for
30                  Lead Document (AQCD) (U.S. EPA. 1977).

31                  The first review of the Pb standards was initiated in the mid-1980s. The scientific
32                  assessment for that review is described in the 1986 Pb AQCD (U.S. EPA. 1986a). the
33                  associated Addendum (U.S. EPA. 1986c) and the 1990 Supplement (U.S. EPA. 1990a).
34                  As part of the review, the Agency designed  and performed human exposure and health
35                  risk analyses (U.S. EPA. 1989).  the results of which were presented in a 1990 Staff Paper
      November 2012                            Ixxiii                  Draft - Do Not Cite or Quote

-------
 1                  (U.S. EPA. 1990c). Based on the scientific assessment and the human exposure and
 2                  health risk analyses, the 1990 Staff Paper presented recommendations for consideration
 3                  by the Administrator (U.S. EPA. 1990c). After consideration of the documents developed
 4                  during the review and the significantly changed circumstances since Pb was listed in
 5                  1976, the Agency did not propose any revisions to the 1978 Pb NAAQS. In a parallel
 6                  effort, the Agency developed the broad, multi-program, multimedia, integrated U.S.
 7                  Strategy for Reducing Lead Exposure (U.S. EPA. 1991). As part of implementing this
 8                  strategy, the Agency focused efforts primarily on regulatory and remedial clean-up
 9                  actions aimed at reducing Pb exposures from a variety of non-air sources judged to pose
10                  more extensive public health risks to U.S. populations, as well as on actions to reduce Pb
11                  emissions to air, such as bringing more areas into compliance with the existing Pb
12                  NAAQS (U.S. EPA. 1991).

13                  The most recent review of the Pb air quality criteria and standards was initiated in
14                  November, 2004 (69 FR 64926) and the Agency's plans for preparation of the Air
15                  Quality Criteria Document (AQCD) and conduct of the NAAQS review were contained
16                  in two documents: Project Work Plan for Revised Air Quality Criteria for Lead (U.S.
17                  EPA. 2005e); and Plan for Review of the 'National Ambient Air Quality Standards for
18                  Lead (U.S. EPA. 2006e). The schedule for completion of this review was governed by a
19                  judicial order mMissouri Coalition for the Environment v. EPA (No. 4:04CV00660
20                  ERW, Sept. 14, 2005; and amended on April 29, 2008 and July 1, 2008), which specified
21                  a schedule for the review of duration substantially shorter than five years.

22                  The scientific assessment for the review is described in the 2006 Air Quality Criteria for
23                  Lead [2006 Pb AQCD; (U.S. EPA. 2006b)1. multiple drafts of which received review by
24                  CASAC and the public. EPA also conducted human exposure and health risk assessments
25                  and a pilot ecological risk assessment for the review, after consultation with CASAC and
26                  receiving public comment on a draft analysis plan (U.S. EPA. 2006d). Drafts of these
27                  quantitative assessments were reviewed by CASAC and the public. The pilot ecological
28                  risk assessment was released in December 2006 (ICF. 2006) and the final health risk
29                  assessment report was released in November 2007 (U.S. EPA. 2007g). The policy
30                  assessment based on both of these assessments, air quality analyses and key evidence
31                  from the AQCD was presented in the Staff Paper (U.S. EPA. 2006f). a draft of which also
32                  received CASAC and public review. The final  Staff Paper presented OAQPS staffs
33                  evaluation of the public health and welfare policy implications of the key studies and
34                  scientific information contained in the 2006 Pb AQCD and presented and interpreted
35                  results from the quantitative  risk/exposure analyses conducted for this review. Based on
36                  this evaluation, the Staff Paper presented OAQPS staff recommendations that the
37                  Administrator give consideration to substantially revising the primary and secondary
38                  standards to a range of levels at or below 0.2 (ig/m3.
      November 2012                             Ixxiv                 Draft - Do Not Cite or Quote

-------
 1                  Immediately subsequent to completion of the Staff Paper, EPA issued an advance notice
 2                  of proposed rulemaking (ANPR) that was signed by the Administrator on December 5,
 3                  2007 (72 FR 71488).:  CASAC provided advice and recommendations to the
 4                  Administrator with regard to the Pb NAAQS based on its review of the ANPR and the
 5                  previously released final Staff Paper and risk assessment reports. The proposed decision
 6                  on revisions to the Pb NAAQS was signed on May 1, 2008 and published in the Federal
 7                  Register on May 20, 2008 (73 FR 29184). Members of the public provided both written
 8                  and, at two public hearings, oral comments and the CASAC Pb Panel also provided
 9                  advice and recommendations to the Administrator based on its review of the proposal
10                  notice. The final decision on revisions to the Pb NAAQS was signed on October 15, 2008
11                  and published in the Federal Register on November 12,  2008 (73 FR 66964).

12                  The November 2008 notice described EPA's decision to revise the primary and
13                  secondary NAAQS for Pb from a level of 1.5 (ig/m3 to a level of 0.15 (ig/m3. EPA's
14                  decision on the level for the primary standard was based on the much-expanded health
15                  effects evidence on neurocognitive effects of Pb in children. The level of 0.15 ug/m3 was
16                  established to protect against air Pb-related health effects, including intelligence quotient
17                  (IQ) decrements  in the most highly exposed children, those exposed at the level of the
18                  standard. Results of the quantitative risk assessment were judged supportive of the
19                  evidence-based framework estimates. The averaging time was revised to a rolling
20                  three-month period with a maximum (not-to-be-exceeded) form, evaluated over a
21                  three-year period. As compared to the previous averaging time of calendar quarter, this
22                  revision was considered to be  more scientifically appropriate and more health protective.
23                  The rolling average gives equal weight to all three-month periods, and the new
24                  calculation method gives equal weight to each  month within each three-month period.
25                  Further, the rolling average yields 12 three-month averages each year to be compared to
26                  the NAAQS versus four averages in each year  for the block calendar quarters pertaining
27                  to the previous standard. The indicator of Pb-TSP was retained, reflecting the evidence
28                  that Pb particles of all sizes pose health risks. The secondary standard was revised to be
29                  identical in all respects to the revised primary standards.2

30                  Revisions to the NAAQS were accompanied by revisions to the data handling
31                  procedures, the treatment of exceptional events, and the ambient air monitoring and
32                  reporting requirements, as well as emissions inventory reporting requirements.3 One
      1 The ANPR was one of the features of the revised NAAQS review process that EPA instituted in 2006. In 2009, this
      component of the process was replaced by reinstatement of the OAQPS policy assessment (previously termed the
      Staff Paper).
      2 The 2008 NAAQS for Pb are specified at 40 CFR 50.16.
      3 The 2008 federal regulatory measurement methods for Pb are specified in 40 CFR 50, Appendix G and 40 CFR
      part 53. Consideration of ambient air measurements with regard to judging attainment of the standards is specified in
      40 CFR 50, Appendix R. The Pb monitoring network requirements are specified in 40 CFR 58, Appendix D, section
      November 2012                             Ixxv                   Draft - Do Not Cite or Quote

-------
 1                  aspect of the new data handling requirements is the allowance for the use of Pb-PM10
 2                  monitoring for Pb NAAQS attainment purposes in certain limited circumstances at
 3                  non-source-oriented sites. The monitoring network requirements resulted in a substantial
 4                  number of new monitors being required as of January 2010, Subsequent to the 2008
 5                  rulemaking, additional revisions were made to the monitoring network requirements,
 6                  which required additional monitors as of December 2011; the complete current
 7                  requirements are described in Section 3.4.

 8                  On February 26, 2010 (75 FR 8934), EPA formally initiated its current review of the air
 9                  quality criteria for Pb, requesting the submission of recent scientific information on
10                  specified topics. Soon after, a science policy workshop was held to identify key policy
11                  issues and questions to frame the review of the Pb NAAQS (75 FR 20843). Drawing
12                  from the workshop discussions, a draft IRP  [Integrated Review Plan for the National
13                  Ambient Air Quality Standards for Lead (U.S. EPA. 2011d)1. was developed and made
14                  available in late March, 2011 for public comment and consultation with CASAC and was
15                  discussed by the CASAC via a publicly accessible teleconference consultation on May 5,
16                  2011 (76 FR 20347, 76 FR21346). The final IRP (U.S. EPA.  201 Ic) was released in
17                  November, 2011 (76 FR 76972).

18                  As part of the  science assessment phase of the current review, EPA held a workshop in
19                  December 2010 (75 FR 69078) to discuss, with invited scientific experts, preliminary
20                  draft materials prepared during the ongoing  development of the Pb ISA. The first external
21                  review draft ISA for Lead was released on May 6,2011 (U.S. EPA. 20116). The  CASAC
22                  Pb Review Panel met at a public meeting  on July 20, 2011 to review the draft ISA
23                  (76 FR 36120). Subsequently, on December 9, 2011, the CASAC panel provided a
24                  consensus letter for their review to the Administrator of the EPA (Frey and Samet 2011).
25                  The second external review draft ISA for Lead, (U.S. EPA. 2012) was discussed  at a
26                  public meeting of the CASAC Pb Review Panel on April 10, 2012. The third external
27                  review draft ISA for Lead will be discussed at a public meeting of the CASAC Pb review
28                  Panel, and timely public comments received will be provided to the CASAC panel. A
29                  future Federal Register notice will inform the public of the exact date and time of that
30                  CASAC meeting.
      4.5. Guidance on the approach for implementation of the new standards was described in the Federal Register
      notices for the proposed and final rules (73 FR 29184; 73 FR 66964).
      November 2012                            Ixxvi                 Draft - Do Not Cite or Quote

-------
References for Legislative and Historical Background
     Frev. C: Samet JM. (2011). CASAC review of the EPA's integrated science assessment for lead (first
       external review draft- May 2011).  (EPA-CASAC-12-002). Washington, DC: U.S. Environmental
       Protection Agency, Clean Air Scientific Advisory Committee.

     ICF  (ICF International). (2006). Lead human exposure and health risk assessments and ecological risk
       assessment for selected areas: Pilot phase: External review draft technical report. Research Triangle
       Park, NC: U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards.

     U.S. EPA (U.S. Environmental Protection Agency). (1977). Air quality criteria for lead [EPA Report].
       (EPA-600/8-77-017). Washington, D.C.
       http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=20013GWR.txt

     U.S. EPA (U.S. Environmental Protection Agency). (1986a). Air quality criteria for lead [EPA Report].
       (EPA/600/8-83/028aF-dF). Research Triangle Park, NC.
       http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1986c). Lead effects on cardiovascular function,
       early development, and stature: An addendum to U.S. EPAAir Quality Criteria for Lead (1986) [EPA
       Report]. (EPA-600/8-83/028aF). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (1989). Review of the national ambient air quality
       standards for lead: Exposure analysis methodology and validation: OAQPS staff report [EPA Report].
       (EPA-450/2-89-011). Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (1990a). Air quality criteria for lead:  Supplement to
       the 1986 addendum [EPA Report]. (EPA/600/8-89/049F). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (1990c). Review of the national ambient air quality
       standards for lead: Assessment of scientific and technical information: OAQPS staff paper [EPA
       Report]. (EPA-450/2-89-022). Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (1991). Strategy for reducing lead exposures  [EPA
       Report]. Washington, D.C. http://www.epa.gov/ttn/naaqs/standards/pb/data/leadstrategv 1991 .pdf

     U.S. EPA (U.S. Environmental Protection Agency). (2005e). Project work plan for revised air quality
       criteria for lead [EPA Report]. (NCEA-R-1465). Research Triangle Park, NC.
       http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=l 13963

     U.S. EPA (U.S. Environmental Protection Agency). (2006b). Air quality criteria for lead:  Volume I of II
       [EPA Report]. (EPA/600/R-05/144aF). Research Triangle Park, NC.
       http ://cfpub. epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823

     U.S. EPA (U.S. Environmental Protection Agency). (2006d). Analysis plan for human health and
       ecological risk assessment for the review of the lead national ambient air quality standards (draft)
       [EPA Report]. Research Triangle Park, NC.
       http://www.epa.gov/to/naaqs/standards/pb/sjb  crjd.html

     U.S. EPA (U.S. Environmental Protection Agency). (2006e). Plan for review of the national ambient air
       quality standards for lead [EPA Report]. Research Triangle Park, NC.
       http://www.epa.gov/to/naaqs/standards/pb/sjb  crjd.html

     U.S. EPA (U.S. Environmental Protection Agency). (2006f).  Review of the national ambient air quality
       standards for lead: Policy assessment of scientific and technical information: OAQPS staff paper - first
       draft [EPA Report]. (EPA-452/P-06-002). Research Triangle Park, NC.
November 2012                             Ixxvii                  Draft - Do Not Cite or Quote

-------
     U.S. EPA (U.S. Environmental Protection Agency). (2007g). Lead: Human exposure and health risk
        assessments for selected case studies: Volume 1: Human exposure and health risk assessments - full-
        scale [EPA Report]. (EPA-452/R-07-014a). Research Triangle Park, NC.
        http://www.ntis. gov/search/product.aspx?ABBR=PB2008102573

     U.S. EPA (U.S. Environmental Protection Agency). (20lie). Integrated review plan for the national
        ambient air quality standards for lead [EPA Report]. (EPA-452/R-11-008). Research Triangle Park,
        NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2011d). Integrated review plan for the national
        ambient air quality standards for lead: External review draft [EPA Report]. (EPA-452/D-11-001).
        Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (20lie). Integrated science assessment for lead [EPA
        Report]. http://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=226323

     U.S. EPA (U.S. Environmental Protection Agency). (2012).  Integrated science assessment for lead (2nd
        external review draft) [EPA Report]. (EPA/600/R-10/075B). RTF, NC.
November 2012                             Ixxviii                  Draft - Do Not Cite or Quote

-------
       1  EXECUTIVE  SUMMARY
      1.1        Introduction

 1                  This Integrated Science Assessment (ISA) is a synthesis and evaluation of the most
 2                  policy-relevant science that forms the scientific foundation for the review of the primary
 3                  (health-based) and secondary (welfare-based) national ambient air quality standard
 4                  (NAAQS) for Lead (Pb). In 2008, the levels of the primary and secondary NAAQS for
 5                  Pb were lowered ten-fold, from the 1978 level of 1.5 ug/m3, to a level of 0.15 ug/m3. The
 6                  averaging time was revised to a rolling three-month period with a maximum (not-to-be-
 7                  exceeded) form, evaluated over a three-year period. EPA's decision on the level for the
 8                  revised primary standard in 2008 was based on the substantive increase in the evidence
 9                  on neurocognitive effects of Pb in children. The revised standard was established to
10                  protect against air Pb-related health effects, including intelligence quotient (IQ) loss, in
11                  the most highly exposed children.

12                  The U.S. Environmental Protection Agency (EPA) has a systematic process for
13                  evaluating the scientific evidence and drawing conclusions and judgments regarding the
14                  causal association of air pollution with health and environmental effects. The ISA process
15                  includes literature search strategies, criteria for selecting and evaluating studies,
16                  approaches for evaluating the weight of the evidence, and a framework for making
17                  causality determinations. The ISA uses this five-level hierarchy that classifies the weight
18                  of evidence for causation:

19                     • Causal relationship
20                     • Likely to be a causal relationship
21                     • Suggestive of a causal relationship
22                     • Inadequate to infer a causal relationship
23                     • Not likely to be a causal relationship
24                  The process and causality framework are described in more detail in the  Preamble to the
25                  ISA. Considerations that are specific to the causal determinations drawn for the health
26                  and ecological effects of Pb are described in Section 2.1 of the document.
      November 2012                              1-1                   Draft - Do Not Cite or Quote

-------
      1.2        Sources, Fate and Transport of Lead (Pb) in the Environment,
                 and the Resulting Human Exposure and Dose

 1                  Emissions of Pb to ambient air have declined by more than two orders of magnitude over
 2                  the period 1970 to 2008 following the ban on alkyl-Pb additives for on-road gasoline and
 3                  tightened industrial emission standards. Emissions in the U.S. were estimated to be 964
 4                  tons in 2008, a small fraction of the total Pb used in production. More than half of these
 5                  emissions were from piston-engine aircraft. Other important sources of ambient air Pb,
 6                  beginning with the next largest, include metals processing, fossil fuel combustion, other
 7                  industrial sources, and roadway-related sources.

 8                  During the same period that saw the dramatic decrease in Pb emissions, ambient air Pb
 9                  concentrations1 also declined. The median annual concentration in 2010,
10                  0.03 micrograms per cubic meter (ug/m3), was approximately thirty times lower than it
11                  was in 1980. The sharpest drop in median Pb concentration occurred from 1980-1990;
12                  concentrations continued to decline up to 2010. Specific levels near Pb sources as well as
13                  away from Pb sources have also shown a sharp decrease (Section 2.2.2).

14                  Atmospheric deposition has led to measurable Pb concentrations in rain, snowpack, soil,
15                  surface waters, sediments, agricultural plants, livestock, and wildlife across the world,
16                  with the highest concentrations near Pb sources, such as smelters. After the phase-out of
17                  Pb from on-road gasoline and declining industrial emissions, Pb concentrations have
18                  decreased considerably in rain, snowpack, and surface waters. Pb is retained in soils and
19                  sediments, where it provides a historical record of deposition and associated
20                  concentrations. The national average Pb concentration in soil was 18.9 milligrams of Pb
21                  per kilogram (mg Pb/kg), measured in over 1,300 non-urban, generally vegetated
22                  sampling locations. The national median fresh surface water Pb concentration was
23                  0.5 micrograms per liter ((ig/L) (Section 2.2.3). In remote lakes, sediment profiles
24                  indicate higher Pb concentrations in near surface sediment as compared to pre-industrial
25                  era sediment from greater depth; sediment profiles indicate peak Pb concentrations
26                  between 1960 and 1980 (when leaded gasoline was at peak use).

27                  The size distribution of Pb-bearing particulate matter (PM), (i.e., PM having Pb as one of
28                  its components) depends on whether there are contributions from industrial sources or
      1 The original indicator for the Pb NAAQS is the mass of the Pb portion of total suspended particles (Pb-TSP). The
      Pb-TSP indicator was retained in 2008 in recognition of the role of all particulate matter (PM) sizes in ambient air
      Pb exposures (Section 2.2.2). The Federal Reference Method (FRM) Pb-TSP sampler's size-selective performance
      is known to be affected by wind speed and direction, and collection efficiency has been demonstrated to decline with
      particle size. Under certain conditions regulatory Pb monitoring can also be performed for ambient Pb sampled
      using the FRM for Pb sampled in particles with an upper 50% cut-point of 10 ± 0.5 micrometer (urn) aerodynamic
      diameter (Pb-PM10). Pb-PM10 is allowed in certain instances where the expected Pb concentration does not approach
      the NAAQS and no sources of ultracoarse Pb particles are nearby.
      November 2012                             1-2                   Draft - Do Not Cite or Quote

-------
 1                   near-road environments (Section 2.2.2). The size distribution of ambient air Pb-bearing
 2                   PM is smaller than the size distribution of soil and dust Pb particles containing PM
 3                   (i.e., soil or dust particles having Pb as one of its components).1 Coarse Pb-bearing PM
 4                   (i.e., approximately 2.5 - 10 (im) deposits to a great extent near its source, contributing to
 5                   local soil Pb contamination, while fine Pb-bearing PM (i.e., smaller than approximately
 6                   2.5 (im) can be transported long distances and possibly deposit in remote areas.
 7                   Depending on local conditions, deposited particles may be resuspended and redeposited
 8                   multiple times before further transport becomes unlikely.

 9                   There are multiple sources of ambient Pb, and human exposure to ambient Pb involves
10                   multiple pathways. Figure  1-1 shows how Pb can cycle through multiple environmental
11                   media. The "air/soil/water" arrows of the figure depict Pb exposures to plants, animals,
12                   and/or humans through contact with Pb-containing media.  Air-related pathways of
13                   ambient Pb exposure are the focus of this assessment. Air-related ambient Pb exposures
14                   include both inhalation of Pb and ingestion of Pb in  dust or soil that originated in the
15                   ambient air. For example, dietary Pb exposure may be air-related if ambient air Pb
16                   deposits on plants or water that become available  for human consumption. Dust and soil
17                   particles containing Pb are typically in the size range that is ingested rather than inhaled.
18                   However, soil can act as a reservoir for deposited Pb emissions, and exposure to soil
19                   contaminated with deposited Pb can occur through resuspended PM as well as hand-to-
20                   mouth contact, which is the main pathway of childhood exposure to Pb. Non-ambient,
21                   non-air-related exposures include hand-to-mouth contact with Pb-containing consumer
22                   goods, hand-to-mouth contact with  dust or chips of peeling Pb-containing paint, or
23                   ingestion of Pb in drinking water conveyed through  Pb pipes.
      1 Pb-bearing PM larger than 10 ^m have a sharp concentration gradient with distance from the source, because
      larger particles have greater settling velocities. Given that wind-related biases strongly affect particles larger than
      10 \im, and given that much of the ambient air Pb fraction is smaller than 10 ^m, the existing TSP samplers
      reasonably capture the airborne fraction of ambient Pb that is available for human exposure.
      November 2012                              1-3                   Draft - Do Not Cite or Quote

-------
                                            Newly Emitted Pb
                                         Historically Emitted Pb
                             OOTDOORSOIL
                                   NDDUST
                  NATURAL WATERS
                   ND SEDIMENTS
 Non-air Pb
   eleases
                                   AIR
                                  SOIL
                                 WATER
    AIR
    SOIL
   WATER
                  AIR
                  SOIL
                 WATER
                                  PLANT
                               EXPOSURE
                  HUMAN
                EXPOSURE
EXPOSURE
                                                                              rinking Water
                                                                 WELR
                                                              COSMETICS
                                                               JOYS etc
OCCUPATION
     Note: This Venn diagram illustrates the passage of Pb through multiple environmental media compartments through which plant,
     animal, and human exposures can occur.

     Figure 1-1      Conceptual model of multimedia Pb exposure.

 1                 The majority of Pb in the body is stored in bone (roughly 90% in adults, 70% in
 2                 children). Much of the remaining Pb is found in soft tissues; only about 1% of Pb is
 3                 found in the blood. Pb in blood is primarily (-99%) bound to red blood cells [RBCs]).
 4                 The small fraction of Pb in blood plasma (<1% of Pb in blood) may be the more
 5                 biologically labile and lexicologically active fraction of the circulating Pb. Both
 6                 Pb uptake and elimination in soft tissues are much faster than they are in bone. Pb
 7                 accumulates in bone regions undergoing the most active calcification at the time of
 8                 exposure. Pb in bone becomes distributed in trabecular (e.g., patella [knee cap]) and the
 9                 more dense cortical bones (e.g., tibia [shin bone]).

10                 Blood Pb is the most common measure used to estimate Pb dose or exposure in
11                 epidemiologic studies of Pb health effects. Overall, blood Pb levels have been decreasing
12                 among U.S. children and adults for the past twenty years. The median blood Pb level for
13                 the U.S. population is 1.1 micrograms per deciliter (ug/dL), with a 95th percentile blood
14                 Pb level of 3.3 ug/dL based on the 2009-2010 National Health and Nutrition
15                 Examination Survey (NHANES) data. Among children aged 1-5 years, the median and
16                 95th percentiles are slightly higher at 1.2 ug/dL and 4.0 ug/dL, respectively. Other
     November 2012
 1-4
                     Draft - Do Not Cite or Quote

-------
 1                  common metrics of Pb dose or exposure used in epidemiologic studies are Pb in bone,
 2                  which measures cumulative exposure over long periods (months to years), and Pb in cord
 3                  blood, which is an indicator of prenatal blood Pb concentration.

 4                  Blood Pb is dependent on both the recent exposure history of the individual, as well as
 5                  the long-term exposure history that determines total body burden and the amount of Pb
 6                  stored in the bone. The contribution of bone Pb to blood Pb changes throughout an
 7                  individual's life time, and depends on the duration and intensity of the exposure, age, and
 8                  various other physiological stressors (e.g., nutritional status, pregnancy, menopause,
 9                  extended bed rest, hyperparathyroidism) that may affect bone remodeling, which
10                  normally and continuously occurs. In children, largely due to faster exchange of Pb to
11                  and from bone, blood Pb is both an index of recent exposure and potentially an index of
12                  body burden. Generally, bone Pb is an index  of cumulative exposure and body burden. Pb
13                  is sequestered in two types of bone compartments; Pb in trabecular bone exchanges more
14                  rapidly with the blood than Pb in denser cortical bone. Therefore, Pb in cortical bone is a
15                  better marker of cumulative exposure, and  Pb in trabecular bone is more likely to be
16                  correlated with blood Pb. During pregnancy,  Pb is transferred from the mother to the
17                  fetus. Transplacental transfer of Pb may be facilitated by an increase in the plasma/blood
18                  Pb concentration ratio during pregnancy. Maternal-to-fetal transfer of Pb appears to be
19                  related partly to the mobilization of Pb from the maternal skeleton.
      1.3        Integrative Overview of Health and Ecological Effects

20                  There is substantial overlap between the ecological and health endpoints related to Pb
21                  exposure, which can be mediated through multiple, interconnected modes of action
22                  (MOAs). The cellular/subcellular effect constituting the principal MOA for human health
23                  and ecological endpoints is altered ion status. Other related MOAs include protein
24                  binding, oxidative stress, inflammation, endocrine disruption, and cell death and
25                  genotoxicity (Figure 1-2). Since the mechanisms of Pb toxicity in some organ systems are
26                  the same or similar across species,  many of the downstream health and ecological effects
27                  are similar across species from invertebrates to vertebrates, including humans
28                  (Section 2.8.1).
      November 2012                             1-5                   Draft - Do Not Cite or Quote

-------
£
Oxidative
Stress
(5.2.4)
/\
/

Cell Death
and
Genotoxicity
(5.2.7)
Inflammation
(5.2.5)

Endocrine
Disruption
(5.2.6)
     Note: The subsections where these MOAs are discussed are indicated in parentheses.
     Figure 1-2       Schematic representation of the relationships between the
                       various MOAs by which Pb exerts its effects.
     1.3.1        Health Effects of Pb

1                  Evidence from epidemiologic and toxicological studies was considered in combination
2                  with the evidence from other disciplines such as exposure sciences and toxicokinetics in
3                  determining the causal relationships for the health endpoints discussed in this assessment.
4                  Detailed discussions of the evidence relating to conclusions regarding the health effects
5                  of Pb are in Section 2.6 and Chapter 5. The major conclusions regarding health effects
6                  from Pb exposure in children and adults are presented in Table 1-1 and summarized
7                  below.
     November 2012
1-6
Draft - Do Not Cite or Quote

-------
Table 1-1     Summary of causal determinations for the relationship  between
                 exposure to Pb and  health effects.
Health Effect
              Causality Determination3
Nervous System Effects (Section 2.6.1)
  Cognitive Function Decrements in Children
              Causal Relationship
  Attention-Related Behavioral Problems in Children
                                                               Causal Relationship
  Conduct Problems in Children and Young Adults
              Likely Causal Relationship
  Internalizing Behaviors in Children
              Likely Causal Relationship
  Sensory Function Decrements in Children
              Likely Causal Relationship
  Motor Function Decrements in Children
                                                               Likely Causal Relationship
  Cognitive Function Decrements in Adults
              Likely Causal Relationship
  Psychopathological Effects in Adults
              Likely Causal Relationship
  Sensory Function Decrements in Adults
              Suggestive of a Causal Relationship
  Neurodegenerative Diseases in Adults
              Inadequate to Infer a Causal Relationship
Cardiovascular Effects (Section 2.6.2)
  Hypertension
              Causal Relationship
  Subclinical Atherosclerosis
                                                               Suggestive of a Causal Relationship
  Coronary Heart Disease
              Causal Relationship
  Cerebrovascular Disease
                                                               Inadequate to Infer a Causal Relationship
Renal Effects (Section 2.6.3)
  Reduced Kidney Function
              Likely Causal Relationship
Immune System Effects (Section 2.6.4)
  Atopic and Inflammatory Responses
              Likely Causal Relationship
  Decreased Host Resistance
                                                               Likely Causal Relationship
  Autoimmunity
              Inadequate to Infer a Causal Relationship
Hematologic Effects (Section 2.6.5)
  Decreased Red Blood Cell Survival and Function
                                                               Causal Relationship
  Altered Heme Synthesis
              Causal Relationship
Reproductive and Developmental Effects (Section 2.6.6)
  Development
              Causal Relationship
  Birth Outcomes
                                                               Suggestive of Causal Relationship
  Male Reproductive Function
              Causal Relationship
  Female Reproductive Function
              Suggestive of Causal Relationship
Cancer (Section 2.6.7)
  Cancer
                                                               Likely Causal Relationship
"Causal determinations were made within approximately 1  order of magnitude of current levels (Preamble, and Section 2.1).
November 2012
1-7
Draft - Do Not Cite or Quote

-------
                     Effects of Pb Exposure in Children

 1                   Multiple epidemiologic studies conducted in diverse populations of children consistently
 2                   demonstrate the harmful effects of Pb exposure on IQ, academic performance, learning
 3                   and memory. Epidemiologic studies also demonstrate the effect of Pb exposure on
 4                   inattention, impulsivity, and hyperactivity in children. The evidence in children is
 5                   supported by findings in animal studies demonstrating both analogous effects and
 6                   biological plausibility at relevant exposure levels. A decrease in cognitive function has
 7                   been observed in populations of children 4 to 11 years old with mean blood Pb levels
 8                   between 2 and 8 ug/dL (Section 2.6.1.1). Evidence suggests that some Pb-related
 9                   cognitive effects may not be reversible and that neurodevelopmental effects of Pb may
10                   persist into adulthood (Section 2.9.4). Pb exposure also causes hematologic effects (such
11                   as effects on blood cells or blood producing organs)  in children and is associated with an
12                   increased risk of internalizing behaviors (e.g., withdrawn behavior and depressive
13                   symptoms), sensory and motor function  decrements, atopic and inflammatory conditions
14                   (e-g-, asthma and allergy) in children, as well as misconduct in older children and young
15                   adults. Uncertainties arising from the lack of information about the specific Pb-exposure
16                   histories which contribute to observed blood Pb levels are greater in adults and older
17                   children than in young children (Section 2.9.5). Despite some uncertainties regarding the
18                   interpretation of blood Pb levels in older children, it is clear that Pb exposure in
19                   childhood presents a risk; further, there is no evidence of a threshold below which there
20                   are no harmful effects from Pb exposure.


                     Effects of Pb Exposure in Adults

21                   A large body of evidence from both epidemiologic studies of adults and experimental
22                   studies in animals  demonstrates the effect of long-term Pb exposure on increased blood
23                   pressure (BP) and  hypertension (Section 2.6.2). In addition to its effect on BP, Pb
24                   exposure leads to coronary heart disease and death from cardiovascular causes and is
25                   likely to cause cognitive function decrements, symptoms of depression and anxiety,
26                   reduced kidney function, and immune effects in adult humans. The extent to which the
27                   effects of Pb on the cardiovascular system are reversible is not well-characterized. It is
28                   also important to note that the frequency, timing, level and duration of Pb exposure
29                   causing the effects observed in adults has not been pinpointed, and higher past exposures
30                   may well have contributed to the development of health effects measured later in life.
31                   However, it is clear that Pb exposure can be harmful to the cardiovascular system and
32                   may also affect a broad array of organ systems in adults.
      November 2012                              1-8                   Draft - Do Not Cite or Quote

-------
      1.3.2        Ecological Effects of Pb

 1                   Ecological effects of Pb are summarized for terrestrial, freshwater and saltwater
 2                   ecosystems, and the ISA discusses endpoints common to plants, invertebrates and
 3                   vertebrates along with considerations of uncertainties in relating atmospheric Pb
 4                   concentrations to ecosystem effects. Effects of Pb in ecosystems are primarily associated
 5                   with Pb deposition onto soil and water, subsequent transport, and exposure through
 6                   environmental media (soil, water, sediment, biota). The 2006 Pb Air Quality Criteria
 7                   Document (AQCD) (U.S.  EPA. 2006b) and previous EPA assessments reported effects of
 8                   Pb exposure on both terrestrial and aquatic organisms that included reduced survival,
 9                   reproduction and growth. Effects on delta-aminolevulinic acid dehydratase (ALAD, an
10                   important rate-limiting enzyme needed for heme production), development, and behavior
11                   were reported in terrestrial organisms (e.g. birds,  mammals), along with decreases in
12                   enzyme activity, heme formation, and behavioral effects in aquatic organisms (e.g. fish,
13                   aquatic invertebrates). Physiological stress and impacts on hematological and
14                   neurobehavioral endpoints may increase susceptibility to other stressors and affect the
15                   fitness of individual organisms, and changes in reproduction, growth, and survival are
16                   likely to lead to changes in communities and ecosystems. Although increasing exposures
17                   generally result in increasing responses in laboratory and field experiments, the
18                   relationship of exposure and responses is difficult to characterize quantitatively in natural
19                   systems because of the influence of multiple environmental variables on both Pb
20                   bioavailability and toxicity, and substantial species and lifestage differences in Pb
21                   sensitivity.

22                   A brief discussion of the conclusions from this assessment and earlier Pb AQCDs
23                   regarding Pb effects on reproduction, growth, and survival is provided below and
24                   summarized in Table 1-2 along with effects of Pb on neurobehavior, hematological, and
25                   stress endpoints. Reproduction, growth, and survival are endpoints commonly used in
26                   ecological risk assessment because they can lead to effects at the population, community,
27                   and ecosystem levels of biological organization. Causal determinations for ecological
28                   effects were based on integration of information on biogeochemistry, bioavailability,
29                   biological effects, and exposure-response relationships of Pb in terrestrial, freshwater,
30                   and saltwater environments. In general, the number of studies available for assessing
31                   causality is greater for freshwater organisms than for marine environments. A detailed
32                   discussion for all relevant welfare effects (i.e. ecological effects) is provided in
33                   Section 2.7 and Chapter_7.
      November 2012                              1-9                   Draft - Do Not Cite or Quote

-------
Table 1-2     Summary of causal determinations for the relationship between Pb
               exposure and effect on plants, invertebrates and vertebrates.
     Level
Effect
Terrestrial3
Freshwater
Saltwater3
Community-
and Ecosystem
1 Population-Level Endpoints
Sub-organismal _ . . . _
R a Organism-Level Responses
Community and Ecosystem Effects (Section 2.7.3.7)
Reproductive and Developmental Effects-Plants
(Section 2.7.3.1)
Reproductive and Developmental Effects-Invertebrates
(Section 2.7.3.1)
Reproductive and Developmental Effects- Vertebrates
(Section 2.7.3.1)
Growth-Plants (Section 2.7.3.2)
Growth-Invertebrates (Section 2.7.3.2)
Growth-Vertebrates (Section 2.7.3.2)
Survival-Plants (Section 2.7.3.3)
Survival- Invertebrates (Section 2.7.3.3)
Survival- Vertebrates (Section 2.7.3.3)
Neurobehavioral Effects-Invertebrates (Section 2.7.3.4)
Neurobehavioral Effects- Vertebrates (Section 2.7.3.4)

Hematoloaical Effects-Invertebrates (Section 2.7.3.5)

Hematological Effects-Vertebrates (Section 2.7.3.5)
Physioloaical Stress-Plants (Section 2.7.3.6)

Physioloaical Stress-Invertebrates (Section 2.7.3.6)

Phvsioloaical Stress-Vertebrates (Section 2.7.3.6)
Likely Causal
Inadequate
Causal
Causal
Causal
Likely Causal
Inadequate
Inadequate
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Causal
Causal
Likely Causal
Causal
Inadequate
Inadequate
Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Inadequate
Suggestive
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Suggestive
Inadequate
Inadequate
Suggestive
Inadequate
"Based on the weight of evidence for causal determination in Table II of the ISA Preamble. Ecological causal determinations are
based on doses or exposures generally within one to two orders of magnitude of the range of Pb currently measured in the
environment (Table 2-1).
November 2012
                           1-10
      Draft - Do Not Cite or Quote

-------
                     Effects on Development and Reproduction

 1                   Reduced reproduction at the level of individual organisms can result in lowered
 2                   population numbers and/or extermination, decreased species diversity, and a decline in
 3                   relative or absolute population numbers at the community level. Effects of Pb on various
 4                   development, fertility, and hormone maintenance endpoints have been documented in
 5                   multiple species of terrestrial and freshwater organisms. In plants, only a few studies
 6                   have addressed reproductive effects of Pb exposure. Among the animal species tested,
 7                   freshwater invertebrates were the most sensitive to Pb with respect to reproduction
 8                   (Section 2.7.3.1).


                     Effects on Growth

 9                   Effects on growth observed at the species level can translate into effects at the ecosystem
10                   level. Exposure to Pb has been shown to have effects on growth in plants and in some
11                   species of invertebrates and vertebrates. Evidence for effects of Pb on growth is strongest
12                   in terrestrial plants. These effects are typically found in laboratory studies with high Pb
13                   exposure concentrations or in field studies near stationary sources where concentrations
14                   are elevated relative to non-polluted locations. Many of those laboratory and field studies
15                   evaluate the effects of increasing levels of Pb exposure, and find that effects on plant
16                   growth increase with increasing  exposure ("biological gradients"). Evidence for Pb
17                   effects on growth in invertebrates has been observed most extensively in freshwater
18                   aquatic species, with growth inhibition in sensitive species occurring in the range of Pb
19                   concentration values available for U.S. surface waters. In general, juvenile organisms are
20                   more sensitive than adults. There are only limited data on growth effects in vertebrates
21                   (Section 2.7.3.2).


                     Effects on Survival

22                   Decreased survival of individuals within a population can have ecosystem-level impacts.
23                   Pb is generally not toxic to aquatic or terrestrial plants at concentrations found in the
24                   environment away from stationary sources, probably due to the fact that plants often
25                   sequester large amounts of Pb in roots, with little translocation to other parts of the plant.
26                   Aquatic invertebrates are generally more sensitive to Pb exposure than other types  of
27                   species, with survival reduced in a few species at concentrations occurring near Pb
28                   sources, as well as at concentrations occasionally encountered in the general environment
29                   (that is, far from major Pb sources). Many terrestrial invertebrates tolerate higher
30                   concentrations of Pb.  Limited studies with vertebrates showed adverse effects of Pb on
31                   survival at concentrations higher than typical ambient Pb  levels in the environment,
32                   although juvenile organisms are  usually more sensitive than adults (Section 2.7.3.3).

      November 2012                              1-11                   Draft - Do Not Cite or Quote

-------
                     Neurobehavioral Effects

 1                   Historical and recent evidence from Pb-exposed animals indicates that Pb affects
 2                   behaviors, such as food consumption, avoidance and escape from predators, behavioral
 3                   regulation of body temperature, and prey capture. Alterations to these behaviors can
 4                   decrease the overall fitness of the organism. Evidence from laboratory studies has shown
 5                   effects of Pb exposure on nervous system endpoints in both terrestrial and freshwater
 6                   animal taxa (Section 2.7.3.4).


                     Hematological Effects

 7                   Changes in hematological characteristics including ALAD activity, blood cell counts, and
 8                   serum profiles are associated with Pb exposure in both aquatic and terrestrial animals. It
 9                   is commonly recognized that ALAD is an indicator of Pb exposure across a wide range of
10                   animals as shown in both field and laboratory studies. Studies conducted over the last two
11                   decades have shown that hematological responses are associated with Pb in the
12                   environment (Section 2.7.3.5).


                     Effects on Physiological Stress

13                   Increased levels of antioxidant enzymes (in response to oxidative stress or altered cell
14                   signaling) and increased lipid peroxidation (the process by which free radicals induce the
15                   oxidation of fatty acids, leading to cell membrane damage) are considered to be reliable
16                   biomarkers of stress. Alterations in these biomarkers are associated with Pb exposure in
17                   plants, invertebrates and vertebrates, and they may be indicative of increased
18                   susceptibility to other stressors,  as well as reduction in individual fitness. Markers of
19                   oxidative damage and antioxidant activity have been observed in field studies in a wide
20                   range of species in terrestrial and aquatic environments when Pb is present, and also
21                   following laboratory exposures  (Section 2.7.3.6).


                     Community and Ecosystem Effects

22                   The effects of Pb on growth, reproduction, and survival at the level of individual
23                   organisms, especially when considered cumulatively, are likely to result in effects on
24                   population, community and ecosystem structure and function. Effects at those higher
25                   levels of biological organization are confirmed by both laboratory and field experiments
26                   in which decreases in abundance, reduced species diversity, and shifts in community
27                   composition have been observed following Pb exposure. However, such ecosystem-wide
28                   effects can only be tested directly in a few of the cases where individual organism effects
29                   are found. Quantitative characterization of exposure-response relationships is difficult at

      November 2012                             1-12                  Draft - Do Not Cite or Quote

-------
 1                  the community and ecosystem levels because potential confounders such as the presence
 2                  of other metals, physico-chemical variables and other stressors cannot be controlled and
 3                  their effects are incompletely characterized (Section 2.7.3.7).
      1.4       Policy Relevant Considerations

                    Public Health Significance

 4                  The concept of population risk is relevant to understanding the public health significance
 5                  of small decrements in IQ and increases in blood pressure caused by Pb exposure. A
 6                  seemingly small decrease in the population mean IQ or increase in systolic blood
 7                  pressure may push the most susceptible group in the population to a critical point on the
 8                  continuum of disease development, such that their condition meets the clinical definition
 9                  of a disease. For example, increases in blood pressure that are caused by Pb exposure can
10                  result in a larger proportion of the population having hypertension. A downward shift in
11                  the mean IQ value can result in a larger proportion of the population at risk for academic
12                  or vocational failure and can also reduce the proportion of the population achieving very
13                  high IQ scores. Moreover, small changes at the population level can translate into large
14                  numbers of individual clinical events.
                    Air Lead(Pb)-to-Blood Lead(Pb) Relationships

15                  A limited number of epidemiological studies evaluated relationships between air Pb and
16                  blood Pb (Section 2.9.2). Regression models are typically used to produce slopes that
17                  estimate the change in blood Pb per change in air Pb concentration ((ig/dL per (ig Pb/m3).
18                  The larger the slope, the larger the estimated contribution of air Pb is to the blood Pb
19                  level in exposed populations.

20                  The range of air-to-blood slope estimates is 2 to 9 (ig/dL per (ig/m3 in studies of children.
21                  The differences in the estimates across studies, at least in part, reflect the choice of model
22                  (e.g., some models predict an increase in the blood Pb-air Pb slope with decreasing air Pb
23                  concentration while other models predict a constant blood Pb-air Pb slope across all air
24                  Pb concentrations) as well as the terms that are included in the model (e.g., soil Pb) that
25                  may account for some of the variation in blood Pb that is attributable to air Pb. Other
26                  factors that may explain the variation in the derived blood Pb-air Pb slope include
27                  differences in the populations examined and Pb sources (e.g., leaded gasoline or smelter).
      November 2012                             1-13                  Draft - Do Not Cite or Quote

-------
                     Concentration-Response Relationships for Health Effects

 1                   Previous assessments found that progressively lower blood Pb levels were associated
 2                   with cognitive deficits in children and newly available evidence is generally consistent
 3                   with findings of the previous review (Section 2.9.3). Compelling evidence for a larger
 4                   effect of Pb on children's IQ at lower blood Pb levels compared to higher blood Pb levels
 5                   was presented in the 2006 Pb AQCD based on the international pooled analysis of seven
 6                   prospective cohort studies. A subsequent reanalysis of these data focusing on the shape of
 7                   the concentration-response function and several recent studies support the findings of the
 8                   original the pooled analysis. The majority of the epidemiologic evidence from stratified
 9                   analyses comparing the lower and the higher ends of the blood Pb distributions also
10                   indicates larger effect of Pb on IQ at lower blood Pb levels. The shape of concentration-
11                   response relationships is not well characterized for effects of Pb in adults (Section 2.9.3).


                     Patterns of Pb Exposure and Neurodevelopmental Deficits in Children

12                   Among the populations included in epidemiologic studies using blood Pb as a metric of
13                   Pb exposure, the relative proportion of blood Pb derived from recent versus past exposure
14                   cannot be fully characterized in the absence of detailed information on exposure history.
15                   Uncertainty regarding the role of recent exposure is greater in adults and older children
16                   than in young children who do not have lengthy exposure histories. Several lines of
17                   evidence inform the interpretation of epidemiologic studies of young children with regard
18                   to the patterns of exposure that contribute to observed health effects (Section 2.9.4). In
19                   summary, epidemiologic studies find associations of cognitive function and/or attention
20                   related behavior problems with several different blood Pb exposure metrics that represent
21                   blood Pb during lifestages from prenatal to adolescence. These findings are generally
22                   consistent with Pb effects reported in experimental animal studies and are consistent with
23                   the fact that the nervous system continues to develop throughout childhood.


                     Potentially At-Risk Populations

24                   The NAAQS are intended to protect public health with an adequate margin of safety. In
25                   so doing, protection provided for both the population as a whole and those groups at
26                   increased risk for health effects  in response to the air pollutant for which each NAAQS is
27                   set. Children are at increased risk for the effects of Pb exposure. Among children, the
28                   youngest age groups were observed to be most at risk of elevated blood Pb levels, with
29                   levels decreasing with increasing age of the children. Evidence related to childhood and
30                   other at-risk factors is described in  Section 2.9.6.
      November 2012                             1-14                  Draft - Do Not Cite or Quote

-------
                     Pb Concentrations Corresponding to Ecological Effects

 1                   There is limited evidence to relate ambient air concentrations of Pb to levels of deposition
 2                   onto terrestrial and aquatic ecosystems and to subsequent movement of atmospherically-
 3                   deposited Pb through environmental compartments (e.g., soil, sediment, water, and biota)
 4                   (Section 2.9.7). The contribution of atmospheric Pb to specific sites is not clear and the
 5                   connection between air concentration of Pb and ecosystem exposure continues to be
 6                   poorly characterized. Furthermore, the level at which Pb elicits a specific effect is
 7                   difficult to establish in terrestrial and aquatic systems, due to the influence of other
 8                   environmental variables (e.g., pH, organic matter) on both Pb bioavailability and toxicity,
 9                   and also to substantial species differences in Pb sensitivity.  Current evidence indicates
10                   that Pb is bioaccumulated in biota; however, the sources of Pb  in biota have only been
11                   identified in a few studies, and the  relative contribution of Pb from all sources is usually
12                   not known.
      1.5       Summary

13                   Overall, the evidence evaluated for the current review expands upon findings of the
14                   2006 Pb AQCD and previous assessments, which concluded that there was a strong body
15                   of evidence substantiating the health effects from Pb exposure as well as strong evidence
16                   of the effects from Pb exposure on some ecological endpoints.

17                   Nervous system effects in children, specifically cognition problems in children, are the
18                   effects that are best substantiated as occurring at the lowest blood Pb concentrations
19                   (Section 2.6.1.1). Causal relationships were also determined for several cardiovascular
20                   effects in adults, for which the evidence strongly suggests that long-term Pb exposure
21                   plays a role. Since Pb exposures were generally higher in the past than they are today,
22                   uncertainties still exist regarding the relative importance of recent versus past exposure in
23                   the development of the Pb-related health effects in the adult populations studied.

24                   With regard to the ecological effects of Pb, uptake of Pb into fauna and subsequent
25                   effects on reproduction, growth and survival are established and are further supported by
26                   more recent evidence. These may  lead to effects at the population, community, and
27                   ecosystem level of biological organization. In both terrestrial and aquatic organisms,
28                   gradients  in response are observed with increasing concentration of Pb and some studies
29                   report effects within the range  of Pb detected in environmental media. Specifically,
30                   observations from controlled studies on reproduction, growth, and survival in sensitive
31                   freshwater invertebrates are well-characterized at concentrations at or near Pb
32                   concentrations occasionally encountered in U.S. surface waters. Hematological and stress
      November 2012                             1-15                  Draft - Do Not Cite or Quote

-------
 1                   related responses were also associated with elevated Pb levels in polluted areas in some
 2                   terrestrial and aquatic species. However, in natural environments, modifying factors
 3                   affect Pb bioavailability and toxicity and there are considerable uncertainties associated
 4                   with generalizing effects observed in controlled studies to effects at higher levels of
 5                   biological organization.  Furthermore, available studies on community and ecosystem-
 6                   level effects are usually  from contaminated areas where Pb concentrations are much
 7                   higher than typically encountered in the environment. The contribution of atmospheric Pb
 8                   to specific sites is not clear and the connection between air concentration of Pb and
 9                   ecosystem exposure continues to be poorly characterized. Furthermore, the level at which
10                   Pb elicits a specific effect is difficult to establish in terrestrial and aquatic systems, due to
11                   the influence of other environmental variables (e.g., pH, organic matter) on both Pb
12                   bioavailability and toxicity, and also to substantial species differences in Pb sensitivity.
      November 2012                              1-16                   Draft - Do Not Cite or Quote

-------
       2 INTEGRATIVE SUMMARY
      2.1        ISA Development and Scope

 1                  This chapter summarizes and synthesizes the recently available scientific evidence and is
 2                  intended to provide a concise synopsis of the ISA conclusions and findings that best
 3                  inform the review of the current NAAQS for lead (Pb). The Integrated Review Plan (IRP)
 4                  for the National Ambient Air Quality Standards for Lead (U.S. EPA. 201 Ic) identifies a
 5                  series of policy-relevant questions (in Chapter_3) that provide the framework for this
 6                  assessment, and which also frames the entire review of the NAAQS for Pb, and thus are
 7                  informed by both science and policy considerations. The plans and underlying questions
 8                  for the ISA are included in the IRP. The ISA organizes, presents, and integrates the
 9                  scientific evidence, which is considered along with findings from any risk analyses and
10                  policy considerations, to help the U.S. Environmental Protection Agency (EPA) address
11                  these questions during the NAAQS review for Pb. The ISA includes:

12                     • An integration of the evidence on the human health effects associated with Pb
13                       exposure, discussion of important uncertainties identified in the interpretation of
14                       the scientific evidence, and an integration across different scientific disciplines
15                       and across individual endpoints within major outcome categories.
16                     • An integration of the evidence on the welfare effects of Pb in terrestrial,
17                       freshwater and saltwater ecosystems, discussion of endpoints common to plants,
18                       invertebrates and vertebrates and consideration of uncertainties in relating
19                       atmospheric Pb concentrations to welfare effects.
20                     • An integration of the effects associated with exposure to Pb across the scientific
21                       disciplines for health and ecology, focusing on common modes of action.
22                     • Discussion of policy relevant considerations, such as potentially at-risk
23                       populations and concentration-response relationships.
24                  EPA has a systematic process for evaluating the scientific evidence and for drawing
25                  conclusions and judgments regarding the causal association of air pollution with health
26                  and environmental effects. The ISA process includes literature search strategies, criteria
27                  for selecting and evaluating studies, approaches for evaluating weight of the evidence,
28                  and a framework for making causality determinations. As part of this process, the ISA is
29                  reviewed by the public and peer reviewed by a formal panel of scientific experts (the
30                  Clean Air Scientific Adviory Committee [CASAC]). The process and causality
      November 2012                             2-1                   Draft - Do Not Cite or Quote

-------
 1                   framework are described in more detail in the Preamble to the ISA. This section provides
 2                   a brief overview of the process for development of this ISA.

 3                   EPA initiated the current review of the NAAQS in February 2010 with a call for
 4                   information from the public (75 FR 8934). Literature searches were conducted routinely
 5                   to identify studies published since the last review, focusing on studies published from
 6                   2006 (close of previous scientific assessment) through September 2011. References that
 7                   were considered for inclusion or cited in this ISA can be found at
 8                   http://hero.epa.gov/lead.

 9                   This ISA evaluates relevant epidemiologic, animal toxicological, and welfare effects
10                   studies, including those related to concentration-response relationships, mode(s) of action
11                   (MOA), and susceptible populations. Additionally, air quality and emissions data, studies
12                   on environmental fate and transport, and issues related to Pb toxicokinetics and exposure
13                   were considered for inclusion in the document. Previous AQCDs (U.S. EPA. 2006b.
14                   1986b. 1977) have included an extensive body of evidence on these topics. In this ISA,
15                   the conclusions and key findings from previous reviews are summarized at the beginning
16                   of each section, to provide the foundation for consideration of evidence from recent
17                   studies. Results of key studies from previous reviews are  included in discussions or tables
18                   and figures, as appropriate, and conclusions are drawn based on the synthesis of evidence
19                   from recent studies with the extensive literature summarized in previous reviews.

20                   The Preamble discusses the general framework for conducting the science assessment
21                   and developing an ISA, including  criteria for selecting studies for inclusion in the ISA
22                   evaluating and integrating the scientific evidence and developing scientific conclusions.
23                   In selecting the studies for inclusion in the Pb ISA, particular emphasis is placed on those
24                   studies most relevant to the review of the NAAQS.

25                   In drawing judgments regarding causality for the criteria air pollutants, evidence of health
26                   effects in the range of relevant pollutant exposures or doses is considered. With regard to
27                   the causal determinations drawn for human health  effects of Pb, population-based
28                   epidemiology studies were emphasized over occupational studies. Recent occupational
29                   studies were considered insofar as they addressed a topic area that was of particular
30                   relevance to the NAAQS review (e.g., longitudinal studies designed to examine recent
31                   versus historical Pb exposure). Evidence from toxicological studies of effects observed in
32                   experimental animals at doses that were relevant to, or somewhat above, those currently
33                   experienced by the U.S. general population were emphasized. Generally studies with
34                   blood Pb levels within one order of magnitude above the upper end of the distribution of
      November 2012                              2-2                   Draft - Do Not Cite or Quote

-------
 1                   U.S. blood Pb levels were considered in forming conclusions1 with the majority of studies
 2                   reporting blood Pb levels below 25-30 (ig/dL. Studies with higher blood Pb levels were
 3                   considered if they informed the evaluation of modes of action, mechanisms, or kinetics.
 4                   For toxicological studies where blood Pb levels were not measured, judgments regarding
 5                   how to distinguish high from the more relevant low doses were made considering the
 6                   range of doses across the available body of evidence and emphasizing studies at the lower
 7                   end of the range.

 8                   Relevant concentrations for drawing causality judgments for the welfare effects of Pb
 9                   were determined considering the range of Pb concentrations in the environment and the
10                   available evidence for concentrations at which effects were observed in plants,
11                   invertebrates, and vertebrates. Effects observed at or near ambient Pb concentrations
12                   measured in soil, sediment and water in the most recent available studies (Table 2-1)
13                   were emphasized and studies generally within one to two orders of magnitude above the
14                   reported range of these values were considered in the body of evidence for terrestrial,
15                   freshwater and saltwater ecosystems. Studies at higher concentrations were used to the
16                   extent that they informed modes of action and illustrated the wide range of sensitivity to
17                   Pb across taxa.

18                   The causal determinations for terrestrial, freshwater and saltwater effects are divided into
19                   two categories. The first category includes endpoints that are commonly used in
20                   ecological risk assessment (reproduction, growth, and survival). It is clear that these
21                   endpoints can lead to population-level (e.g., abundance, production, extirpation),
22                   community-level (e.g., taxa richness, relative abundance) and ecosystem-level effects
23                   (Ankley et al.. 2010; Suter et al.. 2005). The second category includes organism- and sub-
24                   organism-level responses such as physiological stress, hematological effects, and
25                   neurobehavioral effects. As recognized in EPA's Framework for Ecological Risk
26                   Assessment (U.S.  EPA. 1992). and in the adverse outcome pathway (AOP) framework
27                   (Anklev et al.. 2010) endpoints that are measured at one level of biological organization
28                   may be related to an endpoint at a higher level. The AOP conceptual framework was
29                   proposed to link mechanistic data from initiating events at the molecular level through a
30                   series of higher order biological responses to growth, survival and reproductive endpoints
31                   that can be used in ecological risk assessment, i.e., at the population level and higher. In
32                   the case of Pb, sub-organismal responses (i.e., physiological stress, hematological effects)
33                   and organism-level responses (neurobehavioral alterations) may decrease the overall
34                   fitness of an organism, even though their connection to effects at higher levels of
35                   biological organization may not have been characterized. Furthermore, the effects of Pb
      1 For example, the 97.5th percentile of the NHANES distribution of blood Pb level in children 1-5 years old is
      5 ng/dL; however, the proportion of individuals with blood Pb levels that exceed this concentration varies
      depending on factors including age and sex (Section 4.1).
      November 2012                              2-3                   Draft - Do Not Cite or Quote

-------
 1                   on ecosystems necessarily begin with some initial effects at the molecular level of
 2                   specific organisms within the ecosystem (U.S. EPA. 1986b). There are many different
 3                   molecular and cellular level effects, and toxicity of Pb in ecosystems may be attained
 4                   through multiple modes of action.

 5                   The ISA considers evidence of health effects for both short- and long-term pollutant
 6                   exposures. Since biomarkers are typically used as an index of exposure or dose in
 7                   epidemiologic studies, there is uncertainty regarding the timing, frequency, level, and
 8                   duration of the exposure(s) associated with the blood Pb (or other biomarker) levels
 9                   measured in these studies. Some animal toxicological studies provide evidence to inform
10                   the exposure patterns that are needed to induce effects in animals and these studies  are
11                   drawn upon to interpret the human health effects  evidence. Exposure regimens used in
12                   toxicological studies typically include chronic exposure (i.e.,  over 10% of the lifespan of
13                   the animal), long-term exposure (e.g., greater than 4 weeks in rodents) and acute or
14                   short-term exposure (e.g., less than 4 weeks in rodents). For the purpose of this
15                   assessment, short-term human exposures are generally defined to include exposures of
16                   months (e.g., 
-------
 1                     • What exposure conditions (dose or exposure, exposure pathways, duration and
 2                      pattern) are important?
 3                     • What populations and lifestages appear to be differentially affected i.e., at
 4                      greater risk of Pb-related health effects?
 5                     • What elements of the ecosystem (e.g., types, regions, taxonomic groups,
 6                      populations, functions, etc.) appear to be affected or are more sensitive to
 7                      effects?
 8                  This ISA is composed of seven chapters including the Executive (ChapterJJ and
 9                  Integrative Summaries (Chapter_2). Chapter_3 highlights key concepts or issues relevant
10                  to understanding the sources, ambient concentrations, and fate and transport of Pb in the
11                  environment. Chapter_4 summarizes key concepts and recent findings on Pb exposures,
12                  toxicokinetics, and biomarkers reflecting Pb exposure and body burden. Chapter_5
13                  presents a discussion of the MOA of Pb and evaluates and integrates epidemiologic and
14                  animal toxicological information on health effects related to Pb exposure. Chapter_6
15                  summarizes the evidence on potentially susceptible populations. ChapterJ? evaluates
16                  welfare effects evidence that is relevant to the review of the secondary NAAQS for Pb.

17                  This chapter summarizes and integrates the newly available scientific evidence that best
18                  informs consideration of the policy-relevant questions that frame this assessment. The
19                  organization of this chapter generally follows the organization of the document as a
20                  whole, with several additional sections including: a discussion of the assessment
21                  development and scope (Section 2.1); an integration of the evidence across the disciplines
22                  of health and ecology (Section 2.8); a discussion of policy-relevant considerations
23                  (Section 2.9); and, an overall summary (Section 2.10).
      2.2       Ambient Pb: Source to Concentration
      2.2.1       Sources, Fate and Transport of Ambient Pb

24                  The findings of this review build upon those from the 2006 Pb AQCD (U.S. EPA.
25                  2006b), which documented the decline in ambient air Pb emissions following the phase
26                  out on alkyl-Pb additives for on-road gasoline and reductions in industrial facility
27                  emissions of Pb. Pb emissions declined by 98% from 1970 to 1995 and then by an
28                  additional 76% from 1995 to 2008. The 2008 National Emissions Inventory (NEI)
29                  reported ambient air Pb emissions of 964 tons. Air Pb emissions represent just  a small
30                  fraction (by weight) of the Pb used in U.S. production.
      November 2012                             2-5                  Draft - Do Not Cite or Quote

-------
 1                  According to the 2008 NEI (U.S. EPA. 201 la), piston-engine aircraft emissions comprise
 2                  the largest share (57%) of total atmospheric Pb emissions. Other sources of ambient air
 3                  Pb, beginning with the largest, include metal working and mining, fossil fuel combustion,
 4                  other industrial sources, and roadway related sources. Although piston-engine aircraft
 5                  collectively comprise the  largest emissions source, the highest emitting individual
 6                  industrial sites produce more ambient air Pb emissions than individual airports.

 7                  Global atmospheric Pb deposition peaked in the 1970s, followed by a decline
 8                  (Section 3.2). Pb deposition is greater near Pb emission sources. Both wet and dry
 9                  deposition are important mechanisms for removing Pb from the atmosphere, and the
10                  atmosphere is the main environmental transport pathway for Pb which is deposited onto
11                  surface water and soil. Wet deposition is more important for the fine fraction while the
12                  coarse fraction is usually  removed by dry deposition. Pb associated with coarse PM
13                  deposits to a great extent near local industrial sources, contributing to soil Pb
14                  concentrations in those locations, while fine Pb-bearing PM can be transported long
15                  distances, contributing Pb contamination in remote areas. Depending on local conditions,
16                  once deposited particles may be resuspended and redeposited before reaching a site
17                  where further transport is unlikely, especially for dry deposition (Section 3.3). Surface
18                  waters act as an important reservoir, with Pb lifetimes in the water column largely
19                  controlled by deposition and resuspension of Pb in sediments.  Substantial amounts of Pb
20                  from vehicle wear and building materials can also be transported by runoff waters to
21                  surface waters and  sediments without becoming airborne. Pb containing sediment
22                  particles can be remobilized into the water column, and sediment concentrations tend to
23                  follow those in overlying  waters (Section 3.3).
      2.2.2        Monitoring and Concentrations of Ambient Air Pb

24                  The indicator for the Pb NAAQS is Pb in total suspended particles (Pb-TSP). The Federal
25                  Reference Method (FRM) for Pb-TSP specifies that ambient air is drawn through a
26                  high-volume TSP sampler onto a glass fiber filter. The Pb-TSP sampler's size selective
27                  performance is known to be affected by wind speed and direction, and collection
28                  efficiency has been demonstrated to decline with increasing particle size with an
29                  uncertain upper size limit (Wedding etal.. 1977). There have been only a few studies
30                  since the publication of the 2006 Pb AQCD with regard to sampling error in the Pb-TSP
31                  FRM or alternatives to the existing Pb-TSP sampling technology. In addition to monitors
32                  used historically for sampling Pb-PM, several single stage and multi-stage impactors and
33                  inlets used for sampling PM concentrations are also potential options for Pb-PM
34                  monitoring when the majority of particles are smaller than 15 pirn. Given that most sites
35                  with collocated Pb-PMi0 and Pb-TSP monitors have average Pb-PMi0:Pb-TSP ratios


      November 2012                             2-6                    Draft - Do Not Cite  or Quote

-------
 1                  greater than 0.75 (Section 3.5.3). the existing samplers reasonably capture the airborne
 2                  fraction of ambient Pb that is available for human exposure (Section 3.4.1).

 3                  Ambient air Pb monitoring requirements have undergone several changes since
 4                  publication of the 2006 Pb AQCD. The current Pb monitoring network design
 5                  requirements include two types of FRM monitoring sites: source-oriented and
 6                  non-source-oriented (Section 3.4). For the purpose of analyzing data for the ISA,
 7                  monitors reporting to the U.S. EPA Air Quality System (AQS) database were considered
 8                  to be source-oriented if they were designated in AQS as source-oriented, or if they were
 9                  located within 1 mile of a 0.5 ton per year or greater source, as noted in the 2005 NEI
10                  (U.S. EPA. 2008a) or the 2008 NEI (U.S. EPA. 201 la). Source-oriented FRM Pb-TSP
11                  monitoring sites are required near sources of air Pb emissions which are expected to or
12                  have been shown to contribute to ambient air Pb concentrations in excess of the NAAQS.
13                  Non-source-oriented FRM (Pb-TSP or Pb-PMi0) monitoring is also required at national
14                  core multipollutant monitoring network (NCore) sites in Core Based Statistical Areas
15                  (CBSA) with a population of at least 500,000. In addition to FRM monitoring,  Pb is also
16                  routinely measured in smaller particle fractions in the chemical speciation network
17                  (CSN), interagency monitoring of protected visual environment (IMPROVE), and the
18                  national air toxics trends station (NATTS) networks. While monitoring in multiple
19                  networks provides extensive geographic coverage, measurements between networks are
20                  not directly comparable in all cases because there are differences in the methods,
21                  including the different particle size ranges sampled in the different networks. Depending
22                  on monitoring network, Pb is monitored in TSP, PMi0, or PM2 5.

23                  Ambient air Pb concentrations have declined drastically over the period 1980-2010
24                  (Section 3.5). The median annual concentrations have dropped by 97% from 0.87 ug/m3
25                  in 1980 to 0.03 ug/m3 in  2010. The mean  source-oriented Pb maximum 3-month average
26                  concentration was skewed toward the 75th percentile of the data distribution and
27                  exceeded the level of the NAAQS, suggesting that ambient air Pb concentrations are high
28                  near a subset of industrial sources of airborne Pb. Studies in the peer-reviewed literature
29                  have shown slightly elevated Pb concentrations downwind of industrial sources and
30                  airports. Estimates for the natural background Pb concentrations from sources including
31                  volcanoes, sea-salt spray, and biogenic sources are -0.00002 to 0.001  ug/m3. These
32                  estimates indicate that background airborne Pb concentrations are well below current
33                  ambient concentrations.

34                  The size distribution of Pb-bearing PM has changed over time and varies by site
35                  (Section 3.5.3). Recent study results indicate that the size distribution has shifted upward
36                  since the 1980s, with the mode of the size distribution of Pb-PM particles now falling
37                  between 2.5 um and 10 um (Cho et al.. 2011). The Pb-PM size distribution depends on
      November 2012                             2-7                   Draft - Do Not Cite or Quote

-------
 1                  whether there are contributions from industrial sources or near-road environments. In
 2                  contrast to Cho et al. (2011). analysis of the distributional properties of the Pb-PM
 3                  measured by the AQS monitors, which are often sited near sources, suggests that the
 4                  largest proportion of particles is still below 2.5 pirn in diameter.
      2.2.3        Ambient Pb Concentrations in Non-Air Media and Biota

 5                   Releases of Pb to the atmosphere have led to measurable Pb concentrations observed in
 6                   rain, snowpack, soil, surface waters, sediments, agricultural plants, livestock, and wildlife
 7                   across the world, with highest concentrations near Pb sources, such as smelters. After the
 8                   phase-out of Pb from on-road gasoline, Pb concentrations have decreased considerably in
 9                   rain, snowpack, and surface waters.

10                   Declining Pb concentrations in tree foliage, trunk sections, and grasses, as well as surface
11                   sediments and soils, have also been observed (U.S. EPA. 2006b).

12                   Often, Pb is retained in soils and sediments, where it provides a historical record of
13                   deposition. In remote lakes, sediment profiles indicate higher Pb concentrations in near
14                   surface sediment as compared to pre-industrial era sediment from greater depth and
15                   indicate peak concentrations between 1960 and 1980 (when leaded on-road gasoline was
16                   at peak use). Concentrations of Pb in moss, lichens, peat, and aquatic bivalves have been
17                   used to understand spatial and temporal distribution patterns of air Pb concentrations.
18                   Ingestion and water intake are the major routes of Pb exposure for aquatic organisms, and
19                   food, drinking water, and inhalation are major routes of exposure for livestock and
20                   terrestrial wildlife.

21                   Overall, Pb concentrations have decreased substantially in media through which Pb is
22                   rapidly transported, such as air and water. Substantial Pb remains in soil and sediment
23                   sinks. In areas less affected by major local sources, the highest concentrations are below
24                   the surface layers and reflect the phase-out of Pb from on-road gasoline and emission
25                   reductions from other sources.

26                   Information on ambient Pb concentrations in non-air media and biota is reported in
27                   Section 3.6. and concentrations considered in the interpretation of the ecological evidence
28                   are tabulated in Table 2-1. As noted in the preamble, the ecological causal determinations
29                   focus on studies where effects of Pb exposure are observed at or near ambient levels of
30                   Pb and studies generally within the range  of one to two orders of magnitude above
31                   current or ambient conditions were also considered in the body of evidence.
      November 2012                              2-8                   Draft - Do Not Cite or Quote

-------
Table 2-1     Ambient Pb concentrations in non-air media and biota considered for
                 ecological  assessment.
Media
Pb Concentration
Years Data
 Obtained
                                                                         References
Soil
                   National Average:
                   18.9 mg Pb/kg (dry weight)
                   Range of state averages:
                   5-38.6 mg Pb/Kg (dry weight)
                                                        1961-1997
                                                      U.S. EPA (2007d. 2006b. 2003b)
Freshwater
Sediment
Median:
73 mg Pb/kg (dry weight)
                   Median:
                   28 mg Pb/kgb (dry weight)
 1996-2001        Mahler et al. (2006)
                                     1991-2003        U.S. EPA(2006b)
Saltwater Sediment
Fresh Surface
Water
(Dissolved Pb)b
Range:
0.6 to 1 ,050 mg Pb/kga
Median: 0.50 ug Pb/Lb; Max:
30 ug Pb/L, 95th percentile
1.1 ug Pb/L
Dates not available
1991-2003
Sadiq
(1992)
U.S. EPA (2006b)
                   Range: 0.0003-0.075 ug Pb/L
                   (Set of National Parks in western
                   U.S.)
                                     2002-2007        Field and Sherrell (2003).
                                                      U.S. National Park Service (2011)
Saltwater0
                   Range: 0.01-27 ug Pb/L
                                  Dates not available    Sadiq (1992)
Vegetation
Lichens: 0.3-5 mg Pb/kg (dry
weight) (Set of National Parks in
western U.S.)
                                                        2002-2007
                 U.S. National Park Service (2011)
                   Grasses: 31% (percent of soil Pb in
                   grass)
                                                       1980s-2000s
                                                     Vandenhove et al. (2009)
Vertebrates
                   Fish:
                   Geometric Mean:
                   0.59 mg Pb/kg
                   (dry weight) (whole fish)
                   0.15 mg Pb/kg
                   (dry weight) (liver)
                   Range:
                   0.08-22.6 mg Pb/kg
                   (dry weight) (whole fish)
                   0.01-12.7 mg Pb/kg
                   (dry weight) (liver)
                                                        1991-2001
                   Fish (from a set of national parks
                   in western U.S.):
                   0.0033 (fillet) to 0.97 (liver)
                   mg Pb/kg (dry weight)
                   Moosed:
                   0.008-0.029 mg Pb/kg
                   (dry weight) (meat)
                   0.012-0.023 mg Pb/kg
                   (dry weight) (liver)
                                                        2002-2007
                                                      U.S. National Park Service (2011)
aNo information available regarding wet or dry weight
"Based on synthesis of National Water-Quality Assessement (NAWQA) data reported in 2006 Pb AQCD (U.S. EPA. 2006b)
°Data from a combination of brackish and marine saltwater samples. In general, Pb in seawater is higher in coastal areas and
estuaries since these locations are closer to sources of Pb contamination and loading from terrestrial systems.
dThree moose in one Alaskan park
November 2012
                              2-9
                  Draft - Do Not Cite or Quote

-------
      2.3        Exposure to Ambient Pb

 1                  Human Pb exposure is difficult to assess because Pb has multiple sources in the
 2                  environment and passes through various media (Section 4.1). Air-related pathways of Pb
 3                  exposure are the focus of this assessment. In addition to inhalation of Pb in ambient air,
 4                  air-related Pb exposure pathways include inhalation and ingestion of Pb in indoor dust
 5                  and/or outdoor soil that originated from recent or historic ambient air (e.g., air Pb that has
 6                  penetrated into the residence either via the air or tracking of soil), ingestion of Pb in
 7                  drinking water drawn from surface water contaminated from atmospheric deposition or
 8                  contaminated from surface runoff of deposited Pb, and ingestion of Pb in dietary sources
 9                  after uptake by plants or grazing animals. Soil can act as a reservoir for deposited Pb
10                  emissions. Exposure to soil contaminated with deposited Pb can occur through
11                  resuspended PM as well as shoe tracking and hand-to-mouth contact, which is the main
12                  pathway of childhood air-related exposure to Pb. Non-ambient air-related exposures
13                  include hand-to-mouth contact with dust or chips of peeling Pb-containing paint, or
14                  ingestion of Pb in drinking water conveyed through Pb pipes. Several study results
15                  indicate that Pb-containing paint in the home and home age (often a surrogate for the
16                  presence of Pb paint) are important residential factors that increase risk of elevated blood
17                  Pb (Sections 2.9.6 and 6.2.6). Most Pb biomarker studies do not indicate species or
18                  isotopic signature. As a consequence, non-air exposures are reviewed in this section,
19                  because they can also contribute to Pb body burden.

20                  A number of monitoring and modeling techniques have been employed for ambient Pb
21                  exposure assessment. Environmental Pb concentration data can be collected from
22                  ambient air Pb monitors, soil Pb samples, dust Pb samples, and dietary Pb samples to
23                  estimate human exposure. Exposure estimation error depends in part on the collection
24                  efficiency of these methods; collection efficiency for ambient air Pb FRM samplers is
25                  described in Section 3.4. Additionally, high spatial variability of the Pb concentrations in
26                  various media also can contribute to exposure error, as described in the 2009 PM ISA
27                  (U.S. EPA. 2009a). Models, such as the Integrated Exposure Uptake Biokinetic (IEUBK)
28                  model, simulate human exposure to Pb from multiple  sources and through various routes
29                  including inhalation and ingestion. IEUBK model inputs include soil-Pb concentration,
30                  air-Pb concentration, dietary-Pb intake including drinking water, Pb-dust ingestion,
31                  human activity, and biokinetic factors. Measurements and/or assumptions can be utilized
32                  when formulating the model inputs; errors in measurements and assumptions thus have
33                  the potential to propagate through exposure models.
      November 2012                             2-10                  Draft - Do Not Cite or Quote

-------
 1                  The size distribution of dust particles containing Pb differs from the size distribution of
 2                  inhalable ambient Pb-bearing PM (Sections 3.5 and 4.1). Airborne particles containing Pb
 3                  tend to be small (much of the distribution <10 (im) compared with soil or dust particles
 4                  containing Pb (-50 (im to several hundred (im). Ingestion through hand-to-mouth contact
 5                  is the predominant exposure pathway for the larger particles in soil and dust containing
 6                  Pb.
      2.4       Toxicokinetics

 7                  The majority of Pb in the body is found in bone (roughly 90% in adults, 70% in children);
 8                  only about 1% of Pb is found in the blood. Pb in blood is primarily (-99%) bound to red
 9                  blood cells (RBCs). It has been suggested that the small fraction of Pb in plasma (<1%)
10                  may be the more biologically labile and lexicologically active fraction of the circulating
11                  Pb. The relationship between Pb in blood and plasma is pseudo-linear at relatively low
12                  daily Pb intakes (i.e., <10 ug/day-kg) and at blood Pb concentrations <25 (ig/dL, and
13                  becomes curvilinear at higher blood Pb concentrations due to saturable binding to RBC
14                  proteins. As blood Pb level increases and the higher affinity binding sites for Pb in RBCs
15                  become saturated, a larger fraction of the blood Pb is available in plasma to distribute to
16                  brain and other Pb-responsive tissues. See Section 4.2 for additional details.

17                  The burden of Pb in the body may be viewed as divided between a dominant slow
18                  (i.e., uptake and elimination) compartment (bone) and smaller fast compartment(s) (soft
19                  tissues). Pb uptake and elimination in soft tissues is much faster than in bone. Pb
20                  accumulates in bone regions undergoing the most active calcification at the time of
21                  exposure. During infancy and childhood, bone calcification is most active in trabecular
22                  bone (e.g.,  patella); whereas, in adulthood, calcification occurs at sites of remodeling in
23                  cortical (e.g., tibia) and trabecular bone (Aufderheide and Wittmers. 1992). A high bone
24                  formation rate in early childhood results in the rapid uptake of circulating Pb into
25                  mineralizing bone; however, in early childhood bone Pb is also recycled to other tissue
26                  compartments or excreted in accordance with a high bone resorption rate (O'Flahertv.
27                  1995). Thus, much of the Pb acquired early in life is not permanently fixed in the bone.

28                  The exchange of Pb from plasma to the bone surface is a relatively rapid process. Pb in
29                  bone becomes distributed in trabecular and the more dense cortical bone. The proportion
30                  of cortical to trabecular bone in the human body varies by age, but on average is about
31                  80% cortical to 20% trabecular. Of the bone types, trabecular bone is more reflective of
32                  recent exposures than is cortical bone due to the slow turnover rate and lower blood
33                  perfusion of cortical bone. Some Pb diffuses to deeper bone regions where it is relatively
34                  inert, particularly in adults. These bone compartments are much more labile in infants
      November 2012                             2-11                   Draft - Do Not Cite or Quote

-------
 1                   and children than in adults as reflected by half-times for movement of Pb from bone into
 2                   the plasma (e.g., cortical half-time = 0.23 years at birth, 3.7 years at 15 years of age, and
 3                   23 years in adults; trabecular half-time = 0.23 years at birth, 2.0 years at 15 years of age,
 4                   and 3.8 years in adults) (Leggett 1993). See Section 4.2 for additional details.

 5                   Evidence for maternal-to-fetal transfer of Pb in humans is derived from cord blood to
 6                   maternal blood Pb ratios (i.e., cord blood Pb concentration divided by mother's blood
 7                   Pb). Group mean ratios range from about 0.7 to 1.0 at the time of delivery for mean
 8                   maternal blood Pb levels ranging from 1.7 to 8.6 (ig/dL. Some evidence suggests the ratio
 9                   of cord blood Pb to maternal blood Pb may decrease with decreasing maternal blood Pb
10                   (Section 4.2.2.4). Transplacental transfer of Pb may be facilitated by an increase in the
11                   plasma/blood Pb concentration ratio during pregnancy. Maternal-to-fetal transfer of Pb
12                   appears to be related partly to the mobilization of Pb from the maternal skeleton. See
13                   Section 4.2 for additional details.

14                   The dominant elimination phase of Pb kinetics in the blood, exhibited shortly after a
15                   change in exposure occurs, has a half-life of-20-30 days. An abrupt change in Pb uptake
16                   gives rise to a relatively rapid change in blood Pb, to a new quasi-steady state, achieved
17                   in -75-100 days (i.e., 3-4 times the blood elimination half-life).  A slower phase of Pb
18                   clearance from the blood may become evident with longer observation periods following
19                   a decrease in exposure due to the gradual redistribution of Pb among bone and other
20                   compartments. See Section 4.3 for additional details.
      2.5       Pb Biomarkers

21                  Overall, trends in blood Pb levels have been decreasing among U.S. children and adults
22                  over the past 20 years (Section 4.4). The median blood Pb level for the entire U.S.
23                  population is 1.1 ug/dL and the 95th percentile blood Pb level is 3.3 ug/dL, based on the
24                  2009-2010 National Health and Nutrition Examination Survey (NHANES) data (NCHS.
25                  2010). Among children aged 1-5 years, the median and 95th percentiles were slightly
26                  higher, at 1.2 ug/dL and 4.0 ug/dL, respectively.

27                  Blood Pb is dependent on both the recent exposure history of the individual, as well as
28                  the long-term exposure history that determines body burden and Pb in bone. The
29                  contribution of bone Pb to blood Pb changes, depending on the duration and intensity of
30                  the exposure, age, and various other physiological stressors (e.g., nutritional status,
31                  pregnancy, menopause, extended bed rest, hyperparathyroidism) that  may affect bone
32                  remodeling, which normally and continuously occurs. In children, largely due to faster
33                  exchange of Pb to and from bone, blood  Pb is both an index of recent exposure and
34                  potentially an index of body burden. In adults and children, where exposure to Pb has

      November 2012                             2-12                   Draft - Do Not  Cite or Quote

-------
 1                   effectively ceased or greatly decreased, there is a rapid decline in blood Pb over the first
 2                   few months followed by a more gradual, slow decline in blood Pb concentrations over the
 3                   period of years due to the gradual release of Pb from bone. Bone Pb is an index of
 4                   cumulative exposure and body burden. Even bone compartments should be recognized as
 5                   reflective of differing exposure periods with Pb in trabecular bone exchanging more
 6                   rapidly than Pb in cortical bone with the blood. Consequently, Pb in cortical bone is a
 7                   better marker of cumulative exposure, while Pb in trabecular bone is more likely to be
 8                   correlated with blood Pb, even in adults. See Section 4.3 for additional details.

 9                   Sampling frequency is an important consideration when evaluating blood Pb and bone Pb
10                   levels in epidemiologic studies, particularly when the exposure is not well characterized.
11                   It is difficult to determine what blood Pb is reflecting in cross-sectional studies that
12                   sample blood Pb once, whether recent exposure or movement of Pb from bone into blood
13                   from  historical exposures. In contrast, cross-sectional studies of bone Pb and longitudinal
14                   samples of blood Pb concentrations overtime provide more of an index of cumulative
15                   exposure and are more reflective of average Pb body burdens overtime. The degree to
16                   which repeated sampling will reflect the actual long-term time-weighted average blood
17                   Pb concentration depends on the sampling frequency in relation to variability in
18                   exposure. High variability in Pb exposures can produce episodic (or periodic) oscillations
19                   in blood Pb concentration that may not be captured with low sampling frequencies.
20                   Furthermore, similar blood Pb concentrations in two individuals (or populations),
21                   regardless of their age, do not necessarily translate to similar body burdens or similar
22                   exposure histories.

23                   The concentration of Pb in urine  follows blood Pb concentration. There is added
24                   complexity with Pb in urine because concentration is also dependent upon urine flow
25                   rate, which requires timed urine samples that is often not feasible in epidemiologic
26                   studies. Other biomarkers have been utilized to a lesser extent (e.g., Pb in teeth). See
27                   Section 4.3.
      2.6       Health  Effects

28                   This section summarizes and evaluates the evidence from toxicological and
29                   epidemiologic studies of the health effects associated with Pb exposure and integrates
30                   that evidence across these disciplines. The coherence of evidence across toxicological
31                   and epidemiologic findings and across a spectrum of related endpoints, including
32                   evidence for mode of action (MOA), is evaluated to establish biological plausibility and
33                   address uncertainties in the epidemiologic evidence due to biases from factors such as
34                   selective publication of positive results, recruitment or participation of subjects; reverse
      November 2012                             2-13                  Draft - Do Not Cite or Quote

-------
1

2

3

4

5

6

7
causality; or confounding. Information on the frequency, timing, level and duration of

exposure in animal toxicological studies is used to inform the interpretation of

epidemiologic studies regarding the relevant patterns of exposure that are likely

associated with the health effects. The results from the health studies are also considered

in combination with the evidence from other fields (e.g., toxicokinetics, exposure

science) to draw conclusions regarding the causal relationship between Pb exposure and

the health  effects evaluated in this assessment (Table 2-2). A more detailed discussion of

the underlying evidence used to formulate each causal determination can be found in

Chapter_5  of this document.
     Table 2-2    Summary of causal determinations between exposure to Pb and
                     health outcomes.
     Health Outcome
                                            Causality Determination3
                                               (Table with Key Supporting Evidence)
     Nervous System Effects (Section 5.3.16)
       Children
        Cognitive Function Decrements (e.g., Full Scale IQ [FSIQ],
        learning)
                                            Causal Relationship (Table 5-17)
        Attention-Related Behavioral Problems (e.g., inattention,
        impulsivity, hyperactivity)
                                            Causal Relationship (Table 5-17)
        Conduct Problems in Children and Young Adults
        (e.g., parent/teacher conduct ratings, criminal offenses)
                                            Likely Causal Relationship (Table 5-17)
         Internalizing Behaviors (e.g., withdrawn behavior, symptoms
         of depression)
                                            Likely Causal Relationship (Table 5-17)
        Sensory Function Decrements (e.g., hearing threshold,
        electroretinography [ERG])
                                            Likely Causal Relationship (Table 5-17)
        Motor Function Decrements
                                                                 Likely Causal Relationship (Table 5-17)
       Adults
        Cognitive Function Decrements (e.g., executive function,
        visuospatial skills, learning and memory)
                                            Likely Causal Relationship (Table 5-17)
         Psychopathological Effects (e.g., symptoms of anxiety
         and depression)
                                            Likely Causal Relationship (Table 5-17)
        Sensory Function Decrements (e.g., hearing threshold)
                                            Suggestive of a Causal Relationship (Table 5-17)
         Neurodegenerative Diseases (e.g., Parkinson's disease,
         Alzheimer's disease)
                                            Inadequate to Infer a Causal Relationship
                                            (Table 5-17)
     November 2012
                              2-14
Draft - Do Not Cite or Quote

-------
Health Outcome
               Causality Determination3
                  (Table with Key Supporting Evidence)
Cardiovascular Effects (Section 5.4.7)
  Hypertension
               Causal Relationship (Table 5-24)
  Subclinical Atherosclerosis (e.g., peripheral artery disease
  [PAD],intimal media thickness [IMT], atherosclerotic plaque
  presentation)
               Suggestive of a Causal Relationship (Table 5-24)
  Coronary Heart Disease (e.g., Heart Rate Variability [HRV],
  Myocardial Infarction [Ml], Ischemic Heart Disease [IHD],
  cardiovascular mortality; and, in animals, increased thrombosis,
  coagulation, and arrhythmia)
               Causal Relationship (Table 5-24)
  Cerebrovascular Disease (e.g., stroke, transient ischemic attack, and
  subarachnoid hemorrhage)
               Inadequate to Infer a Causal Relationship
               (Table 5-24)
Renal Effects (Section 5.5.5)
  Reduced Kidney Function (e.g., reduced Glomerular Filtration Rate
  [GFR], reduced creatinine clearance, and increased serum creatinine)
               Likely Causal Relationship (Table 5-31)
Immune System Effects (Section 5.6.8)
  Atopic and Inflammatory Responses (e.g., asthma, allergy,
  Immunoglobulin E [IgE], increased Th2 cytokines)
               Likely Causal Relationship (Table 5-34)
  Decreased Host Resistance (e.g., antigen responses, effects on
  macrophages, neutrophil and infection)
               Likely Causal Relationship (Table 5-34)
  Autoimmunity (e.g., auto-antibodies)
               Inadequate to Infer a Causal Relationship
               (Table 5-34)
Hematologic Effects (Section 5.7.4)
  Decreased Red Blood Cell (RBC) Survival and Function
  (e.g., Hemoglobin [Hb], Hematocrit [Hct], and mean corpuscular
  volume [MCV], and measures of oxidative stress)
               Causal Relationship (Table 5-35)
  Altered Heme Synthesis
               Causal Relationship (Table 5-35)
Reproductive and Developmental Effects (Section 5.8.6)
  Development (e.g., delayed onset of puberty, postnatal growth)
               Causal Relationship (Table 5-48)
  Birth Outcomes (e.g., low birth weight, spontaneous abortion)
               Suggestive of Causal Relationship (Table 5-48)
  Male Reproductive Function (e.g., sperm parameters)
               Causal Relationship (Table 5-48)
  Female Reproductive Function (e.g., hormones)
               Suggestive of Causal Relationship (Table 5-48)
Cancer (Section 5.10.5)
  Cancer
                                                                 Likely Causal Relationship (Table 5-50)
aCausal determinations were made within approximately 1 order of magnitude of current levels (Preamble, Section 2.1).
November 2012
2-15
Draft - Do Not Cite or Quote

-------
      2.6.1        Nervous System Effects

 1                  The collective body of epidemiologic and toxicological evidence integrated across that
 2                  reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b) coupled with recently available data
 3                  demonstrates the effects of Pb exposure on a range of nervous system effects. In children,
 4                  these effects include cognitive function, attention-related behavioral problems, conduct
 5                  problems, internalizing behaviors, sensory function, and motor function. In adults,
 6                  nervous system effects examined in relation to Pb exposure include cognitive function,
 7                  psychopathological effects, sensory function, and neurodegenerative diseases.
                     2.6.1.1      Children

                     Cognitive Function Decrements

 8                   Multiple prospective studies conducted in diverse populations consistently demonstrate
 9                   associations of higher blood and tooth Pb levels with lower FSIQ, lower academic
10                   performance, and lower performance on tests learning, memory, and executive function.
11                   These associations remained after adjustment for a range of potential confounding
12                   factors, but most commonly, parental IQ, parental education level, and parental
13                   caregiving quality. In school-aged children, associations were found with early
14                   childhood, childhood average, and concurrent blood Pb levels and substantiated in
15                   populations of children ages 4 to 11 years with mean blood Pb levels between 2 and
16                   8 (ig/dL. Uncertainty remains regarding the lifestage of exposure within childhood that is
17                   associated with the greatest risk of cognitive function decrements. Observation of a
18                   supralinear concentration-response relationship and associations with mean (or quantile)
19                   blood Pb levels <4 (ig/dL do not provide evidence for a threshold for the cognitive effects
20                   of Pb exposure in children. Evidence in children was clearly supported by observations of
21                   Pb-induced impairments in learning, memory, and executive function in juvenile animals.
22                   Several studies in animals indicated learning impairments with prenatal and early
23                   postnatal Pb exposures that resulted in blood Pb levels of 8-26 (ig/dL. The mode of action
24                   for Pb-associated cognitive impairments  is supported by observations of Pb-induced
25                   impairments in neurogenesis, synaptogenesis and synaptic pruning, long term
26                   potentiation, and neurotransmitter function. The associations consistently found in
27                   prospective studies of children with adjustment for Social Economic Status (SES),
28                   parental education and caregiving quality for associations with various indicators of
29                   cognitive function and the biological plausibility provided by evidence in animals for
30                   impairments in learning, memory, and executive function with relevant Pb exposures and
      November 2012                             2-16                  Draft - Do Not Cite or Quote

-------
 1                   evidence describing modes of action is sufficient to conclude that there is a causal
 2                   relationship between Pb exposure and decrements in cognitive function in children.


                     Attention-Related Behavior Problems

 3                   Several prospective studies demonstrate associations of earlier childhood and lifetime
 4                   average blood Pb levels or tooth Pb levels with inattention, impulsivity, and hyperactivity
 5                   in children 7-17 years and young adults ages 19-24 years; as assessed using both
 6                   objective neuropsychological tests and also parent  and teacher ratings of behavior. Most
 7                   studies adjusted for potential confounding by SES  and parental caregiving quality, with a
 8                   few studies also considering substance abuse and nutritional status. Blood Pb-associated
 9                   increases in attention-related behavioral problems were found in populations with
10                   prenatal cord or lifetime average mean blood Pb levels of 6.8 and 14 (ig/dL and groups
11                   with early childhood blood Pb levels >10 (ig/dL. Biological plausibility for observations
12                   in children is provided by consistent findings in animals for increases in impulsivity or
13                   impaired response inhibition with relevant prenatal and early postnatal Pb exposures that
14                   resulted in blood Pb levels of 10 to 31 (ig/dL. The mode of action for Pb-associated
15                   attention-related behavioral problems is supported  by observations of Pb-induced
16                   impairments in neurogenesis, synaptic pruning, and dopamine transmission in specific
17                   regions of the brain (specifically the prefrontal cerebral cortex, cerebellum, and
18                   hippocampus). The consistency of epidemiologic evidence, particularly from prospective
19                   studies, and the biological plausibility provided by evidence for Pb-induced impulsivity
20                   in animals and for underlying modes of action is sufficient to conclude that there is a
21                   causal relationship between Pb exposure and attention-related behavioral problems in
22                   children.


                     Internalizing Behaviors

23                   Prospective studies in a few populations demonstrate associations of higher lifetime
24                   average blood or childhood tooth Pb levels with higher parent and teacher ratings of
25                   internalizing behaviors such as depression, anxiety, and withdrawn behavior in school-
26                   aged children. The lack of selective participation by blood Pb level and associations
27                   found with parental and teacher ratings do not provide strong indication of biased
28                   reporting of behaviors for children with higher blood Pb levels. While results were
29                   adjusted for maternal education and SES-related variables, consideration for potential
30                   confounding by parental caregiving quality was inconsistent. The biological plausibility
31                   for the  effects of Pb on internalizing behaviors is provided by consistent findings  in
32                   animals, with some evidence at blood Pb levels relevant to humans and by evidence
33                   supporting modes of action, including Pb-induced changes in the Hypothalmic-Pituitary-
      November 2012                              2-17                   Draft - Do Not Cite or Quote

-------
 1                   Adrenal (HPA) axis and dopaminergic and GABAergic systems .HPA axis and
 2                   dopaminergic and GABAergic systems. The evidence from prospective studies in a few
 3                   populations and the supporting toxicological evidence is sufficient to conclude that a
 4                   causal relationship is likely to exist between Pb exposure and internalizing behaviors.


                     Conduct Problems in Children and Young Adults

 5                   Prospective studies consistently indicate that early childhood or lifetime average blood
 6                   Pb levels or tooth Pb levels are associated with criminal offenses in young adults ages
 7                   19-24 years and with higher parent and teacher ratings of misconduct in children ages
 8                   7-17 years. These associations were found without any indication of strong selection bias
 9                   and with adjustment for SES, parental education and IQ, parental  caregiving quality,
10                   family functioning, smoking, and substance abuse. Supporting evidence is provided by
11                   cross-sectional evidence of children participating in NHANES and a meta-analysis of
12                   prospective and cross-sectional studies. Evidence for Pb-induced aggression in animals is
13                   mixed. The consistent epidemiologic evidence from prospective and cross-sectional
14                   studies for criminal offenses and ratings of misconduct but lack of clear evidence for
15                   aggression in animals is sufficient to conclude that a causal relationship is likely to exist
16                   between Pb exposure and misconduct in children and young adults.


                     Sensory Function Decrements

17                   Evidence from prospective and cross-sectional studies in a few populations indicates
18                   associations of higher blood Pb levels with increases in hearing thresholds and decreases
19                   in auditory evoked potentials after adjustment for several potential confounding factors,
20                   including SES, parental caregiving quality, and child health. The high participation rates,
21                   particularly in a prospective study with follow-up from birth, reduce the likelihood of
22                   biased participation of children with higher blood Pb levels. Findings in animals indicate
23                   that dietary Pb exposure increases hearing thresholds and latencies for auditory evoked
24                   potentials; however, studies examined higher  Pb exposures than those relevant to
25                   humans. The evidence in children, particularly from prospective studies, with
26                   consideration for potential confounding by factors such as SES, parental caregiving
27                   quality, child health, and nutritional factors, is sufficient to conclude that a causal
28                   relationship is likely to exist between Pb exposure and decrements in sensory function in
29                   children.
      November 2012                             2-18                  Draft - Do Not Cite or Quote

-------
                     Motor Function Decrements

 1                   Evidence from prospective and cross-sectional studies in a few populations indicates
 2                   associations of higher blood Pb level with decrements in fine and gross motor function
 3                   with adjustment for several potential confounding factors, including SES, parental
 4                   caregiving quality, and child health. The high participation rates, particularly in
 5                   prospective studies with follow-up from birth or early infancy, reduce the likelihood of
 6                   biased participation of children with higher blood Pb levels. The biological plausibility
 7                   for associations observed in children is provided by observations of poorer balance in
 8                   male mice with relevant Pb exposures. The evidence in children, particularly from
 9                   prospective studies, and the coherence with findings in mice is sufficient to conclude that
10                   a causal relationship is likely to exist between Pb exposure and decrements in motor
11                   function.
                     2.6.1.2      Adults

                     Cognitive Function

12                   In adults without occupational exposure, prospective studies in the Normative Aging
13                   Study (NAS) and Baltimore Memory Study (BMS) cohorts indicate associations of
14                   higher tibia Pb levels with declines in cognitive function over 2- to 4-year periods. While
15                   the specific factors differed between studies, these tibia Pb-associated cognitive function
16                   decrements were found with adjustment for potential confounding factors such as age,
17                   education, SES, alcohol use, and smoking. Supporting evidence is provided by cross-
18                   sectional studies finding stronger associations with bone Pb level than concurrent blood
19                   Pb level, with more extensive consideration for potential confounding factors, including
20                   dietary factors, physical activity, medication use, and comorbid conditions. The specific
21                   timing, frequency, duration, and magnitude of Pb exposures contributing to the
22                   associations observed with bone Pb levels are uncertain. Also uncertain is the residual
23                   confounding by age. Associations between blood Pb levels and cognitive function
24                   decrements found in adults with occupational Pb exposures provide evidence for the
25                   effects of current Pb exposures, although these studies did not consider potential
26                   confounding by other workplace exposures. The biological plausibility for the effects of
27                   Pb exposure on cognitive  function decrements in adults is provided by findings that
28                   lifetime or postnatal Pb exposure induces learning impairments in adult animals and by
29                   evidence supporting modes of action. The associations between bone Pb level and
30                   cognitive function decrements consistently found in the few prospective and cross-
31                   sectional studies of adults without occupational Pb exposure, the coherence with animal
32                   findings, and evidence supporting mode of action are sufficient to conclude that a causal


      November 2012                              2-19                   Draft - Do Not Cite or Quote

-------
 1                  relationship is likely to exist between long-term cumulative Pb exposure and cognitive
 2                  function decrements in adults.
                     Psychopathological Effects

 3                   Cross-sectional studies in a few populations demonstrate associations of higher
 4                   concurrent blood or tibia Pb levels with self-reports of depression and anxiety symptoms
 5                   in adults. The examination of multiple exposures and outcomes in the available studies
 6                   does not provide strong indication of biased reporting of effects by adults with higher Pb
 7                   exposures. In adults, Pb-associated increases in self-reported depression and anxiety were
 8                   found with adjustment for age, SES, and in the NAS, daily alcohol intake. The biological
 9                   plausibility for the effects of Pb on depression is provided by observations of depression-
10                   like behavior in animals, with some evidence at blood Pb levels relevant to humans and
11                   by evidence supporting modes of action, including Pb-induced changes in the HPA axis
12                   and dopaminergic and GABAergic systems. Although the frequency, timing, level and
13                   duration of exposure contributing to the observed associations is uncertain, the evidence
14                   from  a few populations in adults and the supporting toxicological evidence are sufficient
15                   to conclude that a causal relationship is likely to exist between Pb exposure and
16                   psychopathological effects in adults.


                     Sensory Function Decrements

17                   Evidence was provided by the prospective analysis of the Normative Aging Study of the
18                   U.S. Department of Veteran's Affairs (NAS) men, for associations of higher tibia Pb
19                   level  with a greater rate of elevations in hearing threshold.  Biological plausibility was
20                   provided by the evidence for Pb-induced decreases in auditory evoked potentials in
21                   animals, albeit at higher blood Pb levels than those relevant to humans. Taken together,
22                   the evidence  is suggestive of a causal relationship between Pb exposure and sensory
23                   function decrements in adults.
                     Neurodegenerative Disease
24                   While evidence is inconclusive for Amyotrophic Lateral Sclerosis (ALS)and Alzheimer's
25                   disease, a few case-control studies each found higher blood Pb levels in adults with
26                   essential tremor and higher bone Pb levels in adults with Parkinson's disease. Because of
27                   the inconclusive evidence for some diseases and limitations such as reverse causality for
28                   essential tremor and the lack of consideration for potential confounding by manganese
29                   (Mn) exposure for both essential tremor and Parkinson's disease, the evidence is
      November 2012                             2-20                  Draft - Do Not Cite or Quote

-------
 1                   inadequate to determine that there is a causal relationship between Pb exposure and
 2                   neurodegenerative diseases.
      2.6.2        Cardiovascular Effects

 3                   The 2006 Pb AQCD (U.S. EPA. 2006b) concluded that there was a relationship between
 4                   higher blood Pb and bone Pb and cardiovascular effects in adults, in particular increased
 5                   BP and increased incidence of hypertension , and recent evidence strengthens this
 6                   conclusion. For the evaluation of causal relationships with Pb exposure, evidence was
 7                   grouped in categories using the  U.S. Surgeon General's Report on Smoking as a
 8                   guideline (CDC. 2004). The categories include hypertension (Section 5.4.7.1), subclinical
 9                   atherosclerosis (Section 5.4.7.2). coronary heart disease (Section 5.4.7.3). and
10                   cerebrovascular disease (Section 5.4.7.4).


                     Hypertension

11                   Evidence from epidemiologic and toxicological studies demonstrates consistent effects of
12                   Pb exposure on hypertension. Longitudinal prospective studies clearly support the
13                   association of biomarkers of Pb exposure with hypertension incidence and increased BP.
14                   These high quality studies provide evidence that is replicated across a large number of
15                   studies with different designs, populations, and locations (Section 5.4.2 and
16                   Section 5.4.7.1). Meta-analyses of Pb-associated increases in BP and hypertension
17                   indicate consistency in the strength of associations across studies (Navas-Acien et al.
18                   2008; Nawrot et al.. 2002) found that each doubling of concurrent blood Pb level
19                   (between 1 and >40 ug/dL) was associated with a 1 mmHg increase in systolic BP and a
20                   0.6 mmHg increase in diastolic  BP. Navas-Acien et al. (2008) found that all studies
21                   included in the meta-analysis showed a relationship between higher bone Pb levels,
22                   which is a marker of cumulative exposure, and higher BP.

23                   Further support for a causal relationship between blood and bone Pb levels and increased
24                   BP and hypertension is provided by a multiple cross-sectional analyses [e.g., (Martin et
25                   al.. 2006; Muntner et al..  2005)1. Despite the extensive evidence for associations at
26                   relatively low concurrent blood Pb levels, these cardiovascular outcomes were most often
27                   examined in adults that were exposed to higher levels of Pb earlier in life, and uncertainty
28                   remains concerning the Pb exposure level, timing, frequency, and duration contributing to
29                   the observed effects; however, the majority of animal toxicological studies provide
30                   support for the effects of long-term Pb exposure on BP in a range that is relevant to
31                   humans  10 -30 (ig/dL (Figure 5-21). Prospective epidemiologic studies, in conjunction
32                   with these animal toxicology studies, also demonstrate the temporal relationship between


      November 2012                             2-21                   Draft - Do Not Cite or Quote

-------
 1                   the exposure and effect. Evidence for Pb-induced hypertension and increased BP is
 2                   further supported by prospective epidemiologic studies showing consistent associations
 3                   between Pb biomarkers and related conditions including cardiovascular and all-cause
 4                   mortality (Section 5.4.5). Animal toxicology studies further indicate coherence and
 5                   strengthen the evidence for causality by providing evidence of biological plausibility
 6                   (e-g-, oxidative stress and contractile processes) for the effect of Pb exposure on the risk
 7                   of increased BP and hypertension.

 8                   While the control for specific potential confounders varied by study, factors including
 9                   age, diet, sex, body mass index (BMI), blood pressure lowering medication use, SES,
10                   race/ethnicity, alcohol consumption, cholesterol, smoking, pre-existing disease
11                   (i.e., diabetes), measures of renal function, and copollutant exposures (i.e., cadmium
12                   [Cd]) were considered. Although no  single study adjusted for all potential confounders
13                   and residual confounding by age may be present in studies using bone Pb measurements,
14                   uncertainties related to confounding bias are reduced with consideration of other lines of
15                   evidence that demonstrate the long term effects of Pb in animals and characterize
16                   biologically plausible modes of action. Overall, evidence in epidemiologic of adults and
17                   experimental studies in animals demonstrates the effect of long-term Pb exposure in
18                   increasing BP and hypertension although uncertainty remains concerning the relevant Pb
19                   exposure patterns contributing to the observed effects. Overall the evidence is sufficient
20                   to conclude that there is a causal relationship between Pb exposure and hypertension.


                     Subclinical Atherosclerosis

21                   A limited number of studies have evaluated markers of subclinical atherosclerosis
22                   following Pb exposure in humans or animals.  Epidemiologic findings are limited to
23                   cross-sectional analyses. One previous NHANES analysis reported an association
24                   between Pb and peripheral artery disease (PAD) that was not confounded by cadmium
25                   (Cd) (Navas-Acien et al.. 2004). Another study  reported an increasing trend in the odds
26                   of PAD, which is an indicator of atherosclerosis, and concurrent blood Pb level in adults
27                   within the NHANES population (Muntner et al.. 2005). Although evidence of plausible
28                   biological mechanisms (e.g., oxidative stress, inflammation, endothelial cell dysfunction)
29                   is clearly described in the animal toxicological literature, these studies have provided
30                   limited evidence to suggest Pb exposure may initiate  atherosclerotic vessel disease.
31                   Further, uncertainty remains concerning the Pb  exposure level, timing, frequency, and
32                   duration contributing to the observed association with PAD. The evidence is suggestive
33                   of a causal relationship between Pb exposure and subclinical atherosclerosis.
      November 2012                              2-22                   Draft - Do Not Cite or Quote

-------
                    Coronary Heart Disease

 1                  There was a small number of studies discussed in the 2006 Pb AQCD (U.S. EPA. 2006b)
 2                  that indicated associations between Pb biomarker levels and increased risk of coronary
 3                  heart disease (CHD) endpoints. Recent longitudinal prospective studies that consistently
 4                  report that blood Pb level is associated with risk of mortality from cardiovascular disease,
 5                  specifically MI, IHD, or CHD (Figure 5-30 and Table 5-23) support and expand upon this
 6                  body of evidence. In addition, Weisskopf et al. (2009) found that patella bone Pb levels
 7                  were  associated with increased mortality from IHD (similar magnitude non-statistically
 8                  significant associations were observed with tibia Pb levels) among subjects enrolled in
 9                  the NAS. Several recent studies also report associations between biomarkers of Pb and
10                  incidence of CHD-related outcomes including a prospective analysis reporting an
11                  increased incidence of IHD (physician confirmed MI, angina pectoris) with blood and
12                  bone  Pb levels (Jain et al.. 2007). Uncertainty remains regarding the level, timing,
13                  frequency, and duration of Pb exposure contributing to CHD in adult populations with
14                  higher past than recent exposure. However, coherence for the associations in humans is
15                  supported by the observation of thrombus formation in animals after long term exposure
16                  (Sections 5.4.7.3 and 5.4.3) and mode of action for Pb-induced CHD (i.e., hypertension,
17                  HRV, increased corrected QT (QTc) interval, and corrected QRS complex (QRSc)
18                  duration in electrocardiogram [ECG]) in humans and animals (Sections 5.4.2 and
19                  5.4.3.4). The overall evidence is sufficient to  conclude that there is a causal relationship
20                  between Pb exposure and coronary heart disease.


                    Cerebrovascular Disease

21                  Both  hypertension and atherosclerosis are risk factors for cerebrovascular disease and the
22                  mechanisms for these outcomes also apply to cerebrovascular disease. Despite strong
23                  evidence for hypertension and CHD and Pb exposure, very few studies have  examined
24                  the effects of Pb exposure on cerebrovascular disease (Section 5.4.7).  These few  studies
25                  provide insufficient evidence to inform the presence or absence of a causal relationship
26                  between cerebrovascular disease and Pb exposure. Thus, the current evidence is
27                  inadequate to determine that a causal  relationship exists between Pb exposure and
28                  cerebrovascular disease.
      November 2012                             2-23                   Draft - Do Not Cite or Quote

-------
      2.6.3        Renal Effects

 1                   Recent epidemiologic and toxicological studies evaluated in the current review support
 2                   and expand upon the strong body of evidence presented in the 2006 Pb AQCD (U.S.
 3                   EPA. 2006b) indicating that Pb exposure is associated with reduced kidney function
 4                   (Section 5.5.5). The causal determination for reduced kidney function is informed by
 5                   evidence for reduced GFR, reduced creatinine clearance, and increased  serum creatinine.


                     Reduced Kidney Function

 6                   The epidemiologic evidence from prospective and cross-sectional studies consistently
 7                   demonstrates a relationship between higher blood Pb level and reduced  kidney function
 8                   (e-g-, lower creatinine clearance, higher serum creatinine, and lower GFR) in
 9                   nonoccupationally-exposed adults with mean concurrent or baseline blood Pb levels of
10                   2-10 (ig/dL. Associations were observed after adjustment for multiple potential
11                   confounding factors such as age, sex, comorbid cardiovascular conditions, BMI,
12                   smoking, and alcohol use. However, uncertainties involve the potential  for reverse
13                   causality to play a role in the findings of cross-sectional studies and inconsistent findings
14                   in occupational studies. Further, since the blood Pb level in nonoccupationally-exposed
15                   adults reflects both recent and past Pb exposures, the magnitude, timing, frequency, and
16                   duration of Pb exposure contributing to the observed associations is also uncertain. A few
17                   analyses find higher blood Pb levels to be associated with a greater longitudinal decrease
18                   in kidney function over time (4-15 years), suggesting that past Pb exposures may
19                   contribute to ongoing renal effects and better characterizing the time sequence between
20                   Pb exposure and lower kidney function. Studies in animals with long-term exposure to Pb
21                   report mixed evidence for Pb-induced kidney dysfunction and histopathological changes,
22                   including tubular atrophy and sclerosis at relevant Pb blood and exposure levels. Animal
23                   studies provide biological plausibility for the associations observed in epidemiologic
24                   studies between blood Pb levels and reduced kidney function with evidence for
25                   Pb-induced hypertension, renal oxidative stress, inflammation, apoptosis, and glomerular
26                   hypertrophy. The body of the evidence is sufficient to conclude that a causal relationship
27                   is likely to exist between Pb exposures and reduced kidney function.
      November 2012                             2-24                  Draft - Do Not Cite or Quote

-------
      2.6.4        Immune System Effects

 1                  The cumulative body of epidemiologic and toxicological evidence from the
 2                  2006 Pb AQCD (U.S. EPA. 2006b) and the current assessment describes several effects
 3                  of Pb exposure on the immune system related to a shift from T-derived lymphocyte
 4                  helper (Th)l - to T-derived lymphocyte helper (Th)2 -type responses, including an
 5                  increase in atopic and inflammatory conditions and a decrease in host resistance (Section
 6                  5.6.8). Outcomes related to an increase in atopic and inflammatory conditions include
 7                  asthma, allergy, increased IgE, and mode of action endpoints such as selective
 8                  differentiation of Th2 cells, increased production of Th2 cytokines, B cell activation, and
 9                  inflammation. Outcomes related to decreased host resistance include enhanced
10                  susceptibility to bacterial and viral infection, suppressed delayed type hypersensitivity
11                  (DTH), and those describing mode of action, i.e., decreased production of Thl cytokines,
12                  reduced phagocyte function,  and increased inflammation.  A small body of studies
13                  indicates the effects of Pb exposure on autoimmunity.


                    Atopic and Inflammatory Conditions

14                  Prospective studies in a few populations of children indicate associations of blood Pb
15                  levels with asthma and allergy, with a cross-sectional study providing supporting
16                  evidence (Section 5.6.5.2). Prospective design, lack of selective participation of subjects,
17                  and objective assessment of outcomes reduce the likelihood that findings are explained
18                  by selection bias or reverse causality. Several studies consider potential confounding by
19                  SES and exposure to smoking or allergens. Although adjustment for these factors
20                  increases confidence that the observed associations reflect an independent association
21                  with Pb exposure, the potential for residual confounding by factors such as SES remains.
22                  However, the evidence for asthma and allergy is supported by associations found
23                  between higher concurrent blood Pb levels in children and higher IgE, an important
24                  mediator of asthma and allergy. In addition, biological plausibility for the effects of Pb on
25                  IgE is provided by consistent findings in animals (i.e., Pb-induced increases in Th2
26                  cytokine production and inflammation), with some evidence at blood Pb levels relevant
27                  to humans. In epidemiologic studies, higher IgE and higher asthma prevalence were
28                  examined and found in children with concurrent blood Pb levels >10 (ig/dL. The strong
29                  toxicological evidence from experimental animal studies supporting modes of action for a
30                  shift to a Th2 phenotype combined with the epidemiologic evidence for asthma and
31                  allergy in a few populations is sufficient to conclude that a causal relationship is likely to
32                  exist between Pb exposures and an increase in atopic and inflammatory conditions.
      November 2012                             2-25                   Draft - Do Not Cite or Quote

-------
                     Decreases in Host Resistance

 1                   Much of the evidence on decreased host resistance was available in the 2006 Pb AQCD
 2                   (U.S. EPA. 2006b) and summarized in Section 5.6.5.1 (and Section 5.6.8.2). Decreased
 3                   host resistance is demonstrated by several toxicological observations that dietary Pb
 4                   exposure producing relevant blood Pb levels increased susceptibility to bacterial infection
 5                   and suppressed DTH in rodents and by the coherence with evidence describing modes of
 6                   action, including suppressed production of Thl cytokines and decreased macrophage
 7                   function. Animal studies found that gestational Pb exposures, producing blood Pb levels
 8                   of 6 and 25 (ig/dL, resulted in decreases in Thl cytokines, suppression of DTH, and
 9                   greater susceptibility to bacterial infection. However, these effects related to decreased
10                   host resistance also were affected by neonatal short-term (several days to 2-3 weeks),
11                   long-term (>4 weeks), and lifetime Pb exposures that produced blood Pb levels
12                   1-25 (ig/dL. Thus, the toxicological evidence does not clearly identify a particular
13                   lifestage of Pb exposure that is more strongly associated with decreased host resistance.
14                   Epidemiologic evidence indicates Pb-associated increases in  respiratory infections but
15                   limitations, including the lack of rigorous methodology and consideration for potential
16                   confounding produce uncertainty regarding the effects of Pb on decreased host resistance
17                   in humans.  The consistent toxicological evidence but lack of available informative
18                   epidemiologic evidence is sufficient to conclude that a causal relationship is likely to
19                   exist between Pb exposure and decreased host resistance.


                     Autoimmunity

20                   Toxicological evidence indicates the potential of Pb to increase autoimmunity, with a few
21                   previous studies showing Pb-induced generation of auto-antibodies (discussed in the
22                   2006 Pb AQCD), and with recent studies providing indirect evidence by showing
23                   formation of neoantigens that could result in the formation of auto-antibodies (Sections
24                   5.6.5.4 and 5.6.8.3). Several observations were made in animals injected with Pb, which
25                   is a route of exposure with less relevance to humans. Higher levels of auto-antibodies
26                   also were found in Pb-exposed battery workers; however, implications are limited
27                   because of the high blood Pb levels (range:  10-40 (ig/dL)  of some of the workers and lack
28                   of consideration for potential confounding by several factors, including other
29                   occupational exposures. Because results from available toxicological and epidemiologic
30                   studies do not sufficiently inform Pb-induced generation of auto-antibodies with relevant
31                   Pb exposures, the evidence is inadequate to determine if there is a causal relationship
32                   between Pb exposure and autoimmunity.
      November 2012                             2-26                   Draft - Do Not Cite or Quote

-------
      2.6.5        Hematological Effects

 1                  Recent toxicological and epidemiologic evidence substantiates evidence presented in the
 2                  2006 Pb AQCD (U.S. EPA. 2006b) that exposure to Pb affects hematological endpoints,
 3                  and supports a causal relationship between Pb exposure and decreased red blood cell
 4                  (RBC) survival and function and altered heme synthesis. Outcomes related to decreased
 5                  RBC survival and function included alterations in multiple hematological parameters
 6                  (e-g-, Hb, Hct, MCV), oxidative stress (altered antioxidant enzyme activities, decreased
 7                  cellular glutathione (GSH), and increased lipid peroxidation), increased cytotoxicity in
 8                  RBC precursor cells, and mode of action endpoints such as decreased intracellular
 9                  calcium concentrations [Ca2+]l5 decreased adenosine-triphosphase (ATPase) activity, and
10                  increased phosphatidylserine expression. Outcomes related to altered heme synthesis
11                  included decreased activities of ALAD and ferrochelatase, and decreased levels of Hb.


                    Decreased Red  Blood Cell Survival and Function

12                  Experimental animal studies demonstrate that exposures via drinking water and gavage,
13                  resulting in blood Pb levels relevant to humans, alter several hematological parameters,
14                  increase measures  of oxidative stress and increase cytotoxicity in red blood cell (RBC)
15                  precursor cells. Support for these findings is provided by biologically plausible modes of
16                  action including decreased intracellular calcium concentrations [Ca2+]l5 decreased ATPase
17                  activity, and increased phosphatidylserine expression. Epidemiologic studies demonstrate
18                  evidence in both adults and children that Pb exposure results in altered hematological
19                  endpoints and increased measures of oxidative stress, and altered hematopoiesis.
20                  Although the majority of these studies are limited by the lack of rigorous methodology
21                  and consideration for potential confounding they support the toxicological findings.
22                  While some studies in children did control for or considered potential confounding and
23                  effects in adults and children are coherent with effects observed in exposed animals, there
24                  remains some uncertainty regarding the evidence for altered RBC survival and function
25                  in human populations. Collectively, the strong evidence from toxicological studies that is
26                  supported by findings from mode of action and epidemiologic studies is sufficient to
27                  conclude that there is a causal relationship between Pb exposures and decreased RBC
28                  survival and function.


                    Heme Synthesis

29                  Altered heme synthesis is demonstrated by a small, but consistent, body of experimental
30                  animal studies reporting that exposures via drinking water and gavage (resulting in blood
31                  Pb levels relevant to humans) decreased ALAD and ferrochelatase activities. Supporting
      November 2012                             2-27                  Draft - Do Not Cite or Quote

-------
 1                  this toxicological evidence is a larger body of ecotoxicological studies that demonstrate
 2                  decreased ALAD activity across a wide range of taxa exposed to Pb. Epidemiologic
 3                  studies demonstrate evidence in both adults and children that Pb exposure results in
 4                  decreased ALAD and ferrochelatase activities. However, the majority of these studies are
 5                  limited by the lack of rigorous methodology and consideration for potential confounding.
 6                  While some studies in children did control for or considered potential confounding and
 7                  effects in adults and children are coherent with effects observed in exposed animals, there
 8                  remains some uncertainty regarding the evidence for altered heme synthesis in human
 9                  populations. Evidence for altered heme synthesis is also provided by a large body of
10                  toxicological and epidemiologic studies that report decreased Hb concentrations due to
11                  Pb exposure. Collectively, the strong evidence from toxicological and ecotoxicological
12                  studies that is supported by findings  from epidemiologic studies is sufficient to conclude
13                  that there is a causal relationship between Pb exposures and altered heme synthesis.
      2.6.6        Reproductive and Developmental Effects

14                  Many epidemiologic and toxicological studies of the effects of Pb on reproductive and
15                  developmental outcomes have been conducted since the 2006 Pb AQCD. The evaluation
16                  of causal relationships with Pb exposure focuses on four areas: developmental effects,
17                  birth outcomes, reproductive function among males, and reproductive function among
18                  females.
                    Development

19                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported Pb-associated effects on development in
20                  toxicological studies. Findings from epidemiologic studies of postnatal growth are
21                  inconsistent and findings from animal toxicological studies are mixed, with recent studies
22                  on the effect of Pb on growth showing Pb to elicit adult onset obesity. Multiple recent
23                  epidemiologic studies of Pb and puberty have shown associations between concurrent
24                  blood Pb levels and delayed pubertal onset for girls and boys. In cross-sectional
25                  epidemiologic studies of girls (ages 6-18 years) with mean and/or median concurrent
26                  blood Pb levels less than 5 (ig/dL consistent associations with delayed pubertal onset
27                  were  observed. In boys (ages 8-15 years), fewer epidemiologic studies were conducted
28                  but associations were observed, including associations among boys in a longitudinal
29                  study. These associations are consistently observed in populations with blood Pb levels
30                  <10(ig/dL; however, there is uncertainty with regard to the exposure frequency, timing,
31                  duration and level contributing to the observed effects  in studies of older children and
32                  adolescents. The evidence of delayed pubertal onset among males and females from
      November 2012                             2-28                   Draft - Do Not Cite or Quote

-------
 1                  epidemiologic studies is consistent and coherent with evidence from toxicological studies
 2                  at relevant exposure levels. Potential confounders considered in the epidemiologic studies
 3                  varied. Most studies controlled for age and body mass index (BMI). Other variables, such
 4                  as measures of diet, SES, and race/ethnicity, were included in some of the studies.
 5                  Although, adjustment for nutritional factors was done less often, a NHANES analysis of
 6                  girls (Selevan et al.. 2003) controlled for various dietary factors as well as other potential
 7                  confounders and reported an association between increased blood Pb levels and delayed
 8                  pubertal onset. Overall, the toxicological and epidemiologic evidence together is
 9                  sufficient to conclude that there is a causal relationship between Pb  exposure and
10                  developmental effects.


                    Birth Outcomes

11                  Overall, results of pregnancy outcomes were similar to those of the 2006 Pb AQCD (U.S.
12                  EPA. 2006b); no consistent evidence of a relationship with Pb was available for preterm
13                  birth. The 2006 Pb AQCD included a few studies that reported potential associations
14                  between Pb and neural tube defects, but the recent epidemiologic studies found no
15                  association (Section 5.8.6.2). Some associations were observed between Pb and low birth
16                  weight when epidemiologic studies used  measures of maternal bone Pb or air exposures.
17                  The associations were less consistent when using maternal blood Pb or umbilical cord
18                  and placenta Pb, but some associations between increased Pb levels and decreased low
19                  birth weight/fetal growth were observed.  The effects of Pb exposure during early
20                  development in animal toxicological studies included mixed findings with some studies
21                  showing reduction in litter size, implantation, and birth weight. Based on the observation
22                  of some associations observed in well-conducted epidemiologic studies of preterm birth
23                  and low birth weight/fetal growth, the evidence is suggestive of a relationship between Pb
24                  exposure and birth outcomes.


                    Male Reproductive Function

25                  Evidence from recent experimental studies of animals and supporting evidence from
26                  epidemiologic studies expands upon findings from the 2006 Pb AQCD to indicate that a
27                  causal relationship exists between Pb exposure and male reproductive function.
28                  Toxicological studies in rodents,  non-human primates, and rabbits show detrimental
29                  effects on semen quality, sperm, and fecundity/fertility epidemiologic studies report
30                  detrimental effects on sperm. The effect of Pb on other aspects of male reproduction,
31                  including hormone aberrations is less clear. Findings from epidemiologic studies of
32                  fertility among men were also mixed. Pb  may exert effects on the reproductive system by
33                  affecting the responsiveness  of the hypothalamic-pituitary-gonad axis or by suppressing
      November 2012                             2-29                  Draft - Do Not Cite or Quote

-------
 1                   circulating hormone levels. Recent toxicological studies suggest that oxidative stress is a
 2                   major contributor to the effects of Pb on male reproductive system, providing mode of
 3                   action support. The effects of reactive oxygen species (ROS) may involve interference
 4                   with cellular defense systems leading to increased lipid peroxidation and free radical
 5                   attack on lipids, proteins, and DNA. Several recent studies showed Pb induced an
 6                   increased generation of ROS within the male sex organs and germ cell injury as
 7                   evidenced by aberrant germ cell structure and function. Co-administration of Pb with
 8                   various antioxidant compounds either eliminated Pb-induced injury or greatly attenuated
 9                   its effects. In addition, many studies that observed increased oxidative stress also
10                   observed increased apoptosis which is likely a critical underlying mechanism in
11                   Pb-induced germ cell DNA damage and dysfunction. Based on the consistency and
12                   coherence of findings for effects of Pb exposure on sperm in the toxicological literature,
13                   and the support from epidemiologic studies with biological plausibility provided by mode
14                   of action evidence, the evidence is sufficient to conclude that there is a causal
15                   relationship between Pb exposures and male reproductive  function.


                     Female Reproductive Function

16                   Epidemiologic and toxicological studies of reproductive function among females
17                   investigated whether Pb biomarker levels were associated with hormone levels, fertility,
18                   estrus cycle changes, and morphology or histology of female reproductive organs
19                   including the placenta (Section 5.8.6.4). Toxicological studies of experimental animals
20                   reported in the 2006 Pb AQCD (U.S. EPA. 2006b) demonstrated associations between Pb
21                   exposure and female reproductive function, although little evidence was provided by
22                   epidemiologic studies. Some studies have shown associations with blood Pb levels and
23                   altered hormone levels in adults, with inconsistency across studies likely due to the
24                   different hormones examined and the different timing in the menstrual and life cycles.
25                   There is some evidence of a potential relationship between Pb exposure and female
26                   fertility, but findings are also mixed. The majority of the epidemiologic studies are
27                   cross-sectional, and adjustment for potential confounders varies from study to study, with
28                   some potentially important confounders, such as BMI, not included in all studies. Also,
29                   most of the studies have small samples sizes and are generally of women attending
30                   infertility clinics. The design of animal toxicological studies often employs prenatal or
31                   early postnatal Pb  exposures with Pb contributing to placental pathology and
32                   inflammation, decreased ovarian antioxidant capacity, altered ovarian steroidogenesis and
33                   aberrant gestational hormone levels. Although epidemiologic and toxicological studies
34                   provide information on different exposure periods, both types of studies support the
35                   conclusion that Pb possibly affects at least some aspects of female reproductive function.
36                   Overall, the relationship observed with female reproductive outcomes is sufficient to
      November 2012                             2-30                  Draft - Do Not Cite or Quote

-------
 1                  conclude that there is a suggestive relationship between Pb exposure and female
 2                  reproductive function.
      2.6.7        Cancer

 3                  The toxicological literature from experimental animal studies provides the strong
 4                  evidence for Pb exposure and cancer. The consistent evidence indicating Pb-induced
 5                  carcinogenicity in animal models is substantiated by the mode of action findings from
 6                  multiple high-quality studies from different laboratories. This conclusion is in agreement
 7                  with those of other agencies including the International Agency for Research on Cancer
 8                  (IARC, which has classified inorganic Pb compounds as a probable human carcinogen),
 9                  and the National Toxicology Program (NTP, which has listed Pb and Pb compounds as
10                  "reasonably anticipated to be human carcinogens"). Strong evidence from animal
11                  toxicological studies demonstrates an association between Pb and cancer, genotoxicity,
12                  mutagenicity or epigenetic modification. Carcinogenicity in animal toxicology studies
13                  with relevant routes of Pb exposure has been reported in the kidneys, testes, brain,
14                  adrenals, prostate, pituitary, and mammary gland (Table 5-50). Epidemiologic studies of
15                  cancer incidence and mortality reported inconsistent results; one strong epidemiologic
16                  study demonstrated an association between blood Pb and increased cancer mortality, but
17                  the other studies reported weak or no associations.  In the 2006 Pb AQCD (U.S.  EPA.
18                  2006b). Pb exposure was found to be associated with stomach cancer, and a recent study
19                  on stomach cancer and Pb exposure, reported mixed findings depending on the type of Pb
20                  exposure (organic Pb, inorganic Pb, or Pb from gasoline emissions). Similarly, some
21                  studies  in the 2006 Pb AQCD reported associations between Pb exposure and lung
22                  cancer.  Recent epidemiologic studies of lung cancer focused on occupational exposures
23                  and reported inconsistent associations. The majority of epidemiologic studies of brain
24                  cancer had null results overall, but positive associations between Pb exposure and brain
25                  cancer were observed among individuals with certain genotypes. Overall, despite the
26                  inconsistent findings from epidemiologic studies, the consistent and strong body of
27                  evidence from toxicological studies is sufficient to conclude that there  is a likely causal
28                  relationship between Pb exposure and cancer.
      November 2012                             2-31                  Draft - Do Not Cite or Quote

-------
      2.7        Ecological  Effects of Pb

 1                  Sections 2.7.1 and 2.7.2 are summaries of the evidence evaluated in ChapterJ? in which
 2                  the effects of Pb on terrestrial and aquatic ecosystems are presented separately. The
 3                  evidence supporting ecological causal determinations is synthesized across endpoints
 4                  (reproduction, growth, survival, neurobehavioral effects, hematological effects,
 5                  physiological stress) common to terrestrial, freshwater and saltwater biota in Section
 6                  2.7.3 (Table 2-3). An integration of the evidence across endpoints examined in both
 7                  human health and ecological studies follows (Section 2.8). Consideration of atmospheric
 8                  deposition of Pb as related to ecological effects is discussed under policy relevant
 9                  considerations (Section 2.9.7).
      2.7.1       Summary of Effects on Terrestrial Ecosystems

10                  Historically, Pb poisoning is one of the earliest recognized toxicoses of terrestrial biota,
11                  occurring primarily through ingestion of spent shot by birds (Section 7.3.4.3). At the time
12                  of the 1977 Pb AQCD, few studies of Pb exposure and effects in wild animals other than
13                  birds were available. A limited number of rodent trapping studies and observations from
14                  grazing animals near smelters provided evidence for differences in Pb sensitivity among
15                  species and these findings were further supported in the 1986 and 2006 Pb AQCDs (U.S.
16                  EPA. 2006b. 1986b. 1977). Commonly observed effects of Pb on terrestrial organisms
17                  include decreased survival, reproduction, and growth, as well as effects on development,
18                  behavior, and ALAD activity (U.S. EPA. 2006b. 1986b. 1977).

19                  In plants, Pb effects have been studied for several decades. At the time of the 1977 Pb
20                  AQCD, it was understood that Pb uptake in plants was influenced by plant species and by
21                  the available Pb pool in the soils, and that most of the Pb taken up by plants from soil
22                  remains in the roots, with translocation to other portions of the plant varying with species
23                  (U.S. EPA. 1977). Plant growth was recognized as an endpoint of Pb toxicity in plants in
24                  the 1977 Pb AQCD and additional effects of Pb on growth processes were reported in
25                  subsequent Pb AQCDs (U.S. EPA. 2006b. 1986b. 1977V In the 1977 Pb AQCD evidence
26                  for effects of Pb on forest-nutrient cycling and shifts in arthropod community
27                  composition was found in one study conducted in the vicinity of a smelting complex. In
28                  subsequent AQCDs, other ecosystem-level effects, including decreased species diversity,
29                  changes in floral and faunal community composition, and decreasing vigor of terrestrial
30                  vegetation have subsequently been reported near stationary sources of Pb (U.S. EPA.
31                  2006b. 1986b. 1977: Watson etal.. 1976).
      November 2012                            2-32                  Draft - Do Not Cite or Quote

-------
 1                   Pb is either deposited directly onto plant surfaces, or soil where it can bind with organic
 2                   matter or dissolve in pore water. The amount of Pb dissolved in soil pore water
 3                   determines the impact of soil Pb on terrestrial ecosystems to a much greater extent than
 4                   the total amount present. It has long been established that the amount of Pb dissolved in
 5                   soil solution is controlled by at least six variables: (1) solubility equilibria;
 6                   (2) adsorption-desorption relationship of total Pb with inorganic compounds;
 7                   (3) adsorption-desorption reactions of dissolved Pb phases on soil organic matter; (4) pH;
 8                   (5) cation exchange capacity (CEC); and (6) aging. Since 2006, further details have been
 9                   contributed to the understanding of the role of pH, CEC, organic matter, and aging.
10                   Smolders et al. (2009) demonstrated that the two  most important determinants of both Pb
11                   solubility and toxicity in soils are pH and CEC. However, they had previously shown that
12                   experimental aging, primarily in the form of initial leaching following addition of Pb,
13                   decreases soluble metal fraction by approximately one order of magnitude  (Smolders et
14                   al., 2009).  Since 2006, organic matter has been confirmed as an important influence on
15                   Pb sequestration, leading to longer-term retention in soils with higher organic matter
16                   content, and also creating the potential for later release of deposited Pb. Aging, both
17                   under natural conditions and simulated through leaching, was shown to substantially
18                   decrease bioavailability to plants, microbes, and vertebrates.

19                   Evidence over several decades of research, previously reviewed in Pb AQCDs and in
20                   more recent studies, shows that Pb accumulates in terrestrial plants, invertebrates and
21                   vertebrates. Studies with herbaceous plant species growing at various distances from
22                   smelters added to the existing strong evidence that atmospherically transported Pb is
23                   taken up by those plants. In most species tested, soil Pb taken up by the roots is not
24                   translocated into  the stem and leaves. These studies did not establish the relative
25                   proportion that originated from  atmospheric Pb deposited in the soil, as opposed to that
26                   taken up directly from the atmosphere through the leaves. In trees, studies have found
27                   that soil Pb generally is translocated to other parts, in contrast to herbaceous plants, and
28                   recent studies have shown that the proportion of Pb that is taken up through the leaves
29                   and trunk is likely substantial. One study attempted to quantify this proportion Pb that is
30                   taken up directly from the atmosphere suggested  it amounts to 50% of the Pb contained
31                   in Scots pine (Pinus sylvestris) in Sweden (Klaminder et al., 2005).

32                   Since the 2006 Pb AQCD, various species of terrestrial  snails have been found to
33                   accumulate Pb from both diet and soil. Recent studies with earthworms have found that
34                   both internal concentration of Pb and mortality increase with  decreasing soil pH and
35                   CEC, and the importance of the interaction of those factors with Pb concentration has
36                   been strongly confirmed, but only very partially quantified. Tissue concentration
37                   differences have been found between species of earthworms that burrow in different soil
38                   layers, and the rate of accumulation in each of these species may result from differences
      November 2012                             2-33                  Draft - Do Not Cite or Quote

-------
 1                   in interacting factors such as pH and CEC between layers. Because earthworms often
 2                   sequester Pb in granules, some authors have suggested that earthworm Pb is not
 3                   bioavailable to their predators. There is some evidence that earthworm activity increases
 4                   Pb availability in soil, but it is inconsistent. In various arthropods collected at
 5                   contaminated  sites, recent studies found gradients in accumulated Pb that corresponded to
 6                   gradients in soil with increasing distance from stationary sources.

 7                   There are a few  recent studies of Pb bioavailability and uptake in birds since the
 8                   2006 Pb AQCD. Several found tissue levels in birds that indicated exposure to Pb, but
 9                   none of the locations for these studies was in proximity to stationary sources, and the
10                   origin of the Pb  could not be identified. A study at the Anaconda Smelter Superfund site
11                   found increasing Pb accumulation in gophers with increasing soil Pb around the location
12                   of capture. A study of swine fed various Pb-contaminated soils showed that the form of
13                   Pb determined accumulation. Recent studies were able to measure Pb in the components
14                   of various food chains that included soil, plants, invertebrates, and vertebrates. They
15                   confirmed that trophic transfer of Pb is pervasive, but no consistent evidence of trophic
16                   magnification was found.

17                   Evidence in this review further supports the findings of the previous Pb AQCDs that
18                   biological effects of Pb  on terrestrial organisms vary with species and lifestage, duration
19                   of exposure, form of Pb, and soil characteristics. In photosynthetic organisms,
20                   experimental studies have added to the existing evidence of photosynthesis impairment in
21                   plants exposed to Pb, and have found damage to photosystem II due to alteration of
22                   chlorophyll structure, as well as decreases in chlorophyll content in diverse taxa,
23                   including lichens and mosses. Evidence of oxidative stress in response to Pb exposure has
24                   also been observed in plants. Reactive oxygen species were found to increase in broad
25                   bean and tomato plants  exposed to increasing concentrations of soil  Pb, and a
26                   concomitant increase in superoxide dismutase, glutathione, peroxidases,  and lipid
27                   peroxidation, as well as decreases in catalase were observed in the same  plants. Monocot,
28                   dicot, and bryophytic taxa grown in Pb-contaminated soil or in experimentally spiked soil
29                   all responded to increasing exposure with increased antioxidant activity.  In addition,
30                   reduced growth  was observed in some experiments, as  well as genotoxicity, decreased
31                   germination, and pollen sterility.

32                   In terrestrial invertebrates, evidence for Pb effects has included neurological and
33                   reproductive endpoints. Recently published studies have  shown neuronal damage in
34                   nematodes exposed to concentrations of Pb [2.5  uM (0.5  mg Pb/L)]  in laboratory settings,
35                   accompanied by behavioral abnormalities. Reproductive  adverse effects were found  at
36                   lower exposure in younger nematodes, and effects on longevity and  fecundity were
37                   shown to persist for several generations. Increased mortality was found in earthworms,
      November 2012                             2-34                  Draft - Do Not Cite or Quote

-------
 1                  but was strongly dependent on soil characteristics including pH, CEC, and aging. Snails
 2                  exposed to Pb through either topical application or through consumption of Pb-exposed
 3                  plants had increased antioxidant activity, and decreased food consumption, growth, and
 4                  shell thickness. Effects on arthropods exposed through soil or diet varied with species and
 5                  exposure conditions, and included diminished growth and fecundity, endocrine and
 6                  reproductive anomalies, and body malformations. Within each study, increasing
 7                  concentration of Pb in the exposure medium generally resulted in increased effects, but
 8                  the relationship between concentration and effects varied between studies, even when the
 9                  same medium, e.g., soil, was used. Evidence suggested that aging and pH are important
10                  modifiers.

11                  ALAD was identified in the 1977 Pb AQCD as a sensitive indicator of exposure to Pb in
12                  rats and waterfowl, and became regarded as a biomarker of exposure in many terrestrial
13                  vertebrates. Other effects of Pb on vertebrates reviewed in Pb AQCDs and the current
14                  document include decreased white blood cell counts and behavioral anomalies observed
15                  in amphibians and reptiles. However, large differences in effects were observed at the
16                  same concentration of Pb in soil, depending on whether the soil was freshly amended or
17                  field-collected from contaminated areas. As in most studies where the comparison was
18                  made, effects were smaller when field-collected soils were used. In some birds, maternal
19                  elevated blood Pb level was associated in recent studies with decreased hatching success,
20                  smaller clutch size, high corticosteroid level, and abnormal behavior.  Some species
21                  evidenced little or no effect of elevated blood Pb level. Effects of dietary exposure were
22                  studied in several mammalian species, and cognitive, endocrine, immunological, and
23                  growth effects were observed.

24                  Recent evidence reviewed in Sections 7.3.6 and 7.3.12.7 demonstrates that exposure to
25                  Pb is generally associated with negative effects in terrestrial ecosystems. It also
26                  demonstrates that many factors, including species and various soil physiochemical
27                  properties, interact strongly with  Pb  concentration to modify those effects. In these
28                  ecosystems, where soil is generally the main component of the exposure route, Pb aging
29                  is a particularly important factor, and one that may be difficult to reproduce
30                  experimentally. Without quantitative characterization of those interactions,
31                  characterizations of exposure-response relationships would likely not be transferable
32                  outside of experimental settings.  Since the 2006 Pb AQCD, few studies of
33                  exposure-response have been conducted, and results have been inconsistent.

34                  Recent evidence of effects of Pb  at the community and ecosystem levels of biological
35                  organization include several studies  of the ameliorative effects of mycorrhizal fungi on
36                  plant growth in the presence of Pb, attributed to decreased uptake of Pb by plants,
37                  although both mycorrhizal fungus and plant were negatively affected  at the exposures
      November 2012                             2-35                  Draft - Do Not Cite or Quote

-------
 1                  assessed. Most recently published research on community and ecosystem-level effects of
 2                  Pb has focused on soil microbial communities, which have been shown to be impacted in
 3                  both composition and activity. Many of the recent studies of effects on soil microbial
 4                  communities have taken place in environments contaminated with multiple metals, and
 5                  some have attempted to separate the effects of individual metals when possible. Soil
 6                  microbial activity was generally diminished, but in some cases recovered over time.
 7                  Species and genotype composition were consistently altered, and those changes were
 8                  long-lasting or permanent. Recent studies have addressed differences in sensitivity
 9                  between species explicitly, and have clearly demonstrated high variability between
10                  related species, as well as within larger taxonomic groupings. Mammalian no observed
11                  effect concentration (NOEC) values expressed as blood Pb levels were shown to vary by
12                  a factor of 8, while avian blood NOECs varied by a factor of 50 (Buekers et al. 2009).
13                  Protective effects of dietary Ca2+ have been found in plants, birds, and invertebrates.
      2.7.2        Summary of Effects on Aquatic Ecosystems

14                  Effects of Pb on plants, invertebrates, and vertebrates are reported for both freshwater
15                  and saltwater ecosystems. Although effects of Pb exposure are likely mediated through
16                  common mode(s) of action across freshwater and marine/estuarine taxa, these ecosystems
17                  are considered separately because of different environmental and physiological factors
18                  that influence Pb toxicity such as bioavailability of the metal, form of Pb, water quality
19                  parameters and adaptations in freshwater and saltwater organisms. Toxicity of Pb also
20                  varies by organism, lifestage and duration of exposure. (U.S. EPA. 2006b. 1986a).
21                  Closely related organisms can vary greatly in bioaccumulation of Pb and other
22                  non-essential metals as well as in their susceptibility to Pb. Pb effects on aquatic biota
23                  were previously assessed in the 1977 Pb AQCD, the 1986 Pb AQCD and the
24                  2006 Pb AQCD (U.S. EPA. 2006b. 1986a.  1977).

25                  Exposure of freshwater and estuarine organisms to Pb, and associated effects are tied to
26                  terrestrial systems via watershed processes (Section 7.2). Atmospherically-derived Pb can
27                  enter aquatic systems through runoff from terrestrial systems or via direct deposition over
28                  a water surface. In aquatic ecosystems affected by Pb, exposures are most likely
29                  characterized as low dose, chronic exposures. Once Pb enters surface waters, its fate and
30                  bioavailability are influenced by Ca2+ concentration, pH, alkalinity, total suspended
31                  solids, and dissolved organic carbon (DOC), includinig humic acids. In saltwater, higher
32                  levels of ions additionally affect Pb bioavailability. In sediments, Pb bioavailability may
33                  be influenced by the presence of other metals, sulfides, iron (Fe-) and manganese
34                  (Mn-)oxides, and physical disturbance. Recent studies provide further evidence for the
      November 2012                             2-36                   Draft - Do Not Cite or Quote

-------
 1                   role of modifying factors such as pH, DOC, and hardness. Toxicity of the same
 2                   concentration of Pb can vary greatly under different experimental conditions.

 3                   As recognized in the 2006 Pb AQCD and further supported in this review, uptake of Pb
 4                   by aquatic invertebrates and vertebrates may preferentially occur via exposure routes
 5                   other than direct absorption from the water column such as ingestion of contaminated
 6                   food and water, uptake from sediment pore waters, or incidental ingestion of sediment
 7                   (U.S. EPA. 2006b). Currently available models for predicting bioavailability focus on
 8                   acute toxicity and do not consider all possible routes of uptake. They are therefore of
 9                   limited applicability, especially when considering  species-dependent differences in
10                   uptake and bioaccumulation of Pb. Recent evidence supports the 2006 Pb AQCD
11                   conclusion that processes such as Pb adsorption, complexation, and chelation alter
12                   bioavailability to aquatic organisms.


                     Biological Effects of Pb on Freshwater Plants, Invertebrates and
                     Vertebrates

13                   Recent evidence further supports the findings  of the previous Pb AQCDs that waterborne
14                   Pb is highly toxic to freshwater plants, invertebrates and vertebrates, with toxicity
15                   varying with species and lifestage, duration of exposure, form of Pb, and water quality
16                   characteristics. Concentration-response data from freshwater organisms indicate that
17                   there is a gradient of response to increasing Pb concentration and that some effects in
18                   sensitive species are observed at concentrations of Pb quantified in U.S.  surface waters
19                   (Table 2-1).

20                   The toxicity of Pb to aquatic algae and plants has been recognized in earlier EPA reviews
21                   of this metal. In the 1977 Pb AQCD, differences in sensitivity to Pb among different
22                   species of algae were reported and concentrations of Pb varied within and between
23                   genera. This observation  was subsequently generalized across aquatic taxa (U.S. EPA.
24                   1977). At the time of the  1977 Pb AQCD, the  information available on effects of Pb on
25                   freshwater plants was limited. For plants in general, Pb was recognized to affect
26                   photosynthesis, mitosis, and growth, but at concentrations higher than typically found in
27                   the environment. Effects  of Pb on plants reported in subsequent Pb AQCDs included
28                   decreased growth, deformation of cells, and blocking of the pathways that lead to
29                   pigment synthesis, thus affecting photosynthesis.

30                   Effects of Pb on aquatic plants supported by additional evidence in this review include
31                   oxidative  damage, decreased photosynthesis, and reduced growth. Most recent studies
32                   report effects on growth at concentrations much higher than Pb typically encountered in
33                   the environment, however, some sublethal endpoints such as effects on chlorophyll  were
34                   reported at lower concentrations, albeit still much higher than those typically encountered

      November 2012                             2-37                  Draft - Do Not Cite or Quote

-------
 1                   in U.S. surface waters (in the 100 to 200 (ig Pb/L range). Elevated levels of antioxidant
 2                   enzymes are commonly observed in aquatic plant, algae, and moss species exposed to Pb
 3                   (U.S. EPA. 1977) and recent evidence continues to support this observation. Recent
 4                   studies on uptake of Pb by aquatic plants support the findings of previous Pb AQCDs that
 5                   all such plants with roots tend to sequester larger amounts of Pb in their roots than in
 6                   their shoots, and provide additional evidence for species differences in
 7                   compartmentalization of sequestered Pb and in responses to Pb in water and sediments.
 8                   Exposure-response relationships in which increasing concentrations of Pb leads to
 9                   increasing effects have consistently been reported in freshwater algae and macrophytes,
10                   suggesting that effects on growth and antioxidant activity are also occurring at lower
11                   concentrations, however, most current observations of Pb effects in freshwater plants are
12                   at concentrations that exceed Pb concentration values available for U.S. surface waters
13                   (Table 2-IV

14                   The largest body of evidence for effects of Pb at or near concentrations encountered in
15                   U.S. surface waters is from invertebrates. In the  1986 Pb AQCD (U.S. EPA. 1986a) and
16                   2006 Pb AQCD  (U.S. EPA. 2006b). reduced reproduction, growth, and survival were
17                   reported in various species of freshwater invertebrates. In the 2006 Pb AQCD,
18                   concentrations at which effects were observed in aquatic invertebrates ranged from 5 to
19                   8,000 (ig Pb/L. Recent evidence for effects of Pb on reproduction, growth, and survival
20                   supports findings in previous Pb AQCDs. In a series of 48-hour acute toxicity tests using
21                   a variety of natural waters across North America, LC50 values ranged from 29 to
22                   180 (ig Pb/L tests with the cladoceran Ceriodaphnia dubia (Esbaugh et al.. 2011). In this
23                   same species, increased DOC leads to an increased mean EC50 for reproduction as low as
24                   25 (ig Pb/L. Reproductive and growth effects have also been reported in rotifer, midge
25                   and mayfly species near the range of Pb concentrations encountered in freshwater
26                   habitats. Several studies in this review have provided evidence of growth effects at lower
27                   concentrations. Among the  most sensitive species, growth of juvenile freshwater snails
28                   (Lymnaea stagnalis) was inhibited at an EC2o of <4 (ig Pb/L (Grosell and Brix. 2009;
29                   Grosell et al., 2006b). A chronic value of 10 (ig Pb/L, obtained in 28-day exposures of
30                   2-month-old freshwater mussel (Lampsilis siliquoidea) juveniles, was the lowest
31                   genus-mean chronic value ever reported for Pb (Wang et al., 2010e).

32                   Since the 2006 Pb AQCD, there is additional evidence for Pb effects on antioxidant
33                   enzymes, lipid peroxidation, stress response and osmoregulation in aquatic invertebrates,
34                   as well as additional information on Pb bioaccumulation. Recent studies using stable
35                   isotopes have enabled simultaneous measurement of uptake and elimination in several
36                   aquatic organisms to assess the relative importance of water versus dietary uptake. In
37                   uptake studies of various invertebrates, Pb was mainly found in the gills and digestive
38                   gland/hepatopancreas.
      November 2012                             2-38                  Draft - Do Not Cite or Quote

-------
 1                  Pb effects on freshwater vertebrates were previously assessed in the 1977 Pb AQCD, the
 2                  1986 Pb AQCD and the 2006 Pb AQCD (U.S. EPA. 2006b. 1986a. 1977). Evidence of
 3                  toxicity of Pb and other metals to freshwater organisms goes back to early observations
 4                  of contamination of natural areas by Pb mining leading to extirpation offish from streams
 5                  (U.S. EPA. 1977). Recent evidence supports the findings of effects on survival,
 6                  reproduction, and behavior reported in previous Pb AQCDs for freshwater vertebrates. In
 7                  a series of 96-hour acute toxicity tests with fathead minnow conducted in a variety of
 8                  natural waters across North America, LC50 values ranged from 41 to 3,598 (ig Pb/L
 9                  (Esbaugh et al. 2011). Reproductive effects associated with water quality parameters
10                  were also noted with this species (Mager et al., 2010). In fish, several recent studies on
11                  behavioral effects of Pb indicate decreased prey capture rate, slower swim speed and
12                  decline in startle response and visual contrast with Pb exposure. These reported effects
13                  provide additional evidence for toxicity of Pb to fish. Chronic NOEC and ECio values
14                  reported for trout, a sensitive species, are within the range of Pb occasionally encountered
15                  in U.S. surface waters (Table  7-2).

16                  Observed responses offish to Pb reported in the 1986 Pb AQCD and the 2006 Pb AQCD
17                  included inhibition of heme formation, alterations in brain receptors, effects on blood
18                  chemistry and hormonal systems, and decreases in some enzyme activities (U.S. EPA.
19                  2006b. 1986a). Since the 2006 Pb AQCD, possible molecular targets for Pb neurotoxicity
20                  have been identified in fish and additional mechanisms of Pb toxicity have been
21                  elucidated in the fish gill and the fish renal system. In the 2006 Pb AQCD, amphibians
22                  were considered to be relatively tolerant to Pb. Observed responses to Pb exposure
23                  included decreased enzyme activity (e.g., ALAD reduction) and changes in behavior.
24                  Since the 2006 Pb AQCD, studies conducted at concentrations approaching
25                  environmental levels of Pb have indicated sublethal effects on tadpoles including
26                  deformities and decrements in growth and swimming ability.

27                  In the 2006 Pb AQCD, adverse effects were found in freshwater fish at concentrations
28                  ranging from 10 to >5,400 (ig Pb/L, generally depending on water quality variables
29                  (e.g., pH, hardness, salinity). Additional testing of Pb toxicity under conditions of varied
30                  alkalinity, DOC, and pH has been conducted since the last review. Effects in fish
31                  observed in recent studies fall within the range of concentrations observed in the previous
32                  Pb AQCD. Recent evidence also supports the 2006 conclusions that the gill is a major
33                  site of Pb uptake in fish, and that there are species differences in the rate of Pb
34                  accumulation and distribution of Pb within the organism. The anterior intestine has been
35                  newly  identified as a site of uptake of Pb through  dietary exposure studies. At the time of
36                  the publication of the 2006 Pb AQCD, trophic transfer of Pb through aquatic food chains
37                  was considered to be negligible. Measured concentrations of Pb in the tissues of aquatic
38                  organisms were generally higher in algae and benthic organisms than in consumers at
      November 2012                             2-39                  Draft - Do Not Cite or Quote

-------
 1                  higher trophic levels, indicating that Pb was bioconcentrated but not biomagnified. Some
 2                  studies published since the 2006 Pb AQCD support the potential for transfer of Pb in
 3                  aquatic food webs, while other studies indicate that Pb concentration decreases with
 4                  increasing trophic level.

 5                  Ecosystem-level effects associated with Pb reported in previous Pb AQCDs include
 6                  alteration of predator-prey dynamics, species richness, species composition, and
 7                  biodiversity. Since the 2006 Pb AQCD, additional evidence for community and
 8                  ecosystem level effects of Pb have been observed primarily in microcosm studies or field
 9                  studies near contaminated areas (mining, effluent). Findings from field studies of aquatic
10                  communities in the vicinity of heavily contaminated sites include changes in species
11                  composition and species richness, predator/prey interactions, nutrient cycling and energy
12                  flow; however, Pb is often found coexisting with other metals and other stressors, which
13                  risk confounding the observed effects. Recent studies provide evidence in additional
14                  habitats for these community and ecological-level effects, specifically in aquatic
15                  macrophyte communities and sediment-associated communities. Different species may
16                  exhibit different responses to Pb-impacted ecosystems dependent not only upon other
17                  environmental factors (e.g., temperature, pH), but also on species sensitivity, lifestage, or
18                  seasonally-affected physiological state. Aquatic ecosystems with low pH and low
19                  dissolved organic matter are likely to be the most sensitive to the effects of
20                  atmospherically-deposited Pb.


                    Biological Effects of Pb on Saltwater Plants, Invertebrates and Vertebrates

21                  In general, Pb toxicity to marine/estuarine plants, invertebrates and vertebrates is less
22                  well characterized than toxicity to Pb in freshwater systems due to an insufficient
23                  quantity of studies on saltwater organisms. In marine algae, effects on growth are
24                  observed in the most sensitive species at Pb concentrations that exceed amounts
25                  measured in the open sea or estuaries (Table 2-1). The majority of available studies of Pb
26                  effects on saltwater organisms are for invertebrate species. Evidence for Pb effects on
27                  reproduction, growth and survival as well as neurobehavioral, hematological and
28                  physiological stress endpoints are coherent with findings in freshwater invertebrates
29                  although most effects are observed at concentrations above  100 (ig Pb/L which exceeds
30                  Pb typically encountered in seawater (Table 2-1). Fewer studies are available for Pb in
31                  marine sediments. In the amphipod, Elasmopus laevis, onset to reproduction was
32                  significantly delayed at  118 mg/Pb kg sediment; a concentration that the authors indicate
33                  is below the current marine sediment regulatory guideline for Pb (218 mg Pb/kg
34                  sediment) (Ringenary et al., 2007; NOAA. 1999). In the same study, no effects of Pb on
35                  adult survival in 28-day or 60-day sediment exposures were observed. Additional studies
36                  on reproduction, growth, and survival in marine invertebrates report effects above the

      November 2012                             2-40                  Draft - Do Not Cite or Quote

-------
1                  range considered for causal determinations (Table II Preamble). Several field monitoring
2                  studies with marine bivalves have used ALAD as a biomarker for Pb exposure and
3                  correlated ALAD inhibition to increased Pb tissue content. Field and laboratory studies
4                  provide evidence for antioxidant response to Pb exposure, however, most effects are
5                  observed at concentrations of Pb that are higher than concentrations detected in marine
6                  environments. No recent evidence for effects of Pb on marine vertebrates in controlled
7                  exposures was available for review.
     November 2012                             2-41                   Draft - Do Not Cite or Quote

-------
2.7.3
Determinations of Causality for Effects on Ecosystems
Table 2-3    Summary of Pb causal determinations for plants, invertebrates and
              vertebrates.
     Level
  Effect
Terrestrial3
Freshwater
Saltwater3
Community-
and
Ecosystem-
Level
1 Population-Level Endpoints
Sub-orgamsmal . _
_. a Organism-Level Responses
Responses a r
Community and Ecosystem Effects
(Sections 7.3.12.7, 7.4.12.7, and 7.4.21 .7)

Reproductive and Developmental Effects - Plants
(Sections 7.3.12.1 , 7.4.12.1 , and 7.4.21 .1)
Reproductive and Developmental Effects - Invertebrates
(Sections 7.3.12.1 , 7.4.12.1 , and 7.4.21 .1)
Reproductive and Developmental Effects - Vertebrates
(Sections 7.3.12.1 , 7.4.12.1 , and 7.4.21 .1)
Growth - Plants
(Sections 7.3.12.2, 7.4.12.2, and 7.4.21 .2)
Growth - Invertebrates
(Sections 7.3.12.2, 7.4.12.2, and 7.4.21 .2)
Growth - Vertebrates
(Sections 7.3.12.2, 7.4.12.2, and 7.4.21 .2)
Survival - Plants
(Sections 7.3.12.3, 7.4.12.3, and 7.4.21 .3)
Survival - Invertebrates
(Sections 7.3.12.3, 7.4.12.3, and 7.4.21 .3)
Survival - Vertebrates
(Sections 7.3.12.3, 7.4.12.3, and 7.4.21 .3)
Neurobehavioral Effects - Invertebrates
(Sections 7.3.12.4, 7.4.12.4, and 7.4.21 .4)
Neurobehavioral Effects - Vertebrates
(Sections 7.3.12.4, 7.4.12.4, and 7.4.21 .4)
Hematological Effects - Invertebrates
(Sections 7.3.12.5, 7.4.12.5, and 7.4.21 .5)
Hematological Effects - Vertebrates
(Sections 7.3.12.5, 7.4.12.5, and 7.4.21 .5)
Physiological Stress - Plants
(Sections 7.3.12.6, 7.4.12.6, and 7.4.21 .6)
Physiological Stress - Invertebrates
(Sections 7.3.12.6, 7.4.12.6, and 7.4.21 .6)
Physiological Stress - Vertebrates
(Sections 7.3.12.6, 7.4.12.6, and 7.4.21 .6)
Likely Causal
Inadequate
Causal
Causal
Causal
Likely Causal
Inadequate
Inadequate
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Causal
Causal
Likely Causal
Causal
Inadequate
Inadequate
Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Inadequate
Suggestive
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Suggestive
Inadequate
Inadequate
Suggestive
Inadequate
"Based on the weight of evidence for causal determination in Table II of the ISA Preamble. Ecological causal determinations are
based on doses or exposures generally within one to two orders of magnitude of the range of Pb currently measured in the
environment (Table 2-1).
November 2012
                             2-42
      Draft - Do Not Cite or Quote

-------
                     2.7.3.1      Effects on Development and Reproduction

 1                   Evidence from invertebrate and vertebrate studies from Pb AQCDs and in this review
 2                   indicates that Pb is affecting reproductive performance in multiple species. Various
 3                   endpoints have been measured in multiple taxa of terrestrial and aquatic organisms to
 4                   assess the effect of Pb on development, fecundity, and hormone homeostasis, and they
 5                   have demonstrated the  presence of adverse effects. Reproductive effects are important
 6                   when considering effects of Pb because impaired fecundity at the organism level of
 7                   biological organization can result in a decline in abundance and/or extirpation of
 8                   populations, decreased taxa richness, and decreased relative or absolute abundance at the
 9                   community level (Suter et al.. 2005; U.S. EPA. 2003a). The evidence is sufficient to
10                   conclude that there is a causal relationship between Pb exposures and developmental and
11                   reproductive effects in  terrestrial (Section 7.3.12.1) and freshwater (Section 7.4.12.1)
12                   invertebrates and vertebrates. Although there is less evidence for reproductive and
13                   developmental effects of Pb in marine systems, available evidence is suggestive of a
14                   causal relationship between Pb exposure and reproductive and  developmental effects in
15                   saltwater invertebrates  (Section 7.4.21.1). The evidence is inadequate to conclude that
16                   there is a causal relationship between Pb exposures and developmental and reproductive
17                   effects in saltwater vertebrates and in either terrestrial or aquatic plants.

18                   Recent evidence for developmental and reproductive endpoints in terrestrial invertebrates
19                   shown to be affected by Pb include hatching success in collembolans, increased
20                   development time in fruit flies and aphids, and disrupted hormone homeostasis in moths.
21                   These studies have generally used Pb concentrations that exceed Pb soil concentrations
22                   found at most U.S. locations (Table 2-1). but many of them included multiple
23                   concentrations, and responses increased with increasing concentration. In terrestrial
24                   vertebrates, recent evidence for decreased sperm count and quality in deer at a location
25                   contaminated by mining, and for decreased testis weight in lizards, support previous
26                   associations between Pb exposure and reproductive and developmental  effects. Few
27                   studies are available that specifically address reproductive effects of Pb exposure in either
28                   terrestrial or aquatic plants.

29                   In terrestrial invertebrates, Pb can alter developmental timing, hatching  success, sperm
30                   morphology, and hormone homeostasis.  In fruit flies, Pb exposure increased time to
31                   pupation and decreased pre-adult development. Sperm morphology was altered in
32                   earthworms exposed to Pb-contaminated soils. Pb may also disrupt hormonal
33                   homeostasis in invertebrates as studies with moths have suggested. Evidence of
34                   multi-generational toxicity of Pb is also present in terrestrial  invertebrates, specifically
35                   springtails, mosquitoes, carabid beetles, and nematodes where  decreased fecundity in
      November 2012                             2-43                  Draft - Do Not Cite or Quote

-------
 1                  progeny of Pb-exposed individuals was observed. However, effects have only been
 2                  studied in a small number of species.

 3                  For freshwater invertebrates, exposure to Pb under controlled conditions has provided
 4                  evidence for reproductive effects on sensitive taxa (gastropods, amphipods, cladocerans)
 5                  at or near the range of Pb concentration values available for U.S. surface waters (Table
 6                  2-1). Reproductive effects were reported to begin at 19 (ig Pb/L for the snail Lymnaea
 7                  palustris and 27 (ig Pb/L for Daphnia sp. as reported in the 1986 Pb AQCD (U.S. EPA.
 8                  1986b). In a 42-day chronic study reviewed in the 2006 Pb AQCD, the LOEC for
 9                  reproduction was 3.5 (ig Pb/L in amphipods receiving both waterborne and dietary Pb
10                  (Besser et al.. 2005).  Several recent studies of snails, clams, and rotifers support previous
11                  findings of adverse impacts to embryonic development. Reproductive effects have also
12                  been observed in multi-generational studies with aquatic invertebrates. Larval settlement
13                  rate and rate of population increase was decreased in rotifers and marine amphipods.
14                  Rotifers have a reduced fertilization rate associated with Pb exposure that appears to be
15                  due to decreased viability of sperm and eggs.

16                  In freshwater vertebrates there is evidence for reproductive and developmental effects of
17                  Pb. Recent evidence in frogs and freshwater fish continues to support developmental and
18                  reproductive effects of Pb in aquatic vertebrates reported in earlier Pb AQCDs.
19                  Pb-exposure in tadpoles has been demonstrated to delay metamorphosis, decrease larval
20                  size, produce subtle skeletal malformations, and to result in slower swimming speed.
21                  Previous Pb AQCDs  have reported developmental effects in fish, specifically spinal
22                  deformities in larvae  at a concentration of 120 (ig Pb/L. In the 2006 Pb AQCD, decreased
23                  spermatocyte development in rainbow trout was observed at 10 (ig Pb/L and in fathead
24                  minnow testicular damage occurred at 500  (ig Pb/L. In more recent studies, reproduction
25                  in fathead minnows was affected in breeding exposures following 300-day chronic
26                  toxicity testing. However, reproductive performance was unaffected in zebrafish Danio
27                  rerio exposed to Pb-contaminated prey. In fish, there is recent evidence of Pb effects on
28                  steroid profiles from nominal exposure studies.

29                  In terrestrial vertebrates, evidence from Pb AQCDs and more recent evidence support an
30                  association between Pb exposure and  observed adverse reproductive effects. In mammals,
31                  few studies in the field have addressed Pb specifically: most available data in wild or
32                  grazing animals are from near smelters, where animals are co-exposed to other metals.
33                  Evidence obtained using mammals in the context of human health  research demonstrates
34                  adverse effects of Pb on sperm, and on onset of puberty in males and females (Chapter_5),
35                  which is coherent with the partial evidence from mammals in the wild. Other
36                  reproductive endpoints including spontaneous abortions, pre-term birth, embryo
37                  development, placental development,  low birth weight, subfecundity, hormonal changes,
      November 2012                             2-44                   Draft - Do Not Cite or Quote

-------
 1                   and teratology were also affected, but less consistently. Recent toxicological data from
 2                   animal studies support trans-generational effects, a finding that is also an area of
 3                   emerging interest in ecology.

 4                   Many studies of effects on reproductive and developmental endpoints in terrestrial
 5                   invertebrates and vertebrates have been conducted with soil Pb concentrations exceeding
 6                   those found in most U.S. locations.  Recent and past studies that include multiple,
 7                   increasing concentrations of Pb, from background level to levels greater than those
 8                   associated with  heavily contaminated sites, showed exposure-dependent responses. For
 9                   some aquatic species, recent evidence supports previous findings of reproductive and
10                   developmental effects of Pb and differential lifestage response at or near concentrations
11                   of Pb reported from natural environments.
                     2.7.3.2      Effects on Growth

12                   Alterations in the growth of an organism can impact population, community and
13                   ecosystem level variables. Evidence is sufficient to conclude that there is a causal
14                   relationship between Pb exposures and effects on growth in terrestrial plants (Section
15                   7.3.12.2) and freshwater invertebrates (Section 7.4.12.2). Evidence is sufficient to
16                   conclude that a causal relationship is likely to exist between Pb exposure and effects on
17                   growth in terrestrial invertebrates and freshwater plants. Evidence is inadequate to
18                   establish a causal relationship between Pb exposures and effects on growth in terrestrial
19                   and aquatic vertebrates  and saltwater biota (Section 7.4.21.2).

20                   Evidence for effects of Pb on growth is strongest in terrestrial plants. In invertebrates,
21                   evidence for effects of Pb on growth has been observed most extensively in freshwater
22                   taxa, with inhibition in sensitive species occurring in or near the range of Pb
23                   concentration values found in surveys of U.S. surface waters (Table 2-1). Vertebrates,
24                   particularly terrestrial, have been the object of a comparatively much smaller number of
25                   studies of the effects of Pb on growth. Growth effects observed in both invertebrates and
26                   vertebrates, however, underscore the importance of lifestage to overall Pb susceptibility.
27                   In general, juvenile organisms are more sensitive than adults. Evidence for growth effects
28                   of Pb in freshwater and terrestrial plant species is primarily supported by earlier Pb
29                   AQCDs. In aquatic invertebrates, the weight of the evidence continues to support growth
30                   effects of Pb with several recent studies reporting effects at < 10 (ig Pb/L, specifically in
31                   snails and mussels. Also, growth effects in frogs are reported at lower concentrations in
32                   the current document than in earlier reviews.

33                   There is evidence over several decades of research that Pb inhibits photosynthesis and
34                   respiration in plants, both of which reduce growth (U.S. EPA. 2006b. 1977). Many

      November 2012                              2-45                   Draft - Do Not Cite or Quote

-------
 1                  toxicity studies conducted in laboratory and greenhouse settings have reported effects on
 2                  plants. These effects are typically observed in laboratory studies with high exposure
 3                  concentrations or in field studies near stationary sources and heavily contaminated sites,
 4                  but studies that include multiple concentrations of Pb show increased response with
 5                  increasing concentration. Pb has been shown to affect photosystem II, altering the
 6                  pigment structure, and decreasing the efficiency of visible light absorption by affected
 7                  plants. Decreases in chlorophyll a and b content have been observed in various algal and
 8                  plant species. Most primary producers experience EC50 values for growth in the range of
 9                  1,000 to 100,000 (ig Pb/L with minimal inhibition of growth observed as low as
10                  100 ng Pb/L (U.S. EPA. 2006b).

11                  Growth effects of Pb on aquatic invertebrates are reviewed in the draft Ambient Aquatic
12                  Life Water Quality Criteria for Pb (U.S. EPA. 2008b) and the 2006 Pb AQCD (U.S.
13                  EPA. 2006b). In the 2006 Pb AQCD, the LOEC for growth of freshwater amphipods
14                  Hyalella azteca in 42-day chronic exposure to Pb  was 16 (ig Pb/L (Besser et al.. 2005).
15                  Recent studies provide additional evidence for effects on growth of aquatic invertebrates
16                  at < 10 (ig Pb/L. Growth of juvenile freshwater snails L. stagnalis was inhibited below
17                  the lowest concentration tested resulting in an EC2o <4 (ig Pb/L (Grosell and Brix. 2009;
18                  Grosell et al.. 2006b). In the  same study, the NOEC was 12 jig Pb/L  and the LOEC was
19                  16 (ig Pb/L. The authors indicated that the observed effect level  for Pb was very close to
20                  the current U.S. EPA water quality  criteria for Pb  (3.3 (ig Pb/L normalized to test water
21                  hardness) (Grosell and Brix. 2009). In the freshwater mussel, fatmucket (L. siliquoided)
22                  juveniles were the most sensitive lifestage (Wang et al.. 2010e).  A chronic value of
23                  10 (ig Pb/L in a 28-day exposure of 2-month-old fatmucket juveniles was the lowest
24                  genus mean chronic value ever reported for Pb. Growth effects are also reported in
25                  marine invertebrates at higher concentrations of Pb than sensitive freshwater
26                  invertebrates.

27                  In Pb AQCDs, a few studies have reported growth effects of Pb on vertebrates including
28                  fish (growth inhibition), birds (changes in juvenile weight gain), and frogs (delayed
29                  metamorphosis, smaller larvae). A recent study reviewed in this  ISA supports findings of
30                  growth effects in frogs and suggests that these effects may be occurring at lower
31                  concentrations than previously reported: the growth rate of tadpoles of the northern
32                  leopard frog exposed nominally to 100 (ig Pb/L from embryo to  metamorphosis was
33                  slower than the growth rate of the controls (Chen et al.. 2006b). In this study, Pb
34                  concentrations in the tissues  of tadpoles were quantified and the  authors reported that
35                  they were within the range of reported tissue concentrations reported in wild-caught
36                  populations. Reports of Pb-associated growth effects in fish are inconsistent.
      November 2012                             2-46                  Draft - Do Not Cite or Quote

-------
                     2.7.3.3      Effects on Survival

 1                   Survival is a biologically important response that may have a direct impact on population
 2                   size and can lead to effects at the community and ecosystem level of biological
 3                   organization. The evidence is sufficient to conclude that there is a causal relationship
 4                   between Pb exposures and survival in terrestrial invertebrates (Section 7.3.12.3) and
 5                   freshwater invertebrates and vertebrates (Section 7.4.12.3). Evidence is sufficient to
 6                   conclude that a causal relationship is likely to exist between Pb exposure and survival in
 7                   terrestrial vertebrates (Section 7.3.12.3). The evidence is inadequate to conclude that
 8                   there is a causal relationship between Pb exposure and survival in terrestrial and
 9                   freshwater plants (Section 7.3.12.3. and Section 7.4.12.3). as well as in all saltwater biota
10                   (Section 7.4.21.3). There is evidence for mortality in saltwater organisms at
11                   concentrations that greatly exceed Pb concentrations typically encountered in seawater.
12                   In general, marine  organisms are less sensitive to Pb than freshwater species.

13                   In terrestrial vertebrates, evidence for Pb effects on survival is primarily supported by Pb
14                   AQCDs with no recent studies reporting effects on survival at lower concentrations. For
15                   aquatic invertebrates recent studies support previous associations between Pb exposure
16                   and mortality at concentrations near the range of Pb concentration values available for
17                   U.S. surface waters (Table 2-1) in cladocerans, amphipods, and rotifers.  In aquatic
18                   vertebrates, there is recent evidence  for effects in fish at <100 (ig Pb/L. Pb is generally
19                   not phytotoxic to freshwater or terrestrial plants at concentrations found in the
20                   environment away from stationary sources and heavily contaminated sites, probably due
21                   to the fact that plants often sequester large amounts of Pb in roots, and that translocation
22                   to other parts of the plant is limited.

23                   The relationship between Pb exposure and decreased survival rate has been well
24                   demonstrated in terrestrial and aquatic species,  as presented in Sections 7.3.12.3.
25                   7.4.12.3. 7.4.21.3.  and in previous Pb AQCDs.  Toxicological studies have established
26                   LC50 values for some species of plants, invertebrates, and vertebrates.  From the LC50 data
27                   on Pb in this review and previous Pb AQCDs, a wide range of sensitivity to Pb is evident
28                   across taxa. LC50 values are usually much higher than Pb concentrations near
29                   contaminated sites, although physiological dysfunction that adversely impacts the fitness
30                   of an organism often occurs well below concentrations that result in mortality.

31                   Freshwater aquatic invertebrates  are generally more sensitive to Pb  exposure than other
32                   taxa, with survival adversely impacted in a few species at concentrations near typical
33                   ambient levels. Freshwater biota  that exhibit sensitivity to Pb in the upper range of Pb
34                   concentrations measured in U.S.  waters [median 0.50 (ig Pb/L, range 0.04 to 30 (ig Pb/L
35                   (U.S. EPA. 2006b)1. include some species of gastropods, amphipods, cladocerans, and
36                   rotifers although the toxicity of Pb is highly dependent upon water quality variables such

      November 2012                              2-47                  Draft - Do Not Cite or Quote

-------
 1                  as DOC, hardness, and pH. For example, amphipods tested under various water
 2                  conditions exhibited sensitivity to Pb at <10 (ig Pb/L (U.S. EPA. 2006c) and the present
 3                  document). These impacted species may include endangered species or candidates for the
 4                  endangered species list, such as the freshwater mussel Lampsilis rafmesqueana (Neosho
 5                  mucket). The EC50 for foot movement (a measure of viability) for newly transformed
 6                  juveniles of this species was 188 (ig Pb/L.  Other aquatic invertebrates such as odonates
 7                  may be tolerant of Pb concentrations that greatly exceed Pb detected in aquatic
 8                  ecosystems.

 9                  Terrestrial invertebrates typically tolerate high concentrations of Pb relative to
10                  concentrations found in most uncontaminated soils. In the 1986 Pb AQCD it was reported
11                  that Pb at environmental concentrations occasionally found near roadsides and smelters
12                  (10,000 to 40,000 (ig Pb/g dry weight [mg Pb/kg]) can eliminate populations of bacteria
13                  and fungi on leaf surfaces and in soil. LC50 values for soil nematodes vary from
14                  10-1,550 mg Pb/kg dry weight dependent upon soil organic matter content and soil pH
15                  (U.S. EPA. 2006b). In earthworms, 14 and 28 day LC50 values typically fall in the range
16                  of 2,400-5,800 mg Pb/kg depending upon the species tested.

17                  Data on mortality of saltwater species associated with exposure to Pb is limited; however,
18                  in general, marine organisms are less sensitive to this metal than freshwater organisms
19                  and the highest toxicity is observed in juveniles. In one study, effects of Pb on survival at
20                  environmentally relevant concentrations of Pb in diet have been demonstrated through a
21                  simulated marine food chain in which the primary producer, the microalgae Tetraselmis
22                  suecica, was exposed nominally to 20 (ig Pb/L and subsequently fed to brine shrimp
23                  Artemiafranciscana, (mean Pb content 12 to 15 (ig Pb/g) which were consumed by
24                  white-leg shrimp Litopenaeus vannamei, itself consumed by grunt fish Haemulon
25                  scudderi representing the top of the marine food chain (Soto-Jimenez et al..  201 Ib).
26                  Survival of brine shrimp was 25 to 35% lower than the control and both white shrimp and
27                  grunt fish had significantly higher mortalities than controls.

28                  In vertebrates, toxicity is observed in terrestrial avian and mammalian species in
29                  laboratory studies over a wide range of doses (<1 to >1,000 mg Pb/kg body  weight-day)
30                  as reviewed for the guidance and development of ecological soil screening levels
31                  (Eco-SSLs) (U.S. EPA. 2005b). The NOAELs for survival ranged from 3.5 to 3,200 mg
32                  Pb/kg • day. In freshwater vertebrates there is considerable historic information on Pb
33                  toxicity to fish. Recent studies support earlier AQCD findings of Pb effects on fish
34                  survival and indicate effects at lower concentrations when testing with juvenile lifestages.
35                  In a series of 96-hour acute toxicity tests with fathead minnow conducted in a variety of
36                  natural waters across North America, LC50 values ranged from 41 to 3,598 (ig Pb/L and
37                  no Pb toxicity occurred in three highly alkaline waters (Esbaugh et al., 2011). Thirty day
      November 2012                             2-48                  Draft - Do Not Cite or Quote

-------
 1                  LC50 values for larval fathead minnows ranged from 39 to 1,903 (ig Pb/L in varying
 2                  concentrations of DOC, calcium sulfate (CaSO4), and at various pH values (Grosell et al.
 3                  2006b). In a recent study of rainbow trout fry at 2-4 weeks post-swim up, the 96-hour
 4                  LC50 was 120 (ig Pb/L at a hardness of 29 mg/L as calcium carbonate (CaCO3), a value
 5                  much lower than in testing with older fish (Mebane et al., 2008). In the same study, two
 6                  chronic (>60 day) tests were conducted with rainbow trout and the NOECs for survival
 7                  were 24 and 87 (ig Pb/L and the LOECs were 54 and 125 (ig Pb/L, respectively.
                    2.7.3.4     Neurobehavioral Effects

 8                  Overall, the evidence from terrestrial and freshwater systems is sufficient to conclude that
 9                  a causal relationship is likely to exist between Pb exposures and neurobehavioral effects
10                  in invertebrates and vertebrates (Sections 7.3.12.4 and 7.4.12.4). Evidence is inadequate
11                  to conclude that there is a causal relationship between Pb exposure and neurobehavioral
12                  endpoints in saltwater species (Section 7.4.21.4).

13                  Observations from laboratory studies reported in ChapterJ? and previous Pb AQCDs have
14                  shown adverse effects of Pb on neurological endpoints in both terrestrial and freshwater
15                  animal taxa. Studies that consider mode-of-action and molecular targets of Pb toxicity in
16                  biota are now available for a few species. Recent studies have continued adding to the
17                  evidence from both invertebrate and vertebrate studies that Pb adversely affects behaviors
18                  such as food consumption, avoidance, and escape  from predators, behavioral
19                  thermoregulation, and prey capture. These changes are likely to decrease the overall
20                  fitness of the organism. Recent evidence includes  reports of behavioral responses across a
21                  larger variety of organisms including fish larvae born from Pb-exposed adults and
22                  reptiles, while some impairments in feeding and escaping behaviors were reported for the
23                  first time.

24                  Central nervous system effects in fish recognized in previous Pb AQCDs include effects
25                  on spinal neurons and brain receptors. Recent evidence from this review identifies
26                  possible molecular targets for Pb neurotoxicity in  fish. Additionally, there is recent
27                  evidence for neurotoxic action of Pb in invertebrates with exposure to Pb observed to
28                  cause changes in the morphology of gamma aminobutyric acid (GABA)-motor neurons
29                  in nematodes (Caenorhabditis elegans) (Du and Wang. 2009).

30                  Decreased food consumption of Pb-contaminated  diet has been demonstrated in some
31                  invertebrates (snails) and vertebrates (lizards, pigs, fish). Behavioral effects in grunt fish
32                  H. scudderi, occupying the top level of a simulated marine  food chain included lethargy
33                  and decreased food intake in a 42-day feeding study (Soto-Jimenez et al.. 201 Ib). These
34                  fish were fed white shrimp exposed to Pb via brine shrimp that were initially fed

      November 2012                             2-49                   Draft - Do Not Cite or Quote

-------
 1                  microalgae cultured nominally at 20 (ig Pb/L. In the same study, surfacing, reduction of
 2                  motility, and erratic swimming were observed in the white shrimp after 30 days of
 3                  exposure to Pb via diet. Pb may also decrease the ability of an organism to capture prey
 4                  or escape predation. For example, Pb exposure has been demonstrated to adversely affect
 5                  prey capture ability of certain fungal and fish species, and the motility of nematodes was
 6                  adversely affected in Pb-contaminated soils (Wang and Xing. 2008). Prey capture ability
 7                  was decreased in 10-day-old fathead minnows born from adult fish exposed to
 8                   120 (ig Pb/L for 300 days, then subsequently tested in a 21-day breeding assay (Mager et
 9                  al.. 2010). Altered pattern of escape swimming in larval zebrafish exposed to Pb as
10                  embryos was reported at low nominal concentrations of Pb (2 and 6 (ig  Pb/L). Other
11                  behavioral effects of Pb observed in fish include increased hyperactivity and decreased
12                  ability to detect visual contrast. In a laboratory study, Pb-exposed gull chicks exhibited
13                  abnormal behaviors such as decreased walking, erratic behavioral thermoregulation and
14                  food begging that could make them more vulnerable in the wild (Burger and Gochfeld.
15                  2005). The chicks were exposed to Pb via injection to produce  feather Pb concentration
16                  approximately equivalent to those observed in wild gulls. Lizards exposed to Pb through
17                  diet in the laboratory exhibited abnormal coloration and posturing behaviors. These
18                  findings show strong coherence with  findings from studies in laboratory animals that
19                  show that Pb induces changes in learning and memory (Section 5.3.2.3). as well as
20                  attention (Section 5.3.3) and motor function (Section 5.3.8).
                     2.7.3.5      Hematological Effects

21                   Based on observations in both terrestrial and freshwater organisms and additionally
22                   supported by toxicological and epidemiological findings in laboratory animals and
23                   humans, evidence is sufficient to conclude that there is a causal relationship between Pb
24                   exposures and hematological effects in terrestrial and aquatic vertebrates (Sections
25                   7.3.12.5 and 7.4.12.5). The evidence is sufficient to conclude that a causal relationship is
26                   likely to exist between Pb exposures and hematological effects in freshwater
27                   invertebrates and inadequate to conclude that there is a causal relationship between Pb
28                   exposures and hematological effects in terrestrial invertebrates. Limited evidence from
29                   marine invertebrates is suggestive of a causal relationship between Pb exposures and
30                   hematological effects  (Section 7.4.21.5). The mode of action of Pb on ALAD activity is
31                   likely mediated through a common pathway in terrestrial, freshwater and saltwater
32                   organisms.

33                   Recent studies add support to the strong body of evidence presented in Pb AQCDs that
34                   Pb exposure is associated with hematological responses in terrestrial and aquatic
35                   vertebrates. Lower ALAD activity has been significantly correlated with elevated blood


      November 2012                             2-50                  Draft - Do Not Cite or Quote

-------
 1                  Pb levels in fish and mammals. In the 1986 Pb AQCD, decreases in RBC ALAD activity
 2                  following Pb exposure were well documented in birds and mammals (U.S. EPA. 1986a).
 3                  The draft Ambient Aquatic Life Water Quality Criteria for Pb summarized several studies
 4                  of ALAD activity in fish (U.S. EPA. 2008b). Further evidence from the 2006 Pb AQCD
 5                  and this review suggests that this enzyme is an indicator for Pb exposure across a wide
 6                  range of taxa. Since the 2006 Pb AQCD, evidence of Pb effects on ALAD activity has
 7                  been found in additional species of invertebrates and fish, and has been identified in
 8                  bacteria. ALAD activity has been shown to vary greatly between species. In addition to
 9                  consideration of ALAD activity, there is recent evidence for deceased white blood cell
10                  counts in amphibians affected by Pb exposure. The consistency and coherence of these
11                  findings of effects on ALAD activity in vertebrates are also supported by some evidence
12                  of Pb-induced alterations of blood chemistry in fish reported in the 2006 Pb AQCD (U.S.
13                  EPA. 2006b). This evidence  is strongly coherent with observations from human
14                  epidemiologic and animal toxicology studies where a causal relationship was identified
15                  between Pb exposure and decreased RBC survival and function, and altered heme
16                  synthesis in humans and laboratory animals (Sections 2.6.5 and 5.7).

17                  In environmental assessments of metal-impacted habitats, ALAD is a recognized
18                  biomarker of Pb exposure in invertebrates as well as vertebrates  (U.S. EPA. 2006b).
19                  Recent field studies of ALAD activity include observations in songbirds and owls near
20                  historical mining areas and in bivalves collected from freshwater and estuarine
21                  environments. Evidence for hematological effects of Pb in saltwater invertebrates is
22                  limited primarily to field monitoring studies with bivalves.
                    2.7.3.6     Effects on Physiological Stress

23                  Evidence is sufficient to conclude that there is a causal relationship between Pb
24                  exposures and physiological stress in terrestrial plants  (Section 7.3.12.6). Evidence is
25                  sufficient to conclude that a causal relationship is likely to exist between Pb exposures
26                  and physiological stress in terrestrial invertebrates and vertebrates (Section 7.3.12.6) as
27                  well as freshwater plants, invertebrates and vertebrates (Section 7.4.12.6). Further
28                  evidence in saltwater invertebrates is suggestive of a causal relationship between Pb
29                  exposures and physiological stress (Section 7.4.21.6). Evidence is inadequate to conclude
30                  that there is a causal relationship between Pb exposure and physiological stress responses
31                  in saltwater plants and vertebrates.

32                  Endpoints associated with physiological stress received no consideration prior to the
33                  2006 Pb AQCD. Studies reviewed in the 2006 Pb AQCD reported stress-related effects
34                  including upregulation of antioxidant  enzymes and increased lipid peroxidation (U.S.
      November 2012                             2-51                   Draft - Do Not Cite or Quote

-------
 1                   EPA. 2006b). Recent evidence in additional species of terrestrial and freshwater plants,
 2                   invertebrates and vertebrates support, and expand upon findings in the previous Pb
 3                   AQCD. Some of these studies report findings within one to two orders of magnitude of
 4                   the range of Pb concentrations measured in terrestrial and freshwater environments
 5                   (Table 2-1). Recent studies include evidence for production of reactive oxygen species in
 6                   terrestrial plant species and in freshwater algae and fish in response to Pb exposure.

 7                   In the current document and the 2006 Pb AQCD  (U.S. EPA. 2006b). there is strong
 8                   evidence of upregulation of antioxidant enzymes and increased lipid peroxidation
 9                   associated with Pb exposure in many species of plants, invertebrates and vertebrates. In
10                   plants, increases of antioxidant enzymes with Pb exposure occur in algae, aquatic mosses,
11                   floating and rooted aquatic macrophytes, and terrestrial species. Most observations of
12                   antioxidant responses in plants typically occur at concentrations of Pb higher than found
13                   in the environment. However, in a few terrestrial plant species, increases of antioxidant
14                   enzymes occur at concentrations approaching the average Pb concentrations in U.S. soils
15                   (Table 2-1) and limited transplantation studies with aquatic plants indicate elevated
16                   antioxidant enzyme activity associated with Pb levels measured in sediments at polluted
17                   sites. There is evidence for antioxidant activity in invertebrates, including gastropods,
18                   mussels, and crustaceans, in response to Pb exposure.  Some recent evidence for
19                   invertebrate antioxidant responses in freshwater bivalves, and marine bivalve and
20                   crustacean species indicates effects at Pb concentrations associated with polluted sites.
21                   Markers of oxidative damage are also observed in fish, amphibians, and mammals, both
22                   in the laboratory and in exposed natural environments. Across all biota, there are
23                   differences in the induction of antioxidant enzymes that appear to be species-dependent.

24                   Additional stress responses observed in terrestrial and freshwater invertebrates include
25                   elevated heat shock proteins, osmotic stress and decreased glycogen levels. Heat shock
26                   protein induction by Pb exposure has been observed in zebra mussels and mites. Tissue
27                   volume regulation is adversely affected in freshwater crabs and glycogen levels
28                   decreased in freshwater snails following Pb exposure. Although correlated with Pb
29                   exposure, these responses are non-specific and may be altered with exposure to any
30                   number of environmental stressors.

31                   Upregulation of antioxidant enzymes and increased lipid peroxidation are considered to
32                   be reliable biomarkers of stress, and suggest that Pb exposure induces a stress response in
33                   those organisms, which may increase susceptibility to other stressors and reduce
34                   individual fitness. The oxidative stress responses associated with Pb exposure are
35                   consistent in terrestrial biota and in freshwater organisms. Furthermore, these responses
36                   are also observed in  experimental animal studies.
      November 2012                              2-52                   Draft - Do Not Cite or Quote

-------
                     2.7.3.7      Community and Ecosystem Effects

 1                   More evidence for Pb toxicity to terrestrial and aquatic biota has been reported from
 2                   single-species assays in laboratory studies than from whole ecosystem studies. The
 3                   evidence is strong for effects of Pb on growth, reproduction, and survival in very diverse
 4                   species, but considerable uncertainties exist in generalizing effects observed under
 5                   particular, small-scale conditions, up to the ecosystem level of biological organization. At
 6                   the ecosystem level, the presence of multiple stressors, variability in field conditions, and
 7                   differences in bioavailability of Pb make it difficult to measure the magnitude of effects,
 8                   and to quantify relationships between ambient concentrations of Pb and ecosystem
 9                   response. However, the cumulative evidence that has been reported for Pb effects  at
10                   higher levels of biological organization and for endpoints in single species with direct
11                   relevance to population and ecosystem level effects (i.e., development and reproduction,
12                   growth, survival) is sufficient to conclude  that a causal relationship is likely to exist
13                   between Pb exposures and the alteration of species richness, species composition and
14                   biodiversity in terrestrial and freshwater ecosystems (Sections 7.3.12.7 and 7.4.12.7).
15                   Evidence is inadequate to conclude that there is a causal relationship between Pb
16                   exposure and effects at higher levels  of biological organization in saltwater ecosystems
17                   (Section 7.4.21.7).

18                   Ecosystem-level studies in situ  are complicated by the frequent confounding of Pb
19                   exposure in Pb-polluted sites with other factors such as other trace metals and acidic
20                   deposition. In those natural systems,  Pb is often found co-existing with other stressors,
21                   and observed effects may be due to cumulative toxicity. In laboratory studies and
22                   simulated ecosystems, where it is possible to isolate the effect of Pb, this metal has been
23                   shown to alter competitive behavior of species, predator-prey interactions, and
24                   contaminant avoidance. At higher levels of biological organization, these effects may
25                   change species abundance and community structure. Uptake of Pb into aquatic and
26                   terrestrial organisms and its effects on survival, growth, and reproductive endpoints at the
27                   organism level are expected to have ecosystem-level consequences. Where evidence of
28                   effects is observed  at the ecosystem level of organization, evidence from lower levels
29                   brings consistency  and plausibility for causality.

30                   Most direct evidence of community and ecosystem level effects is from near stationary
31                   sources and contaminated sites  where Pb concentrations are higher than typically
32                   observed in the environment. For terrestrial systems, evidence of impacts on natural
33                   ecosystems near smelters, mines, and other industrial sources of Pb has been assembled
34                   in previous decades. Those impacts include decreases in species diversity and changes in
35                   floral and faunal community composition. For freshwater systems, the literature focuses
36                   on evaluating ecological stress from Pb originating from urban and mining effluents
      November 2012                              2-53                   Draft - Do Not Cite or Quote

-------
 1                  rather than atmospheric deposition. Some organisms exhibit contaminant avoidance
 2                  behaviors when exposed to Pb-contaminated areas. For example, snails and fish avoid
 3                  higher concentrations of Pb while frogs and toads lack avoidance response. Recent
 4                  evidence, published since the 2006 Pb AQCD indicates that some species of worms will
 5                  avoid Pb-contaminated soils (Langdon et al., 2005).These dynamics are likely to change
 6                  species abundance and community structure at higher levels of biological organization.

 7                  Recent studies continue to demonstrate associations between Pb exposures and effects at
 8                  higher levels of biological organization that were shown in field and microcosm studies
 9                  in previous Pb AQCDs. Recent studies on plant and soil microbial communities and
10                  sediment-associated and aquatic plant communities increase the total number of types of
11                  ecological associations impacted by Pb. In terrestrial ecosystems, most studies show
12                  decreases in microorganism abundance, diversity, and function with increasing soil Pb
13                  concentration. Specifically, shifts in nematode communities, bacterial species, and fungal
14                  diversity have been observed. Furthermore, presence of arbuscular mycorrhizal fungi
15                  may protect plants growing in Pb-contaminated soils. Increased plant diversity
16                  ameliorated effects of Pb contamination on a microbial community.

17                  In aquatic ecosystems, Pb effects reviewed  in the 2006 Pb AQCD (U.S. EPA. 2006b)
18                  included reduced species abundance, richness and diversity, decreased primary
19                  productivity, and altered predator-prey interactions. Since the 2006 Pb AQCD, there is
20                  further evidence for effects of Pb in sediment-associated communities in both saltwater
21                  and freshwater systems. Community structure and nematode diversity were altered in a
22                  microcosm study with marine sediments (Mahmoudi et al., 2007). Sediment-bound Pb
23                  contamination appears to differentially affect members of the benthic invertebrate
24                  community, potentially altering ecosystems dynamics in small urban streams
25                  (Kominkova and Nabelkova. 2005). Although surface water Pb concentrations in
26                  monitored streams were determined to be very low, concentrations of the metal in
27                  sediment were high enough to pose a risk to the benthic community
28                  (e-g-, 34-101 mg Pb/kg). These risks were observed to be linked to benthic invertebrate
29                  functional feeding group, with collector-gatherer species exhibiting larger body burdens
30                  of heavy metals than benthic predators and  collector-filterers.

31                  Changes to aquatic plant community composition have been observed in the presence of
32                  elevated surface water Pb concentrations. A shift toward more Pb-tolerant species is also
33                  observed in terrestrial plant communities near smelter sites (U.S. EPA, 2006b). Certain
34                  types of plants such as rooted and submerged aquatic plants may be more susceptible to
35                  aerially-deposited Pb resulting in shifts in Pb community composition. High Pb sediment
36                  concentrations are linked to shifts in amphipod communities inhabiting plant structures.
      November 2012                             2-54                   Draft - Do Not Cite or Quote

-------
      2.8        Integration of Health and  Ecological  Effects

 1                  The health and ecological effects considered for causal determination are summarized in
 2                  Table 2-4. The health outcomes were nervous system, cardiovascular, renal, immune,
 3                  effects on heme synthesis and RBC function, reproductive effects, and cancer. The
 4                  ecological endpoints considered for causal determination were: community and
 5                  ecosystem level effects, reproductive and developmental effects, growth, survival,
 6                  neurobehavioral effects, hematological effects, and physiological stress. The evidence
 7                  relating to specific ecological endpoints is also integrated across aquatic and terrestrial
 8                  habitats. Further, the substantial overlap between the ecological and health endpoints
 9                  considered in the causal determinations allowed the integration of the evidence across
10                  these disciplines.
     November 2012                             2-55                  Draft - Do Not Cite or Quote

-------
Table 2-4     Summary of causal determinations for health and ecological effects.
Outcome/Effect
Human Health
Causal Determination3
Ecological Receptors
Causal Determination3
Nervous System
Effects"
Causal Relationship: Cognition and Attention-
Related Behavior Problems in Children
Likely Causal Relationship: Neurobehavioral Effects in
Terrestrial and Freshwater Invertebrates and
Vertebrates
Cardiovascular Effects
                       Causal Relationship: Hypertension and
                       Coronary Heart Disease
                                                                   N/Ae
Renal Effects
                       Likely Causal Relationship: Reduced Kidney
                       Function
                                                                   N/Ae
Immune System         Likely Causal Relationship: Atopic and
Effects                 Inflammatory Conditions
                                                                   N/Ae
Hematologic Effects0
Causal Relationship: Heme Synthesis and
RBC Function
Causal Relationship: ALAD Activity in Terrestrial and
Freshwater Vertebrates
Likely Causal Relationship: ALAD activity in
Freshwater Invertebrates
Reproductive and
Developmental11
Causal Relationship: Development and Male
Reproductive Function
Causal Relationship: Invertebrates and Vertebrates
Cancer
                       Likely to be a causal relationship
                                                                   N/Ae
Mortality
N/Ae (The strongest evidence of Pb-induced
mortality in humans was observed for
cardiovascular disease related mortality and
this evidence was considered in determining
the causal relationship between  Pb exposure
and coronary heart disease.)
Causal Relationship: Survival Terrestrial Invertebrates
and Freshwater Invertebrates and Vertebrates
                                                                   Likely Causal Relationship: Terrestrial Vertebrates
Growth
N/Ae (There is evidence from toxicological and
epidemiologic studies of Pb effects on
postnatal growth, which was considered in
determining the causal  association between Pb
exposure and developmental effects.)
                                                                   Causal Relationship: Terrestrial Plants and Freshwater
                                                                   Invertebrates
                                                                   Likely Causal Relationship: Freshwater Plants and
                                                                   Terrestrial Invertebrates
Physiological Stress
N/Ae (In Human Health, oxidative stress was
considered as a upstream event in the modes
of action of Pb, leading downstream to various
effects. Ecological literature commonly uses
oxidative stress as a proxy indicator of overall
fitness, and thus treats it as an effect.)
Causal Relationship: Terrestrial Plants
Likely Causal Relationship: Terrestrial and Freshwater
Invertebrates and Vertebrates and Freshwater Plants
Community and
Ecosystem Level
Effects
                       N/Ae
                                            Likely Causal Relationship: Terrestrial and Freshwater
                                            Ecosystems
"Causal determinations were made within approximately 1 -2 orders of magnitude of current levels.
bln ecological receptors, the causal determination was developed considering neurobehavioral effects that can be observed in
toxicological studies of animal models and studies of ecological effects in vertebrates and invertebrates.
°The ecological evidence considered for the causal determination included ALAD activity, blood cell count, and altered serum
profiles.
dFor health effects the strongest evidence was for delayed onset of puberty and effects on sperm. In the ecological literature, a wide
range of endpoints, including embryonic development, multigenerational studies, delayed metamorphosis, and altered steroid
profiles, was considered.
eN/A, not applicable, i.e., Endpoints were not directly comparable the health and ecological evidence.
November 2012
                            2-56
           Draft - Do Not Cite or Quote

-------
      2.8.1        Modes of Action Relevant to Downstream Health and Ecological
                  Effects

 1                  The diverse health and ecological effects of Pb are mediated through multiple,
 2                  interconnected modes of action. This section summarizes the principal
 3                  cellular/subcellular effects contributing to modes of action for human health endpoints
 4                  associated with Pb exposure and the concentrations at which those effects are observed.
 5                  Then, effects of Pb observed in aquatic and terrestrial species (Section 2.7) are evaluated
 6                  along with evidence from epidemiological and laboratory animal studies to determine the
 7                  extent to which common modes of action can be inferred from the observed effects. The
 8                  rationale for this approach is that the mechanism of Pb toxicity is likely conserved from
 9                  invertebrates to vertebrates to humans in some organ systems.

10                  Each of the modes of action discussed in Section 5.2 has the potential to contribute to the
11                  development of a number of Pb-induced health effects (Table 2-4). Evidence for the
12                  majority of these modes of action is observed at low blood Pb levels in humans and
13                  laboratory animals, between 2 and 5 (ig/dL, and at doses as low as the picomolar range in
14                  animals and cells. The concentrations eliciting the modes of action (reported in Table
15                  2-5) are drawn from the available data and do  not imply that these modes of action are
16                  not acting at lower exposure levels or that these doses represent the threshold of the
17                  effect. Also, the data in presented this table does not inform regarding the exposure
18                  frequency and duration required to elicit a particular MOA.
     November 2012                             2-57                  Draft - Do Not Cite or Quote

-------
Table 2-5     MOAs, their related  health effects, and information on concentrations
                  eliciting the MOAs.
Mode of Action
[Related Health Effects
(ISA Section)]
                    Concentrations or Doses (Conditions)3
Blood Pb
Dose
Altered Ion Status
[All Health Effects of Pb]
3.5 ug/dL
(Mean in cord blood; association with cord
blood Ca2+ATPase pump activity)
Hueletal. (2008)
0.00005 uM free Pb2+
(In vitro; 30 minutes; calmodulin activation
assay)
Kern et al. (2000)
Protein Binding
[Renal (5.5).
Heme Synthesis and
RBC Function (5.7)1
17.0ug/dL
(Concurrent mean in adult workers with
wildtype metallothionein expression;
increased BP susceptibility)
Chen et al. (201 Oa)
50 uM Pb glutamate
(In vitro; 24 hours; increased nuclear protein
in neurological cell)
Klann and  Shelton (1989)
Oxidative Stress
[All Heath Effects of Pb
(Chapter^)]
5.4 ug/dL
(Concurrent mean in adult male workers;
decreased CAT activity in blood)
Conterato et al. (In Press)
0.1 uM Pb acetate
(In vitro; 48 hours; decreased cellular GSH in
neuroblastoma cells)
Chetty et al. (2005)
Inflammation
[Nervous System (5.3).
Cardiovascular (5.4).
Renal (5.5).
Immune (5.6).
Respiratory (5.6.5 and 5.9.6),
Hepatic (5.9.1)1
Among males with concurrent blood Pb
> 2.5 ug/dL
(Increased serum TNF-a and blood WBC
count)
Kim et al. (2007)
0.01 uM Pb acetate
(In vitro; 48 hours; increased cellular PGE2 in
neuroblastoma cells)
Chetty et al. (2005)
Endocrine Disruption
[Reproductive and
Developmental Effects (5.8).
Endocrine System (5.9.3),
Bone and Teeth (5.9.4)1
1.7ug/dL
(Lowest blood Pb level at which a
relationship could be detected in adult
women with both ovaries removed;
increased serum follicle stimulating hormone
[FSH])
Krieg (2007)
10 uM Pb nitrate
(In vitro; 30 minutes; displaced GHRH
binding to rat pituitary receptors)
Lau et al. (1991)
Cell Death/Genotoxicity
[Cancer (5.10).
Reproductive and
Developmental Effects (5.8).
Bone and Teeth (5.9.4)1
3.3 ug/dL
(Concurrent median in adult women;
increased rate of hypoxanthine guanine
phospho ribosyltransferase reporter gene
[HPRT] mutation frequency)
Vanetal. (2004)
0.03 uM Pb acetate
(In vitro; 18 hours; increased formation of
micronuclei)
Bonackeretal. (2005)
aThis table provides examples of studies that report effects with low doses or concentration; they are not the full body of evidence
used to characterize the weight of the evidence. In addition, the levels cited are reflective of the data and methods available and
do not imply that these modes of action are not acting at lower Pb exposure or blood Pb levels or that these doses represent the
threshold of the effect. Additionally, the blood concentrations and doses (indicating Pb exposure concentrations from in vitro
systems) refer to the concentrations and doses at which these modes of action were observed. While the individual modes of
action are related back to specific health effects sections (e.g., Nervous System, Cardiovascular), the concentrations and doses
given should not be interpreted as levels at which those specific health effects occur.
November 2012
                    2-58
    Draft - Do Not Cite or Quote

-------
 1                   Ecosystem studies have presented evidence for the occurrence of many of these modes of
 2                   action in animals, and to some degree in plants, however the connection to ecological
 3                   outcomes must usually be inferred because ecological studies are typically not designed
 4                   to address mode of action directly. The level at which Pb elicits a specific effect is more
 5                   difficult to establish in terrestrial and aquatic systems due to the influence of
 6                   environmental variables on Pb bioavailability and toxicity and substantial species
 7                   differences in Pb susceptibility.

 8                   The alteration of cellular ion status (including disruption of Ca2+ homeostasis, altered ion
 9                   transport mechanisms, and perturbed protein function through displacement of metal
10                   cofactors) appears to be the major unifying mode of action underlying all subsequent
11                   modes of action in plants, animals, and humans (Figure 5-1). Pb can interfere with
12                   endogenous cation homeostasis, necessary as a cell signal carrier mediating normal
13                   cellular functions. Pb is able to displace metal ions, such as Zn, Mg, and  Ca2+, from
14                   proteins due to the flexible coordination numbers and multiple ligand binding ability of
15                   Pb, leading to abnormal conformational changes to proteins and altered protein function.
16                   Disruption of ion transport leading to increased intracellular Ca2+ levels is due in part to
17                   the alteration of the activity of transport channels and proteins, such as Na+/K+ ATPase
18                   and voltage-sensitive Ca2+ channels. Pb can interfere with these proteins through direct
19                   competition between Pb and the native metals present in the protein metal binding
20                   domain or through disruption of proteins important in Ca2+-dependent cell signaling, such
21                   as protein kinase C (PKC) or calmodulin.

22                   This competition between metals has been reported not only in human systems, but also
23                   in fish, snails, and plants. Altered Ca2+ channel activity and binding of Pb with
24                   Na /K+ATPase in the gills offish disrupts the Na+ and Cl" homeostasis, which may lead
25                   to ionoregulatory failure and death. Ca2+  influx and ionoregulation has also been shown
26                   to be inhibited by Pb exposure in a sensitive species of snail, leading to a reduction in
27                   snail growth. In plants, substitution of the central atom of chlorophyll, Mg, by Pb
28                   prevents light-harvesting, resulting in a breakdown of photosynthesis. Pb-exposed
29                   animals also have decreased cellular energy production due to perturbation of
30                   mitochondrial function.

31                   Disruption of ion transport not only leads to altered Ca2+ homeostasis, but can also result
32                   in perturbed neurotransmitter function. Evidence for these effects in Pb-exposed
33                   experimental animals and cell cultures has been linked to altered neurobehavioral
34                   endpoints  and other neurotoxicity. Neurobehavioral changes that may decrease the
35                   overall fitness of the organism have also been observed in aquatic and terrestrial
36                   invertebrate and vertebrate studies. There is evidence in tadpoles  and fish to suggest Pb
      November 2012                              2-59                   Draft - Do Not Cite or Quote

-------
 1                   may alter neurotransmitter concentrations, possibly resulting in some of these
 2                   neurobehavioral changes.

 3                   Altered cellular ion status following Pb exposure can result in the inhibition of heme
 4                   synthesis. Pb exposure is commonly associated with altered hematological responses in
 5                   aquatic and terrestrial invertebrates, experimental animals, and human subjects. The
 6                   proteins affected by Pb are highly conserved across species accounting for the common
 7                   response seen in human health and ecological studies. This evolutionarily conserved
 8                   response to Pb is likely the result of the competition of Pb with the necessary metal
 9                   cofactors in the proteins involved in heme synthesis.

10                   Although Pb will bind to proteins within cells through interactions with side group
11                   moieties, thus potentially disrupting cellular function, protein binding of Pb may
12                   represent a mechanism by which cells protect themselves against the toxic effects of Pb.
13                   Intranuclear and intracytosolic inclusion body formation has been observed in the kidney,
14                   liver, lung, and brain following Pb exposure in experimental animals. A number of
15                   unique Pb binding proteins have been detected, constituting the observed inclusion
16                   bodies. The major Pb binding protein in blood is ALAD with carriers of the ALAD-2
17                   allele potentially exhibiting higher Pb binding affinity. Inhibition of ALAD activity  is a
18                   widely recognized response to Pb in environments where Pb is present and is considered
19                   to be biomarker of Pb exposure in both terrestrial and aquatic biota. Additionally,
20                   metallothionein is an important protein in the formation of inclusion bodies and
21                   mitigation of the toxic effects of Pb. Protein binding of Pb is a recognized mechanism of
22                   Pb detoxification in some terrestrial and aquatic biota. For example, plants can sequester
23                   Pb through binding with phytochelatin and some fish have the ability to store
24                   accumulated Pb in heat-stable proteins.

25                   A second major mode of action of Pb is the development of oxidative stress, due in many
26                   instances to the antagonism of normal metal ion functions. Disturbances of the normal
27                   redox  state of tissues can cause toxic effects and is involved in the majority of health and
28                   ecological outcomes observed after Pb exposure. The origin of oxidative  stress produced
29                   after Pb exposure is likely a multi-pathway process.  Studies in humans and experimental
30                   animals provide evidence to conclude that oxidative stress results from oxidation of
31                   5-ALA, NAD(P)H oxidase activation, membrane and lipid peroxidation,  and antioxidant
32                   enzyme depletion. Evidence of increased lipid peroxidation associated with Pb exposure
33                   exists  for many species of plants, invertebrates, and vertebrates. Enhanced lipid
34                   peroxidation can also result from Pb potentiation of Fe2+ initiated lipid peroxidation and
35                   alteration of membrane composition after Pb exposure.  Increased Pb-induced ROS will
36                   also sequester and inactivate biologically active 'NO, leading to the increased production
37                   of the  toxic product nitrotyrosine, increased compensatory NOS, and decreased sGC
      November 2012                              2-60                   Draft - Do Not Cite or Quote

-------
 1                  protein. Pb-induced oxidative stress not only results from increased ROS production but
 2                  also through the alteration and reduction in activity of the antioxidant defense enzymes.
 3                  The biological actions of a number of these enzymes are antagonized due to the
 4                  displacement of the protein functional metal ions by Pb. Increased ROS are often
 5                  followed by a compensatory and protective upregulation in antioxidant enzymes, such
 6                  that this observation is indicative of oxidative stress conditions. A number of studies in
 7                  plants, invertebrates, and vertebrates present evidence of increased antioxidant enzymes
 8                  with Pb exposure. Additionally, continuous ROS production may overwhelm this
 9                  defensive process leading to decreased antioxidant activity and further oxidative stress
10                  and injury.

11                  In a number of organ systems Pb-induced oxidative stress is accompanied by
12                  misregulated  inflammation. Pb exposure will modulate inflammatory cell function,
13                  production of proinflammatory cytokines and metabolites, inflammatory chemical
14                  messengers, and proinflammatory signaling cascades. Cytokine production is skewed
15                  toward the production of proinflammatory cytokines like TNF-a and IL-6 as well as
16                  leading to the promotion of Th2 response and suppression of Thl cytokines and
17                  Thl-related responses.

18                  Pb is a potent endocrine disrupting chemical. Steroid receptors and some endocrine
19                  signaling pathways are known to be highly conserved over a broad expanse of animal
20                  phylogeny. Pb will disrupt the HPG axis evidenced in humans, other mammals, and fish,
21                  by a decrease in serum hormone levels, such as FSH, LH, testosterone, and estradiol. Pb
22                  interacts with the hypothalamic-pituitary level hormone control causing a decrease in
23                  pituitary hormones, altered growth dynamics, inhibition of LH secretion, and reduction in
24                  StAR protein. Pb has also been shown to alter hormone receptor binding likely due to
25                  interference of metal cations in secondary messenger systems and receptor ligand binding
26                  and through generation of ROS. Pb disrupts hormonal homeostasis in invertebrates
27                  necessary for reproduction and development. Pb also may disrupt the HPT axis by
28                  alteration of a number of thyroid hormones, possibly due to oxidative stress. These
29                  studies have been conducted in humans and animals, including cattle; however the results
30                  of these studies are mixed and require further investigation.

31                  Genotoxicity  and cell death has been investigated after Pb exposure in humans, animals,
32                  plants, and cell models. High level Pb exposure to humans leads to increased DNA
33                  damage, however lower blood Pb levels have caused these effects in experimental
34                  animals and cells. Reports vary on the effect of Pb on DNA repair activity, however a
35                  number of studies report decreased repair processes following Pb exposure. There is
36                  some evidence in plants, earthworms, freshwater mussels and fish for DNA damage
37                  associated with Pb exposure.  There is evidence of mutagenesis and clastogenicity in
      November 2012                             2-61                  Draft - Do Not Cite or Quote

-------
 1                  highly exposed humans, however weak evidence has been shown in animals and cells
 2                  based systems. Human occupational studies provide limited evidence for micronucleus
 3                  formation (>10 (ig/dL), supported by Pb-induced effects in both animal and cell studies.
 4                  Micronucleus formation has also been reported in amphibians. Animal and plant studies
 5                  have also provided evidence for Pb-induced chromosomal aberrations. The observed
 6                  increases in clastogenicity may be the result of increased oxidative damage to DNA due
 7                  to Pb exposure, as co-exposures with antioxidants ameliorate the observed toxicities.
 8                  Limited evidence of epigenetic effects is available,  including DNA methylation,
 9                  mitogenesis, and gene expression. Altered gene expression may come about through Pb
10                  displacing Zn from multiple transcriptional factors, and thus perturbing their normal
11                  cellular activities. Consistently positive results have provided evidence  of increased
12                  apoptosis following Pb exposure.

13                  Overall, Pb-induced health and ecological effects can occur through a number of
14                  interconnected and evolutionarily well conserved modes of action that generally originate
15                  with the alteration of ion status.
      2.9        Policy Relevant Considerations


      2.9.1        Public Health Significance

16                  The rationale for establishing the public health significance of the various health
17                  endpoints associated with Pb exposure is multifaceted. The 2006 Pb AQCD (U.S. EPA.
18                  2006b) concluded that neurodevelopmental effects in children and cardiovascular effects
19                  in adults were among the effects best substantiated as occurring at blood Pb levels as low
20                  as 5-10 ug/dL (or possibly lower), and that these categories of effects were clearly of the
21                  greatest public health concern. The evidence reviewed in the current assessment supports
22                  and builds upon this conclusion with recent epidemiologic studies of children reporting
23                  deleterious effects in populations with lower mean blood Pb levels than previously
24                  reported. The supralinear concentration-response relationship, which is widely
25                  documented for Pb and cognitive function in children, does not provide evidence for a
26                  threshold for Pb-associated neurodevelopmental effects in the range of blood Pb levels
27                  examined (Sections 2.9.3 and 5.3.13)

28                  The World Health Organization (WHO) definition of "health" is "the state of complete
29                  physical, mental, and social well-being and not merely the absence of disease or
30                  infirmity" (WHO. 1948). By this definition, decrements in health status that are not
31                  severe enough to result in the assignment of a clinical diagnosis might reflect a decrement
32                  in the well-being of an individual.  Further, deficits in subtle indices of health or

      November 2012                             2-62                  Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5

 6
 7
 8
 9
10
11
12
well-being may not be observable except in aggregate, at the population level, so the
critical distinction between population and individual risk is essential for interpreting the
public health significance of study findings. This concept of population risk is relevant to
the interpretation of findings regarding both IQ and blood pressure in the assessment of
their public health significance.
Weiss et al. (1988) discusses concepts related to understanding the shift in a population
distribution of an IQ Score. The conceptual model described by these authors is not based
on actual data and assumes that the incremental concentration-response between Pb
exposure and IQ is similar in children with high and low intelligence. As shown in Figure
2-1 small shifts in population means are often significant from a public health
perspective. Even a small relative risk for a health effect that is highly prevalent in the
population can translate into a large increase in the number of clinical cases.
                                        Shift in Population Mean
13
14
15
16
17
18
      Figure 2-1
                                                                   Critical
                                                                   Line
                        3
                        Q.
                        O
                        Q.
                        O
                        £
                        o
                        Q.
                       Health Outcome
    The effect of a small shift in population mean on the proportion of
    individuals in the population diagnosed with clinical disease
    (i.e., the proportion to the right of the "Critical Line").

For example, small shift in the population mean IQ may result in a substantial increase in
the number of individuals functioning in the low range of the IQ distribution, which is
associated with increased risk of educational, vocational, and social failure
(Section 5.3.13). A downward shift in the mean IQ value can also reduce the proportion
of the population achieving very high IQ scores. It is also important to note that the
change in a population mean observed in an epidemiologic study may be small compared
     November 2012
                           2-63
Draft - Do Not Cite or Quote

-------
 1                  to the standard error of measurement for the outcome. Measurement error can affect the
 2                  likelihood of detecting an association and is not relevant to the size of the association that
 3                  is detected. If a study is large enough it will have adequate statistical power to detect
 4                  small changes. Bias may be introduced if the measurement error of the outcome is highly
 5                  correlated with the exposure. There is no evidence indicating that individuals with higher
 6                  blood Pb levels test systematically lower than their true IQ.

 7                  Pb-associated changes in blood pressure also increase an individual's risk for health
 8                  effects that are of greater clinical consequence than is suggested by a small individual
 9                  change in blood pressure. Nawrot et al. (2002) found that a doubling of blood Pb was
10                  associated with an approximate 1 mmHg increase in systolic blood pressure. Results  from
11                  the Framingham Heart Study show that higher levels of blood pressure, even within the
12                  nonhypertensive range, impose increased rates of cardiovascular disease  (Kannel. 2000a.
13                  b). A continuous graded increase in cardiovascular risk is observed as blood pressure
14                  increases, with no evidence of a threshold value. Most events arise not in the most severe
15                  cases, but mainly in those with high normal blood pressure (i.e., mild hypertension).
16                  Kannel (2000a)  emphasized that systolic blood pressure exerts a strong influence on  more
17                  serious cardiovascular events, as it is the primary cause of hypertension and its adverse
18                  cardiovascular sequelae. In addition to the small increases in blood pressure associated
19                  with Pb, Pb-associated effects on cardiovascular morbidity outcomes such as ischemic
20                  heart disease (Section 5.4.3.6) and mortality (Section 5.4.5) have been observed. The
21                  high correlation between blood pressure and clinical cardiovascular outcomes combined
22                  with the  high prevalence  of cardiovascular disease in the U.S. adult population translate
23                  into a large increase in the prevalence of conditions in the population. In  addition, some
24                  groups within the population can be at greater risks for cardiovascular effects; as
25                  summarized in Chapter_6, there is evidence for increased cardiovascular effects based on
26                  race/ethnicity and several genetic markers. Overall, while some of the specific health
27                  endpoints that have been associated with Pb exposure are small physiological changes in
28                  an individual, these changes can represent substantial risk at the population level.
      2.9.2        Air-Pb-to-Blood-Pb Relationships

29                   The 1986 Pb AQCD described epidemiological studies of relationships between air Pb
30                   and blood Pb. Much of the pertinent earlier literature for children described in the 1986
31                   Pb AQCD was included in a meta-analysis by Brunekreef (1984). Based on the studies
32                   available at that time, the 1986 Pb AQCD concluded that "the blood Pb versus air Pb
33                   slope (3 is much smaller at high blood and air levels." This is to say that the slope (3 was
34                   much smaller for occupational exposures where high blood Pb levels (>40 ug/dL) and
35                   high air Pb levels (much greater than 10 ug/m3) prevailed relative to lower environmental
      November 2012                             2-64                  Draft - Do Not Cite or Quote

-------
 1                   exposures which showed lower blood Pb and air Pb concentrations (<30 ug/dL and
 2                   <3 ug/m3). For those environmental exposures, it was concluded that the relationship
 3                   between blood Pb and air Pb "... for direct inhalation appears to be approximately linear
 4                   in the range of normal ambient exposures (0.1-2.0 ug/m3)" (pp 1-98 of the 1986 Pb
 5                   AQCD). In addition to the meta-analysis of Brunekreef (1984). more recent studies have
 6                   provided data from which estimates of the blood Pb-air Pb slope can be derived for
 7                   children (Table 2-6. Table 4-12). The range of estimates from these studies is 2-9 (ig/dL
 8                   per (ig/m3, which encompasses the estimate from the Brunekreef (1984) meta-analysis of
 9                   (3-6 (ig/dL per (ig/m3). Most studies have described the blood Pb-air Pb relationship as
10                   either log-log (Schnaas et al.. 2004; Haves et al.. 1994; Brunekreef. 1984). which predicts
11                   an increase in the blood Pb-air Pb slope with decreasing air Pb concentration or linear
12                   (Hilts. 2003; Tripathi et al.. 2001; Schwartz and Pitcher. 1989). which predicts a constant
13                   blood Pb-air Pb  slope across all air Pb concentrations. These differences may simply
14                   reflect model selection by the investigators; alternative models are not reported in these
15                   studies.

16                   The blood Pb-air Pb slope may also be affected in some studies by the inclusion of
17                   parameters (e.g., soil Pb) that may account for some of the variance in blood Pb
18                   attributable to air Pb. Other factors that likely  contribute to the derived blood Pb-air Pb
19                   slope include differences in the populations examined and Pb sources, which varied
20                   among individual studies. See Section 4.5 for a detailed discussion of studies that inform
21                   air Pb-to blood-Pb relationships.
      November 2012                             2-65                  Draft - Do Not Cite or Quote

-------
Table 2-6     Summary of estimated slopes for blood Pb to air Pb  relationships
                 in children.
Reference
Brunekreef (1 984)
Hayes et al. (1 994)
Hilts et al. (2003)
Study Methods
Location: Various countries
Years: 1974-1983
Subjects: Children (varying age ranges, n>190,000)
Analysis: Meta analysis of 96 child populations from
18 study locations
Location: Chicago, IL
Years: 1974-1 988
Subjects: 0.5-6 yr (n = 9,604)
Analysis: Regression of quarterly median blood Pb
and quarterly mean air Pb
Location: Trail, BC
Years: 1 989-2001
Subjects: 0.5-6 yr (Estimated n = 220-460, based on
292-536 blood Pb measurements/yr with 75-85%
participation).
Analysis: Regression of blood Pb screening and
community air Pb following upgrading of a local
smelter
Model Description
Model: Log-Log
Blood Pb: 5-76 ug/dL
(mean range for studies)
Air Pb: 0.1 -24 ug/m3
(mean range for studies)
Model: Log-Log
Blood Pb:1 0-28 ug/dL
(quarterly median range)
Air Pb: 0.05-1 .2 ug/m3
(quarterly mean range)
Model: Linear
Blood Pb: 4.7-11. 5 ug/dL
(annual geometric mean
range)
Air Pb: 0.03-1.1 ug/m3
(annual geometric mean
range)
Blood Pb-
Air Pb Slope3
All children:
4.6(1.5)b
Children
<20 ug/dL:
4.8 (0.54)°
8.2 (0.62)d
7.0 (0.48)e
Schwartz and Pitcher
(1989). U.S. EPA
(1986a)
Location: Chicago, IL
Years: 1976-1980
Subjects: Black children, 0-5 yr (n = 5,476)
Analysis: Multivariate regression of blood Pb with
mass of Pb in gasoline (derived from gasoline
consumption data and Pb concentrations in gasoline
for the U.S.)
Model: Linear
Blood Pb:18-27 ug/dL
(mean range)'
Air Pb: 0.36-1.22 ug/m3
(annual maximum quarterly
mean)h
                                                                                                   8.6 (0.75)9
Tripathi et al. (2001)
Location: Mumbai, India (multiple residential
locations)
Years: 1984-1996
Subjects: 6-10 yr (n = 544)
Analysis: Regression of residential location-specific
average blood Pb and air Pb data
Model: Linear
Blood Pb: 8.6-14.4 ug/dL
(GM range for residential
locations)
Air Pb: 0.11-1.18 ug/m3
(GM range for residential
locations)
                                                                                                   3.6 (0.45)1
"Slope is predicted change in blood Pb (ug/dL per ug/m3) evaluated at ± 0.01 ug/m3 from central estimate of air Pb for the study
(shown in parentheses). The central estimate for the Brunekreef (1984) study, is the median of air Pb concentrations, since it was a
meta-analysis; for all other studies the mean is presented. For multiple regression models, this is derived based only on air Pb
coefficient and intercept.  Depending on the extent to which other variables modeled also represent air Pb, this method may
underestimate the slope attributable to air pathways. In single regression models, the extent to which non-modeled factors,
unrelated to air Pb exposures, exert an impact on blood Pb that co-varies with air Pb may lead to the slope  presented  here to over
represent the role of air Pb.
bln(PbB) = In(PbA) x 0.3485 + 2.853
°ln(PbB) = In(PbA) x 0.2159 + 2.620
dln(PbB) = In(PbA) x 0.24 + 3.17
ePbB = PbA x 7.0
'Observed blood  Pb values not provided; data are for regressed adjusted blood Pb.
9PbB = PbA x 8.6
hBased on data for the U.S. [1986 PbAQCD, (U.S. EPA. 1986a)1.
iPbB=PbAx3.6
GM, geometric mean; GSD, geometric standard deviation; PbB, blood Pb concentration (ug/dL); PbA, air-Pb concentration (ug/m3)
November 2012
                             2-66
      Draft - Do Not Cite or Quote

-------
      2.9.3       Concentration-Response Relationships for Human Health Effects

 1                  Concentration response (C-R) relationships have been examined most extensively in
 2                  studies of neurodevelopmental effects in children. Although relatively few studies
 3                  examined the shape of the concentration-response relationship between Pb in blood or
 4                  bone and effects in adults, several recent studies of adult endpoints (i.e., cognitive
 5                  function, cardiovascular and mortality effects) add to the evidence. Some of the
 6                  populations examined (e.g., NHANES, NAS) are likely to have had higher past than
 7                  recent Pb exposure. Other populations (e.g., worker populations) studied have ongoing
 8                  exposure to Pb. As described elsewhere in the document (Sections 4.3. 5.3.  5.4. and 5.5).
 9                  the interpretation of the study findings depends on the exposure history and the choice of
10                  the biomarker in the context of what is known about that exposure history. There is
11                  uncertainty regarding the frequency, duration, timing and level of exposure contributing
12                  to the blood Pb or bone Pb levels in the adult populations studied.


                    Cognitive and Behavioral Effects  in Children

13                  With each successive Pb AQCD and supplement, the epidemic logic and toxicological
14                  study findings show that progressively lower blood Pb levels or Pb exposures are
15                  associated with cognitive deficits in children (Section 5.3.13). For example, effects were
16                  observed in association with blood Pb levels in the range of 10-15 (ig/dL in the 1986
17                  Addendum (U.S. EPA.  1986c) and 1990 Supplement (U.S. EPA. 1990a).  and 10 (ig/dL
18                  and lower in the 2006 Pb AQCD (U.S. EPA. 2006b). No evidence of a threshold for the
19                  effects of Pb on neurodevelopmental effects has been reported across the range of blood
20                  Pb levels examined in epidemiologic studies.

21                  Compelling evidence for a larger decrement in cognitive function per unit increase in
22                  blood Pb among children with lower mean blood Pb concentrations compared to children
23                  with higher mean blood Pb concentrations was presented in the 2006 Pb AQCD. This
24                  evidence was based on the international pooled analysis of seven prospective cohort
25                  studies by Lanphear et al. (2005). a subsequent reanalysis of these data focusing on the
26                  shape of the concentration response function (Rothenberg and Rothenberg. 2005). and
27                  several individual studies (Section  5.3.16. Figure 5-16. and Table 5-16).

28                  Attenuation of concentration-response (C-R) relationships at higher exposure or dose
29                  levels has been reported in the occupational literature for a range of exposures. Reasons
30                  proposed to explain the attenuation include greater exposure measurement error and
31                  saturation of biological  mechanisms at higher levels as well depletion of the pool of
32                  susceptible individuals at higher exposure levels (Stavner et al.. 2003). Explanations
33                  specific to nonlinear relationships observed in studies of Pb exposure in children include
      November 2012                            2-67                  Draft - Do Not Cite or Quote

-------
 1                  a lower incremental effect of Pb due to covarying risk factors such as low SES, poor
 2                  caregiving environment, and higher exposure to other environmental factors (Schwartz.
 3                  1994). differential activity of mechanisms at different exposure levels, and confounding
 4                  by omitted or misspecified variables. Review of the evidence did not reveal a consistent
 5                  set of covarying risk factors to explain the differences in blood Pb IQ C-R relationship
 6                  across high and low Pb exposure groups observed in epidemiologic studies. Nonlinear
 7                  concentration-response relationships including U- or inverted U-shaped curves for
 8                  various endpoints, including those related to cognitive impairment were demonstrated in
 9                  the toxicological literature. However, these toxicological findings are distinct from
10                  epidemiologic findings of supralinear relationships in that some U- or inverted U-shaped
11                  relationships do not indicate Pb-induced impairments at higher exposure concentrations.

12                  The  supralinear relationship reported in multiple prospective studies does not provide
13                  evidence supporting a threshold for Pb-associated cognitive function decrements. As
14                  detailed in Section 5.3.13. higher age 2 year blood Pb levels were associated with FSIQ
15                  decrements in children aged 10 years whose blood Pb levels were in the range of
16                  1.0-9.3 ug/dL, e.g.(Bellinger. 2008). Supporting evidence was provided by Pb-associated
17                  decrements in academic performance observed in fourth grade children with earlier
18                  childhood blood Pb levels 2 (ig/dL versus 1 (ig/dL (Miranda et al.. 2009: 2007a). The
19                  lack of a reference population with blood Pb levels reflecting pre-industrial Pb exposures
20                  limits the ability to identify a threshold in the current population. Toxicological studies
21                  showed that lower Pb  exposures (e.g., 50 ppm in drinking water) induced learning and
22                  memory impairments in animals compared to control exposures or higher Pb exposures
23                  (e-g-, 150 ppm). Additional toxicological evidence suggests that mechanisms may be
24                  differentially activated at lower and higher Pb exposures, and reduced long-term
25                  potentiation (LTP) and hippocampal glutamate release with lower Pb exposures may
26                  provide explanation for impaired learning and memory with lower Pb exposures.


                    Studies of Pb Effects  in Adults

27                  The  shape of the C-R function (e.g., linear versus non-linear) was not examined in most
28                  studies of the association  of Pb biomarkers with cognitive function in adults
29                  (Sections 5.3.2.7 and 5.3.13). Log-linear models were used to fit the data in NHANES
30                  analyses. Nonlinearity in  the relationship between bone Pb and cognitive function among
31                  participants in the BMS and NAS cohorts was examined with the use of quadratic terms,
32                  penalized splines, or visual inspection of bivariate plots. There was some evidence for
33                  nonlinearity in prospective analyses of the NAS cohort (Figure 5-8 and Figure 5-9). but
34                  not all results indicated a  larger decrement in cognitive function per unit increase in bone
35                  Pb level in lower bone Pb groups.  In the BMS cohort, statistical tests of nonlinearity
      November 2012                             2-68                  Draft - Do Not Cite or Quote

-------
 1                  indicated that a linear model fit the relationship between tibia Pb level and various tests
 2                  of cognitive function.

 3                  A meta-analysis of human studies found that each doubling of blood Pb level (between 1
 4                  and >40 ug/dL measured concurrently in most studies of adults for which past exposures
 5                  were likely higher than current exposures) was associated with a 1 mmHg increase in
 6                  systolic BP and a 0.6 mmHg increase in diastolic BP (Nawrot et al., 2002). In this
 7                  analysis, effect sizes were adjusted for the purpose of pooling them depending on
 8                  whether a linear or log (common or natural) linear model was used. The functional form
 9                  of the C-R relationship was examined in few individual studies of cardiovascular effects
10                  (Section 5.4.2.1).  Weaver (2010). reported that a logarithmic function of blood Pb level
11                  better described data from a cohort of Korean workers than the linear form. Only a small
12                  number of studies that focused on Pb-induced hypertension in experimental animals have
13                  included more than two exposure concentrations; however these studies appear to have a
14                  supralinear concentration-response (Figure 5-21).

15                  Studies investigating both all-cause and cardiovascular mortality report both linear and
16                  non-linear relationships (Section 5.4.5). Although associations a re consistently reported,
17                  findings regarding the shape of C-R relationship between blood Pb level and mortality in
18                  NHANES analyses were mixed. In the NAS cohort, C-R relationships between bone Pb
19                  and mortality were approximately linear for patella Pb on the log(heart rate [HR]) scale
20                  for all cardiovascular disease (CVD), but appear nonlinear for IHD (Weisskopf et al.,
21                  2009). It is important to note the wide confidence  limits, which increase uncertainty at the
22                  lower and upper bounds of patella Pb levels. The strongest associations were observed
23                  between mortality and baseline patella Pb  concentration while tibia Pb levels were more
24                  weakly associated with CVD mortality. Tibia bone Pb level is thought to reflect a longer
25                  cumulative exposure period than is patella bone Pb level because the residence time of Pb
26                  in trabecular bone is shorter than that in cortical bone.
      2.9.4        Patterns of Pb Exposure and Neurodevelopmental Deficits in Children

27                  Blood Pb, which is the most common biomarker of exposure used in epidemiologic
28                  studies of children, is an integrative measure that does not allow aspects of exposure such
29                  as frequency, timing, duration, and level to be distinguished. Exposure metrics based on
30                  blood Pb measurements at different ages in childhood are typically highly correlated.
31                  Analyses of serial blood Pb concentrations measured in longitudinal epidemiologic
32                  studies find relatively strong correlations (e.g., r = 0.5-0.8) among each child's individual
33                  blood Pb concentrations measured after 6-12 months of age. (Section 4.3.2).
      November 2012                             2-69                   Draft - Do Not Cite or Quote

-------
 1                   Consequently, it is difficult to distinguish the relative importance of various exposure
 2                   metrics in both cross-sectional and longitudinal studies of the effect of Pb in children.

 3                   As discussed in Section 4.3.5. blood Pb may reflect both recent exposures as well as past
 4                   exposures since Pb is both taken up by and released from the bone. The relative
 5                   proportion of blood Pb from recent versus past exposure is uncertain in the  absence of
 6                   specific information about the pattern of exposure contributing to observed blood Pb
 7                   levels. This uncertainty is greater in adults and older children than in young children who
 8                   do not have lengthy exposure histories. Several lines of evidence, which are summarized
 9                   below, inform the interpretation of epidemiologic studies of young children with regard
10                   to the patterns of exposure that  contribute to observed health effects. See Section 5.3.12
11                   for additional details on specific studies that form the evidence base on lifestage of
12                   exposure and neurodevelopmental  deficits.

13                   Epidemiologic studies consistently show that blood Pb levels measured during lifestages
14                   throughout childhood, as well as averaged over multiple years during childhood, are
15                   associated with cognitive function  decrements and increases in attention-related
16                   behaviors (Section 5.3.12). Evidence in animals also indicates that Pb exposures  during
17                   multiple lifestages, including prenatal only, prenatal plus lactational, postnatal only,
18                   lifetime are observed to induce  impairments in learning in rodents and monkeys (Rice.
19                   1992b.  1990: Rice and Gilbert.  1990b). These findings are  consistent with the
20                   understanding that the nervous  system continues to develop (i.e., synaptogenesis and
21                   synaptic pruning remains active) throughout childhood and into adolescence.

22                   The international pooled analysis of seven prospective studies found that increments in
23                   concurrent and peak blood Pb levels were associated with a decrease in FSIQ measured
24                   between ages 5 and 10 years (Lanphear et al.. 2005). In individual studies, postnatal
25                   (early childhood and concurrent) blood Pb levels are also consistently associated with
26                   cognitive function decrements in children and adolescents (Figure 5-2. Table 5-3. Table
27                   5-14).
28                   Although concurrent blood Pb levels in children are highly affected by recent exposure,
29                   they are also influenced by past/prenatal exposure due to the rapid growth-related bone
30                   turnover in children. Thus, concurrent blood Pb level in children also may reflect
31                   cumulative dose (Section 4.3.5.1). Animal toxicology data indicate that developmental Pb
32                   exposures creating steady-state blood-Pb concentrations of-10 ug/dL result in behavioral
33                   impairments that persist into adulthood in rats and monkeys. In rats, neurobehavioral
34                   deficits that persisted well into adulthood were observed with prenatal, preweaning, and
35                   postweaning Pb exposure. In monkeys, such impaired learning and short-term memory at
36                   when tested at 7 to 8 years (Rice and Karpinski. 1988) and impairments in attention
37                   when tested at 9 to 10 years (Gilbert and Rice. 1987) were observed with Pb exposure


      November 2012                             2-70                  Draft - Do Not Cite or Quote

-------
 1                  that did not begin until postnatal day 400 and that produced peak blood-Pb levels
 2                  <15 ug/dL and steady-state levels ~11 ug/dL, indicating that postnatal juvenile Pb
 3                  exposures were sufficient to produce neurodevelopmental deficits.

 4                  Pb can cross the placenta to affect the developing fetal nervous system and fetal Pb
 5                  exposure can occur from recent maternal exposure or from mobilization of bone Pb stores
 6                  from past exposures (Section 4.2.2.4). In very young children, ages <2 years, decrements
 7                  in mental development, as assessed with MDI, was associated with higher prenatal
 8                  (maternal and cord) and concurrent blood Pb levels (Section 5.3.2.2). Thus, both
 9                  postnatal child and maternal Pb exposures may contribute to neurodevelopmental effects
10                  in children from infancy to age 2 years.

11                  There is some evidence that the relative influence of maternal Pb levels on postnatal
12                  blood Pb level is substantially reduced soon after birth (Section 4.4). There was also a
13                  good correlation between child blood Pb level  and child hand Pb loading (R2 = 0.70) in a
14                  study following children living in a contaminated area, indicating the influence of
15                  concurrent Pb exposures on blood Pb during the early childhood years (Simon et al..
16                  2007). In another study (Carbone et al.,  1998) blood Pb levels of infants aged 6-12
17                  months were  significantly lower than their neonatal cord blood Pb levels (2.24  ug/dL
18                  versus 4.87 ug/dL). Among infants  born with blood Pb levels of greater than 7  ug/dL,
19                  who were followed for a week, there was a dramatic drop  in the blood Pb (Carbone et al..
20                  1998).
      2.9.5        Reversibility and Persistence of Neurotoxic Effects of Pb

21                  The 2006 Pb AQCD concluded that the human and animal evidence suggest that the
22                  neurotoxic effects of Pb are not generally reversible (U.S. EPA. 2006b). Chelation studies
23                  in humans and animals show that chelation decreases total body Pb burden, but does not
24                  necessarily exert evident effects on Pb-induced cognitive deficits. For example, analysis
25                  of multi-center study data indicates that medical interventions involving chelation therapy
26                  (e.g.,  Succimer use) do not fully reverse cognitive deficits associated with early Pb
27                  exposure (Liu et al., 2002).

28                  The persistence of neurodevelopmental effects from comparatively low-level Pb
29                  exposure was also considered in the 2006 Pb AQCD (U.S. EPA. 2006b). with some
30                  evidence suggesting that the effects of Pb on neurodevelopmental outcomes persisted into
31                  adolescence and young adulthood. The toxicological evidence continues to support a
32                  range of effects with prenatal and early postnatal Pb exposures that persist to adulthood
33                  (Sections 5.3.2.3 and 5.3.3.1). A number of mechanisms, including changes in
34                  neurogenesis, synaptogenesis and synaptic pruning, long term potentiation,  and

      November 2012                            2-71                   Draft - Do Not Cite or Quote

-------
 1                   neurotransmitter function have been identified that provide biological plausibility for
 2                   epidemiologic and toxicological findings of persistent cognitive and behavioral effects
 3                   that result from Pb exposures during prenatal and early childhood periods. Furthermore,
 4                   the normal dynamic and rapid rate of development that occurs early in life in the  CNS
 5                   makes insults early in life especially problematic in that they can permanently change the
 6                   trajectory of brain development such that there are little or no compensatory pathways to
 7                   replace the lost potential for proper brain development (Bayer. 1989).

 8                   The persistence of effects appears to depend on the duration and window of exposure as
 9                   well as other factors that may affect an individual's ability to recover from an insult.
10                   There is evidence that some cognitive effects of prenatal Pb exposure may be transient
11                   and that recovery is greater among children reared in households with more optimal
12                   caregiving characteristics and in children whose concurrent blood Pb levels were low
13                   (Bellinger etal.. 1990); the animal toxicology literature  supports these findings using
14                   studies of Pb-exposed animals that live in enriched environments.

15                   Toxicological studies in the 2006  Pb AQCD highlighted the importance of Pb exposure
16                   during early life in promoting Alzheimer's-like pathologies in the adult rodent brain, with
17                   Pb-induced neurodegeneration and formation of neurofibrillary tangles in aged animals in
18                   which blood Pb levels had returned to control levels after an earlier life Pb exposure
19                   (U.S. EPA. 2006b). Sensitive windows of early life Pb exposure or a Pb biomarker and
20                   have been associated with persistent changes in adulthood as demonstrated with animal
21                   models of neurodegeneration, i.e., neurofibrillary tangle formation. These effects are not
22                   reflective of concurrent blood-Pb  levels at the age of manifestation of the pathology but
23                   instead are associated with an earlier life Pb exposure.
      2.9.6        Populations Potentially At-Risk for Health Effects

24                   The NAAQS are intended to protect public health with an adequate margin of safety. In
25                   so doing, protection is provided for both the population as a whole and those groups
26                   potentially at increased risk for health effects from exposure to the air pollutant for which
27                   each NAAQS is set (Preface to this ISA). To facilitate the identification of populations at
28                   increased risk for Pb-related health effects, studies have evaluated various factors that
29                   may contribute to susceptibility and/or vulnerability to Pb. These characteristics include
30                   various factors, such as genetic background, race and ethnicity, sex, age, diet, pre-
31                   existing disease, SES, and characteristics that may modify exposure or the response to
32                   Pb. Table 2-7 (Table 6-5) provides an overview of the factors examined as potentially
33                   increasing the risk of Pb-related health effects based on the recent evidence integrated
      November 2012                             2-72                   Draft - Do Not Cite or Quote

-------
                    across disciplines. They are classified according to the criteria discussed in the
                    introduction to Chapter_6.
      Table 2-7    Summary of evidence for factors that potentially increase the risk of
                    Pb-related health effects.
Factor Evaluated
Childhood (Sections 6.2.1 , 6.3.1)

Older Adulthood (Sections 6.2.1 .2, 6.3.1 .2)

Sex (Sections 6.2.2, 6.3.2)

Genetics (Sections 6.3.3)
Pre-existing Disease3 (Section 6.3.4)
Smoking Status (Section 6.3.5)
Socioeconomic Status (SES) (Sections 6.2.4, 6.3.6)

Race/Ethnicitv (Sections 6.2.3, 6.3.7)

Proximity to Pb Sources (Section 6.2.5)
Residential Factors (Section 6.2.6)
Body Mass Index (BMI) (Section 6.3.8)
Alcohol Consumption (Section 6.3.9)
Nutrition (Section 6.3.10)
Stress (Section 6.3.11)
Maternal Self-Esteem (Section 6.3.12)
Cognitive Reserve3 (Section 6.3.13)
Other Metals (Section 6.3.14)
3Possible mediator
Classification
Adequate

Suggestive

Suggestive

Suggestive
Suggestive
Inadequate
Suggestive

Adequate

Adequate
Adequate
Inadequate
Inadequate
Adequate
Suggestive
Inadequate
Inadequate
Suggestive

 3
 4
 5
 6
 7
 8
 9
10
11
12
In consideration of the evidence base as a whole (e.g., stratified and longitudinal
analyses) and integrating across disciplines of toxicokinetics, exposure, and health, there
is adequate evidence to conclude that children are an at-risk population. It is recognized
that Pb can cross the placenta and affect the developing nervous system of the fetus
(Section 4.2.2.4). Children may have increased exposure to Pb compared with adults
because children's behaviors and activities (including increased hand-to-mouth contact,
crawling, and poor hand-washing), differences  in diets, and biokinetic factors. There is
evidence of increased risk to the neurocognitive effects of Pb exposure during several
lifestages throughout gestation, childhood, and into adolescence (Section 5.3.12).
Findings from magnetic resonance imaging (MRI) studies indicate that normal brain
     November 2012
                            2-73
Draft - Do Not Cite or Quote

-------
 1                   development remains dynamic throughout adolescence, and epidemiologic studies have
 2                   linked concurrent blood Pb level (as well as other blood Pb metrics) in adolescents to
 3                   decrements in cognitive function and delinquent or criminal behavior (Section 5.3.4).
 4                   Delays in puberty onset (Section 5.8.1). and renal effects (Section 5.5.2.2). are also
 5                   observed in association with concurrent blood Pb  level in cross-sectional studies of
 6                   adolescents. Since the populations of older children in these studies generally had higher
 7                   past exposures, the current evidence does not clearly establish the link between a time
 8                   and duration of exposure during adolescence and the observed health effects in the
 9                   adolescent populations studied. Elevated biomarkers levels, which may be related to
10                   remobilization of stored Pb during bone loss and/or higher historical Pb exposures, are
11                   observed in older adults. Studies of older adults report inconsistent findings for effect
12                   measure modification of Pb-related mortality by age and no modification of other health
13                   effects  studied. However, toxicological studies support the possibility of age-related
14                   differences in Pb-related health effects. The overall evidence, based on limited
15                   epidemiologic evidence but support from toxicological studies and differential exposure
16                   studies, is suggestive that older adults are potentially at risk of Pb effects.  However, there
17                   are uncertainties related to the exposure profile associated with the effects in older
18                   populations.

19                   The evidence regarding the other at-risk factors listed in the table above is summarized in
20                   detail in Section 6.4. Some studies suggest that males at some ages have higher blood Pb
21                   levels than comparably aged females; this was supported by stratifying the total sample
22                   of NHANES subjects. Sex-based differences appeared to be prominent among the
23                   adolescent and adult age groups but were not observed among the youngest age groups
24                   (1-5 years and 6-11 years). Studies of effect measure modification of Pb and various
25                   health endpoints by sex were inconsistent; although it appears that there are some
26                   differences in associations for males and females. This is also observed in toxicological
27                   studies. Overall, there is suggestive evidence to conclude that sex is a potential at-risk
28                   factor, with males, adolescents, and adults typically demonstrating higher blood Pb
29                   levels, although evidence regarding health outcomes is limited due to inconsistencies
30                   between whether males or females are at greater risk of certain outcomes.

31                   Regarding race and ethnicity, recent data suggest that the difference in blood Pb levels
32                   between black and white subjects is decreasing over time, but black subjects still tend to
33                   have higher Pb body burden and Pb exposures than white subjects. Compared to whites,
34                   non-white populations were observed to be more at risk of Pb-related health effects.
35                   Studies of race/ethnicity provide adequate evidence that race/ethnicity is an at-risk factor
36                   based on the higher exposure observed among non-white populations and  some
37                   modification observed in studies of associations between Pb levels and health effects. For
38                   example, Muntner et al. (2005) reported modification by race/ethnicity in  an analysis of
      November 2012                              2-74                   Draft - Do Not Cite or Quote

-------
 1                  hypertension among NHANES III participants. In comparisons of the highest quartile of
 2                  blood Pb to the lowest, the odds ration for hypertension was 1.54 (95% CI: 0.99, 2.39)
 3                  among Mexican Americans, 1.44 (95% CI: 0.89, 2.32) among Non-Hispanic Blacks and
 4                  1.10 (95% 0.87, 1.41) among Non-Hispanic Whites.

 5                  The gap between SES groups with respect to Pb body burden appears to be diminishing.
 6                  However, biomarkers of Pb exposure have been shown to be higher among lower SES
 7                  groups even in recent studies in which differences among SES groups have lessened.
 8                  Studies of SES and its relationship with Pb-related health effects are few and report
 9                  inconsistent finding regarding low SES as a potential at-risk factor. Overall, the evidence
10                  is suggestive that low SES is a potential at-risk factor for Pb-related health effects.

11                  There is adequate evidence that proximity to areas with Pb sources, including areas with
12                  large industrial sources, is associated with increased Pb exposure. Relatively high
13                  concentrations of ambient air Pb have been measured near sources, compared with  large
14                  urban areas without sources and high Pb exposures have been documented near
15                  Superfund sites. NHANES analyses report increased Pb biomarker levels related to
16                  increase house dust Pb levels, homes built after 1950, and renovation of pre-1978 homes.
17                  Thus, there is adequate evidence that residing in a residence with sources of Pb exposures
18                  will increase the risk  of Pb exposure and associated health effects.

19                  There is suggestive evidence to conclude that various genes potentially modify the
20                  associations between Pb and health effects. Epidemiologic and toxicological studies
21                  reported that ALAD variants may increase the risk of Pb-related health effects. Other
22                  genes examined that may  also affect risk of Pb-related health effects were VDR, DRD4,
23                  GSTM1, TNF-a, eNOS, and HFE. Overall the interaction between genes and Pb
24                  exposure were examined in a small number of studies and these types of analysis are
25                  potentially vulnerable to type II error if multiple statistical tests are conducted. However,
26                  there may be a large potential  impact of Pb exposure in specific at-risk populations
27                  carrying specific gene variants. For example, Scinicariello et al. (2010) found that Non-
28                  Hispanic white carriers of the  ALAD2 genetic variant in the highest blood Pb quartile had
29                  a 2-fold higher risk of hypertension compared with ALAD1 homozygous individuals
30                  (OR=2.00 95%CI:  1.12, 3.55). No evidence of effect modification of the association of
31                  Pb with blood pressure by ALAD was observed in an occupational study of Korean
32                  Pb workers, however (Weaver et al.. 2008). NAS subjects with the H63D polymorphism
33                  of the HFE gene an IQR had a larger Pb-associated increase in pulse pressure compared
34                  to those with the C282Y variant [i.e., 3.3 mmHg increase (95%CI: 0.16, 6.46) versus an
35                  0.89 mmHg increase  (95%CI 0-5.24) (Zhang etal. 2010a)1.

36                  Diets sufficient in minerals such as calcium (Ca2+), iron (Fe), and zinc (Zn) offer some
37                  protection from Pb exposure by preventing or competing with Pb for absorption in the GI


      November 2012                             2-75                   Draft - Do Not Cite or Quote

-------
 1                   tract. Additionally, those with iron deficiencies were observed to be an at-risk population
 2                   for Pb-related health effects in both epidemiologic and toxicological studies. Thus, there
 3                   is adequate evidence across disciplines that some nutritional factors contribute to a
 4                   population being at increased risk. Other nutritional factors, such as Ca2+, Zn, and protein
 5                   intake, demonstrated the potential to modify associations between Pb and health effects
 6                   in toxicological studies. Recent epidemiologic studies of these factors were either not
 7                   performed or observed no effect modification. Folate was also examined in an
 8                   epidemiologic study of birth size but no interaction was reported between Pb  and folate.

 9                   There was suggestive evidence for several other factors as potentially increasing the risk
10                   of Pb-related health effects: pre-existing diseases/conditions, stress, and co-exposure with
11                   other metals. Pre-existing diseases/conditions have the potential to affect the risk of
12                   Pb-related health effects. Recent epidemiologic studies did not support modification of
13                   associations  between Pb and health endpoints (i.e., mortality, HRV) by the prevalence of
14                   diabetes; however, past studies have found individuals with diabetes to be an  at-risk
15                   population with regard to renal function. Studies of Pb levels and both renal effects and
16                   heart rate variability demonstrated greater odds of the associations among hypertensive
17                   individuals compared to those that are normotensive. Stress was evaluated as  a factor that
18                   potentially increases the risk of Pb-related effects  on cognitive function in adults and
19                   hypertension and although limited by the small number of epidemiologic studies,
20                   increased stress was observed to exacerbate the effects of Pb. Toxicological studies
21                   supported this finding. High levels of other metals, such as Cd and Mn, were  observed to
22                   result in greater effects for the associations between Pb and various health endpoints
23                   (e-g-, renal function, cognitive function in children) but overall the evidence was limited.
24                   Finally, there was inadequate evidence to conclude that smoking, BMI, alcohol
25                   consumption, maternal self-esteem, and cognitive reserve are potential at-risk factors due
26                   to limited quantities of studies regarding their effect on Pb-related health outcomes.
      2.9.7        Ecological Effects and Corresponding Pb Concentrations

27                   There is limited evidence to relate ambient air concentrations of Pb to levels of deposition
28                   onto terrestrial and aquatic ecosystems and to subsequent movement of
29                   atmospherically-deposited Pb through environmental compartments (e.g., soil, sediment,
30                   water, biota). The proportion of observed effects of Pb attributable to Pb from
31                   atmospheric sources is difficult to assess due to a lack of information not only on
32                   bioavailability, as affected by the specific characteristics of the receiving ecosystem, but
33                   also on deposition, and on kinetics of Pb distribution in ecosystems in long-term exposure
34                   scenarios. Therefore, the connection between air concentration and ecosystem exposure
      November 2012                              2-76                   Draft - Do Not Cite or Quote

-------
 1                  continues to be poorly characterized for Pb, and the contribution of atmospheric Pb to
 2                  specific sites is not clear.

 3                  Furthermore, the level at which Pb elicits a specific effect is difficult to establish in
 4                  terrestrial and aquatic systems, due to the influence of other environmental variables on
 5                  both Pb bioavailability and toxicity,  and also to substantial species differences in Pb
 6                  susceptibility. Current evidence indicates that Pb is bioaccumulated in biota; however,
 7                  the sources of Pb in biota have only been identified in a few studies, and the relative
 8                  contribution of Pb from all sources is usually not known. There are large differences in
 9                  species sensitivity to Pb, and many environmental variables (e.g., pH, organic matter)
10                  determine the bioavailability and toxicity of Pb.
      2.10      Summary

11                  Table 2-8 characterizes the evidence in the 2006 Pb AQCD (U.S. EPA. 2006b) and
12                  previous assessments and compares it to the evidence evaluated in the current
13                  assessment. Evidence regarding both the health and ecological effects of Pb are
14                  summarized. The purpose of the table is to highlight the extent to which recent evidence
15                  may contribute to current conclusions. The critical assessment of body of evidence as a
16                  whole, however, is discussed in Chapter_5 and ChapterJ? of this document, and
17                  summarized in Sections 2.6 and 2.7. For the health effects evidence, population mean
18                  blood Pb levels are included for studies of children, along with the mean age or range of
19                  ages included in the study, because there is less uncertainty regarding the exposure
20                  patterns contributing to the mean blood Pb levels reported in studies of younger
21                  populations. With regard to ecological effects, evidence pointing to responses in species
22                  at ambient or near ambient concentrations is highlighted in the table.
      November 2012                             2-77                  Draft - Do Not Cite or Quote

-------
Table 2-8     Summary of evidence from  epidemiologic, animal toxicological and
                  ecological studies on the effects associated with exposure to  Pb.
Endpoint
Evidence in the 2006 Pb AQCD
Evidence in the (2012-3rd Draft) Pb ISA
Health Outcomes:
Nervous System Effects
  Cognitive Function
  in Children
The "overall weight of the available evidence
provides clear substantiation of neurocognitive
decrements being associated in young children with
blood-Pb concentrations in the range of 5-10 ug/dL,
and possibly lower". Prenatal, early childhood,
lifetime average, and concurrent blood Pb levels
were associated with decrements in IQ, learning
and memory; however, concurrent blood Pb level
was the strongest predictor.
Recent epidemiologic studies in children continue to
demonstrate associations of concurrent blood Pb
level with IQ decrements; most recent evidence
describes associations of concurrent blood Pb
levels with decrements in cognitive abilities related
to memory, executive function, and academic
performance. These associations were found in
populations with mean blood Pb levels 2-7 ug/dL.
  Attention-Related     Several epidemiologic studies reported
  Behavioral           associations between blood and tooth Pb levels
  Problems in          and inattention and impulsivity in children ages 8-17
  Children             years and young adults 19-20 years. Most studies
                      examined blood Pb levels measured earlier in
                      childhood (means-11-14 ug/dL), tooth Pb, or bone
                      Pb. The few studies of concurrent blood Pb levels
                      did not find associations with inattention in children
                      ages 5 years. There were no studies specifically
                      examining ADHD diagnosis. Uncertainty remained
                      regarding whether Pb exposure was an
                      independent predictor of neurobehavioral effects.
                      Prenatal and postnatal Pb exposure was found to
                      reduce ability to inhibit inappropriate responding
                      and increase distractibility in animals.
                                                Recent studies in children continue to support
                                                associations of blood Pb levels with inattention and
                                                hyperactivity in children ages 8-17 years. In several
                                                recent studies, associations were found with
                                                concurrent blood Pb in populations with mean blood
                                                Pb levels 1-5  ug/dL; however, the influence of
                                                higher past Pb exposures in these older children
                                                cannot be excluded. A few case-control studies
                                                found higher concurrent blood Pb levels in children
                                                with ADHD.
  Internalizing          Several prospective studies reported associations
  Behaviors in          of concurrent, childhood average, tooth, and bone
  Children             Pb levels with parent or teacher ratings of
                      withdrawn behavior, depression-like symptoms,
                      fearfulness, and anxiety in children ages 3-13
                      years.
                                                The few recent available studies found associations
                                                between concurrent blood Pb level and higher
                                                ratings of internalizing behaviors in children ages
                                                3-13 years but had limited implications because of
                                                lack of representativeness of populations and/or
                                                limited consideration for potential confounding,
  Misconduct in
  Children and Young
  Adults
Several epidemiologic studies reported
associations between Pb exposure and conduct
problems as rated by parents and teachers and
criminal offenses in children, adolescents, and
young adults. Most studies examined blood Pb
levels measured earlier in childhood (means
-10 ug/dL), tooth Pb, or bone Pb. There was little
examination of concurrent blood Pb levels.
Recent studies in children continue to support
associations of parent and teacher ratings of
conduct problems with early childhood blood Pb
levels and provide new evidence for concurrent
blood Pb  levels. Additional follow-up of previous
cohorts to older ages, support associations of early
childhood blood Pb levels or tooth Pb levels with
criminal offenses  in adults ages 19-24 years.
  Sensory Function     The selective action of Pb on retinal rod cells and
  Decrements in        bipolar cells (e.g., ERG effects) is well documented
  Children             in earlier AQCDs. Developmental Pb exposure
                      reduced visual acuity in animals. There was
                      coherence between the extensive animal and the
                      limited available human literature reporting
                      associations between concurrent blood Pb levels
                      (population means 7-12 ug/dL) and increased
                      hearing thresholds in children.
                                                The few available recent epidemiologic studies on
                                                sensory organ function in children examined
                                                children with high blood Pb levels (means
                                                >30 ug/dL) and did not consider potential
                                                confounding. Early postnatal Pb exposure of
                                                monkeys (blood Pb level 35-40 ug/dL increased
                                                hearing thresholds and decreased visual acuity.
                                                Retinal effects were found in male rodents with
                                                gestation/early postnatal Pb exposure  producing
                                                lower blood Pb levels (-12 ug/dL) than those at
                                                which effects had been previously reported.
  Motor Function       A small number of studies indicated associations of
  Decrements in        earlier childhood average, lifetime average, and
  Children             concurrent blood Pb levels (means: 5-12 ug/dL)
                      with poorer fine and motor function in children ages
                      4-17 years. The few toxicological studies did not
                      consistently find Pb-induced impairments in
                      balance and coordination in animals with blood Pb
                      levels >60 ug/dL.
                                                The few recent epidemiologic studies did not
                                                consistently find associations between concurrent
                                                blood Pb level and decrements in fine motor
                                                function. A toxicological study found poorer balance
                                                in male mice with gestational plus lactational Pb
                                                exposure (blood  Pb level: 33-42 ug/dL).
November 2012
                             2-78
       Draft - Do Not Cite or Quote

-------
Endpoint
Evidence in the 2006 Pb AQCD
Evidence in the (2012-3rd Draft) Pb ISA
  Cognitive Function
  Decrements in
  Adults
Among environmentally-exposed adults, bone Pb
levels but not blood Pb levels were associated with
poorer cognitive performance. These findings point
to an effect of long-term cumulative Pb exposure.
Recent studies support previous evidence. Recent
prospective studies provide new evidence of
associations of bone Pb levels with subsequent
declines in cognitive function in environmentally-
exposed adults over 3-4 year periods. However, as
these outcomes are observed in adults with likely
higher past Pb exposures, uncertainty exists as to
the Pb exposure level, frequency, duration, and
timing contributing to the observed associations
  Psychopathological
  Effects in Adults
Environmentally-exposed adults were not widely
examined; however a study found associations of
concurrent blood and tibia Pb level with self-
reported symptoms of depression and anxiety in
men. Several studies found higher prevalence of
symptoms related to mood disorders and anxiety in
Pb-exposed workers with mean blood Pb levels
15-38ug/dL
Concurrent blood Pb levels were associated with
symptoms of major depressive disorder and general
anxiety disorder among men and women
participating in NHANES.
  Sensory Function     A few studies found blood Pb level or cumulative Pb
  Decrements in       exposure duration to be associated with increased
  Adults               hearing thresholds and hearing loss in Pb-exposed
                      workers.
                                                 A prospective study found higher tibia Pb level to be
                                                 associated with a faster rate of increase in hearing
                                                 threshold in environmentally-exposed men over a
                                                 median of 23 years.
  Neurodegenerative
  Diseases
In the limited body of epidemiologic studies,
occupational Pb exposure and brain Pb levels were
not associated with Alzheimer's disease. Blood  Pb
levels were not consistently associated with
Amyotrophic Lateral Sclerosis among
environmentally-exposed adults. A few studies
found associations between occupational Pb
exposure and Parkinson's disease and blood Pb
levels and essential tremor. Each study had
sufficient limitations. Toxicological studies found
Pb-induced neuronal cell death loss.
The few case-control studies reported associations
of bone Pb levels with Parkinson's disease in
environmentally-exposed adults and blood Pb
levels with Amyotrophic Lateral Sclerosis and
essential tremor. Limitations of previous studies
apply to the recent evidence. Recent toxicological
evidence suggests that early-life, not adult-only Pb
exposure may be associated with
neurodegeneration in adult animals.
Cardiovascular Effects
  Hypertension        A meta-analysis of numerous epidemiologic studies
                      estimated that a doubling of blood Pb level
                      (e.g., from 5 to 10 ug/dL) was associated with a
                      1 mmHg increase in systolic BP and a 0.6 mmHg
                      increase in diastolic BP."
                      Epidemiologic studies consistently demonstrated
                      associations between Pb and incidence of
                      hypertension with suggestive evidence that bone
                      Pb may be associated with hypertension. Animal
                      studies demonstrated that long-term exposure to
                      Pb resulted in hypertension that persisted after
                      cessation of exposure.
                                                 Recent epidemiologic and toxicological studies
                                                 continue to support associations between long-term
                                                 Pb exposure and increased BP. Associations of
                                                 increased  BP with blood and bone Pb
                                                 concentrations are observed in populations with
                                                 lower mean blood Pb levels.
                                                 Recent studies, including those using bone Pb as a
                                                 metric of cumulative exposure, continue to
                                                 demonstrate associations of hypertension with Pb
                                                 levels in adults at lower population Pb
                                                 concentrations.  Recent studies have emphasized
                                                 the interaction of cumulative exposure  to Pb with
                                                 other factors including stress.
  Subclinical
  Atherosclerosis
One NHANES analysis reported an association of
blood Pb with PAD
Limited evidence for Pb-induced subclinical
atherosclerosis, including one high-quality
epidemiologic study that reports an increasing trend
in the odds of PAD and concurrent blood Pb level in
adults. Recent toxicological studies describe a
plausible biological mechanism.
  Coronary Heart        The evidence for an association of Pb with
  Disease              cardiovascular mortality was limited but supportive.
                       A few cross-sectional studies indicated
                       associations between  Pb biomarker levels and
                       increased risk of CHD outcomes (i.e., Ml and left
                       ventricular hypertrophy).
                                                 Recent studies address limitations of previous
                                                 studies and provide additional evidence for an
                                                 association of Pb with cardiovascular mortality in
                                                 adults. Specific causes of mortality that were
                                                 associated with Pb could be related to  increased BP
                                                 and hypertension.
  Cerebrovascular      No evidence available on the risk of
  Disease              cerebrovascular disease from Pb exposure.
                                                 Limited, mixed evidence for increased risk of
                                                 mortality from stroke.
Renal Effects
  Reduced Kidney      Circulating and cumulative Pb was associated with
  Function             longitudinal decline in renal function in adults.
                      Toxicological studies demonstrated that initial
                      accumulation of absorbed Pb occurred primarily in
                      the kidneys and noted a hyperfiltration
                      phenomenon during the first 3 months of exposure,
                      followed by decrements in kidney function.
                                                 Recent epidemiologic and toxicological studies
                                                 evaluated in the current review support or expand
                                                 upon the strong body of evidence indicating that Pb
                                                 exposure is associated with kidney dysfunction
                                                 (e.g., lower creatinine clearance, higher serum
                                                 creatinine, and lower GFR) in nonoccupationally-
                                                 exposed adults.
November 2012
                              2-79
        Draft - Do Not Cite or Quote

-------
Endpoint
Evidence in the 2006 Pb AQCD
Evidence in the (2012-3rd Draft) Pb ISA
Immune System Effects
  Increases in Atopic
  and Inflammatory
  Conditions
                      Children:
Several epidemiologic studies suggested that Pb
exposure may be associated with effects on cellular
and humoral immunity in children. The principal
effects demonstrated were decreases in T cell
abundance and increases in serum immunoglobulin
E (IgE) levels with concurrent blood Pb levels
>10 ug/dL. lexicological evidence supported these
findings with extensive evidence for prenatal and
early postnatal Pb exposures skewing toward Th2
cytokine production and affecting downstream
events such as increases in  IgE and inflammation
Several toxicological studies found a  Pb-induced
shift to Th2 cytokine production and a
hyperinflarnmatory phenotype of macrophages in
animals with long-term (>4 weeks) prenatal or
postnatal Pb exposure.
Recent studies in children added to the evidence for
associations of blood Pb levels with asthma, allergy,
and IgE. The consistency and coherence of findings
among related immune effects that support a shift
from a Th1 to a Th2 phenotype supports the
biological plausibility for epidemiologic observations
of associations with asthma, allergy and
inflammation-related effects in other organ systems.
                      Adults:
                       Pb exposure-associated immune effects were not
                       widely examined in environmentally-exposed
                       adults.
                                                 A small body of recently available studies provide
                                                 new evidence for increases in cytokines and other
                                                 indicators of inflammation in association with higher
                                                 concurrent blood Pb level. Associations were
                                                 observed in populations with a wide range of mean
                                                 concurrent blood Pb levels (3-22 ug/dL)._However,
                                                 as these outcomes are observed in adults with
                                                 likely higher past Pb exposures, uncertainty exists
                                                 as to the  Pb exposure level, frequency, duration,
                                                 and timing contributing to the observed
                                                 associations. A few available toxicological studies
                                                 find Pb-associated increases in cytokines and
                                                 effects on dendritic cells  in adult mice.
  Decreases in Host
  Resistance
Toxicological evidence demonstrated Pb-induced
increases in bacterial and viral infection and
suppressed DTH in animals. These effects were
supported by extensive evidence for prenatal and
early postnatal Pb exposures decreasing Th1
cytokine production, for short-term prenatal Pb
exposure decreasing nitric oxide production by
macrophages), and for long-term (>4 weeks)
exposure Pb exposure inducing a
hyperinflarnmatory phenotype of macrophages in
adult animals.
A few epidemiologic studies found higher
prevalence of respiratory infections in association
with higher blood Pb levels in children and
occupational Pb exposure in adults; however,
studies did not consider potential confounding.
In the large body of studies in occupationally-
exposed adults, the most consistent findings were
reduced neutrophil functionality in workers with
blood Pb levels >30  ug/dL.  Environmentally-
exposed adults were not widely examined.
A small body of recent studies supports previous
findings of decreased bacterial resistance and
decreased IFN-yTh1 cytokine production in animals
and blood-Pb associated decreases in host
resistance in children. Epidemiologic evidence is
limited to an ecological study that lacked
consideration for potential confounding.
  Autoimmunity        A small number of toxicological studies found that
                      prenatal and postnatal Pb treatment, several by i.p.
                      injection, increased generation of auto-antibodies.
                      A study found higher auto-antibodies to neural
                      proteins in Pb-exposed workers with blood Pb
                      levels 10-40 ug/dL
                                                 A recent toxicological study provided indirect
                                                 evidence by showing Pb-induced increases in the
                                                 activation of neo-antigen specific T cells, which
                                                 have the potential to induce formation of auto-
                                                 antibodies.
November 2012
                              2-80
        Draft - Do Not Cite or Quote

-------
Endpoint
Evidence in the 2006 Pb AQCD
Evidence in the (2012-3rd Draft) Pb ISA
Hematologic System
  Red Blood Cell
  Function and Heme
  Synthesis
                      Children:
Pb exposure was associated with disruption in
heme synthesis with increases in blood Pb levels of
approximately 20 ug/dL sufficient to halve ALAD
activity and inhibit ferrochelatase. Risk of clinical
anemia in children becomes apparent at high blood
Pb levels: 10% probability of anemia was estimated
to be associated with -20 ug/dL Pb at 1 year of
age, 50 ug/dLat 3 years of age, and 75 ug/dLat 5
years of age.
Recent epidemiologic studies provide strong
evidence that exposure to Pb is associated with
numerous deleterious effects on the hematological
system in children, including altered hematological
parameters (Hb, MCV, MCH, RBC count), perturbed
heme synthesis mediated through decreased ALAD
and ferrochelatase activities, and oxidative stress.
                      Adults:
                       Pb exposure was associated with disruption in
                       heme synthesis with increases in blood Pb levels of
                       approximately 20 ug/dL sufficient to halve ALAD
                       activity and inhibit ferrochelatase. Exposures to Pb
                       resulting in blood concentrations <40 ug/dL appear
                       to be tolerated without decreases in blood
                       hemoglobin or hematocrit, however changes in
                       erythropoiesis do occur at these blood levels.
                                                 Recent epidemiologic studies provide strong
                                                 evidence exposure to Pb is associated with
                                                 numerous deleterious effects on the hematological
                                                 system, including altered hematological parameters
                                                 (Hb, MCV, MCH, RBC count), perturbed heme
                                                 synthesis mediated through decreased ALAD and
                                                 ferrochelatase activities, decreased erythropoiesis,
                                                 and oxidative stress.
Developmental and Reproductive Effects
  Development
Epidemiologic studies reported effects including
delayed puberty in girls. Animal toxicological
studies reported Pb-associated developmental
effects on teeth, sensory organs, the Gl system, the
liver, and postnatal growth. Delayed puberty was
also observed in both male and female populations
in animal toxicology studies showing associations
with dam blood Pb levels of-40 ug/dl
                      blood Pb levels of 26 ug/dL.
                                                   i ug/dL and pup
Recent toxicological and epidemiologic studies
provide strong evidence for delayed onset of
puberty in males and females. Most studies found
delayed onset of puberty was among children ages
6-18 years with mean/median blood Pb levels less
than 5 ug/dL. These findings were supported by
studies in the toxicological literature showing effects
on puberty onset at blood Pb levels of 3.5-13 ug/dL.
  Birth Outcomes      Toxicological studies reviewed concluded that Pb
                      exposure can increase fetal mortality and produce
                      sublethal effects, smaller litters, reduced birth
                      weight, and fewer implantation sites. Epidemiologic
                      studies on preterm birth and low birth weight/fetal
                      growth reported inconsistent findings.
                      Epidemiologic studies reported the possibility of
                      small associations between increased Pb exposure
                      and birth defects,  and toxicological studies
                      demonstrated associations between exposure to
                      high doses of Pb and increased incidences of
                      teratogenic effects.
                                                 Recent toxicological and epidemiological studies
                                                 have reported inconsistent findings for studies for
                                                 birth defects, preterm birth, and low birth
                                                 weight/fetal growth. A few well-conducted
                                                 epidemiologic studies of preterm birth and low birth
                                                 weight/fetal growth reported associations between
                                                 increased Pb levels and decreased gestational age
                                                 and birth weight/fetal growth.
  Male Reproductive
  Function
Epidemiologic evidence suggested small
associations between Pb exposure and male
reproductive outcomes including perturbed semen
quality and increased time to pregnancy.
Associations between Pb exposure and male
reproductive endocrine status were not observed in
the occupational populations studied. Toxicological
studies provided evidence that Pb produced effects
on male and female reproductive junction and
development and disrupts endocrine function.
Recent toxicological studies provide strong
evidence for effects on sperm (blood Pb levels
34-37 ug/dL). Epidemiologic studies support the
association observed  in toxicological studies of Pb
exposure and detrimental effects on sperm.
  Female              Toxicological studies reported that Pb exposure
  Reproductive        was associated with effects on female reproductive
  Function             function that can be classified as alterations in
                      female sexual maturation, effects on fertility and
                      menstrual cycle, endocrine disruption, and changes
                      in morphology or histology of female reproductive
                      organs including the placenta. Epidemiologic
                      studies on Pb and female reproductive function
                      provided little evidence for an association between
                      Pb biomarkers and effects on female reproduction
                      and fertility.
                                                 Epidemiologic studies of Pb levels and hormones
                                                 demonstrate associations but are inconsistent
                                                 overall and there is a lack of large, well-conducted
                                                 epidemiologic studies examining associations
                                                 between Pb levels and fertility. Toxicological studies
                                                 of Pb and effects on female reproduction
                                                 demonstrate effects in some studies.
November 2012
                              2-81
        Draft - Do Not Cite or Quote

-------
Endpoint
Evidence in the 2006 Pb AQCD
Evidence in the (2012-3rd Draft) Pb ISA
Cancer
  Cancer
                      Epidemiologic studies of highly exposed
                      occupational populations suggest a relationship
                      between Pb and cancers of the lung and the
                      stomach; however the evidence is limited by the
                      presence of various potential confounders,
                      including metal co-exposures (e.g., to As, Cd),
                      smoking, and dietary habits. The 2003 NTP and
                      2004 IARC reviews concluded that Pb and  Pb
                      compounds were probable carcinogens, based on
                      limited evidence in humans and sufficient evidence
                      in animals. Based on animal data and inadequate
                      human data Pb and Pb compounds would be
                      classified as likely carcinogens according to the
                      EPA Cancer Assessment Guidelines for Carcinogen
                      Risk Assessment.
                                                The toxicological literature continues to provide the
                                                strongest evidence for Pb exposure and cancer with
                                                supporting evidence provided by the epidemiologic
                                                literature. Epidemiologic studies of cancer incidence
                                                and mortality reported inconsistent results.
Ecological/Welfare Effects:
Endpoint
Evidence in the 2006 Pb AQCD
Evidence in the (2012 3rd Draft) Pb ISA
Developmental and
Reproductive Effects
Terrestrial Organisms:
                      No information on reproduction in plants.
                                                There is an insufficient number of studies that
                                                consider Pb effects on plant reproduction.
                      Limited evidence in invertebrates and vertebrates.
                                                                      Recent studies in a few taxa expand the evidence
                                                                      for Pb effects on developmental and reproductive
                                                                      endpoints for invertebrates and vertebrates at
                                                                      concentrations that generally exceed Pb levels in
                                                                      U.S. soils. In some organisms, exposure-dependent
                                                                      responses in development and reproductive
                                                                      outcomes are  observed in experiments where
                                                                      exposure increases from background
                                                                      concentrations to concentrations found in heavily
                                                                      exposed sites near stationary sources. Data on
                                                                      terrestrial species is coherent with toxicological data
                                                                      from mammals in the context of human health
                                                                      research.
                      Aquatic Organisms:
                      No reviewed studies on reproductive effects in
                      aquatic plants.
                      Reproductive and developmental effects reported in
                      a few species of invertebrates at <50 ug Pb/L and
                      in fish at <150 ug Pb/L
                                                Recent evidence supports previous findings of
                                                reproductive and developmental effects of Pb in
                                                freshwater invertebrates and vertebrates and
                                                differential lifestage response at near ambient
                                                concentrations of Pb in some organisms.
Growth
                      Terrestrial Organisms:
                      Pb inhibits photosynthesis and respiration in plants.
                      Limited evidence for growth effects in soil
                      invertebrates, avian and mammalian consumers.
                                                Recent studies support previous findings of Pb
                                                effects on plant growth, with some evidence for
                                                exposure-dependent decreases in the biomass of
                                                some plant species grown in Pb-amended or
                                                Pb-contaminated soil.
                                                Recent data for soil invertebrates supports previous
                                                evidence of increasing effects on growth with
                                                increasing exposure.
                                                Limited studies considered effects on growth on
                                                vertebrates.
                      Aquatic Organisms:
                      Evidence for growth effects in algae, aquatic plants
                      and aquatic invertebrates
                      Most primary producers experience EC50 values for
                      growth in the range of 1,000 to 100,000 ug Pb/L
                                                The weight of the evidence continues to support
                                                growth effects of Pb in freshwater plants and
                                                invertebrates. Recent studies on growth in
                                                freshwater invertebrates find effects of Pb at lower
                                                concentrations than previously reported.
                                                Growth inhibition in one species of freshwater snail
                                                was observed at <4 ug Pb/L in juveniles.
                                                Lowest genus mean chronic value for Pb reported
                                                at 10 ug Pb/L in a freshwater mussel.
November 2012
                             2-82
       Draft - Do Not Cite or Quote

-------
Endpoint
Evidence in the 2006 Pb AQCD
Evidence in the (2012-3rd Draft) Pb ISA
Survival
                      Terrestrial Organisms:
                      No information on mortality in plants. Effects of Pb
                      on invertebrates and vertebrates include decreased
                      survival.
                      In terrestrial and avian species toxicity was
                      observed in laboratory studies over a wide range of
                      doses (<1 to >1,000 mg Pb/kg body weight»day)
                      (U.S. EPA. 2005b).
                                                 Recent studies in invertebrates and vertebrates
                                                 support previous associations between Pb
                                                 exposure and mortality.
                      Aquatic Organisms:
                      No studies reviewed on mortality in plants at current
                      concentrations of Pb in the environment.
                      Pb impacted survival of some aquatic invertebrates
                      at <20 ug Pb/L dependent upon water quality
                      variables (i.e., DOC, hardness, pH).
                      Range of 96-hour LC50 values in fathead minnow:
                      810->5,400ug Pb/L
                                                 The weight of evidence continues to support Pb
                                                 effects on survival of freshwater invertebrates and
                                                 vertebrates and indicates that there are effects in a
                                                 few species at lower concentrations than previously
                                                 reported.
                                                 Recent evidence  for effects in a few freshwater
                                                 invertebrates: at <20 ug Pb/L
                                                 Recent evidence  in freshwater fish for impacts to
                                                 survival at <100 ug Pb/L dependent upon water
                                                 quality parameters and lifestage.
                                                 96- hour LC50 values as low as 41 ug Pb/L in
                                                 fathead minnows tested in natural waters from
                                                 across the U.S.
Neurobehavioral
Effects
Terrestrial Organisms:
                      Exposure to Pb in laboratory studies and simulated
                      ecosystems may alter species competitive
                      behaviors, predator-prey interactions, and
                      contaminant avoidance behaviors.
                                                 Recent studies continue to support previous
                                                 evidence that Pb exposure is associated with
                                                 behavioral alterations. Recent studies identify
                                                 possible molecular targets for Pb neurotpxicity in
                                                 invertebrates and there is new evidence in a few
                                                 invertebrate and vertebrate species for behavioral
                                                 effects associated with Pb exposure  (i.e., feeding
                                                 and escape behaviors).
                      Aquatic Organisms:
                      Exposure to Pb has been shown to affect brain
                      receptors in fish and may alter avoidance behaviors
                      and predator-prey interactions.
                                                 Recent studies continue to support previous
                                                 evidence that Pb exposure is associated with
                                                 behavioral alterations. Recent studies identify
                                                 possible molecular targets for Pb neurotoxicity in
                                                 fish and provide additional evidence for Pb effects
                                                 on behaviors in freshwater organisms that may
                                                 impact predator avoidance (swimming).
Hematological Effects   Terrestrial Organisms:
                      Pb effects on heme synthesis were documented in
                      the 1986 Pb AQCD and continue to be studied in
                      terrestrial biota. Changes in ALAD are not always
                      related to adverse effects but may simply indicate
                      exposure. The linkage between effects of Pb on
                      blood parameters is well documented; however,  the
                      linkage between hematological indicators and
                      ecologically relevant effects is less well understood.
                                                 Consistent with previous studies, the weight of the
                                                 evidence in recent studies continues to support
                                                 findings of Pb effects on heme synthesis and ALAD
                                                 enzyme activity. Recent studies in birds near
                                                 historical mining areas and altered serum profiles
                                                 and blood cell counts in vertebrates provide
                                                 evidence for additional species in which
                                                 hematological endpoints are potentially affected by
                                                 Pb.
                      Aquatic Organisms:
                      In metal impacted habitats, ALAD is a recognized
                      biomarker of Pb exposure. Changes in ALAD are
                      not always related to adverse effects but may
                      simply indicate exposure. In fish, Pb effects on
                      blood chemistry have been documented with Pb
                      concentrations ranging from 100 to 10,000 ug Pb/L.
                                                 Consistent with previous studies, the weight of the
                                                 evidence in recent studies continues to support
                                                 findings of Pb effects on ALAD and expands this
                                                 evidence to additional species of bacteria,
                                                 invertebrates, and vertebrates as well as in recent
                                                 studies on altered blood cell counts in vertebrates.
                                                 Additional field studies in aquatic bivalves report a
                                                 correlation between Pb and ALAD activity.
November 2012
                              2-83
       Draft - Do Not Cite or Quote

-------
Endpoint
Evidence in the 2006 Pb AQCD
Evidence in the (2012-3rd Draft) Pb ISA
Physiological Stress     Terrestrial Organisms:
                       Pb exposure may cause lipid peroxidation and
                       changes in glutathione concentrations. There are
                       species differences in resistance to oxidative
                       stress.
                                                 Recent studies continue to support previous
                                                 associations of Pb exposure with physiological
                                                 stress. New evidence includes upregulation of
                                                 antioxidant enzymes, production of reactive oxygen
                                                 species and increased lipid peroxidation associated
                                                 with Pb exposure in additional species of terrestrial
                                                 plants, invertebrates and vertebrates.  Experimental
                                                 exposures increasing from background
                                                 concentrations to concentrations found in heavily
                                                 exposed sites near stationary sources.
                       Aquatic Organisms:
                       Pb exposure associated with alterations in
                       enzymes involved in physiological stress
                       responses.
                                                 Recent studies continue to support previous
                                                 associations of Pb exposure with physiological
                                                 stress. New evidence in freshwater organisms
                                                 includes upregulation of antioxidant enzymes,
                                                 production of reactive oxygen species and
                                                 increased lipid peroxidation associated with Pb
                                                 exposure. Observed effects generally occurred at
                                                 concentrations that typically exceed Pb levels in
                                                 U.S. waters with limited evidence for effects
                                                 associated with Pb at polluted sites.
Community and
Ecosystem Level
Effects
Terrestrial Ecosystems:
Effects of Pb difficult to interpret because of the
presence of other stressors including metals. The
1986 PbAQCD reported shifts toward Pb-tolerant
communities at 500 to 1,000 mg Pb/kg soil.
In the 2006 PbAQCD, decreased species diversity
and changes in community composition were
observed in ecosystems surrounding former
smelters.
Recent evidence for effects of Pb in soil microbial
communities adds to the body of evidence for
effects at higher levels of biological organization. In
addition, effects of Pb uptake on reproduction,
growth, and survival at the  species level are likely to
lead to effects at the population, community, and
ecosystem level. However, most evidence for Pb
toxicity to terrestrial biota is from single-species
assays, and there are important uncertainties in
generalizing from effects observed under  small-
scale, controlled  conditions, up to effects at the
ecosystem level of biological organization. .
                       Aquatic Ecosystems:
                       Most evidence of community and ecosystem level
                       effects is from near Pb sources, usually mining
                       effluents. Effects of Pb difficult to interpret because
                       of the presence  of other stressors including metals.
                       Generally, there is insufficient information available
                       for single materials in controlled studies to permit
                       evaluation of specific impacts on higher levels of
                       organization (beyond the individual organism).
                                                 Recent evidence for Pb effects on sediment-
                                                 associated and freshwater aquatic plant
                                                 communities add to the body of evidence of effects
                                                 at higher levels of biological organization. However,
                                                 most evidence for Pb toxicity to aquatic biota is
                                                 from single-species assays. Uncertainties exist in
                                                 generalizing effects observed under small-scale,
                                                 predicted conditions up to effects at the ecosystem-
                                                 level however, uptake of Pb into  aquatic organisms
                                                 and subsequent effects on reproduction, growth,
                                                 and survival at the species level  are likely to lead to
                                                 effects at the population, community, and
                                                 ecosystem level.
November 2012
                              2-84
        Draft - Do Not Cite or Quote

-------
References for Chapter 2
     Anklev. GT; Bennett. RS: Erickson. RJ; Hoff. DJ; Hornung. MW: Johnson. RD; Mount. PR; Nichols.
        JW: Russom. CL; Schmieder. PK; Serrrano. JA; Tietge. JE; Villeneuve. PL. (2010). Adverse outcome
        pathways: a conceptual framework to support ecotoxicology research and risk assessment [Review].
        Environ Toxicol Chem 29: 730-741. http://dx.doi.org/10.1002/etc.34

     Aufderheide. AC: Wittmers. LE. Jr. (1992). Selected aspects of the spatial distribution of lead in bone
        [Review]. Neurotoxicology 13: 809-819.

     Bayer. SA. (1989). Cellular aspects of brain development [Review]. Neurotoxicology 10: 307-320.

     Bellinger. D; Leviton. A; Sloman. J. (1990). Antecedents and correlates of improved cognitive
        performance in children exposed in utero to low levels  of lead. Environ Health Perspect 89: 5-11.

     Bellinger. DC. (2008). Email from David Bellinger, Harvard Medical School, to Jee-Young Kim, EPA.
        SUBJECT: Request for additional information on Bellinger and Needleman (2003). (EPA-HQ-
        OAR-2006-0735-5156). U.S. Environmental Protection Agency.
        http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OAR-2006-0735-5156

     Besser. JM; Brumbaugh. WG: Brunson. EL; Ingersoll. CG. (2005). Acute and chronic toxicity of lead in
        water and diet to the amphipod Hyalella azteca. Environ Toxicol Chem 24: 1807-1815.
        http://dx.doi.Org/10.1897/04-480R.l

     Bonacker. D; Stoiber. T; Bohm. KJ; Prots.  I; Wang. M; Unger. E; Thier. R; Bolt. HM; Degen. GH. (2005).
        Genotoxicity of inorganic lead salts and disturbance of microtubule function. Environ Mol Mutagen
        45: 346-353. http://dx.doi.org/10.1002/em.20100

     Brunekreef. B. (1984). The relationship between air lead and blood lead in children: A critical review
        [Review]. Sci Total Environ 38: 79-123. http://dx.doi.org/10.1016/0048-9697(84)90210-9

     Buekers. J: Redeker. ES: Smolders. E. (2009). Lead toxicity to wildlife: Derivation of a critical blood
        concentration for wildlife monitoring based on literature data [Review]. Sci Total Environ 407:
        3431-3438. http://dx.doi.Org/10.1016/i.scitotenv.2009.01.044

     Burger. J: Gochfeld. M. (2005). Effects of lead on learning in herring gulls: An avian wildlife model for
        neurobehavioral deficits. Neurotoxicology 26: 615-624. http://dx.doi.Org/10.1016/i.neuro.2005.01.005

     Carbone. R; Laforgia. N; Crollo. E; Mautone. A; lolascon. A. (1998). Maternal and neonatal lead
        exposure in southern Italy. Neonatology 73: 362-366.

     CDC (Centers for Disease Control and Prevention). (2004). The health consequences of smoking: A
        report of the Surgeon General. Washington, DC: U.S. Department of Health and Human Services.
        http ://www. surgeongeneral.gov/librarv/smokingconsequences/

     Chen. HI; Chiu. YW: Hsu. YK; Li. WF; Chen. YC: Chuang. HY. (2010a). The association of
        metallothionein-4 gene polymorphism and renal function in long-term lead-exposed workers. Biol
        Trace Elem Res 137: 55-62. http://dx.doi.org/10.1007/sl2011-009-8564-x

     Chen. TH; Gross. JA; Karasov. WH.  (2006b).  Sublethal  effects of lead on northern leopard frog (Rana
        pipiens) tadpoles. Environ Toxicol Chem 25: 1383-1389. http://dx.doi.Org/10.1897/05-356R.l

     Chettv. CS: Vemuri. MC: Campbell. K; Suresh. C. (2005).  Lead-induced cell death of human
        neuroblastoma cells involves GSH deprivation. Cell Mol Biol Lett 10: 413-423.

     Cho. SH; Richmond-Bryant. J: Thornburg. J: Portzer. J: Vanderpool. R; Cavender. K; Rice. J. (2011). A
        literature review of concentrations and size distributions of ambient airborne Pb-containing paniculate
        matter. Atmos Environ 45: 5005-5015. http://dx.doi.Org/10.1016/i.atmosenv.2011.05.009
November 2012                              2-85                   Draft - Do Not Cite or Quote

-------
     Conterato. GMM; Bulcao. RP; Sobieski. R; Moro. AM; Charao. MF; de Freitas. FA; de Almeida. FL;
        Moreira. APL; Roehrs. M; Tonello. R; Batista. BL; Grotto. D; Barbosa. F. Jr; Garcia. SC: Emanuelli.
        T\ (In Press) Blood thioredoxin reductase activity, oxidative stress and hematological parameters in
        painters and battery workers: Relationship with lead and cadmium levels in blood. J Appl Toxicol.
        http://dx.doi.org/10.1002/iat. 1731

     Du. M; Wang. D. (2009). The neurotoxic effects of heavy metal exposure on GABAergic nervous system
        in nematode Caenorhabditis elegans. Environ Toxicol Pharmacol 27: 314-320.
        http://dx.doi.0rg/10.1016/i.etap.2008.ll.011

     Esbaugh. A.. J.: Brix. K.. V.; Mager. E.. M.: Grosell. M.. (2011). Multi-linear regression models predict
        the effects of water chemistry on acute lead toxicity to Ceriodaphnia dubia and Pimephales promelas.
        Comp Biochem Physiol C Toxicol Pharmacol  154: 137-145.
        http://dx.doi.0rg/10.1016/i.cbpc.2011.04.006

     Field. PM; Sherrell. RM. (2003). Direct determination of ultra-trace levels of metals in fresh water using
        desolvating micronebulization and HR-ICP-MS: Application to Lake Superior waters.  J Anal At
        Spectrom 18: 254-259. http://dx.doi.org/10.1039/b210628k

     Gilbert. SG: Rice. DC. (1987). Low-level lifetime lead exposure produces behavioral toxicity (spatial
        discrimination reversal) in adult monkeys. Toxicol Appl Pharmacol 91: 484-490.

     Grosell. M; Brix. KV. (2009). High net  calcium uptake explains the hypersensitivity of the freshwater
        pulmonate snail, Lymnaea stagnalis, to chronic lead exposure. Aquat Toxicol 91: 302-311.
        http://dx.doi.0rg/10.1016/i.aquatox.2008.10.012

     Grosell. M; Gerdes. RM; Brix. KV. (2006b). Chronic toxicity of lead to three freshwater invertebrates  -
        Brachionus calyciflorus, Chironomus tentans, and Lymnaea stagnalis. Environ Toxicol Chem 25:
        97-104. http://dx.doi.Org/10.1897/04-654R.l

     Gustavsson. P; Plato. N; Hallqvist. J: Hogstedt. C: Lewne. M; Reuterwall. C: Scheele. P. (2001). A
        population-based case-referent study of myocardial infarction and occupational  exposure to motor
        exhaust, other combustion products, organic solvents, lead, and dynamite Stockholm heart
        epidemiology program (SHEEP) study group. Epidemiology 12: 222-228.

     Haves.  EB; McElvaine. MD; Orbach. HG: Fernandez. AM; Lyne. S: Matte. TD. (1994). Long-term trends
        in blood lead levels among children  in Chicago:  Relationship to air lead levels. Pediatrics 93: 195-200.

     Hilts. SR. (2003). Effect of smelter emission reductions on children's blood lead levels.  Sci Total Environ
        303: 51-58. http://dx.doi.org/10.1016/S0048-9697(02)00357-l

     Huel. G: Sahuquillo. J: Debotte. G: Oury. JF; Takser. L. (2008). Hair mercury negatively correlates with
        calcium pump activity in human term newborns  and their mothers at delivery. Environ Health Perspect
        116: 263-267. http://dx.doi.org/10.1289/ehp.10381

     Jain. NB; Potula. V:  Schwartz. J: Vokonas. PS: Sparrow. D: Wright. RO: Nie. H; Hu. H. (2007). Lead
        levels and ischemic heart disease in  a prospective study of middle-aged and elderly men: The VA
        Normative Aging Study. Environ Health Perspect 115: 871-875. http://dx.doi.org/10.1289/ehp.9629

     Kannel. WB. (2000a). Elevated systolic blood pressure as a cardiovascular risk factor [Review]. Am J
        Cardiol 85:  251-255. http://dx.doi.org/10.1016/S0002-9149(99)00635-9

     Kannel. WB. (2000b). Risk stratification in hypertension: New insights from the Framingham  Study. Am
        JHypertens 13: 3S-10S. http://dx.doi.org/10.1016/S0895-7061(99)00252-6

     Kern. M; Wisniewski. M; Cabell. L; Audesirk. G. (2000). Inorganic lead and calcium interact positively
        in activation of calmodulin. Neurotoxicology 21: 353-363.

     Kim. JH: Lee. KH: Yoo. DH: Kang. D:  Cho. SH: Hong. YC.  (2007). GSTM1 and TNF-alpha gene
        polymorphisms and relations between blood lead and inflammatory markers in a non-occupational
        population.  Mutat Res Genet Toxicol Environ Mutagen 629: 32-39.
        http://dx.doi.0rg/10.1016/i.mrgentox.2007.01.004
November 2012                              2-86                   Draft - Do Not Cite or Quote

-------
     Klaminder. J; Bindler. R; Emteryd. O: Renberg. I. (2005). Uptake and recycling of lead by boreal forest
        plants: Quantitative estimates from a site in northern Sweden. Geochim Cosmo Act 69: 2485-2496.
        http://dx.doi.0rg/10.1016/i.gca.2004.ll.013

     Klann. E; Shelton. KR. (1989). The effect of lead on the metabolism of a nuclear matrix protein which
        becomes prominent in lead-induced intranuclear inclusion bodies. J Biol Chem 264:  16969-16972.

     Kominkova. D; Nabelkova. J. (2005). The risk assessment of heavy metals in the ecosystem of urban
        creeks. Water Sci Technol 53: 10. http://dx.doi.org/10.2166/wst.2006.298

     Krieg. EF. Jr. (2007). The relationships between blood lead levels and serum follicle stimulating hormone
        and luteinizing hormone in the third national health and nutrition examination survey. Environ Res
        104: 374-382. http://dx.doi.Org/10.1016/i.envres.2006.09.009

     Langdon. CJ: Hodson. ME; Arnold. RE; Black. S. (2005). Survival, Pb-uptake and behaviour of three
        species of earthworm in Pb treated soils determined using an OECD-style toxicity test and a soil
        avoidance test. Environ Pollut 138: 368-375. http://dx.doi.Org/10.1016/i.envpol.2005.03.002

     Lanphear. BP; Hornung. R; Khoury. J: Yolton. K; Baghurst. P; Bellinger. DC: Canfield.  RL; Dietrich.
        KN; Bornschein. R; Greene. T; Rothenberg. SJ: Needleman. HL; Schnaas. L; Wasserman. G:
        Graziano. J: Roberts. R. (2005). Low-level environmental lead exposure and children's intellectual
        function: An international pooled analysis. Environ Health Perspect 113: 894-899.

     Lau. YS: Camoratto. AM; White. LM; Moriartv. CM. (1991). Effect of lead on TRH and GRF binding in
        rat anterior pituitary membranes. Toxicology 68: 169-179.
        http://dx.doi.org/10.1016/0300-483X(91)90019-W

     Leggett. RW. (1993). An age-specific kinetic model of lead metabolism in humans. Environ Health
        Perspect 101: 598-616.

     Liu. X; Dietrich. KN; Radcliffe. J: Ragan. NB; Rhoads. GG: Rogan. WJ. (2002). Do children with falling
        blood lead levels have improved cognition? Pediatrics 110: 787-791.

     Mager. EM; Brix. KV;  Grosell. M. (2010). Influence of bicarbonate and humic  acid on effects of chronic
        waterborne lead exposure to the fathead minnow (Pimephales promelas). Aquat Toxicol 96: 135-144.
        http://dx.doi.0rg/10.1016/i.aquatox.2009.10.012

     Mahler. BJ; Van Metre. PC: Callender. E. (2006). Trends in metals in urban and reference lake sediments
        across the United States,  1970 to 2001. Environ Toxicol Chem 25: 1698-1709.

     Mahmoudi. E; Essid. N; Beyrem. H; Hedfi. A; Boufahja. F; Vitiello. P; Aissa. P. (2007).  Individual and
        combined effects of lead and zinc on a free-living marine nematode community: Results from
        microcosm experiments. Exp Mar Bio Ecol 343: 217-226.
        http://dx.doi.0rg/10.1016/i.jembe.2006.12.017

     Martin. D; Glass. TA; Bandeen-Roche. K; Todd. AC: Shi. WP; Schwartz. BS. (2006). Association of
        blood lead and tibia lead with blood pressure and hypertension in a community sample of older adults.
        Am J Epidemiol 163: 467-478. http://dx.doi.org/10.1093/aje/kwj060

     Mebane. CA; Hennessv. DP; Dillon. FS. (2008). Developing acute-to-chronic toxicity ratios for lead,
        cadmium, and zinc using rainbow trout, a mayfly, and a midge. Water Air Soil Pollut 188: 41-66.
        http://dx.doi.org/10.1007/sll270-007-9524-8

     Miranda. ML; Kim. D; Galeano. M; Paul. CJ: Hull. AP; Morgan. SP  (2007a). The relationship between
        early childhood blood lead levels and performance  on end of grade tests. Environ Health Perspect 115:
        1242-1247. http://dx.doi.org/10.1289/ehp.9994

     Miranda. ML; Kim. D; Reiter. J: Overstreet Galeano. MA; Maxson. P. (2009). Environmental contributors
        to the achievement gap. Neurotoxicology 30:  1019-1024.
        http://dx.doi.0rg/10.1016/i.neuro.2009.07.012

     Muntner. P; Menke. A; DeSalvo. KB; Rabito. FA; Batuman. V. (2005). Continued decline in blood lead
        levels among adults in the United States - The National Health and Nutrition Examination Surveys.
        Arch Intern Med 165: 2155-2161. http://dx.doi.org/10.1001/archinte.165.18.2155
November 2012                              2-87                   Draft - Do Not Cite or Quote

-------
     Navas-Acien. A; Schwartz. BS: Rothenberg. SJ: Hu. H; Silbergeld. EK; Guallar. E. (2008). Bone lead
        levels and blood pressure endpoints: Ameta-analysis. Epidemiology 19: 496-504.
        http://dx.doi.org/10.1097/EDE.Ob013e31816a2400

     Navas-Acien. A; Selvin. E; Sharrett. AR; Calderon-Aranda. E; Silbergeld. E; Guallar. E. (2004). Lead,
        cadmium, smoking, and increased risk of peripheral arterial disease. Circulation 109: 3196-3201.
        http://dx.doi.org/10.1161/01.CIR.0000130848.18636.B2

     Nawrot. TS: Thijs. L; Den Hond. EM; Roels. HA; Staessen. JA. (2002). An epidemiological re-appraisal
        of the association between blood pressure and blood lead: Ameta-analysis. J Hum Hypertens 16:
        123-131. http://dx.doi.org/10.1038/si/ihh/1001300

     NCHS (National Center for Health Statistics). (2010). National health and nutrition examination survey:
        Questionnaires, datasets, and related documentation. Available online at
        http://www.cdc.gov/nchs/nhanes/nhanes questionnaires.htm (accessed November 19, 2010).

     NOAA (National Oceanic and Atmospheric Administration). (1999). Sediment quality guidelines
        developed for the national status and trends program. Silver Spring, MD: National Ocean Service.

     NFS (U.S. National Park Service). (2011). WACAP database.
        http://www.nature.nps.gov/air/studies/air  toxics/WACAPreport.cfm

     O'Flahertv. EJ.  (1995). Physiologically based models for bone-seeking elements: V. Lead absorption and
        disposition in childhood [Review]. Toxicol Appl Pharmacol 131: 297-308.
        http://dx.doi.org/10.1006/taap.1995.1072

     Rice. DC. (1990). Lead-induced behavioral impairment on a spatial discrimination reversal task in
        monkeys exposed during different periods of development. Toxicol Appl Pharmacol 106:  327-333.
        http://dx.doi.org/10.1016/0041-008X(90)90251-O

     Rice. DC. (1992b). Lead exposure during different developmental periods produces different effects on FI
        performance in monkeys tested as juveniles and adults. Neurotoxicology 13: 757-770.

     Rice. DC: Gilbert. SG. (1990b). Sensitive periods for lead-induced behavioral impairment (nonspatial
        discrimination reversal) in monkeys. Toxicol Appl Pharmacol 102:  101-109.
        http://dx.doi.org/10.1016/0041-008X(90)90087-B

     Rice. DC: Karpinski.  KF. (1988). Lifetime  low-level lead exposure produces deficits in delayed
        alternation in adult monkeys. Neurotoxicol Teratol 10: 207-214.
        http://dx.doi.org/10.1016/0892-0362(88)90019-0

     Ringenary. MJ; Molof. AH; Tanacredi. JT;  Schreibman. MP; Kostarelos. K. (2007). Long-term sediment
        bioassay of lead toxicity in two generations of the marine amphipod Elasmopus laevis, SI Smith,
        1873. Environ Toxicol Chem 26: 1700-1710. http://dx.doi.Org/10.1897/06-303Rl.l

     Rothenberg. SJ: Rothenberg. JC. (2005). Testing the dose-response specification in epidemiology: Public
        health and policy consequences for lead. Environ Health Perspect 113: 1190-1195.
        http://dx.doi.org/10.1289/ehp.7691

     Sadiq. M. (1992). Lead in marine environments. In Toxic metal chemistry in marine environments. New
        York, NY: Marcel Dekker, Inc.

     Schnaas. L; Rothenberg.  SJ: Flores. MF; Martinez. S: Hernandez. C: Osorio. E; Perroni. E. (2004). Blood
        lead secular trend  in a cohort of children in Mexico City (1987-2002). Environ Health Perspect 112:
        1110-1115. http://dx.doi.org/10.1289/ehp.6636

     Schwartz. J. (1991). Lead, blood pressure, and cardiovascular disease in men and women. Environ Health
        Perspect 91: 71-75. http://dx.doi.org/10.1289/ehp.919171

     Schwartz. J. (1994). Low-level lead exposure and children's IQ: Ameta-analysis and search for a
        threshold. Environ Res 65: 42-55. http://dx.doi.org/10.1006/enrs.1994.1020

     Schwartz. J: Pitcher. H. (1989).  The relationship between gasoline lead and blood lead in the United
        States. J Offic Stat 5:  421-431.
November 2012                              2-88                   Draft - Do Not Cite or Quote

-------
     Selevan. SG: Rice. DC: Hogan. KA; Euling. SY; Pfahles-Hutchens. A; Bethel. J. (2003). Blood lead
        concentration and delayed puberty in girls. N Engl J Med 348: 1527-1536.
        http://dx.doi.org/10.1056/NEJMoa020880

     Simon. PL; Maynard. EJ; Thomas. KD. (2007). Living in a sea of lead changes in blood- and hand-lead
        of infants living near a smelter. J Expo Sci Environ Epidemiol 17: 248-259.
        http://dx.doi.org/10.1038/si.ies.750Q512

     Smolders. E; Ports. K; van Sprang. P; Schoeters. I; Janssen. CR; McGrath. SP;  McLaughlin. MJ. (2009).
        Toxicity of trace metals in soil as affected by soil type and aging after contamination: Using calibrated
        bioavailability models to set ecological soil standards. Environ Toxicol Chem 28: 1633-1642.
        http://dx.doi.0rg/10.1897/08-592.l

     Soto-Jimenez. MF; Arellano-Fiore. C: Rocha-Velarde. R; Jara-Marini. ME; Ruelas-Inzunza. J: Voltolina.
        D; Frias-Espericueta. MG: Quintero-Alvarez. JM; Paez-Osuna. F.  (201 Ib). Biological responses of a
        simulated marine food chain to lead addition. Environ Toxicol Chem 30: 1611-1617.
        http://dx.doi.org/10.1002/etc.537

     Stayner. L; Steenland. K; Dosemeci. M; Hertz-Picciotto.  I. (2003). Attenuation of exposure-response
        curves in occupational cohort studies at high exposure levels. Scand J Work  Environ Health 29:
        317-324.

     Suter. GW: Norton. SB; Fairbrother. A. (2005). Individuals versus organisms versus populations in the
        definition of ecological assessment endpoints. Integr Environ Assess Manag 1: 397-400.
        http://dx.doi.org/10.1002/ieam.5630010409

     Tripathi. RM; Raghunath. R; Kumar. AV; Sastry. VN; Sadasivan. S. (2001). Atmospheric and children's
        blood lead as indicators of vehicular traffic and other  emission sources in Mumbai, India. Sci Total
        Environ 267: 101-108. http://dx.doi.org/10.1016/S0048-9697(00)00770-l

     U.S. EPA (U.S. Environmental Protection Agency). (1977). Air quality criteria for lead [EPA Report].
        (EPA-600/8-77-017). Washington, D.C.
        http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=20013GWR.txt

     U.S. EPA (U.S. Environmental Protection Agency). (1986a). Air quality criteria for lead  [EPA Report].
        (EPA/600/8-83/028aF-dF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1986b). Air quality criteria for lead: Volume I of IV
        [EPA Report]. (EPA-600/8-83/028aF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1986c). Lead effects on cardiovascular function,
        early development, and stature: An addendum to U.S. EPA Air Quality Criteria for Lead (1986) [EPA
        Report]. (EPA-600/8-83/028aF). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (1990a). Air quality criteria for lead: Supplement to
        the 1986 addendum [EPA Report]. (EPA/600/8-89/049F). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (1992). Framework for ecological risk assessment
        [EPA Report]. (EPA/630/R-92/001). Washington, D.C.
        http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=30004UKJ.txt

     U.S. EPA (U.S. Environmental Protection Agency). (2003a). Generic ecological assessment endpoints
        (GEAEs) for ecological risk assessment [EPA Report]. (EPA-630/P-02/004F). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2003b). Guidance for developing ecological soil
        screening levels (Eco-SSLs): Review of background concentration for metals - Attachment 1-4  [EPA
        Report]. (OSWER Directive 92857-55). Washington,  D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2005b). Ecological soil screening levels for lead:
        Interim final [EPA Report]. (OSWER Directive 9285.7-70). Washington, D.C.
November 2012                              2-89                   Draft - Do Not Cite or Quote

-------
     U.S. EPA (U.S. Environmental Protection Agency). (2006b). Air quality criteria for lead: Volume I of II
        [EPA Report]. (EPA/600/R-05/144aF). Research Triangle Park, NC.
        http ://cfpub. epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823

     U.S. EPA (U.S. Environmental Protection Agency). (2006c). Air quality criteria for lead: Volume II of II
        [EPA Report]. (EPA/600/R-05/144bF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/CFM/recordisplav.cfm?deid=158823

     U.S. EPA (U.S. Environmental Protection Agency). (2007d). Guidance for developing ecological soil
        screening levels [EPA Report]. (OSWER Directive 92857-55). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2008a). 2005 National Emissions Inventory data and
        documentation. Available online at http://www.epa.gov/ttn/chief/net/2005inventory.html (accessed
        November 1, 2010).

     U.S. EPA (U.S. Environmental Protection Agency). (2008b). Ambient aquatic life water quality: Lead
        (draft) [EPA Report]. (EPA Contract No. 68-C-04-006). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for paniculate
        matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546

     U.S. EPA (U.S. Environmental Protection Agency). (2011a). 2008 National Emissions Inventory data and
        documentation. Available online at http://www.epa.gov/ttn/chief/net/2008inventory.html (accessed
        February 17,2011).

     U.S. EPA (U.S. Environmental Protection Agency). (20lie). Integrated review plan for the national
        ambient air quality standards for lead [EPA Report]. (EPA-452/R-11-008).  Research Triangle Park,
        NC.

     Van Larebeke. N; Koppen. G: Nelen. V; Schoeters. G: Van Loon. H; Albering. H; Riga. L; Vlietinck. R;
        Kleinjans. J. (2004). Differences in HPRT mutant frequency among middle-aged Flemish women in
        association with area of residence and blood lead levels.  Biomarkers 9: 71-84.
        http://dx.doi.org/10.1080/13547500310001652160

     Vandenhove. H; Olyslaegers. G: Sanzharova. N; Shubina. O: Reed. E; Shang.  Z; Velasco. H. (2009).
        Proposal for new best estimates of the soil-to-plant transfer factor of U, Th, Ra, Pb and Po. J Environ
        Radioact 100: 721-732. http://dx.doi.Org/10.1016/i.ienvrad.2008.10.014

     Wang. DY; Xing. XJ. (2008). Assessment of locomotion behavioral defects induced by acute toxicity
        from heavy metal exposure in nematode Caenorhabditis elegans. J Environ Sci 20: 1132-1137.
        http://dx.doi.org/10.1016/S1001-0742(08)62160-9

     Wang. N; Ingersoll. CG: Ivev. CD; Hardestv. DK; May. TW: Augspurger. T; Roberts. AD; van Genderen.
        E; Barnhart. MC. (2010e). Sensitivity of early life stages of freshwater mussels (Unionidae) to acute
        and chronic toxicity of lead, cadmium, and zinc in water. Environ Toxicol Chem 29: 2053-2063.
        http://dx.doi.org/10.1002/etc.250

     Watson. AP; Van Hook. RI; Jackson. PR; Reichle. DE. (1976). Impact of a lead mining smelting complex
        on the forest- floor litter arthropod fauna in the new lead belt region of southeast Missouri.
        (ORNL/NSF/EATC-30). Oak Ridge, Tennessee: Oak Ridge National Laboratory.

     Weaver. VM. (2010). Further analysis of Korean lead worker data set with a focus on determining the
        functional form of the exposure-response relationship. Available online

     Weaver. VM: Ellis. LR: Lee. BK: Todd. AC: Shi. W: Ahn. KD: Schwartz. BS. (2008). Associations
        between patella lead and blood pressure in lead workers. Am J Ind Med 51: 336-343.
        http://dx.doi.org/10.1002/aiim.20573

     Wedding. JB; McFarland. AR; Cermak. JE. (1977). Large particle collection characteristics of ambient
        aerosol samplers. Environ Sci Technol 11: 387-390. http://dx.doi.org/10.1021/es60127a005

     Weiss. B. (1988). Neurobehavioral toxicity as a basis for risk assessment [Review].  Trends Pharmacol Sci
        9: 59-62.
November 2012                              2-90                   Draft - Do Not Cite or Quote

-------
     Weisskopf. MG: Jain. N; Nie. HL; Sparrow. D; Vokonas. P; Schwartz. J; Hu. H. (2009). A prospective
        study of bone lead concentration and death from all causes, cardiovascular diseases, and cancer in the
        department of veterans affairs normative aging study. Circulation 120: 1056-1064.
        http://dx.doi.org/10.1161/circulationaha.108.827121

     WHO (World Health Organization). (1948). Preamble to the Constitution of the World Health
        Organization as adopted by the International Health Conference, New York, 19-22 June, 1946. In
        Constitution of the World Health Organization (pp. 2). Geneva, Switzerland.
        http://whqlibdoc.who.int/hist/official records/constitution.pdf

     Zhang. A; Park. SK; Wright. RO: Weisskopf. MG: Mukherjee. B; Nie. H; Sparrow. D;  Hu. H. (2010a).
        HFE H63D polymorphism as a modifier of the effect of cumulative lead exposure on pulse pressure:
        The Normative Aging Study. Environ Health Perspect 118: 1261-1266.
        http://dx.doi.org/10.1289/ehp. 1002251
November 2012                              2-91                   Draft - Do Not Cite or Quote

-------
       3AMBIENT  LEAD:  SOURCE  TO  CONCENTRATION
      3.1        Introduction

 1                  This chapter reviews concepts and findings in atmospheric sciences that provide a
 2                  foundation for the detailed presentation of evidence of Pb exposure and Pb-related health
 3                  and ecological effects in subsequent chapters. Information in this chapter builds on
 4                  previous Pb AQCDs using more recent data and studies. This includes new knowledge of
 5                  Pb fate and transport, the latest developments in monitoring  and analysis methodologies,
 6                  and recent data describing Pb concentrations as a function of size range. The chapter
 7                  focuses on Pb concentrations in the U.S. but includes non-U.S. studies to the extent that
 8                  they are informative regarding current conditions in the U.S. Description of the chemical
 9                  forms of Pb is not provided here, however, because this information is well established.
10                  The reader is referred to the 2006 Pb AQCD for a description of the chemical forms of Pb
11                  (U.S. EPA. 2006R

12                  Section 3.2 provides an overview of the sources of ambient air Pb.  Section 3.3 provides a
13                  description of the fate and transport of Pb in air, soil, and aqueous media. Descriptions of
14                  Pb measurement methods, monitor siting requirements, and monitor locations are
15                  presented in Section 3.4. Ambient Pb concentrations, their spatial and temporal
16                  variability, size distributions of Pb-bearing particulate matter (PM), associations with
17                  copollutants and background Pb concentrations are characterized in Section 3.5.
18                  Concentrations of Pb in non-air media and biota are described in Section 3.6.
      3.2        Sources of Atmospheric Pb

19                  The following section reviews emissions estimates from the 2008 National Emissions
20                  Inventory (NEI) data from 2008 (U.S. EPA. 201 la), augmented with information on
21                  sources not included in the 2008 NEI version 21 and compares these emissions data with
22                  those from previous years. This section also reviews updated information from the peer-
23                  reviewed literature regarding sources of ambient Pb. Detailed information about
24                  processes for anthropogenic emissions and naturally-occurring emissions can be found in
25                  the 2006 Pb AQCD (U.S. EPA. 2006b). The papers cited herein generally utilized PM
      1 For presentation in this ISA, 2008 NEI, version 2 emissions estimates have been augmented with emissions
      estimates from the 2008 Toxics Release Inventory for a number of additional facilities that had been estimated to
      emit more than 0.5 tons per yr in 2005 and that are not included in the 2008 NEI, v2 for reasons related to the Pb
      emissions reporting threshold. This issue is described (and the facilities identified) as "NEI Identified Issue" #29
      under "Point Data Category" in the 2008 NEI version 2 issues file accessible from
      ftp://ftrj.epa.gov/EmisInventorv/2008v2/doc/2008neiv2 issues.xlsx.
      November 2012                             3-1                  Draft - Do Not Cite or Quote

-------
 1                   sampling data, because a majority of ambient airborne Pb readily condenses to PM. The
 2                   mobile source category included combustion products from organic Pb antiknock
 3                   additives used in piston-engine aircraft (hereafter referred to piston-engine aircraft
 4                   emissions).
      3.2.1        National Emissions Inventory

 5                  The 2006 Pb AQCD (U.S. EPA. 2006b) listed the largest sources to be (in order):
 6                  industrial-commercial-institutional boilers and process heaters (17%), coal utilities
 7                  boilers (12%), mobile sources (10%), iron and steel foundries (8%), and miscellaneous
 8                  sources from industrial processes, incineration, and utilities, each contributing less than
 9                  5% (53%). The sources listed in the 2006 Pb AQCD were based on the 2002 NEI (U.S.
10                  EPA. 2006a). Subsequent correction of computational errors prior to completion of the
11                  2008 NAAQS review provided corrected estimates for the 2002 inventory which
12                  indicated the largest sources to be (in order): mobile sources from leaded aviation gas
13                  usage in piston-engine aircraft (45%), metallurgical industries (23%), manufacturing
14                  (14%), incineration (8%), boilers (6%), and miscellaneous sources contributing less than
15                  5% (U.S. EPA. 2007h). The 2002 and prior year inventories discussed in this document
16                  reflect the corrected information.

17                  Emissions of Pb have dropped substantially over the past forty years, as shown in Figure
18                  3-1 and Figure 3-2. The reduction before 1990 is largely due to the phase-out of Pb as an
19                  anti-knock agent in gasoline for on-road automobiles, as discussed in the 2006 Pb AQCD
20                  (U.S. EPA. 2006b). This  action resulted in a 98% reduction in Pb emissions from
21                  1970-1995. Total Pb emissions over the period 1995-2008 decreased an additional 76%,
22                  from 4,100 tons in 1995 to 964 tons in 2008. Additional emissions reductions are related
23                  to enhanced control of the metals processing industry. In  1995, metals processing
24                  accounted for 42% (2,200 tons) of total Pb emissions. By 2008, metals processing
25                  accounted for 17% (168 tons) of total emissions. This represented more than an order of
26                  magnitude decrease in Pb emissions from metals processing. Emissions from piston-
27                  engine aircraft decreased 34% over this time period. In 1990, nonroad Pb emissions were
28                  990 tons, 830 tons of which were generated from piston-engine aircraft, and represented
29                  19% of total Pb emissions. In 2008, nonroad Pb emissions from piston-engine aircraft
30                  were slightly lower at 550 tons,1 which represented 57% of all Pb emissions. 2008 piston-
31                  engine aircraft emissions were comprised of 254 tons of Pb from emissions at or near
32                  airports and 296 tons of Pb  emitted in flight (i.e., outside the landing and take-off cycles).
33                  "Miscellaneous" emissions from other industrial processes, solvent utilization,
      1 This reflects EPA's best estimates of piston-engine aircraft emissions. The piston-engine aircraft emissions
      inventory can be obtained from the following site: http://www.epa.gov/ttn/chief/net/2008neiv2/2008_neiv2_tsd_draft.pdf.
      November 2012                              3-2                   Draft - Do Not Cite or Quote

-------
              agriculture, and construction constituted 3% of emissions (25 tons) in 2008 (U.S. EPA.
              2011a.2008a).
     250
     200
   o
   8
   r£
   I/I
   O
     150  -
   E 100
   LLJ
   J)
   a.
      50
                                                 Highway Vehicles
                                                I Metal Working and Mining
                                                 Fuel Combustion
                                                I Piston Engine Aircraft
                                                I Miscellaneous
           1970
1975
1980
1985
1990
1995
1999
2002
2005
2008
Source: U.S. EPA (2011 a. 2008a)
Figure 3-1      Trends in Pb emissions (thousand tons) from stationary and
                 mobile sources in the U.S., 1970-2008.
November 2012
                     3-3
                                 Draft - Do Not Cite or Quote

-------
          6 -i
                                                                            Highway Vehicles
                                                                            Metal Workingand Mining
                                                                            Fuel Combustion
                                                                            Piston Engine Aircraft
                                                                            Miscellaneous
                  1990
1995
1999
2002
2005
2008
     Source: U.S. EPA (2011 a, 2008a)

     Figure 3-2      Trends in Pb emissions (thousand tons) from stationary and
                       mobile sources in the U.S., 1990-2008.

 1                  Direct emissions of Pb into the atmosphere primarily come from piston-engine aircraft,
 2                  fuel combustion, and industrial activities. Direct Pb emissions estimated by the 2008 NEI
 3                  are shown in Figure 3-3. Piston-engine aircraft produced 57% of all emissions (550 tons).
 4                  Metal working and mining contributed 168 tons (17%) of Pb emissions in 2008, followed
 5                  by fuel combustion (12%), other industry (10%), and miscellaneous contributions from
 6                  agriculture, solvent utilization, and operation of commercial marine vessels and
 7                  locomotives (3%) (U.S. EPA. 201 la). Pb emissions from the "metal working and
 8                  mining" category include the single primary Pb smelter in the U.S., the Doe Run facility
 9                  in Herculaneum, MO; secondary Pb smelters, mostly designed to reclaim Pb for use in
10                  Pb-acid batteries: and smelters for other metals.
     November 2012
               3-4
                     Draft - Do Not Cite or Quote

-------
                                                                                           551
                                     100        200         300        400
                                                    2008 Emissions (tons)
                                   500
                          600
     Source: U.S. EPA (2011 a)

     Figure 3-3      Nationwide stationary and mobile source Pb emissions (tons) in
                       the U.S. by source sector in 2008.

 1                  There is substantial variability in Pb emissions across U.S. counties, as shown in Figure
 2                  3-4 for the continental U.S. The emissions levels, shown in units of tons, vary over
 3                  several orders of magnitude. Ninety-five percent of U.S. counties, territories,  and tribal
 4                  areas had 2008 emissions below 1 ton; 50% of counties, territories, and tribal areas had
 5                  2008 emissions below 0.041 tons. Jefferson County, MO was the highest emitting single
 6                  county, with over 20 tons of airborne Pb emissions in 2008. Jefferson County is home to
 7                  the Doe Run primary Pb smelting facility, which is the only remaining operational
 8                  primary Pb smelter in the U.S. and is planning to cease the existing smelter operations at
 9                  this site by April, 2014 (DRRC. 2010). Pb emissions from piston-engine aircraft
10                  operating on leaded fuel are estimated to occur at approximately 20,000 airports across
11                  the U.S.  Many of the more active airports are more numerous in highly populated
12                  metropolitan regions, which suggests that emissions from piston-engine aircraft may be
13                  higher in these locations compared with rural areas. In twenty-five counties, piston-
     November 2012
3-5
Draft - Do Not Cite or Quote

-------
                    engine aircraft are estimated to emit cumulatively greater than one ton of Pb in 2008 U.S.
                    EPA (201 la).

        2008 NEI Pb Emissions by County (tons)
        ^H 0.00 -0.14
        ^•0.15-0.41
            0.42 - 0.87
          J 0.88-1.65
        |   | 1.66-2.88

        ^H 4.88 - 8.36
        ^•8.37-20.10
      Source: U.S. EPA (2011 a)

      Figure 3-4       County-level Pb emissions (tons) in the U.S. in 2008.

 3                  Figure 3-5 illustrates the locations and relative magnitude of Pb emissions for 182
 4                  facilities in the U.S. emitting 0.5 tons or more in 2008 (U.S. EPA. 201 la). One facility,
 5                  Doe Run in Herculaneum (Jefferson Co.), MO, emitted more than 10 tons in 2008.
 6                  Additionally, the map illustrates several locations where there is a confluence of point
 7                  sources (not to be confused with total sources including non-point), including Jefferson
 8                  Co., MO, Lake Co., IN, Iron Co., MO, and Gila, AZ, in each of which are facilities that
 9                  were estimated to cumulatively emit more than 10 tons. Among the facilities shown, 124
10                  are non-airport facilities; nine of these were estimated to emit more than 5 tons, 52 to
11                  emit between 1 and 5 tons, and 63 to emit between 0.5 and 1 tons in 2008. Figure 3-5
     November 2012
3-6
Draft - Do Not Cite or Quote

-------
                    additionally includes 58 airports, the six largest of which were estimated to emit between
                    1 and 1.3 tons.

                                                          Facilities with estimates <1 0 and =->0.50 tpy**
                                                          Facilities with estimates <5 0 and =>1.0 tpy*
                                                          Facilities with estimates => 5.0 tpy
                                                          Airports with estimates <1 0 and =>0.50 tpy"
                                                       -^ Airports with estimates <50 and =>1 0 tpy*
                                                          Study Airports
     Figure 3-5
Pb facilities estimated to emit 0.5 tons or more in 2008.
     3.2.2        Anthropogenic Sources

3                   Anthropogenic Pb source categories are organized below in order of magnitude with
4                   regard to the sum of emissions nationally reported on the 2008 NEI (U.S. EPA. 201 la).
5                   Pb sources were reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b) by species. Forms of
6                   Pb commonly observed in the environment are carried forward from the 2006 Pb AQCD
7                   (U.S. EPA. 2006b) and are presented in Table 3-1 to serve as a reference for the
8                   categories of Pb sources described in Sections 3.2.1 and 3.2.2.
     November 2012
                         3-7
Draft - Do Not Cite or Quote

-------
Table 3-1    Pb compounds observed in the environment.
Emission Source
Minerals
Smelting aerosols
Coal combustion aerosols
Coal combustion flue gases
Wood combustion
Waste incineration aerosols
Soils near mining operations
Piston-engine aircraft emissions, racing vehicle exhaust
(combustion of leaded fuel)
Roadside dust
Brake wear, wheel weights
Aircraft engine wear
Observed Pb Compounds
PbS (Galena)
PbO (Litharge, Massicot)
Pb3O4 ("Red Pb")
PbSO4 (Anglesite)
Pb°, PbS
PbSO4, PbO
PbCO3
Pb silicates
PbS
PbSe
Pb°, PbO, Pb02 (Above 1 ,150 K)
PbCI2 (Low rank coals, above 1,150 K)
PbSO4 (Below 1,1 50 K)
PbCO3
PbCI2, PbO
PbCO3
PbSO4
[PbFe6(S04)4(OH)12]
[Pb5(P04)3CI]
[Pb4S04(C03)2(OH)3]
PbS-Bi2S3
Pb oxides, silicates
PbBr2
Alkyl Pb
PbBrCI-NH4CI, PbBrCI-2NH4CI
PbS04, Pb°, PbS04(NH4)S04, Pb3O4, PbO-PbSO4 and
2PbC03-Pb(OH)2
Pb°
Pb°
Source: Biggins and Harrison (1980. 1979): U.S. EPA (2006b)
November 2012
Draft - Do Not Cite or Quote

-------
1
2
3
4
5
6
7
Pb emissions in the U.S. derive from a combination of mined, processed, and imported
Pb. Figure 3-6 illustrates trends in the origin of Pb used in the various sectors described
below over the period 1991-2010. Over this time period, the amount of Pb used in
secondary Pb processing increased by 37%, while exports of Pb increased by 103%.
Primary Pb processing decreased by 67%. Pb mining and imports fluctuated over
1991-2010 without a clear increasing or decreasing trend. In 2008, 1.28 million tons of
Pb were introduced to the market by primary and secondary processing combined. 964
tons of Pb were emitted to the ambient air. Hence, 99.9% of Pb produced in 2008
remained in products or was emitted directly to soil or water following disposal.
                                                                          • Mine
                                                                          • Primary Production
                                                                          Secondary Production
                                                                          • Imports
                                                                          • Exports
     Note: Exports are shown by negative numbers to illustrate that the Pb was leaving the U.S. Data were aggregated into five-year
     totals to stabilize the data shown.
     Source: U.S. Geological Survey (2012. 2006: 2001: 1996)

     Figure 3-6       Five-year totals for Pb mining, primary and secondary production,
                       imports, and exports, 1991-2010.
     November 2012
                            3-9
Draft - Do Not Cite or Quote

-------
                    3.2.2.1      Pb Emissions from Piston-engine Aircraft Operating on
                                 Leaded-Aviation Gasoline and Other Non-road Sources

 1                  The largest source of Pb in the NEI, in terms of total emissions nationally, is emissions
 2                  from piston-engine aircraft operating on leaded aviation gasoline (U.S. EPA. 201 la). As
 3                  outlined in Table 3-1, there are several forms of Pb emitted from engines operating on
 4                  leaded fuel. Dynamometer testing has indicated that Pb emissions from piston engines
 5                  operating on leaded fuel can occur in the particulate and gaseous forms. For example,
 6                  Gidney et al. (2010) performed dynamometer testing on automobiles operating on
 7                  standard gasoline and on gasoline with low levels of organometallic additives. Tetraethyl
 8                  Pb was included since it is still used in piston-engine aircraft fuel. Gidney et al. (2010)
 9                  point out that, where tetraethyl Pb is used as an additive in piston-engine aircraft fuel, the
10                  fuel also contains ethylene dibromide, which reacts with Pb to form Pb bromide and Pb
11                  oxybromides. Pb bromides and Pb oxybromides are more volatile than elemental Pb at
12                  combustion temperatures and are therefore exhausted from the engine. After being
13                  exhausted, the brominated Pb compounds cool to ambient temperatures and condense to
14                  form solid particles. In contrast, emissions of organic Pb would remain largely in the
15                  vapor phase at ambient temperatures. Studies of Pb emissions within enclosed
16                  microenvironments where automobiles were the dominant Pb source cited within the
17                  1986 Pb AQCD (U.S. EPA.  1986a). reported that organic Pb vapors contributed less than
18                  20% of total vehicular Pb emissions. A more recent study  supports this (Shotyk et al.,
19                  2002). The 20% estimate of organic Pb emissions from the previous studies of on-road
20                  Pb emissions may potentially provide an upper bound for organic Pb emissions from
21                  current piston-engine aircraft.

22                  Pb emission rates from piston aircraft vary with fuel consumption rates, which depend on
23                  the engine/airframe combination and the mode of operation of the aircraft. The ASTM
24                  specification for the maximum Pb content in "100 Low Lead", the most commonly used
25                  leaded piston-engine aircraft fuel, is 2.12 g of elemental Pb/gallon (ASTM. 2007). Fuel
26                  consumption rates can be obtained for some engine/aircraft combinations by running
27                  FAA's Emissions and Dispersion Modeling System (FAA. 2011). Fuel consumption for
28                  piston-engine aircraft operating at one airport in the U.S. were estimated to range from
29                  1.6 g/sec of fuel during taxi-out to 15.3 g/sec of fuel during run-up preflight check for
30                  single-engine aircraft and 5.1 g/sec during taxi and 50 g/sec during preflight run-up check
31                  for twin-engine aircraft (Carr et al.. 2011). Fuel consumption rates for aircraft listed in
32                  FAA's Emissions and Dispersion Modeling System were used to develop the Pb
33                  emissions inventory for piston aircraft that are  discussed in Section 3.2.1. EPA estimates
34                  that on average, 7.34 g of Pb is emitted during  a landing and take-off cycle conducted by
35                  piston-engine aircraft (ERG. 2011).
      November 2012                             3-10                  Draft - Do Not Cite or Quote

-------
                    3.2.2.2      Emissions from Metals Processing and Mining

 1                  High Pb emissions were observed in the 2008 NEI (U.S. EPA. 201 la) in Herculaneum,
 2                  MO, where the Doe Run Pb smelter is operated. Although it is set to cease smelting
 3                  operations in 2014 (DRRC. 2010). it is of interest to consider studies of primary smelter
 4                  emissions in the context of the data analyzed in this ISA. Batonneau et al. (2004) and
 5                  Sobanska et al. (1999) found that the Pb content in PM emitted from a primary Pb
 6                  smelter was 56.6% by weight, and the Pb content in PM from a Pb/Zn smelter was 19.0%
 7                  by weight. Choel et al. (2006) confirmed that Pb was strongly associated with sulfur in
 8                  Pb-Zn smelter emission PM, and that Pb sulfates and Pb oxy-sulfates were the most
 9                  abundant species, with important contributions from Pb oxides. Pb concentrations
10                  1,800 meters downwind of the smelter (0.625-0.880 (ig/m3) were roughly thirty-five
11                  times higher than a monitor 1,800 meters upwind (0.017-0.026  (ig/m3).

12                  Fugitive emissions (i.e., unaccounted ambient air Pb emissions) from secondary Pb
13                  processing (e.g., Pb recovery from batteries) can be substantial  over the course of a year,
14                  but they are difficult to estimate. Thurston et al. (2011) performed source apportionment
15                  of PM2 5 found that Pb-PM2 5 concentrations from the Chemical Speciation Network
16                  (CSN) were associated with the metals industry along with Zn-PM25. Goyal et al. (2005)
17                  estimated fugitive emissions using concentration data obtained  from samplers sited in
18                  close vicinity of secondary Pb processing facilities and meteorological data from nearby
19                  weather monitoring stations. Regression modeling and Bayesian hierarchical modeling
20                  were both used to estimate fugitive and stack emissions from secondary Pb processing
21                  facilities in Florida, Texas, and New York. Depending on the model used, median
22                  fugitive emissions were estimated to be 1.0 x 10"6to4.4 x 10"5 g Pb/m2-sec at the Florida
23                  site,  9.4 x 10"7to2.0 x 10"6 g/m2-sec for the Texas site, and 8.8 x 10"7to
24                  1.1 x 10"6 g/m2-sec at the New York site. Median stack emissions estimates varied widely
25                  among the models, with the Florida site median ranging from 1.4 x  10"6to 1.4 x 10"1 g
26                  Pb/sec, the Texas site median ranging  from 8.4 x  10"2to 8.6 x 10"2 g/sec, and the New
27                  York site ranging from 8.4 x 10"3to 1.0 x 10"2 g/sec. Additionally, the Bayesian
28                  hierarchical model was used to estimate fugitive Pb emissions from secondary Pb
29                  processing facilities nationwide using  concentration data as prior information.
30                  Nationwide median fugitive emissions from secondary Pb processing facilities were
31                  estimated to be 9.4 x  10"7 to 3.3 x 1Q"6 g/m2-sec. Recently, speciation of emissions from a
32                  battery recycling facility indicated that PbS was most abundant, followed by Pb sulfates
33                  (PbSO4 and PbSO4-PbO),  PbO and Pb° (Uzu et al.. 2009).

34                  In addition to secondary Pb smelting, Pb emissions occur from processing of other
35                  metals. For example,  a recent study examined Pb emissions from a sintering plant, a
36                  major component of the steel making process in southern France (Sammut et al.. 2010).
      November 2012                            3-11                  Draft - Do Not Cite or Quote

-------
 1                  Cerussite, a Pb carbonate (PbCO3-2H2O), was observed to be the most abundant species
 2                  and contributed 20 g Pb/kg measured PM. In another example, Reinard et al. (2007) used
 3                  a real-time single particle mass spectrometer to characterize the composition of PM]
 4                  collected in Wilmington, Delaware in 2005 and 2006. Strong Pb-Zn-K-Na associations
 5                  were observed within 13% of PM samples. Comparison with stack emissions revealed
 6                  that a nearby steel manufacturing facility was an important source of Pb. Ambient PM
 7                  classes containing only a subset of such elements, e.g., Zn only, Pb-K only were
 8                  non-specific and so could not be mapped to individual sources. Ogulei et al. (2006)
 9                  observed that 6% of Pb in  PM25, along with some O3, Cu, and Fe, was attributed to steel
10                  processing in  Baltimore, MD. Murphy et al. (2007) conducted a detailed study of the
11                  distribution of Pb in single atmospheric particles during the fifth Cloud and Aerosol
12                  Characterization Experiment in the Free Troposphere campaign at the Jungfraujoch High
13                  Altitude Research Station in Switzerland and found that the predominant type of urban
14                  Pb-bearing aerosols contained Pb together with K and Zn. The mode of the size
15                  distribution for this type was around 200 nm.

16                  Waste from current or defunct mines has been shown to present an additional fugitive
17                  source of Pb. For example, distribution of Pb along a haul road connecting an active mine
18                  to a port has been documented in Alaska (see Section 3.6.6). Additionally, Zheng et al.
19                  (2009) applied source apportionment in three northeastern Oklahoma towns to identify
20                  the influence of "chat", or waste piles from formerly operational Pb-Zn mines, on PMi0.2.5
21                  and PM2 5. They estimated that mine waste was responsible for 88% of Pb in PMi0-2.5
22                  samples and 40% of Pb in PM25 samples.
                    3.2.2.3      Fossil Fuel Combustion

23                  Murphy et al. (2007) found that the volatility of Pb and its compounds such as PbO may
24                  result in its presence at high concentration in the submicron fraction of PM emitted from
25                  coal emissions. PbSO4, also derived from coal combustion, has low water solubility
26                  (Barrett etal. 2010). PbSO4 was estimated to comprise 37% of Pb in PM10 from a
27                  2002-2003 study of PM in Shanghai, China (Tan et al.. 2006) and 0.6% of total  PM10
28                  from a study of PM in Taiyuan City, China (Xie et al.. 2009). Murphy  et al. (2007)
29                  presented an estimated U.S. mass budget for Pb emitted from consumption of select fuels
30                  and crude oil. Fuel consumption estimates for 2005 were employed (Freme. 2004). Based
31                  on an annual consumption of 1.0 x  109 metric tons coal with an average Pb concentration
32                  of 20 mg/kg (range: 5 to 35 mg/kg) and using an emission factor (airborne fraction) of
33                  approximately 0.01, coal contributed approximately 200 metric tons Pb/yr to the
34                  atmosphere. At the time of the Murphy et al. (2007) study, there were no emission factors
35                  available to estimate airborne Pb emissions for crude oil or residual oil, but these


      November 2012                             3-12                   Draft - Do Not Cite or Quote

-------
 1                  represent potentially large sources (with total Pb in these sources estimated by Murphy et
 2                  al. (2007) to be as much as 100-500 metric tons/year and 25-700 metric tons/year,
 3                  respectively). These calculations imply that there is substantial uncertainty in estimates of
 4                  Pb emissions resulting from fuel combustion. It is important to note that Murphy et al.
 5                  (2007) state that the crude oil estimates are based on a limited number of samples and
 6                  that there was uncertainty in the estimates of Pb content in residual oil. Furthermore,
 7                  Murphy et al. (2007) was based on data ranging back in time from 1972 to 2005.
 8                  Therefore, the Murphy et al. (2007) findings do not necessarily conflict with reported
 9                  ambient air Pb emissions from the NEI. As part of recent rulemaking, EPA has developed
10                  a draft Pb emission factor of 1.3xlO"5 Ib/MMBtu for boilers larger than 25  MW that use
11                  #2 or #6 fuel oil (U.S. EPA. 2011b).The amounts of Pb emitted from these U.S. sources,
12                  however, are several orders magnitude  smaller than those estimated to arise from coal
13                  combustion in China.

14                  Coal combustion is considered to be a major source of Pb in the atmosphere now that
15                  leaded gasoline has been phased out for use in on-road vehicles (Diaz-Somoano et al..
16                  2009). Global Pb estimates are considered here to inform understanding of U.S. Pb
17                  emissions from coal combustion. McConnell and Edwards (2008) examined correlations
18                  of Pb with BC, Cd, Ce, sea salt Na, non-sea salt S, and Tl in a Greenland ice core and
19                  observed high correlations for BC, Cd,  non-sea salt S and Tl during the period
20                  1860-1940, when coal combustion was the predominant energy source. With the
21                  exceptions of non-sea salt S and Tl, the high correlations were not maintained into the
22                  years 1940-2003, when oil combustion was the most prevalent energy source. This
23                  suggests common industrial sources of PbS or PbSO4. Rauch and Pacyna (2009)
24                  constructed global metal cycles using anthropogenic data from 2000. They confirmed that
25                  the largest anthropogenic airborne Pb emissions arise from fossil fuel combustion, and
26                  they quantified Pb emissions at 85,000  tons/year worldwide. Globally, Pb  emissions from
27                  stationary sources have been increasing and the north-south gradient in aerosol Pb
28                  concentrations over the Atlantic Ocean has disappeared  as a result of industrialization of
29                  the southern hemisphere (Witt et al.. 2006; Pacvna and Pacyna. 2001). The Pb isotope
30                  ratio values (mainly 206Pb/207Pb) for coal from around the world have been compared with
31                  those for atmospheric aerosols. In most parts of the  world, there has been a difference
32                  between the signature for aerosols and that for coal, where the atmospheric 206Pb/207Pb
33                  ratio values are lower, indicative of additional contributions from other sources. Zhang et
34                  al. (2009a) used single particle aerosol mass spectrometry (ATOFMS) to find that PM
35                  containing Pb along with OC and/or EC was attributed to coal combustion processes in
36                  Shanghai, China; this accounted for roughly 45% of Pb-bearing  PM.

37                  Seasonal effects of the contributions of Pb emissions from coal combustion have been
38                  observed. For example, in Tianjin, northern China, the winter heating period starts in
      November 2012                             3-13                  Draft - Do Not Cite or Quote

-------
 1                  November, and the contribution from coal combustion to the Pb aerosol becomes high
 2                  during the winter. This leads to both a high Pb content and a high 206Pb/207Pb ratio. Coal
 3                  consumption and Pb-bearing PM concentrations declined during the summer months, and
 4                  Pb from other sources, mainly vehicle exhaust emissions, became relatively more
 5                  pronounced (Wang et al., 2006c). This seasonal relationship contrasts with observations
 6                  for the U.S. when power stations are more active in summer months (EIA. 2012). The
 7                  increased energy use in summer periods in the U.S. may be attributable to increased
 8                  requirements for air-conditioning.
                    3.2.2.4     Waste Incineration

 9                  Waste incineration studies suggest that the Pb content vary by industrial or municipal
10                  waste stream. For example, Ogulei et al. (2006) performed positive matrix factorization
11                  of PM25 and gaseous copollutants for Baltimore, MD and observed that 63% of Pb in
12                  PM2 5 was attributed to waste incineration during the six day study duration. Other
13                  prevalent compounds associated with incineration included NO3", EC, Cd, Cu, Fe, Mn,
14                  Se, Zn,  O3, and NO2 (note that Cl was not observed in this study). Likewise, Song et al.
15                  (2001) used PMF to deduce sources of PM2 5 measured at Washington, B.C., Brigantine,
16                  NJ, and Underbill, VT during the years 1988-1999. They observed a waste incineration
17                  source loaded with OC, EC, Pb, and Zn at all three sites. A study by Moffet  et al. (2008b)
18                  found that Pb-Zn-Cl-containing particles in PM2 5 samples collected from an industrial
19                  area in Mexico City represented as much as 73% of fine PM. These were mainly in the
20                  submicron size range and were typically mixed with elemental carbon (EC), suggesting a
21                  combustion source. Zhang et al. (2009a) also observed high correlation between Pb and
22                  Cl associated with waste incineration in Shanghai, China. Several Pb isotope studies have
23                  also been used to distinguish contributions to incineration from industrial sources. Isotope
24                  analysis is discussed in more detail in  Section 3.4.1.5. Novak et al. (2008) evaluated
25                  changes in the amounts and sources of Pb emissions in the U.K. and Czech Republic
26                  during the 19th and 20th centuries and found uncertainty in the amount and the isotope
27                  composition of Pb emanating from incineration plants. The isotopic signature of Pb
28                  recycled into the atmosphere by incineration of various industrial wastes could have
29                  shifted from relatively high 206Pb/207Pb ratios consistent with local Variscan  ores to lower
30                  values reflecting imported Precambrian ores. However, other environmental studies
31                  concerning incineration have given highly consistent values for the Pb isotope ratio for
32                  European incineration sources. For example, Cloquet et al. (2006) showed that the Pb
33                  isotopic composition of urban waste incineration flue gases in northeastern France was
34                  ~1.16. De la Cruz et al. (2009) reported that waste incineration was an important source
35                  of Pb and showed that the 206Pb/207Pb and 208Pb/207 Pb ratios for waste incineration Pb
      November 2012                             3-14                  Draft - Do Not Cite or Quote

-------
 1                  emitted in European countries were 1.14-1.16 and 2.43 respectively (de la Cruz et al.,
 2                  2009).
                    3.2.2.5     Wood Burning

 3                  Another potentially uncontrollable source is Pb deposited historically in forests and
 4                  remobilized during forest fires. Section 4.1.3.1 describes residential Pb-PM
 5                  concentrations related to in-home burning of wood contaminated with Pb of ambient
 6                  origin, while this section describes ambient air measurements of Pb attributed to wood
 7                  burning. The 2006 Pb AQCD (U.S. EPA. 2006b) presented data by Nriagu (1989)
 8                  estimating that 1,900 metric tons of Pb were emitted globally each year from wildfires.
 9                  Wildfire Pb emissions were not included in the NEI. Murphy et al. (2007) observed that a
10                  fraction of particles contained small quantities of Pb on biomass particles measured using
11                  ATOFMS to sample directly from forest fire plumes in northwest Canada and eastern
12                  Alaska in July, 2004; these particles also typically contained SO4"2. Several studies
13                  illustrate moderate-to-long range transport of biomass burning plumes containing Pb.
14                  Using positive matrix factorization, Ogulei et al. (2006) estimated that 20% of Pb in
15                  PM2 5 measured in Baltimore, MD was attributed to a July, 2002 episode of wildfires in
16                  Quebec, Canada in his 6-day study. Other components strongly associated with the
17                  Quebec wildfires included NO3", OC, EC, Cd, Mn, Zn, O3, and CO. Qureshi et al. (2006)
18                  also observed a spike up to 42 ng/m3 in Pb-PM2 5 concentration in Queens, NY coinciding
19                  with the Quebec wildfires; for comparison, the authors provide the 3-month average from
20                  July to September of 5.1 ng/m3 for Pb-PM2 5 in Queens. Similarly, Anttilla et al. (2008)
21                  measured PMi0 in Virolahti, Finland during a wildfire in Russia and observed average
22                  Pb-PMio concentrations during the forest fire episodes to be 1.7-3.0 times higher than the
23                  reference concentration of 3.5 ng/m3. Hsu et al. (2009c) observed Pb concentrations in
24                  Taiwan attributed to biomass burning in Northeastern China; Pb was highly correlated
25                  with K attributed to biomass burning during these episodes. Odigie and Flegal (2011)
26                  studied remobilization of Pb during the 2009 wildfires in Santa Barbara, CA. Pb
27                  concentrations in ash samples obtained after the wildfire ranged from 4.3 to 51 mg/kg.
28                  Isotopic analysis of the ash suggested that the remobilized Pb was initially emitted by a
29                  mix of contemporary and previous industrial sources and historic combustion of leaded
30                  gasoline. Grouped with "miscellaneous" Pb emissions, fires from agricultural field
31                  burning and prescribed fires accounted for 2.4 tons of U.S. Pb emissions in 2008 (U.S.
32                  EPA. 2011 a). Polissar et al. (1998) used positive matrix factorization to apportion PM2 5
33                  and found small Pb signals attributed to  the forest fire factor at two of six Alaskan sites
34                  where a forest fire factor was detected; the forest fire factor was dominated by a
35                  combination of BC, FT, and K.
      November 2012                             3-15                  Draft - Do Not Cite or Quote

-------
 1                  Several studies have explored the chemical properties of biomass emissions. Obernberger
 2                  et al. (2006) simulated biomass combustion in a laboratory setting to assess emissions.
 3                  They reported pre-combustion mean Pb content in wood, bark, and logging residues to
 4                  range from 2-5 mg/kg dry basis. They reported volatilization and subsequent
 5                  condensation of Pb emissions from combustion. Van Lith et al. (2008; 2006) studied the
 6                  inorganic element content of wood chips and particle board and the release of inorganic
 7                  elements during combustion of those materials in laboratory experiments. They measured
 8                  a Pb content of 16 mg/kg dry basis in particle board and of 0.44 mg/kg dry basis in
 9                  spruce wood chips. Using three different types of combustion for different materials, they
10                  found that up to 10% of Pb was released at a combustion temperature of 500 °C and up to
11                  85% was released at a temperature of 850 °C. At temperatures greater than 650 °C, PbO
12                  gas was released under oxidizing conditions; under reducing  conditions, Pb gas, PbCl
13                  gas, and PbS gases were released at temperatures above 500 °C. Jimenez et al. (2008)
14                  performed laboratory experiments of olive tree combustion and concluded that Pb
15                  vaporizes upon combustion and then condenses between 900 °C and 560 °C. Jimenez  et
16                  al. (2008) also observed that Pb concentration in PM changes with oxygen content and
17                  temperature, with concentrations converging toward 2,000 mg/kg for increasing percent
18                  available oxygen and increasing temperature.

19                  Pb deposition on trees has been documented in Acadia National Park in Maine with mean
20                  foliar concentrations ranging from <0.5 to 3.1 mg/kg (Wiersma et al.. 2007). Tree ring
21                  core samples obtained in the Czech Republic illustrate that the amount of Pb deposited on
22                  trees from coal and leaded gasoline combustion sources tended to increase over the depth
23                  of the core, with maximum concentrations corresponding to time periods of 1969-1972,
24                  1957-1960, and 1963-1966 in three samples (Zunaet al.. 2011).
                    3.2.2.6      Roadway-Related Sources

                    Contemporary Emissions from Vehicle Parts
25                  Contemporary Pb emissions from motor vehicles may occur because several vehicle parts
26                  still contain Pb. Wheel weights, used to balance tires, are clipped to the rims of tire
27                  wheels in order to balance the tires, and may become loose and fall off. Pb wheel weights
28                  have been banned in several states including Washington, Maine, and Vermont with
29                  legislation considered in Iowa,  California, and Maryland. However, Pb wheel weights are
30                  a source in most states for the period of time covered in this assessment. Ambient air Pb
31                  concentrations near heavily trafficked areas may be related to use of Pb-based wheel
32                  weights that are prone to dislodgement. Root (2000) and Aucott and Caldarelli (2012)
33                  estimated that 2.7-5% of the mass of wheel weights is deposited from cars to the road

      November 2012                             3-16                  Draft - Do Not Cite or Quote

-------
 1                  daily. Aucott and Caldarelli (2012) extrapolated their results for Mercer County, NJ to
 2                  the U.S. to estimate that 480 tons of Pb are deposited to roadways each year. On
 3                  pavement they may be ground into dust by the pounding forces of traffic (Root. 2000).
 4                  For example, Aucott and Caldarelli (2012) estimated that 13.8 ± 5.0% of the deposited
 5                  mass of wheel weights are dispersed each year through abrasion and grinding by traffic.
 6                  Schauer et al. (2006) measured Pb emissions in two traffic tunnels and found that the
 7                  Pb-PM2 5 concentration did not exceed 17% of the Pb-PM10 concentration in any of the
 8                  runs. Schauer et al. (2006) suggested that enrichment in the coarse fraction may have
 9                  been related to wheel weights. Additionally, Schauer et al. (2006) measured PM10 and
10                  PM2.5 composition from brake dust and found concentrations that were low but
11                  statistically significantly greater than zero for Pb in PMi0 (0.02 ± 0.01 mg/g) and Pb in
12                  PM25 (0.01 ± 0.00 mg/g) for semi-metallic brake pads and for Pb in PM10
13                  (0.01 ± 0.00 mg/g) for low-metallic brake pads.  Song and Gao (2011) speciated coarse
14                  and fine PM samples obtained next to the New Jersey Turnpike in winter and summer of
15                  2007-2008. Using principal component analysis, they found that Pb was prevalent in the
16                  factor including automobile exhaust and brake wear. Pb was observed to have a similar
17                  size distribution as Zn in the winter and Zn and Cd in the summer, with higher
18                  concentrations in the fine  fraction at a mode of 0.18-0.32 (im. Fauser (1999) observed
19                  that 90% of particles generated by tire abrasion are smaller than 1 pirn. Similarly, Maher
20                  et al. (2008) observed that vehicle-derived Pb was observed with fine particles smaller
21                  than 1  nm; however, they concluded based on the submicrometer size of the particles that
22                  fuel combustion, not tire wear, was the primary source  of Pb. Additionally, Hjortenkrans
23                  et al. (2007) used material metal concentrations, traffic volume, emissions factors, and
24                  sales data to estimate the quantity of Pb emitted  from brake wear and tires in Stockholm,
25                  Sweden in 2005. They observed that 24 kg (0.026 ton)  of Pb were emitted from brake
26                  wear each year, compared with 2.6 kg (0.0029 ton) of Pb from tire tread wear; an
27                  estimated 549 kg (0.61 ton) was estimated to have been emitted from brake wear in 1998.
28                  McKenzie et al. (2009) determined the composition of various vehicle components
29                  including tires and brakes and found that tires were  a possible source of Pb in stormwater,
30                  but no identification of Pb-containing  PM in stormwater was carried out. However, PM
31                  from tire abrasion is usually found in coarser size ranges (Chon etal., 2010). while those
32                  in the submicron range are more typically associated with combustion and incineration
33                  sources.


                    Unleaded Fuel

34                  Unleaded fuel contains Pb as an impurity within crude  oil (Pacyna et al., 2007). Schauer
35                  et al. (2006) measured Pb in PM2s from tailpipe emissions and observed quantities in
36                  on-road gasoline emissions that were statistically significantly different from zero


      November 2012                            3-17                   Draft - Do Not Cite or Quote

-------
 1                  (83.5 ± 12.80 mg/kg), whereas emissions of Pb from diesel engines were not statistically
 2                  significantly different from zero. Hu et al. (2009a) investigated the heavy metal content
 3                  of diesel fuel and lubricating oil. They found 
-------
 1                  In a recent paper, Laidlaw and Filipelli (2008) analyzed Interagency Monitoring of
 2                  Protected Visual Environments (IMPROVE) data to explore conditions under which
 3                  PM25 particles estimated to be of crustal origins that may contain Pb may become
 4                  airborne. They observed a seasonal pattern in the concentration of PM25 of crustal origins
 5                  in the atmosphere, and they also found that at one IMPROVE site in central Illinois, 83%
 6                  of the variability in concentrations of crustal PM2s was predicted by variability in
 7                  meteorology and soil moisture content. The authors concluded that seasonality and
 8                  climate parameters could not be eliminated in relation to ambient air Pb concentrations.
 9                  Such mechanisms are described in more detail in Section 3.3. As described in
10                  Sections 3.2.2.6 and 3.6.1, there are many contemporary contributions of Pb to soil in
11                  urban areas, and studies summarized here have not quantitatively differentiated the
12                  contributions of these various sources to Pb concentrations in urban areas.


                    Pb from Industrial Activities

13                  Several studies have indicated elevated levels of Pb are found in soil exposed to industrial
14                  emissions, including brownfield sites (Dermont et al., 2010; Verstraete and Van
15                  Meirvenne. 2008; Jennings and Ma. 2007; van Herwijnen et al.. 2007; Deng and
16                  Jennings. 2006). Pb in industrial soils is described in  Section 3.6.1. Recent Pb speciation
17                  results also indicate a contribution from resuspended  soils in areas with previous major
18                  emission sources, but without current major sources.  Data from airborne PM in the
19                  vicinity of an inactive smelter in El Paso, TX were described as consistent with
20                  Pb-humate as the major form of Pb in airborne PM, which the authors suggest relates to
21                  soil resuspension since the local near-surface  soils appeared to have had high humic
22                  content (Pingitore et al.. 2009).


                    Pb from Paint and  Building Materials

23                  Exterior structures painted with Pb-based paint have long been known to be a source of
24                  Pb in outdoor dust or grit (U.S. EPA. 2006b). Recent studies support earlier findings.
25                  Mielke and Gonzales (2008) sampled exterior paint chips from paint applied prior to
26                  1992 on 25 homes in New Orleans, LA, and they found elevated Pb levels in 24 of the 25
27                  tested exterior paints. Weiss et al. (2006) studied the distribution of Pb concentration in
28                  roadway grit in the vicinity of steel structures in New York City and contrasted those data
29                  with roadway grit concentration data where no steel structure was nearby. In each case,
30                  the difference was significant (p <0.006 at one site and p <0.0001 at 4 other sites), with
31                  median Pb concentrations in the grit under the steel structures (median: 1,480 mg/kg)
32                  collectively being 4.4 times higher than median Pb concentrations in the roadway grit not
33                  near a  structure (median: 340 mg/kg).
      November 2012                             3-19                  Draft - Do Not Cite or Quote

-------
 1                  The studies described above considered paint as a source of Pb in outdoor dust through
 2                  gradual abrasion of the painted surfaces. However, atmospheric conditions may break
 3                  down polymers in aging paint, causing previously bound Pb-based pigments to be
 4                  released from the surface more readily. Edwards et al. (2009) performed experiments to
 5                  simulate one week of exposure of Pb-based paints to highly elevated levels of O3
 6                  (11.3 ± 0.8 mg/kg or 150 times the level of the 8-hour NAAQS) and NO2
 7                  (11.6 ± 0.9 mg/kg, or 220 times the level of the annual NAAQS). Following NO2
 8                  exposure, the Pb in wipe samples increased by a median of 260% (p <0.001), and
 9                  following O3 exposure, the Pb in wipe samples increased by a median of 32%
10                  (p = 0.004).

11                  Building demolition was listed as a source of Pb in urban dust in the 2006 Pb AQCD
12                  (U.S. EPA. 2006b). In a follow-up study to previous work cited therein, Farfel et al.
13                  (2005) observed that surface loadings of dust containing Pb increased by 200% in streets,
14                  by 138% in alleys, and by 26% in sidewalks immediately following demolition of an old
15                  building compared with surface loadings of dust containing Pb prior to demolition. One
16                  month later, Pb dust loadings were still elevated in alleys (18% higher than pre-
17                  demolition) and sidewalks (18% higher than pre-demolition), although they had
18                  decreased in streets by 29% compared with loadings prior to demolition. However, Farfel
19                  et al. (2005) did not provide  detailed time series samples from before or after demolition
20                  to judge whether the observations made one month following demolition were within the
21                  normal conditions of the urban area. These results suggest that building demolition may
22                  be a short-term source of Pb in the environment, but it is unclear if demolition is related
23                  to long-term Pb persistence in the environment.


                    Pbfrom Historic Automobile Emissions

24                  Historic Pb emissions, or Pb emitted from on-road vehicles prior to the ban on use of
25                  leaded automobile gasoline,  deposited onto soil and in some areas may serve as a
26                  potential source of airborne Pb.  The historical use of leaded on-road gasoline has been
27                  estimated from documents submitted by Ethyl Corporation to the U.S. Senate (1984) and
28                  a report by the U.S. Geological Survey (USGS. 2005): see Mielke et al. (201 Ic). These
29                  estimates are presented in Figure 3-7. The peak U.S. use of Pb additives  occurred
30                  between 1968 and 1972 with an annual amount of over 200,000 metric tons. According to
31                  Ethyl Corporation, the 1970  use of Pb additives was 211,000 metric tons. By 1980, the
32                  annual use of Pb additives to on-road gasoline decreased to about 91,000 metric tons or a
33                  57% reduction from its 1970 peak. From 1970 to 1990 there was a 92% decline in Pb
34                  additive use. In 1990, the annual U.S. use of Pb additives decreased to
35                  16,000 metric tons, a further 82% decline in Pb additive use from 1980. The final U.S.
36                  ban on the use of Pb additives for highway use in on-road gasoline occurred in 1996.

      November 2012                             3-20                   Draft - Do Not Cite or Quote

-------
                    After that time, Pb additives were only allowed in nonroad applications, including piston-
                    engine aircraft fuel, racing fuels, farm tractors, snowmobiles, and boats.
                                       Lead Additives in U.S. Gasoline
                   250,000
                   200,000
                   150,000
                   100,000
                    50,000
                       (^SfV>fV>fV>fV>fV>fV>fV>fV>fV>fV>fV>fV>fV>»V>
                                                         Year
      Note: Estimates were derived from the proceedings of the U.S. Senate hearings on the Airborne Pb Reduction Act of 1984, S. 2609
      (1984) and the U.S. Geological Survey Pb end use statistics (USGS. 2005).
      Source: Reprinted with permission of Pergamon Press,  Mielke et al. (2011c).

      Figure 3-7       Total U.S. Pb additives in on-road gasoline used in on-road
                        vehicles, 1927-1995.

 3                   Pb emissions from on-road sources were estimated by the U.S. EPA (1986a). which
 4                   indicated that 75% of Pb additives were emitted as exhaust, while the remainder were
 5                   retained within the engine. The tonnages of relatively large >10 (im mass median
 6                   aerodynamic diameter (MMAD) Pb-PM probably settled locally. EPA Q986a) indicated
 7                   that 35% of the Pb-PM at that time were <0.25 um in MMAD. In high traffic urbanized
 8                   areas, soil Pb from historic emissions as well as contemporary sources, are elevated
 9                   adjacent to roadways and decrease with distance away from roadways (Laidlaw and
10                   Filippelli. 2008).

11                   The use of Pb additives resulted in a national scale of influence. For example, variously
12                   sized urbanized areas of the  U.S. have different amounts of vehicle traffic associated with
13                   Pb (Mielke etal., 2010). Figure 3-8  illustrates the national scale of the estimated vehicle-
14                   derived Pb aerosol emissions. Note that the estimated 1950-1982 Pb aerosol emissions in
15                   the 90 cities below vary from 606 metric tons for Laredo,  Texas, to nearly
      November 2012
3-21
Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
10
11
12
13
150,000 metric tons for the Los Angeles-Long Beach-Santa Anna urbanized area.
Although this figure might imply that the soil Pb concentration in these areas would be
proportional to the magnitude of historic on-road emissions in each city, it is recognized
that the atmospheric dispersion of emissions, as well as the atmospheric deposition and
subsequent distribution associated with surface runoff, will have varied substantially
among the cities illustrated. Additionally, the amount of soil turnover since 1982 may
have varied substantially among the cities illustrated in Figure 3-8. depending on the
amount of highway construction in  those cities. As noted in Section 3.2.2.6 , there have
historically been, and are currently, many additional sources of Pb contributing to near-
roadway soil Pb concentrations. Data are lacking that quantify the range of airborne Pb
concentrations originating from historic Pb in resuspended soil particles, but data on
airborne concentrations near roadways indicate measured air Pb concentrations (from all
contributing sources) to be generally less than  0.02 (ig/m3 (Section 3.5.3.2).
              JP"
                                                                                   i    70.  I
                                                                 -•  *   t-          V
                                                                  *     *«*      *?
                                                      •                    « « •»   a W
                                                                        r
                 U.S. Urbanized Areas
                 Pb 1950 -1982 (metric tons)
                  •  608
                  •  652 • 4.570
                  •  4.662 - 8.108
                  0 8,403 - 16.614
                  0 16.623-91.878
                  A 149.938
      Note: The numbers on the map are rankings of each urbanized area (UA). The size of each dot refers to the magnitude of motor
      vehicle gasoline-related emissions for each group of UAs.
      Source: Reprinted with permission of Pergamon Press, Mielke et al. (2011c)

      Figure 3-8        Estimated Pb aerosol inputs from on-road gasoline into 90 U.S.
                         urbanized areas (UAs), from 1950 through 1982.
      November 2012
                            3-22
Draft - Do Not Cite or Quote

-------
     3.3       Fate and Transport of Pb
 i
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
There are multiple routes of exposure to Pb, including direct exposure to atmospheric Pb,
exposure to Pb deposited in other media after atmospheric transport, and exposure to Pb
in other media that does not originate from atmospheric deposition. As a result, an
understanding of transport within and between media such as air, surface water, soil, and
sediment is necessary for understanding direct and indirect impacts of atmospheric Pb as
well as the contribution of atmospheric Pb to total Pb exposure. Figure 3-9 describes
relevant Pb transport pathways through environmental media discussed in this chapter
and their relationship to key environmental and human exposure pathways for which
some or all of the Pb is processed through the atmosphere. This discussion includes
recent research on atmospheric transport of Pb, atmospheric deposition and resuspension
of Pb, Pb transport in surface waters and sediments, and Pb transport in soil. Facets of
fate and transport relevant to the ecological effects of Pb are also summarized in Section
7.2.
                     Source
                                                  Biota
                                                                            Human
                                                                            Exposure
                                                     Drinking Water

                                                     Transport
                                                     to Sea
      Note: Media through which Pb is transported and deposited are shown in bold.
      Figure 3-9       Fate of atmospheric Pb.
     November 2012
                           3-23
Draft - Do Not Cite or Quote

-------
      3.3.1       Air

 1                  The 2006 Pb AQCD (U.S. EPA. 2006b) concluded that Pb was primarily present in
 2                  submicron aerosols, but that bimodal size distributions were frequently observed. Pb-PM
 3                  in the fine fraction is transported long distances, found in remote areas, and can be
 4                  modeled using Gaussian plume models and Lagrangian or Eulerian continental transport
 5                  models as reported by several studies. Good agreement between measurements and these
 6                  models have been reported. Historical records of atmospheric deposition to soil,
 7                  sediments, peat, plants, snowpacks, and ice cores have provided valuable information on
 8                  trends and characteristics of atmospheric Pb transport. Numerous studies using a variety
 9                  of environmental media indicated a consistent pattern of Pb deposition peaking in the
10                  1970s, followed by a more recent decline. These findings indicated that the elimination of
11                  leaded gasoline for motor vehicles and systematic reductions in emissions from other Pb
12                  sources has not only led to lower atmospheric concentrations in areas impacted by
13                  vehicles  (Section 3.5). but a pervasive pattern of decreasing atmospheric Pb deposition
14                  and decreasing concentrations in other environmental media even at great distances from
15                  sources.
                    3.3.1.1      Transport

16                  Recent research on long range transport as well as transport of Pb in urban areas has
17                  advanced the understanding of Pb transport in the atmosphere. While the 2006 Pb AQCD
18                  described long range Pb transport as essentially a process of submicron PM transport
19                  (U.S. EPA. 2006b), much of the recent research on Pb transport has focused on
20                  interactions between anthropogenic and coarser geogenic PM that leads to incorporation
21                  of Pb into coarse PM as well as subsequent transformation on exposure to mineral
22                  components of coarse PM. Using scanning electron microscopy (SEM), Schleicher et al.
23                  (2010) observed interactions of anthropogenic soot and fly ash particles on the surfaces
24                  of coarse geogenic mineral particles in Beijing, China and concluded that toxic metals
25                  were often associated with TSP. Murphy et al. (2007) found that PM released from wild
26                  fires and transported over long distances scavenged and accumulated Pb and sulfate
27                  through coagulation with small Pb rich PM during transport and that Pb was associated
28                  with PM over a wide size range. Erel et al. (2006) also found that Pb enrichment factors
29                  calculated for PMi0 from dust storms collected in Israel were much greater than those
30                  sampled at their north African source, suggesting that the dust samples had picked up
31                  pollutant Pb in transit between the  Saharan desert and Israel. Marx et al. (2008)
32                  characterized dust samples collected from the surface of glaciers and in dust traps on the
33                  remote west coast of New Zealand's South Island and observed that most of the dust
      November 2012                             3-24                  Draft - Do Not Cite or Quote

-------
 1                  samples were enriched in metals, including Pb, compared with their source area
 2                  sediments.

 3                  Pb accumulated on mineral dusts is also subject to atmospheric transformations. PbSO4 is
 4                  one of the main constituents of Pb-containing aerosols resulting from coal combustion
 5                  (Giere et al.. 2006) and it has been  shown to react with calcite, CaCO3, a PM mineral
 6                  component, to form Pb3(CO3)2(OH)2, Pb(CO3) and Ca(SO4)2-H2O on the surface of the
 7                  calcite (Falgayrac et al.. 2006). In laboratory experiments, (Ishizaka et al.. 2009) also
 8                  showed that PbSO4 could be converted to PbCO3 in the presence of water. Approximately
 9                  60-80% was converted after only 24 hours for test samples immersed in a water droplet.
10                  This compared with only 4% conversion for particles that had not been immersed. As a
11                  result of recent research, there is considerable evidence that appreciable amounts of Pb
12                  can accumulate on coarse PM during transport, and that the physical and chemical
13                  characteristics of Pb can be altered by this process due to accompanying transformations.
                    3.3.1.2     Deposition

14                  The 2006 Pb AQCD (U.S. EPA. 2006b) documented that soluble Pb was mostly removed
15                  by wet deposition, and most of the insoluble Pb was mostly removed by dry deposition.
16                  As a result, dry deposition was the major removal mechanism for Pb in coarse PM (which
17                  is mainly insoluble and settles faster than fine PM), and wet deposition was the most
18                  important removal mechanism for fine PM and Pb halides (which were more soluble).
19                  Numerous studies reported that Pb dry deposition velocities in the U.S. were mostly
20                  within a range of 0.05 to 1.0 cm/sec and dry deposition fluxes ranging from 0.04 to
21                  4 mg/m2-yr. Precipitation concentrations ranged mostly from 0.5 to 60 ug/L, but with
22                  considerably lower concentrations in remote areas, and wet deposition fluxes in the U.S.
23                  ranged from 0.3 to  1.0 mg/m2-yr. Wet deposition was linked to precipitation intensity,
24                  with slow even rainfalls usually depositing more Pb than intense rain showers. Rain
25                  concentrations decreased dramatically between the early 1980s and the 1990s, reflecting
26                  the overall decreasing trend in Pb emissions due to elimination of leaded motor vehicle
27                  gasoline. A summary of studies investigating total deposition including both wet and dry
28                  deposition indicated typical deposition fluxes of 2-3 mg/m2-yr and dry to wet deposition
29                  ratios ranging from 0.25 to 2.5. Seasonal deposition patterns can be affected by both
30                  variations in local source emissions and vegetation cover, and as a result a consistent
31                  seasonal pattern across studies has not been observed, although there have been only a
32                  few investigations. The 2006 Pb AQCD (U.S. EPA. 2006b) concluded that resuspension
33                  by wind and traffic contribute to airborne Pb near sources.
      November 2012                             3-25                  Draft - Do Not Cite or Quote

-------
                    Wet Deposition

 1                  The 2006 Pb AQCD (U.S. EPA. 2006b) documented that dry deposition was the major
 2                  removal mechanism for Pb in coarse PM and wet deposition as the most important
 3                  removal mechanism for fine PM. Which process is most important for atmospheric
 4                  removal of metals by deposition is largely controlled by solubility in rain water. Metal
 5                  solubility in natural waters is determined by a complex multicomponent equilibrium
 6                  between metals and their soluble complexes and insoluble ionic solids formed with
 7                  hydroxide, oxide, and carbonate ions. This equilibrium is strongly dependent on pH and
 8                  ionic composition of the rain water. As pH increases, Pb solubility is reduced. Theodosi
 9                  et al. (2010) found that solubility of Pb was near 100% when rain water pH was
10                  measured to be less than 4.5. As a consequence, it is possible that efforts to reduce acidity
11                  of precipitation could also reduce wet deposition of Pb. Recent research confirms the
12                  general trend described in the 2006 Pb AQCD (U.S.  EPA. 2006b) that Pb associated with
13                  fine PM is usually more soluble in rain water than Pb associated with coarse PM, leading
14                  to a relatively greater importance of wet deposition for fine Pb and of dry deposition for
15                  coarse Pb. Theodosi et al. (2010) concluded that larger particles were less soluble,
16                  because Pb solubility decreased with increasing dust loading. Likewise, Preciado and Li
17                  (2006) observed that solubility decreased with increasing particle size. Moreover,
18                  Theodosi et al. (2010) observed that 53% of wet deposition samples were comprised of
19                  particulate Pb, not soluble Pb. This finding suggests  that wash-out can be equally
20                  important to wet deposition as solubility.

21                  Although recent observations are consistent with previous findings, they also indicate
22                  considerable spatial and seasonal variability. Birmili et al. (2006) found that Pb solubility
23                  varied between the two main Pb-containing size fractions, <0.5 (im (-40%) and
24                  1.5-3.0 (im (-10%), indicative of a different chemical speciation. However, the
25                  observation that the amount of soluble Pb was higher in their U.K.  samples than in an
26                  analytically  identical study carried out in Seville, Spain (Fernandez Espinosa et al..
27                  2004). led them to conclude that Pb solubility in fine PM may vary on a regional basis
28                  (Birmili et al.. 2006). For PM10 from Antarctica, 90 to  100% of the Pb was insoluble at
29                  the beginning of the summer season (November), but by the end of the summer
30                  (January), approximately 50% was soluble. Most of  the Pb was from long range transport
31                  (Annibaldi et al.. 2007). These  studies illustrate the variable nature of atmospheric Pb
32                  solubility.
      November 2012                             3-26                   Draft - Do Not Cite or Quote

-------
                    Dry Deposition

 1                  Recent research on dry deposition has focused differences between urban or industrial
 2                  sites and rural or less industrial areas. For locations outside of industrial areas, new
 3                  measurements of Pb dry deposition fluxes are similar to those reported in the
 4                  2006 Pb AQCD (U.S. EPA. 2006b). but in industrialized urban areas, they are
 5                  considerably greater than in nonindustrialized areas. Deposition is typically documented
 6                  by measurements of Pb concentrations on surface material or by measurements of flux.
 7                  For example, Hasselbach et al. (2005) documented Pb concentration in moss, and the
 8                  spatial distribution of Pb concentration in moss, as evidence of Pb deposition from truck
 9                  traffic between an Alaskan Zn-Pb mine and a port. Additionally, Maher et al. (2008)
10                  measured Pb deposition onto leaves near a road in Norwich, U.K. Their results are
11                  described in detail in Section 3.6.6.

12                  Several studies presented measurements of dry deposition flux obtained by capturing
13                  deposited particles onto a sampling substrate. Hence, these measurements did not provide
14                  information on net deposition following resuspension of deposited material.
15                  Resuspension processes and measurements thereof are described in Section 3.3.1.3.  For
16                  example, Yi et al. (2006) calculated dry  deposition fluxes for trace elements including Pb
17                  in New York-New Jersey harbor and observed much greater dry deposition fluxes for this
18                  urban industrial site in Jersey City (mean: 50 (ig/m2-day) than for suburban
19                  New Brunswick (mean: 8 (ig/m2-day). Sabin and Schiff (2008) measured dry Pb
20                  deposition flux along a transect from Santa Barbara to San Diego, CA in 2006 and
21                  observed a range of 0.52-14 (ig/m2-d for the median values across the eight sites. The
22                  highest median Pb flux was observed at  Los Angeles Harbor, which is downwind of a
23                  harbor with a mix of industrial (harbor-related) and urban activities (14 (ig/m2-day). The
24                  second highest median Pb flux was observed at San Diego Bay, a military port
25                  (3.3 (ig/m2-day). This is consistent with  similar observations of dry deposition fluxes that
26                  were more than ten times greater in urban Chicago than in rural South Haven, Michigan
27                  (Paode et al.. 1998). These results illustrate the strongly localized nature  of atmospheric
28                  Pb deposition in source rich areas.

29                  Elements from anthropogenic sources, including Pb, were generally associated with fine
30                  PM. In a study of Tokyo Bay (Sakata and Asakura. 2008). reported an average dry
31                  deposition velocity of 1.06 cm/sec, which is at the upper end of dry deposition velocities
32                  reported in the 2006 Pb AQCD (U.S. EPA. 2006b). They also reported that dry
33                  deposition fluxes were greater in industrially impacted urban areas, ranging from
34                  12-17 mg/m2-yr, more than  10 times the  upper bound of the range reported in the
35                  2006 Pb AQCD (U.S. EPA. 2006b).
      November 2012                             3-27                  Draft - Do Not Cite or Quote

-------
 1                  Recent results also confirmed the trend of decreasing overall deposition fluxes after
 2                  removal of Pb from on-road gasoline, as described in the 2006 Pb AQCD (U.S. EPA.
 3                  2006b). Watmough and Dillon (2007) found that the bulk annual deposition of Pb in a
 4                  central Ontario forested watershed during 2002-2003 was 0.49 mg/m2-yr; this was lower
 5                  than the value of 1.30-1.90 mg/m2-yr for 1989-91 and represented a 75% decline in Pb
 6                  deposition. It was consistent with the decline more generally observed for the
 7                  Northeastern U.S. as a consequence of the restrictions to alkyl-Pb additives in on-road
 8                  gasoline. From previously published work, and in agreement with the precipitation data
 9                  described above, most of the decline in Ontario Pb deposition took  place before the start
10                  of the Watmough and Dillon (2007) study.

11                  Within-day variation in deposition fluxes was observed to be related to the urban
12                  boundary layer. Lim et al. (2006) observed higher deposition fluxes of Pb and other
13                  transition metals during the nighttime in Los Angeles, when inversions are frequent
14                  occurrences. Deposition mass was  also greater for particles larger than 10 nm in the
15                  urban areas where measurements occurred. Larger particle deposition flux was greater
16                  during the day, indicating that the source was  anthropogenic.

17                  Several important observations can be highlighted from the few studies of atmospheric
18                  Pb deposition carried out in the past several years. Deposition fluxes have greatly
19                  declined  since the removal of Pb additives from on-road gasoline. However, more recent
20                  results in industrial areas indicate that local deposition fluxes there  are much higher than
21                  under more typical conditions. In general, wet deposition appears to be more important
22                  for Pb in fine PM, which is relatively soluble; and dry deposition appears to be generally
23                  more important for Pb in coarse PM, which is relatively insoluble. However, the relative
24                  importance of wet and dry deposition is highly variable with respect to location and
25                  season, probably reflecting both variations in Pb speciation and variations in external
26                  factors such as pH and rain water composition. Although industrial Pb emissions are
27                  mainly associated with fine PM, and wet deposition is likely to be more important for this
28                  size range, a substantial amount of Pb is apparently removed near industrial sources.
                    3.3.1.3     Resuspension of Pb from Surface Soil to Air after
                                 Deposition

29                  The following information focuses on issues regarding the transport processes affecting
30                  resuspended soil Pb and dust Pb in urban environments. As described in Section 3.2.1.
31                  the greatest point source Pb emissions in the U.S. occur in locations near specific major
32                  facilities, such as secondary smelters, and other industrial operations involving large
33                  scale metal processing or fuel combustion. However, in the absence of such sources and
      November 2012                             3-28                  Draft - Do Not Cite or Quote

-------
 1                  in the vicinity of previous major sources, the 2006 Pb AQCD (U.S. EPA. 2006b)
 2                  concluded that resuspension by mechanical stressors such as traffic, construction, and
 3                  wind can be a source of airborne Pb above background levels near sources, with Pb
 4                  accounting for between 0.002 to 0.3% of the mass of resuspended PMi0. Reentrainment
 5                  of deposited Pb complicates air related pathways of human (Section 4.1.1) and ecosystem
 6                  exposure (Section 7.2.2).

 7                  Results from several studies have suggested minor contributions from resuspension to
 8                  airborne Pb concentration from various sources, including city centers (Laidlaw and
 9                  Filippelli. 2008). major freeways (Sabin et al.. 2006b). and steel structures with abrading
10                  paint (Weiss et al.. 2006). Recently, Laidlaw et al. (2012) modeled concurrent
11                  measurements of (log-transformed) air Pb-PM2 5 as a function of (log-transformed)
12                  airborne soil measured in PM2 5. They observed a marginal but statistically significant
13                  increase in air Pb of 0.84% with a 1% increase in  airborne soil (p <0.01). As noted in the
14                  2006 Pb AQCD  (U.S. EPA. 2006b). the contribution of resuspended soil and dust to the
15                  airborne burden  may be significant from highly contaminated sites (e.g., active  or
16                  abandoned industrial facilities and Superfund sites). In contrast, as summarized in
17                  Section 3.5.3. Pb concentrations near roads in  urban areas are one to two orders of
18                  magnitude below the current Pb NAAQS.

19                  The urban environment can be considered quite different from natural landscapes because
20                  it has been highly modified by human activity, including above- and below-ground
21                  infrastructure, buildings, and pavement, and a  high density of motorized transportation.
22                  These factors may influence the distribution and redistribution of Pb-bearing PM. As
23                  shown in Figure 3-10. urban turbulence occurs on several scales. Transport and
24                  dispersion of urban grit is subject to air movement within the urban canopy layer, where
25                  air movement is driven by air velocity within the urban boundary layer and urban
26                  topographical conditions  such as building shape, building facade, and street canyon
27                  aspect ratio (Fernando. 2010). Within a street canyon, air circulates and tends to form
28                  counter-rotating eddies along the height of the canyon (Figure 3-10). which result in
29                  lower mean components of air movement, higher  turbulence components, and higher
30                  shear stress within the canyon compared with open field conditions (Kastner-Klein and
31                  Rotach. 2004; Britter and Hanna. 2003). Recirculation around intersection corners and
32                  two-way traffic conditions can also enhance turbulence levels, while one-way traffic
33                  conditions increase  air velocity along the street (Soulhac et al.. 2009; Kastner-Klein et al..
34                  2003; Kastner-Klein et al.. 2001). Sedefian et al. (1981) measured the length scales of
35                  turbulent eddies  resulting from passing 50 mph (22.2 meters/seconds) traffic on a test
36                  road and observed scales of 0.6-2.7 meters when winds were perpendicular to the test
37                  road and scales of 1.8-2.7 meters when winds were parallel to the road.
      November 2012                             3-29                  Draft - Do Not Cite or Quote

-------
                                                                           ABL
               ABL. -  Atmospheric boundary layer
               UBL -  Urban boundary layer
               RSL -  Roughness sublayer
                    (transition layer, wake layer, inlerfacial layer)
               UCL -  Urban canopy layer
               USL -  Urban surface layer
               ML •  Mixed layer
               CFL -  Constant flux layer
                    (Lejnertial sublayer- ISL)
                                                       Street
                                                      Canyon
                                                    Recirculation
Note: Top: multiple scales within the atmospheric boundary layer. Bottom: illustration of airflow recirculation within a single street
canyon located in the urban canopy layer.
Source: Reprinted with permission of Annual Reviews, Fernando (2010)

Figure 3-10      Scales of turbulence within an  urban environment.
November 2012
3-30
Draft - Do Not Cite or Quote

-------
 1                   Recent research on urban PM transport is highly relevant to Pb transport and dispersion
 2                   because Pb is most prevalently particle-bound. Relevant results for Pb exposure in these
 3                   areas include observations that PM concentration peaks dissipate more rapidly on wider
 4                   streets than in narrow street canyons (Buonanno et al.. 2011); concentrations are typically
 5                   low next to a building because either less source material is available or less material
 6                   penetrates the boundary layer of the building (Buonanno et al.. 2011); and there are
 7                   stronger inverse relationship between mean wind speed and PM concentration fluctuation
 8                   intensities at middle sections of urban street blocks compared with intersections (Hahn et
 9                   al.. 2009). Patra et al. (2008) conducted experiments in London, U.K. in which a "tracer"
10                   grit (i.e., rock salt) was applied to a road and then the grit's dispersion by traffic was
11                   measured over time to simulate resuspension and transport of road dust. During the
12                   experiments, 0.039% of the tracer grit was measured to move down the road with each
13                   passing vehicle, 0.0050% was estimated to be swept across the road with each passing
14                   vehicle, and 0.031% was estimated to become airborne when a vehicle passed.

15                   Harris and Davidson (2008) developed a model of resuspension of single particles
16                   initially at rest on a solid surface based on the balance of lift, drag, gravity, torque, and
17                   adhesion forces on the particle in addition to turbulent wind fluctuations within a
18                   simulated urban boundary layer. In their model simulations showed 2.5 (im and 10 (im
19                   particles to reach a maximum height of 0.04-0.06 meter above ground level (AGL), while
20                   50 (im particles reached a maximum of 0.2 meter AGL and 75 (im particles reached at
21                   least 0.4 meter AGL, depending on friction velocity. Empirical analysis has shown that
22                   lift force is proportional to particle diameter to the power of approximately  1.5, so that
23                   large particles actually have larger initial vertical displacement than smaller particles. At
24                   the same time, lateral travel distance following resuspension tended to decrease linearly
25                   with increasing particle size, reflecting the counteracting force of gravity. For all cases
26                   simulated, the resuspension and deposition were estimated to occur over time frames on
27                   the order of seconds.

28                   Early work described resuspension as an important process for wind erosion for particles
29                   up to 100 (im, but indicated that particles larger than this rarely became suspended, and
30                   that the tendency of particles to remain airborne long enough for appreciable transport
31                   decreases sharply beyond a size of 10 to 20 (im  (Nicholson. 1988; Gillette et al..  1974).
32                   As a result, long range transport of dust is usually limited to particles smaller than 10 (im
33                   (Prospero. 1999).

34                   In urban environments the transport distance that must be traversed to penetrate indoors
35                   can be very short, and at the same time resuspension and dispersion of larger particles
36                   may be caused by motor vehicles. Resuspension of road dust by traffic becomes more
37                   difficult with decreasing particle size because adhesive forces  are stronger than shear
      November 2012                              3-31                   Draft - Do Not Cite or Quote

-------
 1                   force that is imparted by traffic-induced turbulent air movement (Harris and Davidson.
 2                   2008). The critical diameter at which resuspension occurs when a particle's settling
 3                   velocity becomes lower than the friction velocity of air needed to move the particle from
 4                   rest. The work of Gillette et al. (1974). in which a critical diameter of roughly 20 (im was
 5                   estimated, is based on wind in an open landscape. It would be reasonable to expect that
 6                   friction velocity would be higher for urban environments with traffic-induced turbulence
 7                   (Britter and Hanna. 2003). Hence, it is possible that larger particles are resuspended in a
 8                   heavily-trafficked urban setting (Nicholson and Branson. 1990).

 9                   Particle  size determines the distance particles can travel and the height which they can
10                   achieve  before they are removed by gravitational settling. Song and Gao (2011) observed
11                   that coarse mode Pb concentration was negatively correlated with wind speed (Dp =14:
12                   p = -0.62;  Dp = 7.8; p = -0.76), which suggests that coarse Pb may be dispersed by  wind.
13                   Observations in near road environments indicate that roughly 15% of Pb in airborne dust
14                   in areas  impacted by heavy traffic is greater than 10 (im (Cho et al.. 2011; Lough et al..
15                   2005; Zereini et al.. 2005). Sabin et al. (2006b) also collected three size fractions greater
16                   than 11  (im and found that approximately 25% of all Pb mass was associated  with
17                   particles larger than 29 (im at a site  10 meters from a freeway, but only a very small
18                   percentage of Pb mass was in this size fraction at an urban background site. These results
19                   suggest that both size distribution and concentrations in the immediate vicinity of
20                   roadways might differ from estimates based on concentrations from monitoring sites at
21                   some distance from roads or on elevated rooftops. In these studies, only one size fraction
22                   slightly greater than 10 (im was collected, but another study of road dust (not specific to
23                   Pb) reported size fractions extending up to 100 (im with a mass median diameter of
24                   greater than 60 (im (Yangetal..  1999). Although the Yang et al. (1999) study did not
25                   include Pb, the results suggest that resuspended dust can be larger than PMi0.
26                   Collectively, the size distribution of Pb-containing resuspended dust is uncertain.

27                   Recent resuspension studies complement previous research indicating street dust half-
28                   lives on  the order of one-hundred days (Allott et al.. 1989). with resuspension and street
29                   run-off as major sinks (Vermette et al.. 1991) as well as observations of a strong
30                   influence of street surface pollution on resuspension (Bukowiecki et al.. 2010).
31                   observations of greater resuspension of smaller PM than coarser PM, leading  to
32                   enrichment of metal concentrations in resuspended PM relative to street dust (Wong et
33                   al.. 2006) and observations of wind  speed, wind direction, vehicular traffic, pedestrian
34                   traffic, agricultural activities, street sweeping and construction operations as important
35                   factors determining resuspension.
      November 2012                              3-32                   Draft - Do Not Cite or Quote

-------
      3.3.2        Water

 1                  As described in the 2006 Pb AQCD (U.S. EPA. 2006b). atmospheric deposition, urban
 2                  runoff, and industrial discharge have been identified as major sources of Pb in surface
 3                  waters. Water columns have been described as transient reservoirs with Pb residence
 4                  times in lakes typically several months long, and shorter residence times expected in
 5                  turbulent waterways. Because dispersal in waterways is a relatively rapid process,
 6                  concentrations in surface waters are highest near sources of pollution before substantial
 7                  Pb removal by flushing, evaporation, and sedimentation occurs. Transport in surface
 8                  water is largely controlled by exchange with sediments, and the cycling of Pb between
 9                  water and sediments is governed by chemical, biological,  and mechanical processes that
10                  are affected by many factors, including salinity, organic complexation, oxidation-
11                  reduction potential, and pH; ecological impacts of these factors are described in Section
12                  7.4.9. As described in the 2006 Pb AQCD (U.S. EPA. 2006b). metals in waterways are
13                  transported primarily as soluble chelates  and ions, or adsorbed on colloidal surfaces,
14                  including secondary clay minerals,  iron and manganese oxides or hydroxides, and
15                  organic matter, and adsorption on organic or inorganic colloids is particularly important
16                  for Pb. The extent of sorption strongly depends on particle size as smaller particles have
17                  larger collective surface areas. Aqueous Pb concentrations also increase with increasing
18                  salinity. Pb is found predominantly as PbO or PbCO3 in aqueous ecosystems. Pb is
19                  relatively stable in sediments, with  long residence times and limited mobility. However,
20                  Pb-containing sediment particles can be remobilized into the water column. As a result
21                  trends in sediment concentration tend to follow those in overlying waters. Fe and Mn
22                  oxides are especially susceptible to recycling with the overlying water column. Although
23                  resuspension of sediments into overlying waters is generally small compared to
24                  sedimentation, resuspension of contaminated sediments is often a more important source
25                  than atmospheric deposition. Organic matter (OM) in sediments has a high capacity for
26                  accumulating trace elements. In an  anoxic environmental  removal by sulfides is
27                  particularly important. The following section highlights recent literature regarding the
28                  fate and transport of Pb in water systems. Section 7.2 synthesizes this information with
29                  ecosystem exposure data.

30                  Runoff from  storms was identified as an important source of Pb in aquatic systems (U.S.
31                  EPA. 2006b'). Runoff from atmospheric deposition, buildings due to paint, gutters,
32                  roofing materials and other housing materials were identified as major contributors to Pb
33                  in runoff waters. Investigations of building material contributions indicated runoff
34                  concentrations ranging from 2 to 88 mg/L, with the highest concentrations observed from
35                  more than 10-year-old paint and the lowest concentrations from residential roofs. There
36                  was some indication that Pb from roofing materials, siding, and piping could be due to
37                  dissolution of Pb carbonate (cerussite)  or related compounds. In several studies Pb in

      November 2012                             3-33                   Draft - Do Not Cite or Quote

-------
 1                  runoff was consistently mostly PM, with a relatively small dissolved fraction. Runoff
 2                  release was dependent on storm intensity and length of dry periods between rain events,
 3                  with greater runoff of Pb associated with more intense storms and with longer periods
 4                  between rain events. Several studies indicated a "first flush effect," with highest runoff
 5                  concentrations observed at the beginning of a rain event.
                    3.3.2.1       Pb Transport in Water and Sediment

 6                  Recent publications provide additional detail regarding Pb adsorption on iron rich and
 7                  organic rich colloids. Correlation between Pb concentration in unfiltered water with total
 8                  Fe was observed (Hassellov and von der Kammer. 2008). which is consistent with
 9                  previous research using cross flow filtration (Pokrovsky and Schott. 2002; Ross and
10                  Sherrell. 1999) and SEM examination of single particles (Taillefert et al.. 2000).

11                  Two distinct colloidal phases, one organic-rich (0.5-3 nm in diameter) and the other Fe-
12                  rich (>3 nm in diameter), have been observed to coexist in both soil isolates and river
13                  water (Stolpe and Hassellov, 2007). Pb was observed to be predominantly associated with
14                  Fe-oxide PM in river water but also associated with the organic colloids in the soil
15                  isolates (Hassellov and von der Kammer. 2008). Investigation of Pb binding onto
16                  ferrihydrite showed Pb binding data were consistent with Pb being held at the surface by
17                  sorption processes, rather than enclosed within the particle structure (Hassellov and von
18                  der Kammer. 2008).

19                  Observations in boreal rivers and soil pore waters in permafrost dominated areas of
20                  Central Siberia indicated that Pb was transported with colloids in Fe-rich waters. Trace
21                  elements that normally exhibited limited mobility (including Pb) had 40-80% of their
22                  annual flux in the nominal dissolved phase, operationally defined as material that passes
23                  through a 0.45 (im pore-size filter, and that these metals had a higher affinity for organo-
24                  mineral Fe-Al colloids (Pokrovsky  et al.. 2006). Pokrovsky et al. (2006) postulated that
25                  during the summer, rainwater interacts with degrading plant litter in the top soil leading
26                  to the formation of Fe-Al-organic colloids with incorporated trace elements.  Migration of
27                  trace element-Fe-Al-OM colloids may result in export of Pb and other elements to
28                  riverine systems. Most of the transport occurred after thawing had commenced. This
29                  contrasts with permafrost free areas where trace elements such as Pb are incorporated
30                  into iron colloids during OM-stabilized Fe-oxyhydroxide formation at the redox
31                  boundary of Fe(II)-rich waters and  surficial DOC-rich horizons.  Similarly, during a
32                  spring flood (May) that exported 30-60% of total annual dissolved and suspended flux of
33                  elements including Pb, Pb  was mainly in the nominal dissolved phase, operationally
34                  defined as material that passes through a 0.45 (im pore-size filter (Pokrovsky et al..
      November 2012                             3-34                  Draft - Do Not Cite or Quote

-------
 1                  2010). Pb adsorbed on colloidal surfaces rather than incorporated into particle structure is
 2                  likely to be more readily dissolved because dissolution of the entire particle is not
 3                  required.

 4                  Recent research on retention of Pb in water bodies and sediments has focused on the
 5                  estuarine and marine environment, where considerable retention of Pb was observed in
 6                  estuarine sediments. For a large riparian system, the Trinity River, Texas, Warnken and
 7                  Santschi (2009) found that 80% of riverine Pb was retained in Lake Livingston, an
 8                  estuarine region, while an additional 16% was removed to estuarine sediments, and only
 9                  about 4% eventually reached the ocean. Geochemical (sorption by Fe oxyhydroxides),
10                  biological (seasonal uptake by sinking algae in Lake Livingston) and hydrological
11                  (dilution effects by increasing flow rates) processes were mainly responsible for
12                  controlling dissolved trace metal concentrations rather than pollution sources.

13                  Overall, recent research on Pb transport in aquatic systems has provided a large body of
14                  observations confirming that Pb transport is dominated by iron and organic rich colloids.
15                  In addition, new results indicated that although the 2006 Pb AQCD (U.S. EPA. 2006b)
16                  described rivers and lakes as temporary reservoirs with Pb lifetimes of months or less,
17                  estuaries can present a substantial barrier to transport into the open ocean.
                    3.3.2.2     Deposition of Pb within Bodies of Water and in Sediment

18                  As described in the 2006 Pb AQCD (U.S. EPA. 2006b). in general Pb is relatively stable
19                  in sediments, with long residence times and limited mobility. As described in previous
20                  sections, Pb enters and is distributed in bodies of water largely in PM form. In rivers,
21                  particle-bound metals can often account for > 75% of the total load, e.g., (Horowitz and
22                  Stephens. 2008). Areas near historically Pb emitting industries and urbanized areas tend
23                  to have greater aquatic Pb loads than areas more remote from Pb sources, as several
24                  studies have shown the strong positive correlation between population density and river
25                  or lake sediment Pb concentrations (Horowitz et al.. 2008; Chalmers et al.. 2007). Indeed,
26                  Chalmers et al. (2007) revealed that in river and lake sediments in New England, there
27                  was an order of magnitude difference  between Pb  sediment concentrations in rural versus
28                  urbanized areas.

29                  The fate of Pb in the water column is determined by the chemical and physical properties
30                  of the water (pH, salinity, oxidation status, flow rate and the suspended sediment load
31                  and its constituents, etc). Desorption, dissolution, precipitation, sorption and
32                  complexation processes can all occur concurrently and continuously, leading to
33                  transformations and redistribution of Pb. The pH of water is of primary importance in
34                  determining the likely chemical fate of Pb in terms of solubility, precipitation or organic

      November 2012                             3-35                  Draft - Do Not Cite or Quote

-------
 1                  complexation. In peatland areas, such as those in upland areas of the U.K., organic acids
 2                  draining from the surrounding peatlands can lower stream water pH to below 4. Under
 3                  these conditions, Pb-PM can be desorbed and released into solution, leading to elevated
 4                  dissolved Pb concentrations (Rothwell et al.. 2008). At the other end of the pH scale, Pb
 5                  tends to remain or become complexed, precipitated or sorbed to suspended sediments in
 6                  water, as observed by Das et al. (2008) who studied trace metal geochemistry in a South
 7                  African lake with water pH of 9. They also found marked differences in Pb
 8                  concentrations associated with  increasing depth in the water column [e.g., the surface
 9                  Pb-PM concentration of 2 (ig/L increased to 60 (ig/L at depth and the Pb concentration in
10                  the <0.45 (im fraction increased from 2 (ig/L at the surface to 19 (ig/L at depth (Das et
11                  al.. 2008)]. This is suggestive of a settlement process  in action.

12                  In estuarine and wider marine environments the processes may be more complex because
13                  of the additional perturbation caused by tidal action and the strong effects of salinity.
14                  Again, PM forms of Pb are important in determining Pb distribution and behavior. Li et
15                  al. (2010a) reported that PM Pb accounted for 85 ± 15% and 50 ± 22% in Boston Harbor
16                  and Massachusetts Bay, respectively, while Lai et al.  (2008b) reported a solid (acid
17                  soluble): dissolved Pb ratio of 2.6 for areas of the Australian sector of the Southern
18                  Ocean.

19                  The accurate modeling of Pb behavior in marine waters (including estuaries) requires
20                  consideration of many parameters such as hydrodynamics, salinity, pH, suspended PM,
21                  fluxes between PM and dissolved phases (Hartnett and Berry. 2010). Several new
22                  advances in the study of Pb cycling in these complex environments have been described
23                  in recent publications. Li  et al.  (2010a) used particle organic carbon (POC) as a surrogate
24                  for the primary sorption phase in the water column to describe and model the partitioning
25                  of Pb between PM and dissolved forms. Huang and Conte (2009) observed that
26                  considerable change in the composition of PM occurs as they sink in the marine
27                  environment of the Sargasso Sea, with mineralization of OM resulting in increased
28                  Pb-PM concentration with increased depth. As a result of this depletion of OM in sinking
29                  particles, geochemical behavior at depth was dominated by inorganic processes,
30                  e.g., adsorption onto surfaces, which were largely independent of Pb source. Sinking
31                  rates in marine  environments can vary, but a rate approximating 1 meter/day has been
32                  used in some models of Pb transport and distribution in aquatic-sediment systems (Li et
33                  al.. 2010a). Surface  sediment Pb concentrations for various continental shelves were
34                  collated and compared by Fang et al. (2009) (Table 3-2).
      November 2012                             3-36                  Draft - Do Not Cite or Quote

-------
Table 3-2 Surface sediment
Location
East China Sea
Mediterranean, Israel coast
Aegean Sea
Bane d'Arguin, Mauritania
Campeche shelf, Gulf of Mexico
Laptev Sea, Siberia
Pechora Sea, Russia
Pb concentrations for various
Digestion solution
HCI/HNOs/HF
HN03
HCI/HNOs/HF
HCI/HNOs/HF
HCI/HNOs
HCI/HNOs/HF
Not reported
continental shelves.
Pba (mg/kg)
1 0-49 (27)a
9.9-20
21-44(34)
2.8-8.9
0.22-20 (4.3)
12-22
9.0-22(14)
      "Values in parentheses are the average, where calculable
      Source: Data from Fang et al. (2009) and references therein.
                     3.3.2.3      Flux of Pb from Sediments

 1                   Sediments can be either a source or a sink for metals in the aquatic environment. Release
 2                   can be via re-suspension of the sediment bed via wind, wave and tidal action or by
 3                   dissolution from sediment to the water column. When external Pb inputs to bodies of
 4                   water are decreased by environmental improvement actions or regulations, contributions
 5                   of Pb to the water column from the existing sediments can become an increasingly
 6                   important source. Roulier et al. (2010) determined that Pb flux from sediments originated
 7                   mostly from organic fractions, but also partially from Mn and Fe components undergoing
 8                   reductive dissolution. The rate of release was controlled by OM content, particle size,
 9                   clay type and content, and silt fraction (Roulier et al.. 2010). The importance of sediment
10                   particle size, OM content and acid volatile sulfide concentration in relation to metal
11                   release was similarly identified (Cantwell et al.. 2008): ecosystem effects related to
12                   sulfide concentration are described in  Section 7.4.2.5. The effect of pH change on Pb
13                   release from lake sediments has also been examined, revealing that 1.8 protons (FT) were
14                   exchanged per divalent metal cation released (Lee et al.. 2008a). Processes governing Pb
15                   release from lake sediments, including microbial reductive dissolution of Fe, biogenic
16                   sulfide production and metal sorption-desorption, have been investigated in a basin
17                   heavily contaminated by historical precious metal mining activities, and results indicated
18                   that release of Pb from suboxic and anoxic zones of the heavily contaminated sediment
19                   act as a Pb source to the overlying water of the lake (Sengor et al..  2007). Bacardit and
20                   Camarero (2010a, b) performed a mass balance of Pb, Zn, and As for three lakes in the
21                   Central Pyrenees in France to identify dominant metals distribution processes. They
22                   estimated that flux from the catchment accounted for 91-99% of the lakes' Pb inputs,
      November 2012                              3-37                   Draft - Do Not Cite or Quote

-------
 1                  while sediment flux accounted for 98-99% of Pb outputs. In this paper, sediment was
 2                  only modeled as an output.

 3                  Disturbance of bed sediments also occurs by tidal action contributing to re-suspension of
 4                  sediments. Benthic fluxes of dissolved metals released from sediments measured in
 5                  Boston Bay were calculated as strong enough that in the absence of Pb inputs such
 6                  benthic flux would reduce sediment Pb concentrations in Boston Bay to background
 7                  levels in 30-60 years (Li et al.. 2010a). In a related way, a half-life for sediment Pb
 8                  (considering benthic flux alone as the loss mechanism) of 5.3 years was estimated for
 9                  marine sediments off the Belgian coast (Gao et al.. 2009). Atkinson et al. (2007)
10                  conducted experiments in an area contaminated by metal smelters, Lake Macquarie,
11                  Australia, to assess the factors that influence flux of metals from marine sediment. Low
12                  pH (pH = 6 ± 1), bioturbation, and other mixing processes were found to have stronger
13                  influence over flux than binding to sulfides, which were thought to be sequestered in
14                  deeper sediments.

15                  Radakovitch et al. (2008) investigated the riverine transport of PM including Pb to the
16                  Gulf of Lion, France, and also concluded that a major part of annual fluxes could be
17                  delivered over a short time period. From budget calculations, riverine inputs were  more
18                  important than atmospheric deposition and Pb concentrations in the prodelta sediments
19                  showed a strong correlation with OM content. These sediments, however, were not
20                  considered to be a permanent sink, as resuspension in  these shallow areas was an
21                  important process. OM, Pb and other metals were enriched in resuspended PM compared
22                  with the sediment.

23                  In a heavily contaminated, high salinity embayment upstream from Sydney Harbor in
24                  Australia, Birch and O'Hea (2007) reported higher total suspended solids, turbidity and
25                  total water metal concentration in surface compared with bottom water as well as a
26                  difference in suspended PM metal concentrations between surface water and bottom
27                  sediments, demonstrating that storm water discharge was the dominant process of metal
28                  transfer during high rainfall events. Total suspended sediments (and total water metals) in
29                  bottom water were higher than in the surface water plume, indicating that resuspension of
30                  bottom sediment is a greater contributor of total suspended sediments than stormwater
31                  during such events, especially in shallower regions of the bay. Soto-Jimenez and Paez-
32                  Osuna (2010) determined diffusive and advective fluxes, geochemical partitioning of Pb
33                  and Pb-isotopic signatures in a study of mobility and behavior of Pb  in hypersaline salt
34                  marsh sediments. They determined that sulfides were the main scavengers for Pb that was
35                  diagenetically  released Pb.

36                  Overall, recent research on Pb flux from sediments in  natural waters provided greater
37                  detail on resuspension processes than was available in the 2006 Pb AQCD (U.S. EPA.


      November 2012                            3-38                  Draft - Do Not Cite or Quote

-------
 1                  2006b), and confirms previous findings that resuspended Pb is largely associated with
 2                  OM or Fe and Mn particles, but that anoxic or depleted oxygen environments in
 3                  sediments play an important role in Pb cycling. This newer research confirms previous
 4                  findings that resuspension and release from sediments largely occurs during discrete
 5                  events related to storms. It has also confirmed that resuspension is an important process
 6                  that strongly influences the lifetime of Pb in bodies of water.
                    3.3.2.4     Pb in Runoff

 7                  Runoff is a major source of Pb in surface waters. This complicates any evaluation of the
 8                  contribution of atmospheric Pb to watersheds, which must take into account direct
 9                  atmospheric deposition, runoff of atmospherically deposited Pb, and runoff of Pb from
10                  sources such as mine tailings or paint chips that are shed from outdoor structures. The
11                  2006 Pb AQCD (U.S. EPA. 2006b) concluded that runoff was consistently mostly PM,
12                  with a relatively small dissolved fraction, and that dissolution of carbonate and related
13                  compounds were important contributors to Pb pollution in runoff waters. It also described
14                  runoff Pb release into runoff as dependent on storm intensity and length of dry periods
15                  between rain events, and a "first flush effect," with highest runoff concentrations
16                  observed at the beginning of a rain event. Subsequent research has provided considerable
17                  new information about the flux of Pb from roadway and urban runoff and snow melt to
18                  watersheds.

19                  Severe contamination due to export of anthropogenic Pb to adjacent ecosystems via urban
20                  runoff and domestic wastewater  discharge and to a lesser extent by direct atmospheric
21                  deposition has been documented (Soto-Jimenez and Flegal 2009). Recent investigations
22                  also confirm roof runoff as an important contributor to Pb pollution. Huston et al. (2009)
23                  measured Pb concentrations in water from urban rainwater tanks and found Pb
24                  concentrations in bulk deposition were consistently lower than in water in the rainwater
25                  tanks, but that sludge in the tanks had a high Pb content, indicating that not all major
26                  sources of Pb are from atmospheric deposition. Pb levels frequently exceeded drinking
27                  water standards. Pb flashing on the roofs was implicated as the source of Pb in the
28                  rainwater tanks although other possible sources include old paint and Pb stabilized PVC
29                  drain pipes (Lasheen et al.. 2008; Weiss et al.. 2006; Al-Malack. 2001).

30                  New research has improved the understanding of suspended PM size ranges, speciation,
31                  and impacts of Pb runoff from urban soil and road dust. Soil and road dust have been
32                  identified as major sources of Pb pollution to near-coastal waters,  leading to high Pb
33                  concentrations in stormwater runoff that became associated with dissolved and suspended
      November 2012                             3-39                  Draft - Do Not Cite or Quote

-------
 1                  PM phases as well as bedload, material moved by rolling, sliding, and saltating along the
 2                  bottom of a stream (Birch and McCreadv. 2009).

 3                  Several recent studies reported that the size distribution of PM transported in runoff is
 4                  relatively uniform. Characterization of the roadside dust in Australia showed that Pb in
 5                  PM was approximately uniformly distributed among PM size fractions of up to 250 (im.
 6                  The Pb-containing particles had the potential to be dispersed to some distance into
 7                  sensitive ecosystems (Pratt and Lottermoser. 2007). Pb in roadside dusts in Thessaloniki,
 8                  Greece was characterized by Ewen et al. (2009) and no difference in Pb concentration
 9                  was found between <75 (im and 75-125 (im PM size ranges, although a difference in the
10                  chemical form of Pb between slightly versus highly contaminated areas was observed.

11                  Ewen et al. (2009) reported that Pb was mainly in a more exchangeable form (similar to
12                  that in an old auto-catalyst reference material) in small particles, but in the residual, or
13                  least mobile fraction in larger particles. In urban road dust from Manchester U.K.,
14                  Pb-bearing Fe-oxides were observed to be dominant in most of the size fractions, and
15                  PbCrO4 comprised 8-34% of total Pb with the highest concentrations being found in the
16                  largest and smallest size fractions. Pb(CO3)2 and Pb(OH)2 were measured in the two
17                  middle size fractions, while PbO and PbSO4 were present in the largest and smallest size
18                  fractions (Barrett etal. 2010).

19                  Murakami et al. (2007) also emphasized the importance of PbCrO4 as an important
20                  species of Pb from road surfaces. That study identified individual particles containing
21                  high levels of Pb and Cr (> 0.2%), most likely from the yellow road line markings. The
22                  identified PM constituted 46% of Cr and Pb in heavy traffic dust and 7-28% in dust from
23                  residential areas and soakaway sediments. The presence of such particles in soakaway
24                  sediments is consistent with their low environmental solubility.

25                  Recent research also continues to document the first flush effect described in the
26                  2006 Pb AQCD. Flint and Davis (2007) reported that in 13 % of runoff events, more than
27                  50% of Pb was flushed in the first 25% of event water. A second flush occurred less
28                  frequently (4% of runoff events for Pb). In agreement with the 2006 Pb AQCD (U.S.
29                  EPA. 2006b). most recent studies have concluded that, during storm events, Pb is
30                  transported together with large PM. Some studies, however, found that Pb was
31                  concentrated in the fine PM fraction and, occasionally, Pb was found predominantly in
32                  the dissolved fraction. Tuccillo (2006) found that Pb was almost entirely in the >5 (im
33                  size range and, indeed,  may be associated with PM larger than 20 (im. (Sansalone et al..
34                  2010) compared Pb-containing PM size distributions from Baton Rouge, LA, New
35                  Orleans, LA; Little Rock, AR; North Little  Rock, AR; and Cincinnati, OH and found no
36                  common distribution pattern. Pb was associated with Cincinnati PM mainly in the
37                  <75 (im fractions, at Baton Rouge and Little Rock Pb mainly in the 75-425 (im PM
      November 2012                             3-40                   Draft - Do Not Cite or Quote

-------
 1                  fractions, and at North Little Rock Pb predominantly in the >425 (im PM fractions. New
 2                  Orleans Pb was almost uniformly distributed among the smaller size PM fractions.
 3                  McKenzie et al. (2008) found that Pb was enriched in the finest PM (0.1-0.3 (im) in
 4                  stormwater samples collected in California, particularly for storms that occurred during
 5                  and after an extended dry period.

 6                  Guo et al. (2006a) investigated the effect of engineered partial exfiltration reactor (PER)
 7                  systems on the partitioning and speciation of Pb in rainfall-runoff at the upstream end of
 8                  an urban source area catchment that is part of the much larger urbanized and industrial
 9                  Mill Creek watershed in Hamilton County, Ohio. The catchment is paved to a large
10                  extent with asphalt and is used for transportation. Guo et al. (2006a) investigated a
11                  catchment that drained toward a wide grassy area and found that Pb was mainly
12                  associated with dissolved organic matter (DOM) in the influent. The study suggested that
13                  interaction of the  rainfall-runoff with the grassy area may have resulted in removal of
14                  PM-bound Pb. PM amount and size can also be influenced by the runoff surface. Guo et
15                  al. (2006a) found that Pb entering the engineered PER system was mainly in the
16                  dissolved fraction with -76%.

17                  There were several recent observations of a relationship between road traffic volume and
18                  runoff Pb concentration, although a clear relationship was not always observed. At a
19                  relatively clean location, Desta et al. (2007) studied highway runoff characteristics in
20                  Ireland and found that although as expected, Pb was strongly correlated with total
21                  suspended solids, no relationship between total suspended solids and rainfall, rain
22                  intensity, antecedent dry days or runoff event duration were observed. They concluded
23                  that runoff composition from site to site could be highly variable. Most other studies did
24                  find a relationship between traffic volume and Pb concentration. A California study of
25                  highway runoff by Kayhanian et al. (2007) reported that 70-80% Pb was in particulate
26                  form for both non-urban and urban highways,  and that the concentration of Pb in runoff
27                  from low traffic flow (30,000-100,000 vehicles/day) urban highways was 50% higher
28                  than that from non-urban highways (total Pb mean =16.6 (ig/L). Additionally, the
29                  concentrations in runoff from high traffic flow (>100,000 vehicles/day) urban areas  were
30                  five times higher than those from non-urban highways. Helmreich et al. (2010)
31                  characterized road runoff in Munich, Germany, with an average daily traffic load of
32                  57,000 vehicles. The mean total Pb concentration, 56 (ig/L (maximum value = 405 (ig/L),
33                  lay in between the values for low traffic flow and high traffic flow runoff from urban
34                  areas in California, i.e., there was good agreement with Kayhanian et al. (2007). There
35                  was no detectable dissolved Pb, i.e., 100% in PM form. Seasonal effects of highway
36                  runoff have also been observed recently. Hallberg et al. (2007) found that summer
37                  particle-bound Pb concentrations in runoff water in Stockholm ranged from
38                  1.37-47.5 (ig/L while, in winter, the range was 1.06—296 (ig/L. There was a strong
      November 2012                             3-41                  Draft - Do Not Cite or Quote

-------
 1                  correlation between Pb (and most other elements) and total suspended solids (R2 = 0.89).
 2                  Helmreich et al. (2010) also found higher metal concentrations during cold seasons in
 3                  Stockholm but Pb concentrations increased only slightly during the snowmelt season.
 4                  There was no change in the distribution of Pb between dissolved and PM forms for the
 5                  rain and snowmelt periods. Runoff from urban snowmelt has been intensively
 6                  investigated since the 2006 Pb AQCD was published. The relocation of snow means that
 7                  the area receiving the snowmelt is not necessarily the same area that which received the
 8                  snowfall. Magill and Sansalone  (2010) also noted that plowed snowbanks alongside
 9                  roadways form a temporary linear reservoir for traffic generated constituents such as
10                  metals and PM. Snowmelt concentrations of metals such as Pb can therefore be several
11                  orders of magnitude higher than those in rainfall runoff (Sansalone and Buchberger.
12                  1996). The melt process usually occurs in a sequence: pavement melt, followed by
13                  roadside (impervious) and finally pervious area melt. As part of this sequence, rain-on-
14                  snow can transport high loads of PM-associated pollutants (Oberts. 2000). Westerlund
15                  and Viklander (2006) investigated differences in PM and Pb concentrations between
16                  rainfall events occurring during  snowmelt and rain periods. Runoff events occurring
17                  during the snowmelt period (i.e., rain-on-snow) had about five times higher numbers of
18                  particles (in the size range 4 to 120 (im)/Liter of runoff. The first rain-on-snow event was
19                  characterized by an increase in the number of particles in the 4 to 25 (im size range. The
20                  rain-on-snow gave a "flush" through the  snow but this was still not sufficient to transport
21                  the larger sized particles.  Only the highest energy rain-on-snow events increased
22                  transport of PM across the entire size spectrum. There was no difference in particle size
23                  distributions  between snowmelt and rain on snow events, although more was transported
24                  during snowmelt. Pb concentrations were most strongly associated with the smaller PM
25                  size fractions.

26                  Overall, there was a significant difference between the melt period and the rain period in
27                  terms of concentrations, loads, transportation and  association of heavy metals with
28                  particles in different size fractions (Westerlund and Viklander. 2006). Over a 4-year
29                  period, Magill and Sansalone (2010) analyzed the distribution of metal in snow plowed to
30                  the edge of roads in the Lake Tahoe catchment in  Nevada, and concluded that metals
31                  including Pb  were mainly associated with large PM (179-542 (im). The PM-associated
32                  metal could be readily separated from runoff water (e.g., in urban drainage systems), but
33                  there  is potential for leaching of metals from the PM within storage basins (Ying and
34                  Sansalone. 2008). For adsorbed  species that form  outer sphere complexes, a decrease in
35                  adsorption and an increase in aqueous complexes  for pollutant metals is a likely
36                  consequence  of higher deicing salt concentrations. If metals form inner-sphere complexes
37                  directly coordinated to adsorbent surfaces, background deicing salt ions would have less
38                  impact. It is thought that physical and outer-sphere complexes predominate for coarse
      November 2012                             3-42                  Draft - Do Not Cite or Quote

-------
 1                  PM, as was the case in Nevada, and so leaching would be likely to cause an increase in
 2                  dissolved phase Pb concentrations.

 3                  Rural runoff has also been extensively studied since publication of the 2006 Pb AQCD
 4                  (U.S. EPA. 2006b). including several recent publications on a forested watershed (Lake
 5                  Plastic) in central Ontario (Landre et al.. 2010. 2009; Watmough and Dillon. 2007) and
 6                  nearby Kawagama Lake, Canada (Shotyk and Krachler. 2010).  Results indicated that
 7                  bulk deposition substantially decreased to 0.49 mg/m2 in 2002 from 1.30-1.90 mg/m2 in
 8                  1989-91. The upland soils retained >95% of the Pb in bulk deposition, i.e., leaching
 9                  losses to stream water were small. The wetland area was, however, a net source of Pb
10                  with annual Pb concentrations in stream water ranging from 0.38 to 0.77 (ig/L. Lake
11                  sediments were efficient sinks for atmospherically deposited Pb with 80-91% of the Pb
12                  input being retained.  Up to 68% of the Pb entering the lake was derived from the
13                  terrestrial catchment. Overall, the watershed effectively retained atmospherically
14                  deposited Pb, but some Pb was then redistributed from the catchment to the lake
15                  sediments; and the Pb in the near-surface lake sediments reflected terrestrially transported
16                  soil material, rather Pb being deposited from the atmosphere. The highest concentrations
17                  of dissolved organic carbon (DOC), Fe and Pb in the wetland draining stream occurred in
18                  summer when Pb concentration frequently exceeded 1 (ig/L (Landre et al.. 2009).

19                  Graham et al. (2006) observed two temporally separated mechanisms occurring during
20                  storm events in a rural organic rich upland catchment. At the beginning of an event, Pb
21                  was transported together with large particles in the >25 (im size range, but after several
22                  hours Pb was mainly transported with colloidal or DOM (<0.45 (im), and the remaining
23                  30-40% of storm related Pb was transported in this form. This indicated that rapid
24                  overland flow rapidly transported Pb-PM into the receiving streams at the very beginning
25                  of the event, and this was followed within a few hours by transport of organic-colloidal
26                  Pb  via near-surface throughflow. The authors used a conservative estimate of Pb removal,
27                  based on their observations that the catchment was continuing to act as a sink for Pb.
28                  These observations about the transport and  fate of Pb  agree well with those of Watmough
29                  and Dillon (2007) and Shotyk et al. (2010).

30                  Soil type was also found to have a strong influence on runoff contributions. Dawson et al.
31                  (2010) found that for organic-rich soils Pb was mobilized from  near-surface soils together
32                  with DOC, but for more minerogenic soils percolation of water allowed Pb bound to
33                  DOC to be retained in mineral horizons and combine with other groundwater sources.
34                  The resulting Pb in stream water had been had a more geogenic signature (Dawson et al..
35                  2010). The findings of both Graham et al. (2006) and Dawson et al. (2010) were
36                  important because the provenance and transport mechanisms of Pb may greatly affect the
37                  net export to receiving waters, particularly since high concentrations of previously
      November 2012                             3-43                   Draft - Do Not Cite or Quote

-------
 1                  deposited anthropogenic Pb are usually found in the near-surface sections of upland U.K.
 2                  soils [e.g., (Farmer etal. 2005)1.

 3                  In another study, Rothwell et al. (2007b) observed stormflow Pb concentrations in a peat
 4                  catchment in Southern Pennines, U.K. almost three times higher than those reported by
 5                  Graham et al. (2006) for northeastern Scotland. The generally high dissolved Pb were due
 6                  to high soil Pb pools and high stream water DOC concentrations (Rothwell et al.. 2007b).
 7                  In a separate study, Rothwell et al. (2007a) showed that OM was the main vector for Pb
 8                  transport in the fluvial system. Some seasonal variability was observed: declining Pb
 9                  concentrations in autumn stormflow may indicate the exhaustion of DOC from the
10                  acrotelm (the hydrologically active upper layer of peat which is subject to a fluctuating
11                  water table and is generally aerobic) or a dilution effect from an increasing importance of
12                  overland flow.

13                  Erosion of agricultural soils and the effects of different types of storm events on soil
14                  particle and Pb losses from these soils was characterized by Quinton and  Cart (2007). A
15                  close link between metal concentration and the silt, or clay and organic content of stream
16                  sediments was consistent with enrichment of metals as a consequence of  small erosion
17                  events. They also noted that short intense events could produce the same  amount of
18                  sediment as longer low-intensity events. More intense events, however, could mobilize a
19                  wider range of particle sizes whereas  low intensity events mobilized finer but more
20                  metal-rich material. Smaller events accounted for 52% of Pb losses from  the agricultural
21                  soil.

22                  The Tinto River in Spain drains one of the largest polymetallic massive sulfide regions in
23                  the world: the Iberian Pyrite Belt. Evaporitic sulfate salts, formed as a result of acid mine
24                  drainage processes, are considered to be a temporary sink for many heavy metals. Upon
25                  the arrival of rainfall, however, they rapidly dissolve, releasing acidity and contaminant
26                  metals into receiving waters. Thus rivers in semi-arid climate regions such as the Tinto
27                  River which alternate between long periods of drought and short but intense rainfall
28                  events, can experience quick acidification and increases in metal concentration. In a study
29                  of such events, Canovas et al. (2010)  found that while many element concentrations
30                  decreased during events, the concentrations of Fe, Cr, Pb and As increased. This was
31                  attributed to the redissolution and transformation of Fe oxyhydroxysulfates and/or
32                  desorption processes.

33                  Dunlap et al. (2008) studied a large (>160,000 km2) riparian system (the Sacramento
34                  River, CA) and showed that the present day flux of Pb into the river was dominated by Pb
35                  from historical  anthropogenic sources, which included a mixture of high-ratio hydraulic
36                  Au mining-derived Pb and persistent historically-derived Pb from leaded  on-road
37                  gasoline. Outside of the mining region, 57-67% Pb was derived from past on-road


      November 2012                             3-44                   Draft - Do Not Cite or Quote

-------
 1                  gasoline emissions and 33-43% was from hydraulic Au mining sediment. Periods of high
 2                  surface runoff mobilize additional fluxes of Pb from these two sources and carry them
 3                  into the river.

 4                  Rothwell et al. (2007b) commented that although there have been substantial reductions
 5                  in sulfur deposition to U.K. uplands over the last few decades (Fowler etal.. 2005).
 6                  anthropogenic acidification of upland waters is still possible if there is nitrogen leaching
 7                  from the surrounding catchment and this may increase with nitrogen saturation (Curtis et
 8                  al., 2005). Rothwell et al. (2007b) predicted that if an increase in surface water
 9                  acidification is coupled with further increases in DOC export from organic-rich
10                  catchments, metal export from peatland systems will increase.  The deterioration of peat
11                  soils by erosion is considered to be  exacerbated by climatic change.  Rothwell et al.
12                  (2010) used digital terrain analysis to model suspended Pb concentrations in
13                  contaminated peatland catchments.  The peat soils of the Peak District are characterized
14                  by extensive eroding gullies and so  they were combined in an empirical relationship
15                  between sediment-associated Pb concentrations and mean upslope gully depth with fme-
16                  resolution mapping of the gully areas. This model will enable prediction of metal
17                  contamination in receiving waters.

18                  Klaminder et al. (2010) investigated the environmental recovery of sub-arctic lakes in
19                  response to reduced atmospheric deposition over the last few decades. They found that
20                  there had been no improvement in surface sediments and indeed the reduction in Pb
21                  contamination had been much less than the 90% reduction in emissions over the last four
22                  decades. The weak improvement in the 206Pb/207Pb ratio together with the Pb contaminant
23                  concentrations suggests that catchment export processes of previously-deposited
24                  atmospheric contaminants have had a considerable impact on the recent contaminant
25                  burden of sub-arctic lakes. In Arctic regions, soil export of contaminants to surface
26                  waters may dramatically increase in response  to climate change if it triggers thawing of
27                  frozen soil layers. It is thought that thawing may generate accelerated soil erosion, altered
28                  hydrological flow paths, increased runoff and exposure of soluble compounds that had
29                  previously been in the frozen layers. At this stage, however, the links between catchment
30                  export and climate change have not yet been clearly established.

31                  Coynel et al. (2007) also considered the effects of climate change  on heavy metal
32                  transport.  In this case, the scenario of flood-related transport of PM in the Garonne-
33                  Gironde fluvial-estuarine system was investigated. Export of suspended PM during a
34                  five-day flood in December 2003 was estimated at -440,000 tons, accounting for -75%
35                  of the annual suspended PM fluxes. Sediment remobilization accounted for -42% of the
36                  total suspended particulate matter (SPM) flux during the flood event (-185,000 tons
37                  suspended PM) and accounted for 61% of the 51 tons Pb that was exported. Coynel et al.
      November 2012                             3-45                   Draft - Do Not Cite or Quote

-------
 1                  (2007) postulate that flood hazards and transport of highly polluted sediment may
 2                  increase as a result of climate change and/or other anthropogenic impacts (flood
 3                  management, reservoir removal).

 4                  In heavily contaminated catchments [e.g., that of the Litavka River, Czech Republic (Zak
 5                  et al.. 2009)1. the flux of heavy metals to the river during storm events can be substantial.
 6                  Even during a minor 4-day event, 2,954 kg of Pb was transported, and the majority was
 7                  associated with suspended PM. For the Adour River in a mountainous area of France, Pb
 8                  pollution predominantly originated from mining activities, and Point et al. (2007) showed
 9                  that 75% of annual soil fluxes into the river were transported in 30-40 days.

10                  The consequences of flood management (dam flushing) practices on suspended PM and
11                  heavy metal fluxes in a fluvial-estuarine system (Garonne-Gironde, France) were
12                  considered by Coynel et al. (2007). Dam flushing enhanced mobilization of up to
13                  30-year-old polluted sediment from reservoir lakes. Sediment remobilization accounted
14                  for -42% of the total suspended PM fluxes during the flood and strongly contributed to
15                  PM-bound metal transport (61% for Pb). They concluded that flood management will
16                  need to be taken into consideration in future models for erosion and pollutant transport.

17                  Bur et al. (2009) investigated the associations of Pb in stream-bed sediments of the
18                  French Gascony region. They found that Pb enrichment in stream sediments was
19                  positively correlated with catchment cover and increasing organic content whereas Pb
20                  concentration was strongly linked with Fe-oxide content in cultivated catchments. For the
21                  low-OM, anthropogenic Pb was associated with carbonates and Fe-oxides (preferentially,
22                  the amorphous fraction). Fe-oxides became the most efficient anthropogenic Pb trapping
23                  component as soon as the carbonate content is reduced. They noted, however, that OM
24                  was always weakly involved. N'Guessan et al. (2009) also studied trace elements in
25                  stream-bed sediments of the French Gascony region. They used enrichment factors to
26                  show that only -20-22% of Pb was from anthropogenic sources with the remainder
27                  originating from natural weathering processes.

28                  Overall, research results from the last several years have greatly expanded the extent of
29                  the knowledge concerning Pb from runoff. Substantial Pb input to estuarine and marine
30                  ecosystems has been well documented. More detail concerning the origin of Pb from roof
31                  runoff has led to the  conclusion that roof flashing could be especially important. Research
32                  on road runoff has provided valuable insight into PM size and composition, indicating
33                  that size distributions for Pb-containing PM in runoff water varies from study to study
34                  and from location to location. Recent studies confirmed the "first flush" effect, releasing
35                  more Pb at the beginning of rainfall than subsequently, and documented size distributions
36                  of Pb-containing PM also vary considerably when water from the first flush is isolated.
37                  Influence of road traffic volume on runoff has also been more fully documented in recent
      November 2012                            3-46                  Draft - Do Not Cite or Quote

-------
 1                  years. The role of urban snowmelt and rain-on-snow events is also better understood, and
 2                  it has been observed that greater runoff occurs from snowmelt and in rain-on-snow events
 3                  than when snow is not present, and that metals, including Pb, are often associated with
 4                  coarse PM under these circumstances. Runoff in rural areas is strongly controlled by soil
 5                  type and the presence of vegetation, with less runoff and greater retention in mineral soils
 6                  or when grass is present, and more runoff for soils high in OM. Runoff also follows a
 7                  two-step process of transport of larger particles at the beginning of an event, followed
 8                  within hours by transport of finer colloidal material.  Some initial research on the effects
 9                  of climate change on runoff has focused  on documenting the association between
10                  increased runoff and more intense rain events and greater thawing. Overall, recent
11                  research has provided greater detail on amounts, particle size distributions, composition,
12                  and important processes involving Pb transport, and the understanding of Pb runoff has
13                  become more complete since publication of the 2006 Pb AQCD (U.S. EPA. 2006b).
      3.3.3        Soil

14                   The 2006 Pb AQCD (U.S. EPA. 2006b) summarized that Pb has a relatively long
15                   retention time in the organic soil horizon, although its movement through the soil column
16                   also suggests potential for contamination of groundwater. Leaching was consistently
17                   observed to be a slower process for Pb than for other contaminants because Pb was only
18                   weakly soluble in pore water, but anthropogenic Pb is more available for leaching than
19                   natural Pb in soil. Pb can bind to many different surfaces and Pb sorption capacity is
20                   influenced by hydraulic conductivity, solid composition, OM content, clay mineral
21                   content, microbial activity, plant root channels, animal holes, and geochemical reactions.
22                   As a result of Pb binding to soil components, leaching is retarded by partitioning to soils,
23                   which is not only influenced by sorption capacity, but leaching also increases with
24                   proximity to source, increasing pH, and increasing metal concentrations. Leaching is also
25                   strongly influenced by pore water flow rates, with more complete sorption contributing to
26                   slower leaching at lighter flows. Leaching rates are especially high in soils with a high Cl
27                   content, but typically the most labile Pb fraction is adsorbed to colloidal particles that
28                   include OM, clay, and carbonates. Transport through soils is enhanced by increasing
29                   amount of colloidal suspensions, increasing colloidal surface charge, increasing organic
30                   content of colloids, increasing colloidal macroporosity, and decreasing colloidal size.
31                   Acidity and alkalinity have a more complex influence, with sorption maximized at
32                   neutral pH between pH = 5 and pH = 8.2, and greater mobility at higher and lower pH.
33                   High Pb levels have been observed in leachates from some contaminated soils, but this
34                   effect appears to be pH dependent. In several studies of contaminated soils a substantial
      November 2012                             3-47                  Draft - Do Not Cite or Quote

-------
 1                  fraction of Pb was associated with Mn and Fe oxides or carbonate. Influence of soil
 2                  chemistry on Pb effects in ecosystems is described in Section 7.3.2.
                    3.3.3.1      Deposition of Pb onto Soil from Air

 3                  As described in the 2006 Pb AQCD (U.S. EPA. 2006b). a considerable amount of Pb has
 4                  been deposited from air onto soils in urban areas and near stationary sources and mines.
 5                  Removal and translocation of Pb in soil is an ongoing process.

 6                  High Pb soil concentrations were observed near stationary sources such as smelters and
 7                  battery disposal operations, and soil Pb concentrations decreased rapidly with distance
 8                  from the source. Several recent studies continue to document high concentrations of Pb in
 9                  soil. A study of soil Pb concentrations in Queensland, Australia described atmospheric
10                  transport and deposition of Pb in soils (due to ongoing emissions from nearby mining
11                  [which began in 1924] and smelting [which began in 1931] activities) are continuing to
12                  impact the urban environment, resulting in elevated soil Pb concentrations at urban
13                  property sites within 2 km of the mines (Taylor et al., 2010). Similarly, sediment cores
14                  from four remote Canadian Shield headwater lakes located  along a transect extending
15                  300 km from a non-ferrous metal smelter generated useful information about distance of
16                  Pb transport from the smelter prior to deposition (Gallon et al., 2006).  Shotyk and
17                  Krachler (2010) postulated that long-range transport of Pb from a smelter at Rouyn-
18                  Noranda may still contribute to deposition on these lakes. Recent measurements of
19                  deposition fluxes to soil in rural and remote areas have ranged from approximately
20                  0.5 mg/m2-yr to about 3 mg/m2-yr with fair agreement between locations in Canada,
21                  Scandinavia, and Scotland and showed a substantial decrease compared to when leaded
22                  on-road gasoline was in widespread use (Shotbolt et al., 2008; Watmough and Dillon.
23                  2007: Fowler etal.. 2006: Graham et al.. 2006).

24                  There has been considerable interest in the response of soils to the decreasing aerosol Pb
25                  concentrations and Pb deposition rates that have been recorded in recent years. Kaste et
26                  al. (2006) resampled soils at 26 locations in the Northeast U.S. (during a 2001-2002
27                  survey of soil sites originally sampled in 1980), and found no significant change in the
28                  amount of Pb in the O-horizon at high altitude sites, suggested to be related to reduced
29                  microbial  activity at altitude. However, the amount of Pb in the  O-horizon had decreased
30                  at some locations in the southern part of the survey region (Connecticut, New York,
31                  Pennsylvania), where the forest soils have typically thinner O-horizons, the reasons for
32                  which are discussed further in Section 3.3.3.2.  Relatively high Pb concentrations were
33                  also found in Japan, especially above 600 meters altitude compared with lower altitude
34                  soils sampled during the study (Takamatsu et al.. 2010).
      November 2012                             3-48                  Draft - Do Not Cite or Quote

-------
 1                   Further support for the use of mosses as bioindicators or monitors for atmospheric Pb
 2                   inputs to peat bogs have recently been published by Kempter et al. (2010) who found that
 3                   high moss productivity did not cause a dilution of Pb concentrations in peat bogs. They
 4                   also found that productive plants were able to accumulate particles from the air and that
 5                   rates of net Pb accumulation by the mosses were in excellent agreement with the
 6                   atmospheric fluxes obtained by direct atmospheric measurements at nearby monitoring
 7                   stations. In addition, Bindler et al. (2008) used Pb isotopes to compare the distribution of
 8                   Pb in the forest soils with that of lake sediments where no "plant pumping" processes
 9                   could be invoked, and used Pb isotope ratios to demonstrate that observations were
10                   consistent with anthropogenic Pb deposition to the soils rather than intermixing of natural
11                   Pb from underlying mineral soil horizons.

12                   Overall, recent studies provided deposition data that was consistent with deposition
13                   fluxes reported in the 2006 Pb AQCD (U.S. EPA. 2006b). and demonstrated consistently
14                   that Pb deposition to soils has decreased since the phase-out of leaded on-road gasoline.
15                   Follow-up studies in several locations at high altitude indicated little change in soil Pb
16                   concentrations since the phase-out of leaded on-road gasoline, although reductions in
17                   surface soil Pb concentrations have been documented in some areas.
                     3.3.3.2      Sequestration of Pb from Water to Soil

18                   The 2006 Pb AQCD described Pb as being more strongly retained in soil than other
19                   metals because of its weak solubility in pore water, but that anthropogenic Pb was more
20                   available for leaching than natural Pb (U.S. EPA, 2006b). It also described a complex
21                   variety of factors that influence Pb retention, including hydraulic conductivity, solid
22                   composition, OM content, clay mineral content, microbial activity, plant root channels,
23                   geochemical reactions, colloid amounts, colloidal surface charge, and pH.

24                   Recent research in this area has provided more insight into the details of the Pb
25                   sequestration process. Importance of leaf litter was further investigated, and it was
26                   observed that the absolute Pb content can be substantial because rain events cause
27                   splashing of the leaf litter with soil thus placing the litter in direct contact with soil
28                   metals. The resulting increase in leaf litter metal concentrations suggests that the litter
29                   can act as a temporary sink for metals from the soil around and below leaves on the
30                   ground. The low solubility of Pb in the leaf litter indicates that the Pb is tightly bound to
31                   the decomposing litter, making the decomposing leaves act as an efficient metal storage
32                   pool (Scheid et al.. 2009). Differences between throughfall (i.e.,  water depositing onto the
33                   soil following collection on leaves) and litterfall (i.e., deposition of leaves, bark,  and
34                   other vegetative debris onto soil) in forested areas have been investigated in forested
      November 2012                              3-49                   Draft - Do Not Cite or Quote

-------
 1                  areas, and the combined input of Pb to the forest floor from throughfall and litterfall was
 2                  approximately twice that measured in bulk deposition (Landre et al.. 2010). The
 3                  difference was attributed to a substantial contribution from internal forest cycling and
 4                  indicates that bulk deposition collectors may underestimate the amount of Pb reaching the
 5                  forest floor by about 50% (Landre etal.. 2010).

 6                  New research has also provided details about the complexity of Pb sequestration during
 7                  soil OM decomposition. Schroth et al. (2008) investigated Pb sequestration in the surface
 8                  layer of forest soils and the transformation of Pb speciation during soil OM
 9                  decomposition. The pH range for forest floor soils in the Northeast U.S. is typically 3.5-5
10                  and, under these conditions, dissolved Pb would adsorb strongly to soluble OM and to
11                  Fe/Al/Mn oxides and oxyhydroxides. It had been thought that the high affinity of Pb for
12                  organic ligands meant that sequestered atmospheric Pb would be preferentially bound to
13                  soluble OM. As a consequence, decomposition of OM would lead to Pb migration to the
14                  underlying mineral layers where it would be precipitated with the dissolved OC or
15                  adsorbed to pedogenic mineral phases. However, recent research has revealed a more
16                  complicated picture of gasoline-derived Pb associations in the forest floor. More recent
17                  research indicates that, as decomposition progresses, Pb and Fe become more
18                  concentrated in "hotspots" and Pb likely becomes increasingly distributed on surfaces
19                  associated with Fe and Mn (and to some extent Ca). It was postulated  that Pb was
20                  initially bound to labile organic but, following decomposition, the Pb was adsorbed at
21                  reactive sites on pedogenic mineral phases (Schroth et al., 2008). Differences in litter
22                  types were also reported, with more rapid  decomposition of OM in deciduous litter
23                  mobilizing more Pb initially bound to labile OM than coniferous litter, and producing
24                  more pedogenic minerals that could readily sequester the  released Pb (Schroth et al..
25                  2008). In the next stage of the study, the speciation of Pb  in the O-horizon soils of
26                  Northern Hardwood, Norway spruce and red pine forest soils were compared. In general
27                  there was good agreement between the Pb speciation results for the soils and those  for the
28                  laboratory decomposition experiments. Specifically, for the Northern Hardwood forest
29                  soil, a little more than 60% of the Pb was bound to  soil organic matter (SOM) and this
30                  percentage increased to -70% and -80% for the Norway spruce and red pine soils,
31                  respectively. In all three cases, however, most of the remainder of the  Pb was bound to
32                  ferrihydrite rather than to birnessite. This was not considered to be surprising because of
33                  the well-known leaching and cycling behavior of Mn that would be expected in the
34                  natural  system. Thus the prevalence of Mn phases in the field based samples would be
35                  lessened (Schroth et al..  2008).

36                  More generally, other studies have observed Pb sorption to Mn and Fe phases in soils.
37                  For example, Boonfueng et al. (2006) investigated Pb sequestration on Mn oxide-coated
38                  montmorillonite. Pb formed bidentate corner-sharing complexes. It was found that Pb
      November 2012                             3-50                  Draft - Do Not Cite or Quote

-------
 1                   sorption to MnO2 occurred even when MnO2 was present as a coating on other minerals,
 2                   e.g., montmorillonite. Although their importance in the near-surface phases has clearly
 3                   been demonstrated by Schroth et al. (2008), ferrihydrite surfaces may not be a long-term
 4                   sink for Pb since reductive dissolution of this Fe(III) phase may release the surface-bound
 5                   Pb into the soil solution. Sturm et al. (2008) explored the fate of Pb during dissimilatory
 6                   Fe reduction. Pb was indeed released but was then incorporated into less reactive phases.
 7                   These phases could not, however, be identified. Even so, it was asserted that Pb should be
 8                   largely immobile under Fe-reducing conditions due to its incorporation into refractory
 9                   secondary minerals.

10                   Kaste et al. (2006) found that Pb species currently in the soil O-horizons of the Northeast
11                   U.S. differed considerably from those that were originally deposited from fossil fuel
12                   combustion (including on-road gasoline). PbSO4 was considered to be the main form of
13                   Pb that had been delivered from the atmosphere to the surface of the Earth and it was
14                   postulated that the presence of sulfate may have facilitated the adsorption of Pb to
15                   colloidal Fe phases within the organic-rich horizons.

16                   Altogether, these new results enhance the understanding of Pb sequestration in forest
17                   soils. First, the role of leaf litter as a major Pb reservoir is better understood. Second, the
18                   effect of decomposition on Pb distribution and sequestration on minerals during OM
19                   decomposition has been further characterized, and finally, the relative importance of Mn
20                   and Fe in sequestration is better understood.

21                   Recent research also addressed roadsides soils. Jensen et al. (2006) found that Pb was
22                   retained by an organic-rich blackish deposit with a high OM content and elevated soil Pb
23                   concentrations, originating from total suspended solids in road runoff and from aerial
24                   deposition. Hossain et al.  (2007) observed that after long dry periods,  OM oxidation may
25                   potentially result in the release of Pb. Microbial activity may also breakdown OM and
26                   have similar consequences (i.e., Pb release).  Bouvet et al. (2007) investigated the effect
27                   of pH on retention of Pb by roadside soils  where municipal solid waste incineration
28                   (MSWI) bottom ash had been used for road construction. They found  that the Pb that had
29                   leached from the road construction materials was retained by the proximal soils under the
30                   prevailing environmental  conditions (at pH = 7, <2% was released, but at pH = 4,  slightly
31                   more Pb (4-47%) was released) and the authors speculated that the phase from which Pb
32                   had been released may have been Pb(CO3)2(OH)2, indicating that sequestration of Pb via
33                   formation of oxycarbonate minerals is only effective at near-neutral to alkaline pH values
34                   (Figure 3-11 in Section 3.3.3.3).

35                   Other recent research on Pb sequestration focused on microbial impacts and soil
36                   amendments. There have been few if any previous observations of microbial
37                   sequestration of Pb in soil. Perdrial et al. (2008) observed bacterial Pb sequestration and


      November 2012                              3-51                  Draft - Do Not Cite or Quote

-------
 1                  proposed a mechanism of Pb complexation by polyphosphate. They also postulated that
 2                  bacterial transport of Pb could be important in sub-surface soil environments. Wu et al.
 3                  (2006) also and concluded that Pb adsorption to the bacterial cell walls may be important
 4                  with respect to Pb transport in soils. This new area of research suggests that bacteria can
 5                  play an important role in both sequestration and transport of Pb. Phosphate addition to
 6                  immobilize Pb-contaminated soils has often been used to immobilize Pb in situ through
 7                  the formation of Pb phosphate minerals such as chloropyromorphite. Recent research
 8                  investigated factors affecting the long-term stability of such products, which depends on
 9                  the equilibrium solubility and the dissolution  rate of the mineral, trace impurities, such as
10                  Pb(OH)2, the presence of complexing agents,  and pH (Xie and Giammar. 2007).  Overall,
11                  in agreement with the 2006 Pb AQCD (U.S. EPA. 2006b). the addition of phosphate can
12                  enhance immobilization of Pb under certain conditions in the field but may cause
13                  desorption and mobilization of anionic species of As, Cr and Se.
                    3.3.3.3      Movement of Pb within the Soil Column

14                  The 2006 Pb AQCD summarized studies that demonstrated that Pb has a long retention
15                  time in the organic soil horizon, it also has some capacity to leach through the soil
16                  column and contaminate groundwater more than other contaminants do, because Pb is
17                  only weakly soluble in pore water (U.S. EPA. 2006b). The fate of any metal transport in
18                  soil is in response to a complex set of parameters including soil texture, mineralogy, pH
19                  and redox potential, hydraulic conductivity, abundance of OM and oxyhydroxides of Al,
20                  Fe, and Mn, in addition to climate, situation and nature of the parent material. As a
21                  consequence, it is impossible  to make general conclusions about the final fate of
22                  anthropogenic  Pb in soils. Indeed, Shotyk and LeRoux (2005) contend that the fate of Pb
23                  in soils may have to be evaluated on the basis of soil type. Some generalizations are,
24                  however, possible: Pb migration is likely to be greater under acidic soil conditions
25                  (Shotyk and Le Roux. 2005).  In this respect, it would be expected that there should be
26                  considerable mobility of Pb in the surface layers of certain types of forest soils. This
27                  section reviews recent research on movement of Pb through soil types by first focusing
28                  on forest soils, followed by a  broader treatment of a more diverse range of soils.


                    Forest Soils  and Wetlands

29                  Several studies confirmed the slow downward movement of Pb within the soil column.
30                  Kaste et al. (2006) found that the amount of Pb in O-horizon soils had remained constant
31                  at 15 of 26 sites in remote forested areas of the Northeast U.S. that had been re-sampled
32                  after a 21 -year time period had elapsed, but that measured soil Pb concentrations were
      November 2012                            3-52                  Draft - Do Not Cite or Quote

-------
 1                  lower than predicted concentrations from total deposition, strongly suggesting that the
 2                  O-horizon had not retained all of the atmospheric Pb, and that a proportion of the
 3                  atmospheric deposition must have leached into the underlying mineral layers. At some
 4                  sites, mainly those at the southern latitudes and lower altitude sites, the proportion of Pb
 5                  that had been leached downward from the O-horizon was quite considerable. Relative
 6                  retention of Pb was influenced by the rate of OM decomposition, depth of soil O-horizon,
 7                  and pH. For soils where Pb was strongly retained by the O-horizon, a relationship
 8                  between Pb and Fe-rich phase was observed, but Pb was also significantly correlated with
 9                  other metals. XANES data suggested a possible interaction with an amorphous Fe oxide,
10                  but spectra were not entirely explained by Fe and oxygen and an additional spectral
11                  feature suggested the presence of a sulfur (S) or phosphorus (P) atom, which could result
12                  if OM functional groups were binding to Pb. Kaste et al. (2006) concluded that a
13                  substantial fraction of Pb was associated with amorphous Fe-hydroxides. The strong
14                  binding of Pb coupled with the low solubility of Fe phases under oxic conditions, helped
15                  to explain the relatively long residence time of gasoline-derived Pb in forest floors which
16                  had thick O-horizons and were well-drained. In the situations where Pb was leached
17                  downward to a large extent, mobility was likely governed by OM decomposition and
18                  colloidal transport of Pb associated with colloidal Fe  and OM.

19                  Klaminder et al.  (2006b) also considered the transfer of Pb from the O-horizon to the
20                  underlying mineral horizons (including the C-horizon). They concluded that atmospheric
21                  pollution-derived Pb migrated at a rate about 10-1,000 times slower than water. They
22                  assumed that Pb was mainly transported by dissolved OM and so the mean-residence-
23                  time of Pb in the O-horizon depended on OM transport and turnover. The retardation rate
24                  was a reflection of the slow mineralization and slow downward transport rates of organic-
25                  Pb complexes, due to sorption and desorption reactions involving mineral surfaces.

26                  In a study involving stable Pb isotopes, Bindler et al.  (2008) showed that Pb with a
27                  different isotopic composition could be detected in the soil down to a depth of at least
28                  30 cm and sometimes down to 80 cm in Swedish soils. In comparison, in North American
29                  podzols, pollution Pb is typically only identified to a depth of 10-20 cm (even with the
30                  aid of isotopes).  This difference is attributed to the longer history of metal pollution in
31                  Europe (as has been traced using lake sediments).

32                  Several research groups have attempted to determine the mean residence time of Pb in the
33                  O-horizon of forest soils. Klaminder et al. (2006a) used three independent methods to
34                  estimate a mean  residence time of about 250 years for Pb in the O-horizon of boreal
35                  forests in Sweden, indicating that deposited atmospheric Pb pollution is stored in the
36                  near-surface layers for a considerable period and, consequently, will respond only slowly
37                  to the reduction in atmospheric inputs. It should be noted, however, the OM in the upper
      November 2012                             3-53                   Draft - Do Not Cite or Quote

-------
 1                  parts of the O-horizon is continually being replaced by fresh litter and the mean residence
 2                  time of Pb in these horizons is only 1-2 years. Thus, the uppermost layer will respond
 3                  more quickly than the rest of the O-horizon to the decreases in Pb inputs.

 4                  Klaminder et al. (2008a) considered the biogeochemical behavior of atmospherically
 5                  derived Pb in boreal forest soils in Sweden (Figure 3-11). The estimated annual losses via
 6                  percolating soil water were -2 mg/m2-yr (Klaminder et al., 2008a) and so the annual loss,
 7                  assumed to be  from the mor layer, was greater than the atmospheric input of
 8                  ~0.5 mg/m2-yr. The upward transport of Pb did not compensate for the losses either. In
 9                  contrast, the amount of Pb being stored in the mineral soil layers was increasing. The
10                  mean residence time of Pb in the mor layer was estimated to be -300 years, in reasonable
11                  agreement with their earlier work (Klaminder et al., 2006a). These values were greater
12                  than the values of 2-150 years determined for U.S. forest soils, e.g., (Watmough et al..
13                  2004; Kaste et al.. 2003) but the difference was attributed to the lower decomposition
14                  rates of OM within the northern boreal forests of Sweden. They concluded that more
15                  research  was needed to determine the processes occurring within the mor layer that
16                  control the release of Pb from this horizon.

17                  Klaminder et al. (2008b) investigated in more detail the distribution and isotopic
18                  signature of Pb persisted within the O-horizon (mor layer) of boreal forest soils. They
19                  found that the mor layer preserved a record of past Pb emissions from a nearby smelter.
20                  Minimal animal burrowing activity and low leaching rates observed at the sampling
21                  location were important factors contributing to the preservation of this record. They
22                  concluded that temporal changes in atmospheric fallout in addition to adsorption
23                  processes need to be considered when interpreting Pb concentrations changes within the
24                  mor layer.
      November 2012                             3-54                   Draft - Do Not Cite or Quote

-------
                                            0,05
                                                    0.05
                  Bs
                 B/C
                                                                  Loss of lead
                    Buildup of
                      lead
                      0.1
                                                i—> Present atmospheric deposition (mg m-J yr1)
                                                - - - * Plant uptake (mg rrr2 yr1)
                                                  ~* Soil water flux (mgm^yr1)
     Notes The atmospheric deposition rate is from (Klaminder et al.. 2006a). the plant uptake rates from (Klaminder et al.. 2005) and
     estimated soil-water fluxes from (Klaminder et al.. 2006b).

     Figure 3-11     Schematic model summarizing the estimated flux of Pb within a
                       typical podzol profile from northern Sweden using data from
                       Klaminder et al. (2006a).

 1                  Significantly higher O-horizon Pb concentrations have been observed in coniferous than
 2                  deciduous forest soils (McGee et al.. 2007). Steinnes et al. (2005) noted evidence for
 3                  downward migration of Pb from the O-horizon to the E-horizon of most soils and in some
 4                  cases the upper B-horizon. They found that the downward transport of Pb differed
 5                  considerably between the sites, e.g., from almost no anthropogenic Pb in the B-horizon at
 6                  some sites to -70% at other sites. The greater downwards transport in some locations was
 7                  attributed to climatic variations, with more extensive leaching and possibly a greater
 8                  turnover of OM at sites where higher mean annual temperatures were experienced.
 9                  Higher atmospheric deposition of acidifying substances in these locations was considered
10                  the most important factor in Pb transport, causing release of Pb from exchange sites in the
11                  humus layer and promoting downward leaching.
     November 2012
3-55
Draft - Do Not Cite or Quote

-------
 1                   Seasonal variation in Pb mobility has also been observed in forest soil. Other research
 2                   indicated that Pb concentrations correlated with DOC concentrations in the soil solution
 3                   from the O-horizon, and were lower during late winter and spring compared with summer
 4                   months (Landre et al. 2009). The degradation of OM in the O-horizon produced high
 5                   DOC concentrations in the soil solution. It was also shown that Pb was associated with
 6                   the DOC, and concluded that DOC production is a primary factor enhancing metal
 7                   mobility in this horizon. In the underlying mineral horizons, DOC concentrations
 8                   declined due to adsorption and cation exchange processes. The B-horizon retained most
 9                   of the DOC leached from the O-horizon and it has also been observed that Pb is similarly
10                   retained.


                     Non-forested  Soils

11                   In contrast with forest soils, most non-forested soils are less acidic and so most studies of
12                   Pb behavior in non-forested soils have focused on Pb immobility. However, there are
13                   acid soils in some locations that are not forested. For these soils, as for forest soils, Pb
14                   mobility is weak but correlated with OM. For example, Schwab et al. (2008) observed
15                   that low molecular weight organic acids added to soil enhanced Pb movement only
16                   slightly. Citric acid and tartaric acid enhanced Pb transport to the greatest degree but the
17                   extent of mobilization was only slightly higher than that attained using deionized water
18                   even at high concentrations. While the formation of stable solution complexes and more
19                   acidic conditions favored mobilization of Zn and Cd, Pb remained strongly sorbed to soil
20                   particles and so the presence of complexing agents and low pH (2.8-3.8) did not
21                   substantially enhance Pb mobility.  Similarly, limited penetration and leaching was
22                   observed in an extremely complex temperate soil profile, with highest concentrations of
23                   Pb (-80 mg/kg) found in the top 0-5 cm section of soil. For this uppermost soil section,
24                   there was a  strong correlation between Pb concentration and OC content, both for the
25                   total soil fraction and the acid-extractable fraction. The Pb migration rate was calculated
26                   to be 0.01 cm/yr and it was estimated that Pb would be retained in the soil column for
27                   20,000 years, with no evidence of rapid movement of anthropogenic Pb from the top
28                   0-5 cm soil  section into the soil profile Kylander et al. (2008).

29                   Other recent studies also reported strong retention in non-forest soils and enhanced
30                   mobility of  Fe and OM colloids. Pb was strongly retained on acidic Mediterranean soil
31                   columns, and association of Pb with the exchangeable OM and crystalline Fe oxide
32                   fractions appeared to favor mobility, while Pb association with Mn oxides and
33                   amorphous  Fe oxides was linked with semi-irreversible retention of Pb in the solid phase
34                   (Garrido et  al.. 2008). In another study of Pb mobility within Mediterranean soils, Pb
35                   infiltration velocity was measured to be 0.005 meter/year (Erel. 1998). The authors
36                   attributed Pb movement within the soil column to advection and concluded that the soil

      November 2012                             3-56                  Draft - Do Not Cite or Quote

-------
 1                  profile of Pb is similar to the anthropogenic air Pb emissions trend. Pedrot et al. (2008)
 2                  studied colloid-mediated trace element release at the soil/water interface and showed that
 3                  Pb was mobilized by Fe nanoparticles that were bound to humic acids.

 4                  Soil pH value  is probably the single most important factor affecting solubility, mobility
 5                  and phytoavailability but reducing conditions also results in increased Pb mobility, with
 6                  the release of Pb into an anoxic soil solution due to the combined effect of Fe(III)
 7                  reductive dissolution and dissolved OM release. Dissolved OM is more important than Fe
 8                  oxyhydroxides in determining Pb mobility. Under oxic conditions, Fe-Mn-hydroxides
 9                  often play an important role in the sorption of Pb to the solid phase soil (Schulz-Zunkel
10                  and Krueger. 2009). In an agricultural soil, fate of Pb in soils is related to agricultural
11                  management. Although Pb was found to be strongly sorbed to the soil, downward
12                  migration was observed and the movement of Pb to deeper soils was due to the soil
13                  mixing activities of earthworms (Fernandez et al., 2007). Thus in relatively unpolluted
14                  non-forested soils, as in  forested soils, colloidal Fe and OM, pH, and biophysical
15                  transport all enhance Pb mobility in soil. Pb transport in more highly contaminated soils
16                  has also been the subject of recent research. In a vegetated roadside soil, Pb was leached
17                  from the upper 50 cm of the soil even though the pH was  7.2. Pb was transported on
18                  mobile particles and colloids in the soil solution. Some of the colloids may have formed
19                  from OM produced by roots and decaying shoots. The transport process was enhanced by
20                  preferential flow triggered by intense rainfall events. This study suggested that the value
21                  of the effective sorption coefficient estimated under dynamic conditions was unrelated to
22                  values measured in conventional batch studies. This indicates that the use of batch studies
23                  to derive input values for sorption coefficients in transport models requires caution. It
24                  was concluded that the primary control of Pb transport in the long term was the degree of
25                  preferential flow in the system (Roulier et al., 2008b).

26                  Other studies also noted similarly low Pb mobility, but with substantial variation between
27                  soil types and  locations. A decline in O-horizon Pb concentrations and Pb accumulation
28                  in mineral horizons was also observed for forest soils by Watmough and Dillon (2007),
29                  but did not hold for nearby wetland areas from which a large amount of DOC is exported,
30                  with approximately 10 times more Pb being associated with a given amount of DOC in
31                  the leachate from the LFH-horizon of the wetland soil than with the DOC in the stream
32                  water draining the wetland. This may reflect greater retention of Pb by the wetland and/or
33                  a change in structure of DOC leading to a change in complexing capacity possibly
34                  because of changes in pH or competition with Al and Fe.

35                  Williams et al. (2006) characterized Pb speciation in a mine waste-derived fertilizer,
36                  ironite. It was  thought that PbS would be the main form of Pb, but instead was the
37                  predominant form was PbSO4, which may move more easily through soil and enter
      November 2012                             3-57                  Draft - Do Not Cite or Quote

-------
 1
 2
 3

 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
proximal waters. In contrast, Courtin-Nomade et al. (2008) showed that Pb was
incorporated into barite rather than goethite in waste rock pile materials. The high-
stability phase formed was an anglesite-barite solid solution.

In weathering flotation residues of a Zn-Pb sulfide mine, more Pb was mobile in
weathered topsoil than in the unweathered subsoil (Schuwirth et al.. 2007). The topsoil
had a very high OM content and the Pb enrichment was attributed to an interaction with
soil OM. Overall, the results contrast strongly with most other studies but the
interpretation was supported by the  sequential extraction results which showed that there
was a very large exchangeable Pb component in these surface soils. Scheetz and Rimstidt
(2009) characterized shooting range soils in Jefferson National Forest, VA, in which the
metallic Pb shot rapidly became corroded and developed a coating of hydrocerussite,
which dissolved at the pH values of 8-9; see Figure 3-12. which shows an Eh-pH diagram
indicating the solubility, equilibrium, and stability of these corroded Pb molecules in
terms of the activity of hydrogen ions (pH) versus the activity of electrons (Eh [in volts]).
The solubilized Pb was largely re-adsorbed by the Fe and Mn oxides and carbonate soil
fractions. The minimum solubility of hydrocerussite lies in the pH range 8-9 but
solubility increases by several orders of magnitude at pH below 6 (Scheetz and Rimstidt
2009).
                                                               Pb3(C03)2(OH)2
                                                               Hydrocerussite
                        0    1   2    3   4    5   6    7   8    9   10   11   12   13   14
      Source: Reprinted with permission of Elsevier Publishing, Scheetz and Rimstidt (2009)
      Figure 3-12     Eh-pH diagram for Pb in shooting range soils, Jefferson National
                        Forest, VA.
     November 2012
                            3-58
Draft - Do Not Cite or Quote

-------
 1                  Rooney et al. (2007) also investigated the controls on Pb solubility in soils contaminated
 2                  with Pb shot. Again, corrosion crusts were found to develop on Pb pellets. The
 3                  concentrations of Pb in the soil solution were, however, much lower than if they were
 4                  controlled by the solubility of the dominant crustal Pb compounds (mainly
 5                  hydrocerussite). Instead it was suggested that the concentrations were being controlled by
 6                  sorption of Pb by the soil solid phase. The pH range in this study was 4.5-6.5 and so
 7                  again dissolution of hydrocerussite would be expected. Sorption to solid phases in the soil
 8                  is also consistent with the findings of Scheetz and Rimstidt (2009). Overall, in contrast to
 9                  less polluted forested and non-forested soils, considerable mobility was often, but not
10                  always observed in soils near roadways and  mines and on shooting ranges, with colloid
11                  transport and soil pH playing an important role in Pb mobility. Although there have been
12                  steep declines in Pb deposition, surface soils in have been slow to recover (Bindler et al..
13                  2008: Kaste et al.. 2006). As was concluded in the 2006 Pb AQCD (U.S. EPA. 2006b).
14                  soils continue to act as a predominant sink for Pb.

15                  While in some studies the flux of Pb, from the soil through aquatic ecosystems to lakes
16                  has peaked and declined. In other studies, no recovery of lake sediments in response to
17                  emission reductions was observed (Norton. 2007). For example, Klaminder et al. (2010)
18                  has shown that the Pb concentrations in sub-Arctic lake sediments remain unchanged in
19                  recent years, with the lack of recovery linked to the effects of soil warming, which affect
20                  Pb-OM transport from soil to the receiving lake systems. Shotyk and Krachler (2010)
21                  also reported a disconnect between atmospheric deposition and recent changes in Pb
22                  concentration and isotope ratios in the lake sediments. Simulations of future metal
23                  behavior suggest that the more strongly sorbing metals such as Pb will respond to
24                  changes in metal inputs  or acidification status only over centuries to millennia (Tipping et
25                  al.. 2006).

26                  Overall, recent research confirms the generally low mobility of Pb in soil. This limited
27                  mobility is strongly dependent on both colloid amount and composition, as well as pH,
28                  and may be greater in some contaminated soils. Mobility is so low that soils continue to
29                  act as a sink for atmospheric Pb even though atmospheric Pb concentrations peaked
30                  several decades ago.
      November 2012                             3-59                  Draft - Do Not Cite or Quote

-------
      3.4       Monitoring of Ambient Pb
      3.4.1        Measurement Techniques
                    3.4.1.1      Sample Collection

                    Federal Reference Methods

 1                  The indicator for the Pb NAAQS is Pb in total suspended particles (Pb-TSP) (73 FR
 2                  66964). In order to be used in regulatory decisions judging attainment of the Pb NAAQS,
 3                  ambient Pb concentration data must be obtained for this indicator using either the Federal
 4                  Reference Method (FRM) or a Federal Equivalent Method (FEM) defined for this
 5                  purpose. Accordingly, for enforcement of the air quality standards set forth under the
 6                  Clean Air Act, EPA has established provisions in the Code of Federal Regulations under
 7                  which analytical methods can be designated as FRM or FEM. Measurements for
 8                  determinations of NAAQS compliance must be made with FRMs or FEMs. FRMs and
 9                  FEMs for the Pb NAAQS exist for both sample collection and sample analysis.

10                  There are two FRMs for sample collection in the Pb NAAQS monitoring network
11                  (described in Section 3.4.2 below): (1) Reference Method for the Determination of Lead
12                  in Suspended Particulate Matter Collected From Ambient Air (40 CFR part 50 Appendix
13                  G), and (2) Reference Method for the Determination of Lead in Particulate Matter as
14                  PM10 Collected From Ambient Air (40 CFR part 50, Appendix G). The  Pb-TSP FRM
15                  sample collection method is required for all source-oriented NAAQS monitors, and the
16                  FRM for Pb-PMi0 is accepted for Pb NAAQS monitoring at non-source-oriented
17                  monitors in specified situations.

18                  The Pb-TSP FRM sample collection method specifies use of a high-volume TSP sampler
19                  that meets specified design criteria (40 CFR part 50 Appendix B). Ambient airborne PM
20                  is collected on a glass fiber filter for 24 hours using a high volume air sampler. It has
21                  long been recognized that there is notable variability in high-volume TSP sample
22                  measurements associated with the effects of wind speed and wind direction on collection
23                  efficiency. This variability is predominantly associated with the capture efficiency for
24                  particles larger than 10 (im, but the  sampler's size selective performance is known to be
25                  affected by wind speed and direction. For example, at a simulated wind speed of 4.6
     November 2012                            3-60                  Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
             meters/second, a directional difference of 45 degrees can result in a nearly two-fold
             difference in 15 (im particle collection efficiency and a nearly five-fold difference in
             50 (im particle collection efficiency (Wedding et al., 1977). Effective D50 (size at 50%
             efficiency) was observed to decrease from 50 (im at a 2 km/h wind speed to 22 (im at
             24 km/h (Rodes and Evans. 1985). Figure 3-13 illustrates the effect of sampler
             orientation on collection efficiency as a function of particle size.
           00
       z
       LJ
       oc
       LJ
       LJ
LLJ

uJ
CL

'J-)
           60
           20
                                              U<*  = 15  fps ,    8  %  TURBULENCE
                                          --O	    MODIFIED  I CFM  ANDERSEN
                                          -•-T  -45"-,
                                          __fc	QOJHI-VOLUME  SAMPLER
                                                      PROTOTYPE DICHOTOMOUS  SAMPLER
                                                      ORIGINAL  1 CFM ANDERSEN
                                                                •D—..
                                                                                                •—D
                              10
                                       20               30
                                    PARTICLE   DIAMETER ,   (
                            50
      Different TSP sampler types: (1) Modified Andersen Sampler [open circles]; (2) Hi-volume Sampler (for different incident wind
      direction (45° [open squares], 0° [closed squares); (3) Prototype 15 urn Outpoint Dichotomous Sampler [open triangles]; and
      (4) Original Andersen Sampler [closed circles].
      Source: Reprinted with permission of the American Chemical Society (Wedding et al.. 1977)

      Figure 3-13     Comparison of particle collection efficiency among different
                        TSP sampler types.
 9
10
11
12
13
14
             Some existing commercially available sampler inlets are designed to collect particles
             larger than 10 (im with greater than 50% efficiency (Kenny et al., 2005). and these inlets
             can be tested as potential replacements for TSP sampling. Efficient collection of particles
             much larger than 10 (im is considerably more challenging because their greater inertia
             and higher settling velocities hinder their efficient intake by samplers. The sampling
             difficulties and the long history of research to develop adequate sampling technology for
             large particles have been thoroughly reviewed (Garland and Nicholson. 1991). High
             intake velocities and large inlet openings are necessary to minimize sampling bias for
      November 2012
                                         3-61
Draft - Do Not Cite or Quote

-------
 1                   sampling ultra-coarse particles. At this time, no alternative to the FRM TSP sampler has
 2                   been identified that has been adequately characterized. As such, there is a continued need
 3                   to assess the feasibility of a revised TSP sampler design with improved control on
 4                   collection efficiency over a wider range of particle sizes, including ultra-coarse particles
 5                   (which are not captured with PM10 samplers).

 6                   The spatial scale for which ambient air Pb samples are representative varies depending on
 7                   particle sizes present, as discussed further in Section 3.5.3. Concentrations of particles
 8                   larger than 10 (im are likely to be very spatially and temporally heterogeneous, with
 9                   higher concentrations in the vicinity of their emissions sources. Under typical conditions,
10                   PMio-2.5 particles travel much shorter distances before settling out than finer particles
11                   (U.S. EPA. 2009a). As a result, spatial and temporal heterogeneity is greater for PMi0_2 5
12                   than for PM2 5, because coarser particles have greater settling velocities (Hinds. 1999).
13                   and settling velocities are even greater for particles larger than 10 (im. Thus, spatial
14                   gradients are steepest near sources, such that measured concentrations of larger particle
15                   sizes tend to be most representative of the ambient air in areas in close proximity to the
16                   monitor, with higher concentrations likely to occur closer to the source and decreasing
17                   concentrations with increasing distance from the source. This issue has been thoroughly
18                   discussed in the 2006 Pb AQCD  (U.S. EPA. 2006b). It has also been acknowledged in
19                   previous Pb AQCDs, with a lengthy discussion appearing in  the 1977 AQCD (U.S. EPA.
20                   1986b. 1977).

21                   The low-volume Pb-PMi0 FRM sample collection method specifies use of a low-volume
22                   PMio sampler that meets specified design criteria (40 CFR part 50, Appendix Q).
23                   Ambient airborne PM is collected on a polytetrafluoroethylene (PTFE) filter for 24 hours
24                   using active sampling at local conditions with a low-volume  PMi0 sampler and analyzed
25                   by X-ray fluorescence (XRF). In recognition of the steep spatial gradients associated with
26                   sources of ultracoarse particles, ambient Pb sampled using the FRM for Pb-PM10 is
27                   allowed in certain instances where the expected Pb concentration does not approach the
28                   NAAQS and no sources of ultracoarse Pb are nearby.


                     Alternative Sample Collection Methods

29                   In addition to the FRMs for ambient Pb sample collection, a  range of other PM sampling
30                   methods are available for collecting samples for Pb analysis. These include FRM
31                   sampling methods for PM that have also been used for collection of samples for Pb
32                   analysis, sampling methods in use in other sampling networks such as the CSN,
33                   IMPROVE and National Air Toxics Trends Stations (NATTS) networks described in
34                   Section 3.4.2. and other sampling methods that have been used to measure airborne Pb
35                   concentrations in research studies unrelated to network applications; these methods are
      November 2012                             3-62                  Draft - Do Not Cite or Quote

-------
1
2
3
4
5
6
7
listed in Table 3-3. Most of these methods have been described in considerable detail in
the 2004 PM AQCD (U.S. EPA. 2004V Table 3-3 also lists key conditions of capture for
each method, including particle size, inlet type, collection medium, and flow rate. Not all
methods included in Table 3-3 and in subsequent paragraphs have been applied to Pb-PM
collection, but these methods represent potential alternative methods for this purpose. It
should also be noted that not all of the samplers listed in Table 3-3 have been wind tunnel
tested for a variety of aerodynamic particle sizes, wind speeds, and wind directions. In
addition, some of these samplers are no longer commercially available.
Table 3-3
Sampler
High Volume
TSP
Low Volume
PM10
PM2.5
Met One
SASS
IMPROVE
MOUDI
Noll
Impactor
SEAS
PM10SSI
HiVol
Andersen
Inhalable PM
Sampler
Sierra
Dichotomous
Sampler
TEOM
Airborne PM sampling methods potentially applicable for
Pb sampling.
Network
Pb-FRM
PM-FRM,
NATTS
PM-FRM
CSN
IMPROVE
None
None
None
FRM
FRM
FRM
FRM
Sampler
Type
Single
Channel
Single
Channel
Single
Channel
Multiple
Channel
Multiple
Channel
Multistage
Impactor
Multistage
Impactor
Slurry
Single
Channel
Single
Channel
Dichotomous
Single
Channel
Mass Median
Aerodynamic
Diameter
TSP
<10 urn
<2.5 urn
<2.5 urn
<2.5 urn
8 stages
0.056-18 urn
4 stages
<108 urn
<1.2 urn
<10 urn
<10 urn
<10 urn,
<2.5 urn
<10 urn
Inlet or
Fractionator
Type
Gabled,
rectangular
Louvered Inlet
+ PM10
Impactor
WINS
Impactor or
(VSCC)
Cyclone
Cyclone
Impactor
Impactor
Impactor
PM10Size
Selective Inlet
RAAS10 Inlet
10 urn Inlet +
Virtual
Impactor
R&P PM10
Inlet or
Louvered Inlet
Collection
Medium
Glass
Teflon
Teflon
Teflon
Teflon
Teflon
Coated Mylar
Slurry
8"x 10" filter
paper
Teflon
Teflon
Teflon Coated
Glass Fiber
Typical
Flow
Rate
1130 L/min
16.67
L/min
16.67
L/min
6.7 L/min
22.8 L/min
30 L/min
Rotating
arm
90 L/min
1130 L/min
16.67
L/min
16.7 L/min
16.7 L/min
Reference
U.S. EPA
(2011ft
U.S. EPA
(2011ft
U.S. EPA
(20111)
MetOne
(2009)
CNL(2001)
Marple et
al. (1991)
Noll (1970)
Pancras et
al. (2006).
Ondov et al.
(2006)
U.S. EPA
(2011ft
U.S. EPA
(20111)
U.S. EPA
(20111)
U.S. EPA
(2011ft
     November 2012
                            3-63
Draft - Do Not Cite or Quote

-------
Sampler
Louvered
Inlet TSP
Texas A&M
Lo-Vol TSP
Andersen
Multistage
Impactor
UIUC
Isokinetic
TSP
Sampler
Airmetric
MiniVol
Stacked
Filter Units
ELPI
Wagner &
Leith
Passive
Sampler
Network
None
None
None
None
None
None
None
None
Sampler
Type
Single
Channel
Single
Channel
Multistage
Impactor
Single
Channel
Single
Channel
Two stage
impactor
Multistage
Impactor
Passive
Mass Median
Aerodynamic
Diameter
<15 urn
TSP
8 stages
0.4-10 urn and
above
TSP
TSP or <10 urn
or <2.5 urn
<10 urn,
<2.5 urn
1 3 stages
0.007-10 urn
Not
characterized
Inlet or
Fractionator
Type
Louvered Inlet
None
Inlet Cone
Isokinetic
Sampling
Head
None, PM10
Impactor, or
PM2.s Impactor
PM10Size
Selective Inlet
Berner
Impactor
Mesh screen
Collection
Medium
Teflon
Teflon
Aluminum
Teflon
Teflon
Nuclepore
Aluminum or
polycarbonate
SEM stub
Typical
Flow
Rate
16.67
L/min
16.67
L/min
28.3 L/min
20 L/min
5 L/min
17 L/min
10 or 30
L/min
0 L/min
Reference
Kenny et al.
(2005)
Wang et al.
(2005b):
Wanjura et
al. (2005)
Mercer et
al. (1970)
Jerez et al.
(2006)
Chen et al.
(2011 a)
IAEA
(1 993)
Keskinen et
al. (1992)
Leith et al.
(2007):
Wagner
and Leith
(2001)
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
Size discrimination is usually accomplished with impactors or cyclones. With impactors,
PM is forced through a jet at high speed, and particle inertia carries particles above a
given size into a collection surface downstream of the jet, while smaller particles follow
the air stream around the collector. In multistage impactors, a series of successive stages
of jets are used to collect a range of particle sizes. The micro-orifice uniform deposit
impactor (MOUDI) is a widely used multistage impactor. The impaction process and
performance of various impactors, including the WINS and MOUDI, has been described
in detail in the 2004 PM AQCD (U.S. EPA. 2004). The biggest concern in collection by
impaction is particle  bounce, which occurs when particles collide with the  collection
surface but bounce off the collection stage into the air stream and are not actually
collected. Considerable effort has been devoted to minimizing errors due to bounce in
FRM samplers, and this has been thoroughly discussed in the 2004 PM AQCD (U.S.
EPA. 2004). An alternative to impaction that also eliminates particle bounce is the use of
an air sampling cyclone. In the  CSN and IMPROVE networks, cyclones are used to
remove particles larger than 2.5 (im. An air sampling cyclone brings air into a tangential
jet and directs flow against a circular wall, where particles larger than a given size are
removed by centrifugal and gravitational forces.
     November 2012
                            3-64
Draft - Do Not Cite or Quote

-------
 1                  Collection medium and flow rate are two other key features of a sampling method. One
 2                  advantage of low volume sampling is its suitability for collection of samples for XRF
 3                  analysis. Because Pb in PM2 5 is analyzed by XRF in the CSN and IMPROVE networks,
 4                  sampling methods that employ Teflon filters suitable for XRF analysis have been
 5                  developed for these networks. In practice, this restricts sampling for airborne Pb to low
 6                  volume samplers with a convenient filter size. This also holds true for the Pb-PMi0 FRM
 7                  sampling, which is also restricted to low volume PM10 samplers because XRF has been
 8                  designated as the FRM for Pb-PMi0 analysis. An additional practical advantage of
 9                  available low volume samplers over the existing high volume Pb-TSP FRM is that
10                  established low volume PM2s and PMi0 sampling methods are not dependent on wind
11                  direction. However, this has to do with sampler design rather than flow rate, and there are
12                  high volume PMi0 sampling methods, including the PMi0 FRMs, that are also free of
13                  wind direction bias. These would be suitable for Pb measurement with other analytical
14                  methods, such as ICPMS, and could have a potential advantage of providing more
15                  material in locations with very low concentrations.

16                  Alternatives to TSP sampling have been developed that collect a particle size range that
17                  extends beyond 10 jam. Early examples include samplers for "inhalable particulate
18                  matter" that were designed to have cut points for 50% sampling efficiency of 15 jam
19                  aerodynamic diameter. These included the Andersen Inhalable Particulate Sampler
20                  (Model 7000, Thermo Electron, Smyrna, GA) and the Sierra Dichotomous Sampler
21                  (Series 244, Sierra Instruments, Monterey, CA), which were evaluated and compared to
22                  each other and to TSP sampling in both co-located field comparisons (Solomon et al..
23                  1982) and wind tunnel studies (Watson et al., 1983). with the result that poor agreement
24                  was observed for low or high wind speeds, or when much coarse particulate matter was
25                  present. For example, the Dichotomous Sampler collected on average  only 73 ± 18% as
26                  much mass as the Inhalable Particulate Sampler, and differences were attributed to
27                  differences in the efficiency of large particle collection (Solomon et al.. 1982).

28                  More recently, a variety of inlets have been developed for low volume TSP sampling.
29                  The omnidirectional TEOM TSP Inlet (Model 10-002929, Rupprecht & Pataschnik Co.,
30                  Inc, Albany, NY) was designed to sample 100 jam particles in still air with the suction
31                  velocity equal to the terminal velocity of a 100 jam diameter unit density sphere.
32                  However, substantial PM mass loss was reported for this inlet and attributed to
33                  anisokinetic sampling conditions that led to  inefficient sampling of large particles (Jerez
34                  et al., 2006). The inefficient sampling of larger particles by the TSP inlet was also
35                  observed by Kenny et al. (2005). who carried out wind tunnel tests of 1) a commercially
36                  available omnidirectional low volume (16.67 liters/minute) TSP inlet, and 2) a louvered
37                  dichotomous inlet designed to select particles from a moving airstream and transmit them
38                  to a downstream PM10 impactor.  They observed that the TSP inlet exhibited low

      November 2012                            3-65                  Draft - Do Not Cite or Quote

-------
 1                  sampling efficiency even for larger particles within the PM10 range, and concluded that it
 2                  is likely to give biased PM concentrations that vary with external winds when large
 3                  particles are present. However, for the louvered dichotomous inlet, Kenny et al. (2005)
 4                  reported high sampling efficiencies with little influence of wind speed across the full
 5                  PMio particle size range, and a 50% cutpoint at around 15 jam. For the full scale TSP,
 6                  inlet sampling efficiencies for 46 jam particles were  17% for a 1 meter/second wind speed
 7                  and 28% at 2 meters/second.

 8                  Evaluation of other available TSP samplers also reveals variability among sampler
 9                  models. The Texas A&M Low Volume TSP Sampler (Coulter Counter Multisizer III,
10                  Beckman Coulter, Inc., Fullerton, CA) was designed based on applicable guidelines for
11                  high volume TSP samplers for use in sampling particulate matter from agricultural
12                  sources (Wang et al.. 2005b: Wanjura et al.. 2005). and mass collected with this sampler
13                  has been compared to an Andersen multistage impactor (Model 20-800, Thermo Electron
14                  Co., Smyrna, GA) in an intercomparison exercise (Park et al.. 2009a). Measurements by
15                  the Anderson Impactor were 97% higher for the PM4 fraction and 14% higher for the
16                  PMio fraction compared with the Texas A&M TSP sampler. Jerez et al. (2006) compared
17                  the University of Illinois at Urbana Champaign (UIUC) isokinetic TSP sampler with a
18                  tapered element oscillating microbalance (TEOM) TSP sampler (Series 1400a, Rupprecht
19                  and Pataschnik Co., Inc., Albany, NY) when measuring dust concentrations in a swine
20                  and chicken houses in Illinois, Indiana, Minnesota, and Texas between September to
21                  December, 2003. The TEOM measured concentrations that were 26-117% of the UIUC
22                  sampler; for 86 of 90 measurements, the TEOM measurements were lower than the
23                  UIUC measurements.

24                  Other new approaches to high volume sampling include "saturation samplers" or low
25                  volume (5 liters/minute) samplers designed for high spatial coverage  of PMio and PM2 5,.
26                  Chen et al. (2011 a) provide an intercomparison among portable MiniVol (Airmetric,
27                  Eugene, OR), Omni (BGI, Inc., Waltham, MA), and dichotomous samplers (Model 2025,
28                  Rupprecht and Pataschnik, Albany, NY) and two FRMs (RAAS-100, Andersen, Smyrna,
29                  GA; Partisol-FRM 2000, Rupprecht and Pataschnik, Albany, NY) for PM10 and PM25
30                  measurements. Chen et al. (2011 a) observed R2 of 0.95-0.98 with average mass
31                  concentrations within 4% among the PMio measurements. They observed R2 of 0.96-0.99
32                  and average mass concentrations within 9% among the monitors for PM2 5 measurements.
33                  Hitzenberger et al.  (2004) found more variability among portable PM2 5 and PMio
34                  monitors when performing intercomparisons in Melpitz,  Germany. PM2 5 was measured
35                  with stacked filter units (SFU) (developed at Ghent University), a TEOM (Model 1400A,
36                  Rupprecht and Pataschnick, Albany, NY), a Digitel (Model  DHA-80) high volume
37                  sampler, an electrical low pressure impactor (ELPI) (Outdoor-ELPI, Dekati, Ltd.), a TSP,
38                  and a MOUDI. PM10 was measured with SFUs, a TSP (manufacturer not provided) with


      November 2012                             3-66                  Draft - Do Not Cite or Quote

-------
 1                  30 L/min and 70 L/min Berner impactors, and a MOUDI (model and manufacturer not
 2                  provided). Based on Hitzenberger et al.'s (2004) reported average mass concentrations,
 3                  the PM25 samplers ranged from 73% (TEOM) to 180% (ELPI) of the average. When
 4                  trimming the extrema, the average PM25 concentrations measured within 10% of the
 5                  overall average. For the PM10 monitors, Hitzenberger et al.'s (2004) reported average
 6                  mass concentrations ranged from 83% (one SFU, MOUDI) to 123% (two SFUs) of the
 7                  overall average. When trimming the extrema, the average PM10 concentrations were
 8                  within 13%.

 9                  Passive samplers have also recently been used for capturing spatial variability of ambient
10                  air Pb concentrations. Field testing of the Wagner and Leith Passive Sampler (University
11                  of North Carolina) illustrated very good agreement with a cascade impactor (difference
12                  within 5% for three sampling events) for PM2 5 in certain cases and very poor agreement
13                  (difference of -51 % and -110%) in others; Wagner and Leith (2001) attributed the poor
14                  agreement to low PM2 5 mass events. PMi0 agreement was 14-65% and was attributed to
15                  potential phenomena such as agglomeration on the passive sampler and water
16                  evaporation from the impactor sample. However, when comparing the Wagner and Leith
17                  Passive Sampler for PMi0.2 5 with PMi0-2.5 calculated by differencing measurements
18                  obtained from co-located or dichotomous FRMs for PMi0 and PM2 5, Leith et al. (2007)
19                  observed that Passive Sampler PMi0.2 5 measurements integrated over at least one week
20                  were within one standard deviation of the PMi0.2 5 obtained from co-located or
21                  dichotomous FRMs. Kumar et al. (2012) and Lagudu et al. (2011) illustrated how the
22                  Wagner and Leith Passive Sampler can be coupled with computer-controlled scanning
23                  electron microscopy (CCSEM) to produce concentrations of Pb-PMi0.2 5 in samples taken
24                  across the cities of Syracuse, NY and Rochester, NY, respectively.

25                  These results illustrate that alternative sampling options to TSP are available to capture
26                  ambient air Pb concentrations for particles with a cutpoint of approximately 15 urn or
27                  higher, and the state of the science for sampling in this particle size range is progressing,
28                  including better documentation of performance and field intercomparison data. In
29                  general, both the historical and recently available alternatives to the traditional TSP
30                  sampler illustrate that samplers designed to  collect particles up to a size range greater
31                  than 10 nm have not performed or compared as well as samplers designed to collect
32                  smaller particles, and the  challenge of achieving good performance for collection of
33                  particles size ranges greater than 10 pirn has not been limited to the TSP sampler. This is
34                  expected given the inherent difficulties associated with large particle sampling (Garland
35                  and Nicholson.  1991). In  spite of this, good performance by recently developed samplers,
36                  including the louvered TSP inlet evaluated by Kenny et al. (2005) show promise for good
37                  performance in collection of a particle size range extending beyond 10 pirn. The primary
38                  route of Pb exposure is hand-to-mouth contact with deposited Pb on soil or dust having


      November 2012                             3-67                   Draft - Do Not Cite or Quote

-------
 1                  substantially larger size fractions compared with airborne Pb particles, as described in
 2                  Section 4.1. The relevant particle size distribution for ambient sampling is smaller than
 3                  the size distribution of the settled dust. Particles larger than about 20 nm are generally
 4                  considered too large to be transported for more than a few seconds under typical
 5                  conditions; see Section 3.3.1.3. If preliminary results concerning the TSP louvered inlet
 6                  and other TSP alternatives are verified, this may be very close to the practical limit for
 7                  good sampling data quality. It follows that 15 to  20 urn may be a practical limit for both
 8                  good sampling data quality and representative sampling in a limited area.
                    3.4.1.2      Sample Analysis: Federal Reference and Federal
                                 Equivalence Methods

 9                  As described in Section 3.4.1.1. measurements for determinations of NAAQS  compliance
10                  must be made with FRMs or FEMs. As of October 12, 2011, 1 manual reference method
11                  and 25 manual equivalent methods for sample analysis had been approved for Pb
12                  (http://www.epa.gov/ttn/amtic/files/ambient/criteria/reference-equivalent-methods-
13                  list.pdf). The FRM for Pb (Pb-TSP) was promulgated in 1979 and is based on flame
14                  atomic absorption spectroscopy (AAS) (40 CFR Part 50, Appendix G). Ambient air
15                  suspended in PM is collected on a glass fiber filter for 24 hours using a high volume air
16                  sampler. Pb in PM is then solubilized by extraction with nitric acid (HNO3), facilitated by
17                  heat, or by a mixture of HNO3 and hydrochloric acid (HC1) facilitated by ultrasonication.
18                  The Pb content of the sample is analyzed by atomic absorption spectrometry using an air-
19                  acetylene flame, using the 283.3 or 217.0 nm Pb absorption line, and the optimum
20                  instrumental conditions recommended by the manufacturer. Several FEMs have been
21                  approved based on a variety of principles of operation have been approved, including
22                  inductively coupled plasma optical emission spectrometry, and inductively-coupled
23                  plasma mass spectrometry (ICPMS).


                    Atomic Absorption Spectrometry

24                  AAS is the basis for the existing FRM. Atomic  absorption spectrometry was first
25                  developed in the 19th century, and became widely used in the  1950s. More than 70
26                  elements can be analyzed by AAS. Typically a liquid sample is nebulized into a flame
27                  with sufficient heat for elements to be atomized. The liquid specified by the FRM is a
28                  nitric acid extract of a glass fiber filter used for collection of suspended PM with a high
29                  volume sampler. The atomized sample is then irradiated with visible light at a specific
30                  wavelength to promote an electronic transition to a short-lived excited state, resulting  in
31                  absorption of the light. Elemental selectivity is achieved because light absorption is
      November 2012                            3-68                  Draft - Do Not Cite or Quote

-------
 1                  specific to a particular electronic transition in a particular element. As a result, absorption
 2                  of light at a given wavelength generally corresponds to only one element. The flame is
 3                  irradiated with a known quantity of light and intensity of light is measured on the other
 4                  side of the flame to determine the extent of light absorption in the flame. Using the Beer-
 5                  Lambert law the concentration of the element is determined from the decrease in light
 6                  intensity due to sample absorption.

 7                  A more sensitive variation of atomic absorption spectrometry for most elements is
 8                  graphite furnace atomic absorption spectrometry (GFAAS). Instead of introducing the
 9                  sample into a flame, the liquid sample is deposited in a graphite tube that is then heated to
10                  vaporize and atomize the sample.


                    Inductively-Coupled Plasma Mass Spectrometry

11                  Inductively coupled plasma mass spectrometry (ICPMS) is a sensitive method of
12                  elemental analysis developed in the late 1980s. Argon (Ar) plasma (ionized gas) is
13                  produced by transmitting radio frequency electromagnetic radiation into hot argon gas
14                  with a coupling coil. Temperatures on the order of 10,000 K are  achieved, which is
15                  sufficient for ionization of elements. Liquid samples can be introduced into the plasma by
16                  extracting samples in an acid solution or water, and nebulizing dissolved elements.
17                  Resulting ions are then separated by their mass to charge ratio with a quadrupole and
18                  signals are quantified by comparison to calibration standards. While solid samples can be
19                  introduced by laser ablation, nebulization of liquid extracts of PM collected on Teflon
20                  filters is more typical. One major advantage of ICPMS over AAS is the ability to analyze
21                  a suite of elements simultaneously. An additional advantage is low detection limits of
22                  50-100 parts/trillion for Pb.


                    Inductively-Coupled Atomic Emission Spectroscopy

23                  Inductively coupled atomic emission spectroscopy (ICP-AES) also generates ions from
24                  elements with a hot Ar plasma, similar to ICPMS. Excited atoms and ions are produced,
25                  and these emit electromagnetic radiation with frequencies characteristic of a particular
26                  element. Intensity of emission is used to determine the concentration of an element in the
27                  sample. Elements are extracted from filter samples and nebulized into the plasma.


                    Energy Dispersive X-ray Fluorescence

28                  In energy dispersive X-ray fluorescence spectrometry a beam of X-ray photons from an
29                  external excitation source is applied to a sample, causing ejection of inner shell electrons
      November 2012                            3-69                  Draft - Do Not Cite or Quote

-------
 1                  from elements in the sample. Because inner shell electrons have higher electron binding
 2                  energies than outer shell electrons, the ejection of the inner shell electron induces an
 3                  energetically favorable electronic transition of an outer shell electron to replace the
 4                  ejected electron. The energy released as a result of this transition is in the form of
 5                  electromagnetic radiation, corresponding to the difference in electronic binding energies
 6                  before and after the transition. The energy released is typically in the X-ray portion of the
 7                  electromagnetic spectrum. The release of electromagnetic radiation as a result of an
 8                  electronic transition is defined as fluorescence. Fluorescence energies associated with
 9                  electronic transitions depend on atomic structure, and vary between elements. As a result,
10                  X-ray fluorescence energy is uniquely characteristic of an element, and X-ray intensity at
11                  a given energy provides a quantitative measurement of elemental concentration in the
12                  sample. The X-rays are detected by passing them through a semiconductor material,
13                  resulting in an electrical current that depends on the energy of the X-ray.
                     3.4.1.3     Other Analysis Methods for Total Pb

14                   Several other methods that have not been designated as FRM or FEM methods have also
15                   been frequently used to obtain atmospheric Pb measurements. These include proton
16                   induced x-ray emission (PIXE), X-ray photoelectron spectroscopy (XPS), and other
17                   methods
                     PIXE

18                   Proton-induced X-ray emission (PIXE) spectroscopy has been widely used to measure Pb
19                   in atmospheric PM. In PIXE, a high-energy proton beam passes through the sample,
20                   causing electrons to be excited from inner shells. The x-rays emitted when electronic
21                   transition occur to replace the inner shell electrons are characteristic of an element and
22                   can be used to identify it. Development of PIXE for analysis of airborne PM was
23                   reviewed by Cahill et al. (1981). Numerous applications of PIXE to analysis of airborne
24                   Pb-PM have been reported in the past five years (Cohen etal.. 2010; Waheedet al.. 2010;
25                   Sanchez-Ccovllo et al.. 2009; Chan et al.. 2008; Johnson et al.. 2008; Cong et al.. 2007;
26                   Ariola etal.. 2006; Johnson et al.. 2006; Wahlin et al.. 2006).


                     XPS

27                   X-ray photoelectron spectroscopy (XPS), also called electron spectroscopy for chemical
28                   analysis  (ESCA) has been used to determine Pb concentrations on materials surfaces,
29                   including atmospheric PM (Finlavson-Pitts and Pitts. 2000). A fixed frequency X-ray
      November 2012                             3-70                  Draft - Do Not Cite or Quote

-------
 1                  beam causes inner shell electrons to be emitted and kinetic energy of ejected electrons is
 2                  measured. Binding energy characteristic of an element can be calculated from the
 3                  measured kinetic energy, allowing identification of the element. XPS can also provide
 4                  information about an element's chemical environment or oxidation states because of
 5                  chemical shifts in binding energy caused by differences in chemical environment. There
 6                  have been some recent applications of XPS to airborne PM, concluding that Pb was
 7                  mostly in the form of Pb sulfate (Qi et al., 2006). XPS analysis is a surface technique that
 8                  is suitable only to a depth of 20-50A.


                    Other Total Pb Methods

 9                  Anodic stripping voltammetry, atomic emission spectroscopy, and colorimetry have also
10                  been used for measurement of atmospheric Pb (Finlayson-Pitts and Pitts. 2000). In anodic
11                  stripping voltammetry, metal ions are reduced to metallic form and concentrated as an
12                  amalgam on a suitable electrode (e.g., a mercury (Hg) amalgam on a mercury (Hg)
13                  electrode). This is followed by re-oxidation in solution, which requires "stripping" the
14                  reduced metal from the electrode. Emission spectroscopy can be compared to the existing
15                  FRM for Pb based on AAS. In atomic absorption spectroscopy radiation absorbed by
16                  non-excited atoms in the vapor state is measured. In emission spectroscopy, radiation due
17                  to the transition of the electron back to ground state after absorption is measured, and the
18                  energy of the transition is used to uniquely identify an element in a sample. Colorimetric
19                  methods are wet chemical methods based on addition of reagents to a Pb containing
20                  solution to generate measurable light absorbing products. These methods are less
21                  sensitive than ICPMS, XRF, and PIXE and their use is declining as more sensitive
22                  methods become more widely used, but have advantages regarding simplicity and cost.
                    3.4.1.4     Sequential Extraction

23                  Sequential extraction has been widely used to further classify Pb for various purposes,
24                  including bioavailability, mobility, and chemical speciation. In general the more easily
25                  extractable Pb is considered more mobile in soil and is more bioavailable to organisms.
26                  This approach has also been used widely in characterization of airborne PM. In its
27                  original application (Tessier et al.. 1979) metals extraction solvents were selected to
28                  correspond to common species present in soil, and metals were classified as
29                  exchangeable, bound to carbonates, bound to iron and manganese oxides, bound to OM,
30                  and residual. Extraction was carried out with successively stronger solutions, starting
31                  with magnesium chloride (MgCl2) for removal of exchangeable metals and ending with
32                  hydrofluoric and perchloric acids for removal of residual metals. Pb was one of the
      November 2012                             3-71                   Draft - Do Not Cite or Quote

-------
 1                  elements originally studied by Tessier et al. (1979) as well as one the elements analyzed
 2                  when Tessier's scheme was first applied to airborne PM (Fraser and Lum. 1983).

 3                  Tessier's scheme was modified and optimized for airborne PM over time (Fernandez
 4                  Espinosa et al., 2002) and additional extraction schemes were also developed (Chester et
 5                  al.. 1989). including the simplest case of two fractions corresponding to soluble and
 6                  insoluble fractions (Falta et al., 2008; Canepari et al., 2006; Voutsa and Samara. 2002).
 7                  The variety of methods in current use was recently thoroughly reviewed by Smichowski
 8                  et al. (2005). With the recognition that biological processes involving deposited PM
 9                  metals were related to their solubility (U.S. EPA. 2009a).  sequential extraction methods
10                  or simpler schemes to divide metals into water and acid soluble fractions were
11                  increasingly applied to PM samples to obtain data not just on total metal concentration
12                  but also on water soluble concentration (Graney et al.. 2004; Kyotani and Iwatsuki. 2002;
13                  Wang et al., 2002). Compared to other elements, a large fraction of total Pb is soluble
14                  (Graney et al.. 2004). Recent advances in this area have included application to size
15                  fractionated PM (Dos Santos et al., 2009; Birmili et al., 2006). time resolved
16                  measurements (Perrino et al.. 2010). and an extensive comparison of different
17                  fractionation schemes (Canepari et al.. 2010). Sequential extraction with two or more
18                  fractions is becoming more widely used for characterization of Pb-PM in a variety of
19                  sources (Canepari et  al.. 2008; Smichowski et al.. 2008; Poykio et al.. 2007; Sillanpaa et
20                  al.. 2005) and locations (Perrino et al.. 2010; Dos Santos et al.. 2009; Cizmecioglu and
21                  Muezzinoglu. 2008; Dahl et al.. 2008; Sato et al.. 2008; Annibaldi et al.. 2007; Richter et
22                  al.. 2007; Al-Masri et al.. 2006; Canepari et al.. 2006; Fuiiwara et al.. 2006; Wang et al..
23                  2006c; Yadav and Raiamani. 2006; Gutierrez-Castillo et al.. 2005; Heal et al.. 2005).
24                  leading to a better understanding of mobility characteristics of Pb in airborne PM.
                    3.4.1.5     Speciation Techniques

                    X-ray absorption fine structure (XAFS)
25                  There have been few attempts to speciate Pb in atmospheric PM. However, recently X-
26                  ray absorption fine structure (XAFS) has been applied to PM and road dust to obtain Pb
27                  speciation data from direct analysis of particle surfaces. In XAFS the absolute position of
28                  the absorption edge can be used to determine the oxidation state of the absorbing atom,
29                  and scattering events that dominate in the near edge region provide data on vacant orbital
30                  energies, electronic configurations, and site symmetry of the absorbing atom that can be
31                  used to determine the geometry of the atoms surrounding the absorbing atom. XAFS can
32                  be divided into two spectral regions. X-ray absorption near edge structure (XANES) is
33                  the region of the x-ray absorption spectrum up to 50 eV above the absorption edge

      November 2012                             3-72                 Draft - Do Not Cite or Quote

-------
 1                  observed when an inner shell electron is electronically excited into unoccupied states, and
 2                  Extended X-ray Absorption Fine Structure (EXAFS) up to 1 keV above the absorption
 3                  edge. Both have been applied recently to Pb in PM. XANES spectra of Pb coordination
 4                  complexes with a wide range of environmentally relevant ligands have been reported
 5                  (Swarbrick et al..  2009). XANES has been used to show that several different Pb species
 6                  are probably present in urban airborne PM (Tunasaka et al.. 2008) and urban road dust
 7                  (Barrett et al., 2010). XANES has been used to differentiate between Pb chromate,
 8                  Pb-sorbed minerals, Pb chloride, Pb oxide, Pb carbonate, Pb sulfide and Pb sulfate are
 9                  probably present in urban PM and road dust samples (Barrett et al.. 2010; Funasaka et al..
10                  2008; Tan et al.. 2006). XANES has also been used to investigate Pb in air samples
11                  thought to be complexed with humic substances  from soil (Pingitore et al.. 2009) and to
12                  investigate the speciation of atmospheric Pb in soil after deposition (Guo et al., 2006b).
13                  EXAFS has been  applied to emission sources to  show Pb from a sinter plant was mainly
14                  carbonate (Sammut et al., 2010). XAFS has only been applied to airborne PM very
15                  recently and shows promise for chemical speciation of airborne metals, including Pb.


                    GC-ICPMS and HPLC-ICPMS

16                  Environmental  analytical methods for organolead compounds prior to 2000 were
17                  generally time consuming and costly, requiring extraction, derivatization, and detection
18                  (Quevauviller. 2000). These have been thoroughly reviewed (Pyrzynska. 1996) and
19                  method intercomparison studies have been conducted (Quevauviller.  2000). More
20                  recently, speciation of organometallic compounds in environmental samples has usually
21                  carried out by coupling a chromatographic separation step with a mass spectrometry-
22                  based multi-element detection system capable of analyzing a wide range of elements
23                  along with Pb, and these approaches have also been recently reviewed (Hirner.  2006).
24                  Chromatographic systems in common use are  gas chromatography (GC) and high
25                  performance liquid chromatography (HPLC). Detection systems most commonly used are
26                  an inductively coupled plasma mass spectrometer (ICP-MS [ICPMS]), electron impact
27                  ionization mass spectrometry (El-MS), and electrospray ionization mass spectrometry
28                  (ESI-MS) (Hirner. 2006). Using these techniques, organometallic species are separated
29                  from each other based on differences in retention times on chromatographic columns, and
30                  elemental Pb is determined by the ICPMS used as a detector downstream of the column
31                  to measure elemental Pb in the pure compounds  after chromatographic separation. Pb
32                  speciation analysis has benefited from the development of HPLC-ICPMS in particular
33                  (Quevauviller. 2000). Recent advances in metal speciation analysis in environmental
34                  samples by HPLC-ICPMS have been extensively reviewed (Popp etal.. 2010).  HPLC-
35                  ICPMS has been used for analysis of Pb complexes with humic substances (Vogl and
36                  Heumann. 1997). which could be relevant for  resuspended soil and road dust. GC-ICPMS
      November 2012                             3-73                  Draft - Do Not Cite or Quote

-------
 1                  has been more widely used for separation and analysis of methyl and ethyl Pb species in
 2                  atmospheric PM (Poperechna and Heumann. 2005; Jitaru et al.. 2004; Leal-Granadillo et
 3                  al.. 2000).


                    Pb-lsotope Ratio Analysis

 4                  Classifying Pb by its relative isotopic abundance has also proved useful for a variety of
 5                  purposes, including the determination of its geochemical origins in natural samples and
 6                  the relative contributions of coal burning, mining, smelting, and motor vehicle emissions
 7                  in polluted samples (Farmer et al.. 1996). Typically, isotopes of Pb (208Pb, 207Pb, 206Pb,
 8                  and 204Pb) are measured in a sample using mass spectrometry, and then ratios of the
 9                  isotopes are calculated to obtain a "signature." Isotopes of 208Pb, 207Pb, and  206Pb are
10                  substantially more abundant than 204Pb, but they vary depending on the geologic
11                  conditions  under which the ore was produced through decay of different isotopes of
12                  uranium and thorium (Cheng and Hu. 2010). Isotope ratio analysis was first applied to
13                  airborne PM in 1965 to identify the impact of motor vehicle exhaust on marine and
14                  terrestrial Pb deposition in the Los Angeles area (Chow and Johnstone. 1965). More
15                  recently, high resolution ICPMS has also proved to be a sensitive tool for isotope ratio
16                  analysis. High resolution ICPMS was first applied to geological samples (Walder and
17                  Freedman.  1992). and has since been widely used for determination of Pb isotope ratios
18                  in airborne PM samples. Pb isotope ratios have been measured in a number of recent
19                  studies in a variety of locations to investigate the origin of airborne Pb (Knowlton and
20                  Moran. 2010; Noble et al.. 2008; Hsu et al.. 2006; Widory. 2006). Shotyk and Krachler
21                  (2010) also used Pb isotopes to demonstrate that the fate of Pb from runoff can be
22                  different from Pb with different origins. They observed that humus PM impacted by
23                  leaded on-road gasoline that are derived from soil surfaces are likely to be more easily
24                  transferred to sediments than Pb of other origins, with substantial amounts retained by
25                  lakes.

26                  Recent studies have examined the use of Pb isotope ratios as a tool for source
27                  apportionment. Duzgoren-Aydin and Weiss (2008) provide caveats for using isotope ratio
28                  analyses. They point out that Pb isotope ratios  may vary when Pb from several sources of
29                  different geological origins are introduced to the same location. Duzgoren-Aydin (2007)
30                  warned that the presence of a complex mixture of contaminants containing common Pb
31                  isotopes can lead to an overestimation of the contribution of one source (e.g., soil
32                  contaminated by Pb emissions from on-road gasoline) and an underestimate of another
33                  source, such as that from industry. For this reason, Cheng and Hu (2010) suggest that Pb
34                  isotope analysis only be used when the investigators are confident that the isotopic
35                  signatures of various sources differ substantially. Pb recycling and international trading
36                  may cause  more blending of Pb from various sources, so that there is less heterogeneity

      November 2012                             3-74                   Draft - Do Not Cite or Quote

-------
 1                  in the Pb isotopic signatures sampled. Additionally, Cheng and Hu (2010) point out that
 2                  the isotopic signature of Pb in air or soil may change over time with changing source
 3                  contributions, but historical Pb isotope data are lacking. Duzgoren-Aydin and Weiss
 4                  (2008) suggest the use of geographical information systems (GIS) mapping of Pb isotopic
 5                  information to help distinguish potential sources based on location of sources in addition
 6                  to the sources' isotopic signature.

 7                  Gulson et al.  (2007) examined the relationships between Pb isotope ratios and source
 8                  apportionment metrics at urban and rural sites in New South Wales, Australia. In this
 9                  study, Gulson et al. (2007) performed source apportionment with both principal
10                  component analysis (PCA) and a neural network technique called the self-organizing map
11                  (SOM) and compared results from each method with 206Pb/204Pb, 207Pb/206Pb, and
12                  208Pb/206Pb obtained from PM samples, although only 206Pb/204Pb results were presented
13                  in detail. Wintertime "fingerprints" from both the PCA and  SOM methods produced
14                  similarly linear relationships with 206Pb/204Pb, with linearly decreasing relationships
15                  between the isotope ratios and the "secondary industry," "smoke," "soil," and "seaspray"
16                  source categories. However, the relationships of the isotope ratios with the SOM
17                  fingerprints and PCA factors,  respectively, were very similar. This finding may have
18                  been due to the presence of elements such as black carbon and sulfur in several SOM
19                  fingerprints and PCA factors.  The authors suggest that this might be  related to the
20                  presence of several sources, which in combination result in a weak atmospheric signal.
21                  Additionally, both PM2 5 and TSP samples were utilized for this study, and it was found
22                  that similar results were obtained for either size cut. At the urban site, they observed that
23                  the 206Pb/204Pb ratio decreased overtime with increasing  contributions of industrial, soil,
24                  smoke, and sea spray sources. For the most part, these sources were not substantial
25                  contributions to Pb-PM25 for the rural site. As for the Tan et al. (2006)  speciation study
26                  described above, no notable differences were observed between the size fractions with
27                  regard to isotopic signature.
                    3.4.1.6     Continuous Pb Monitoring

28                  Development of high time resolution measurement capabilities has advantages for
29                  determining peak exposure concentrations and diurnal exposure trends. High time
30                  resolution samplers suitable for analysis after sampling by XRF and ICPMS have been
31                  developed and applied. The eight-stage Davis Rotating Unit for Monitoring (DRUM)
32                  impactor (Raabe et al.. 1988; Cahill et al.. 1987) collects PM samples with a cascade
33                  impactor on Mylar film substrate on a slowly rotating drum, with samples analyzed by
34                  XRF. It has been used to measure size and time resolved Pb and other elements with a
35                  time resolution of less than 6 hours using x-ray fluorescence (Cahill. 2003; Bench et al..


      November 2012                             3-75                  Draft - Do Not Cite or Quote

-------
 1                  2002). The University of Maryland Semi-continuous Elements in Aerosol Sampler
 2                  (Kidwell and Ondov. 2004. 2001) uses direct steam injection to promote condensational
 3                  growth of sampled PM at a high flow rate, and accumulates resulting droplets in a slurry
 4                  by impaction. It has been successfully applied to measurement of Pb and other elements
 5                  by AAS (Pancras et al.. 2006; Pancras et al.. 2005) with a 30-minute time resolution. This
 6                  approach is also suitable for ICPMS analysis. A gas converter apparatus has also been
 7                  developed to improve transfer of ions to the ICPMS, including Pb, and successfully
 8                  tested with outdoor air (Nishiguchi et al.. 2008). Other high time resolution methods
 9                  suitable for Pb analysis in PM are under development, including near real-time XRF
10                  analysis.

11                  Much of the recent progress in ambient aerosol instrumentation has been related to the
12                  development and improvement of single particle mass spectrometry (Prather et al.. 1994).
13                  Preferential loss as a function of particle size is a concern with this method, but
14                  considerable effort has been devoted to optimizing transfer from atmospheric pressure
15                  down to time of flight operating pressures with minimal particle loss (Prather et al..
16                  1994). This technique can also be considered as an effective method for real time Pb
17                  measurement in PM, including size-resolved measurements from 0.1 to 4.0 (im (Silva and
18                  Prather. 1997). Progress has continued in the development of single particle mass
19                  spectrometry to quantify elements and organic  ion fragments and a number of recent
20                  applications that included (Snyder et al.. 2009; Johnson et al.. 2008; Bein et al.. 2007;
21                  Reinard et al.. 2007; Pekney et al.. 2006) or specifically targeted (Salcedo etal..  2010;
22                  Moffet et al.. 2008a; Murphy et al.. 2007) Pb measurements.
      3.4.2        Network Design

23                  Four national monitoring networks collect data on Pb concentrations in ambient air and
24                  report it to the Air Quality System (AQS).: State and local agencies carry out the
25                  monitoring at state and local monitoring stations (SLAMS) using FRMs and FEMs and
26                  report data to these national networks, which have been established for various purposes.
27                  Although these data may be used for other scientific purposes, the SLAMS network is
28                  designed primarily with the goal of evaluating compliance with the Pb NAAQS. In
29                  addition to FRM monitoring, Pb is also measured within the Chemical Speciation
30                  Network (CSN), IMPROVE, and the NATTS networks as described in Section 3.4.2.2.
31                  Measurements among these networks are not directly comparable in all cases because of
32                  method differences, including the PM size range sampled (TSP, PM10, or PM2 5).
      1 The Air Quality System (AQS) is EPA's repository of ambient air quality data. AQS stores data from over 10,000
      monitors, 5,000 of which are currently active (http://www.epa.gov/ttn/airs/airsaqs/').

      November 2012                             3-76                  Draft - Do Not Cite or Quote

-------
                     3.4.2.1      NAAQS Monitoring Network

 1                   Monitors in the SLAMS network include predominantly those sited in compliance with
 2                   regulatory requirements for the purposes of judging attainment with the NAAQS. For this
 3                   purpose, these sites employ FRM samplers coupled with FRM/FEM analysis methods. At
 4                   the time of the last review, there were approximately 250 sites operating in this network,
 5                   although analyses at the time indicated incomplete coverage of the larger stationary
 6                   sources of Pb (U.S. EPA. 2007h). As a result of the review, the Pb NAAQS monitoring
 7                   requirements were revised. These revisions, some aspects of which were finalized in
 8                   2008 and the remainder in December 2010, call for expanded monitoring at both source-
 9                   oriented and non-source-oriented sites (75 FR 81126, 40 CFR part 58, Appendix D,
10                   Section 4.5 to Part 58).: Source-oriented monitoring sites are required near sources of Pb
11                   air emissions which are expected to or have been shown to contribute to ambient air Pb
12                   concentrations in excess of the NAAQS. At a minimum there must be one source-
13                   oriented site located to measure the maximum Pb concentration in ambient air resulting
14                   from each non-airport Pb source estimated to emit Pb at a rate of 0.50 or more tons/year
15                   and in locations near those airports at which activities associated with the use of leaded
16                   aviation fuel are estimated to result in Pb emissions at a rate of 1.0 or more tons
17                   per year.2 The emission monitoring threshold was established to ensure monitoring near
18                   Pb air sources with the greatest potential to cause ambient air concentrations to exceed
19                   the Pb NAAQS. The Pb NAAQS measurements required at these sites may be as Pb-TSP
20                   orPb-PM10(75FR81126).

21                   Monitoring agencies are also required to conduct non-source-oriented Pb monitoring at
22                   each National Core multipollutant monitoring network (NCore)3 site in a Core Based
23                   Statistical Area (CBSA) with a population of 500,000 or more. While non-source-
24                   oriented monitoring data can be used for purposes of NAAQS attainment designations,
25                   the main objective for non-source-oriented monitoring is to gather information on
      1 EPA Regional Administrators may require additional monitoring beyond the minimum requirements where the
      likelihood of Pb air quality violations is significant. Such locations may include those near additional industrial Pb
      sources, recently closed industrial sources, airports where piston-engine aircraft emit Pb and other sources of re-
      entrained Pb dust (40 CFR, part 58, Appendix D, Section 4.5(c).
      2The requirement for monitoring near sources emitting 0.5 tons/year or more may be waived if it can be shown that
      the source will not contribute to a maximum 3-month average Pb concentration in ambient air in excess of 50
      percent of the NAAQS level based on historical monitoring data, modeling, or other means (40 CFR, part 58,
      Appendix D, Section 4.5(a)(ii)).
      3 NCore is a new network of multipollutant monitoring stations intended to meet multiple monitoring objectives.
      The NCore stations are a subset of the SLAMS network are intended to  support long-term trends analysis, model
      evaluation, health and ecosystem studies, as well as NAAQS compliance. The complete NCore network consists of
      approximately 60 urban and 20 rural stations, including some existing SLAMS sites that have been modified for
      additional measurements. Each state will contain at least one NCore station, and 46 of the  states plus
      Washington, D.C., will have at least one urban station.
      November 2012                              3-77                   Draft - Do Not Cite or Quote

-------
 1                  neighborhood-scale Pb concentrations that are typical in urban areas so to better
 2                  understand ambient air-related Pb exposures for populations in these areas.

 3                  Spatial scales defined for Pb monitoring range from microscale to neighborhood scale,
 4                  with the most important spatial scales for source-oriented sites to effectively characterize
 5                  emissions from point sources being microscale and middle scale, and the most important
 6                  scale for non-source-oriented sites to characterize typical Pb concentrations in urban
 7                  areas being neighborhood scale [40 CFR Part 58, Appendix D, 4.5(d)]:

 8                     • Microscale: This scale is intended to typify areas in close proximity to Pb point
 9                       sources where it may represent an area impacted by the emissions plume with
10                       dimensions ranging from several meters up to about 100 meters.
11                     • Middle Scale: This scale is described as generally representing Pb air quality
12                       levels in areas up to several city blocks in size with dimensions on the order of
13                       approximately 100 meters to 0.5 km.
14                     • Neighborhood Scale: This scale is to characterize concentrations throughout
15                       some relatively uniform land use areas with dimensions in the 0.5 to 4.0 km
16                       range. Where a neighborhood site is located away from immediate Pb sources,
17                       the site may be very useful in representing typical air quality values for a larger
18                       residential area, and therefore suitable for population exposure and trends
19                       analyses.
20                  Source oriented monitors near sources estimated to emit 1.0 tons/year Pb were required to
21                  be operational by January 1, 2010, and the remainder of the newly required monitors,
22                  including the non-source-oriented NCore sites, were required to be operational by
23                  December 27, 2011 (75 FR 81126). When the December 2010 Pb network requirements
24                  are fully implemented, the Pb NAAQS monitoring network is expected to consist of
25                  approximately 270 required monitors including approximately 210 source-oriented
26                  monitors and 60 non-source-oriented monitors. Figure 3-14 shows the estimated
27                  geographic distribution of Pb NAAQS monitors in the current Pb NAAQS monitoring
28                  network. This includes monitors that previously existed and are still in operation, along
29                  with those that are newly required.

30                  With the December, 2010 regulations, EPA also required one year of Pb-TSP (FRM)
31                  monitoring near 15 airports in order to gather additional information on the likelihood of
32                  NAAQS exceedances near airports due to the combustion of leaded aviation gasoline
33                  (75 FR 81126). These airports were selected based on three criteria: annual Pb inventory
34                  between 0.50 tons/year and 1.0 tons/year, ambient air within 150 meters of the location of
35                  maximum emissions (e.g., the end of the runway or run-up location), and airport
      November 2012                            3-78                  Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
10
11
configuration and meteorological scenario that leads to a greater frequency of operations
from one runway. These characteristics were selected because they are expected,
collectively, to identify airports with the highest potential to have ambient Pb
concentrations approaching or exceeding the Pb NAAQS. Data from this monitoring
study will be used to assess the need for additional Pb monitoring at airports. These  15
sites (Figure 3-15 and Table 3-4) were required to be operational no later than December
27, 2011. Evaluating the air quality impact of piston aircraft operations includes
consideration of the seasonal variation in activity by these aircraft. At some of the most
active general aviation airports in the country, spring and summer operations (by piston
aircraft) can increase as much as 73% over operations in the fall and winter, while at
other airports, piston aircraft activity is more consistent throughout the year.
                                                Previously existing sites and any newly required 1 tpy source-oriented sites
                                                Potential new 0.50 tpy source-oriented sites1
                                                NCore (non-source-oriented) sites
                                              Based on 2008 National Emission Inventory lead emission estimates
      Figure 3-14      Map of monitoring sites in current Pb NAAQS monitoring
                         network.
      November 2012
                            3-79
Draft - Do Not Cite or Quote

-------
Quality assured results of this study were not available in time for this assessment. Note that the two Santa Clara Co., CA airports
are not distinguishable on the map.


Figure 3-15     Fifteen U.S. locations where a study is currently being performed

                  on airport Pb emissions.
November 2012
3-80
Draft - Do Not Cite or Quote

-------
      Table 3-4     List of 15 airports included in the airport study

      Airport                                       County, State
         Merrill Field                                  Anchorage, AK
         Pryor Field Regional                            Limestone, AL
         Palo Alto Airport of Santa Clara County              Santa Clara, CA
         Reid-Hillview                                 Santa Clara, CA
         McClellan-Palomar                             San Diego, CA
         Gillespie Field                                San Diego, CA
         San Carlos                                   San Mateo, CA
         Nantucket Memorial                            Nantucket, MA
         Oakland County International                     Oakland, Ml
         Republic                                    Suffolk, NY
         Brookhaven                                  Suffolk, NY
         Stinson Municipal                              Bexar, TX
         Northwest Regional                            Denton, TX
         Harvey Field                                 Snohomish, WA
         Auburn Municipal                              King, WA
                    3.4.2.2      Other Pb Monitoring Networks

 1                  In addition to FRM monitoring, Pb is also measured within the Chemical Speciation
 2                  Network (CSN), Interagency Monitoring of Protected Visual Environments (IMPROVE),
 3                  and the National Air Toxics Trends Station (NATTS) networks. Pb in PM2 5 is monitored
 4                  as part of the CSN and IMPROVE networks, and Pb in PM10 as a part of the National Air
 5                  Toxics Trends (NATTS) networks (Figure 3-16 and Figure 3-17). These networks are
 6                  designed to meet different objectives than those of the Pb NAAQS monitoring network.
 7                  The purpose of the CSN is to monitor PM2 5 species to assist in understanding PM2 5
 8                  chemistry and for spatial and temporal analyses including annual, seasonal, and sub-
 9                  seasonal trends (http://www.epa. gov/ttn/amtic/specgen.html). The CSN consists of about
10                  50 long-term trends sites (commonly referred to as the Speciation Trends Network or
11                  STN sites) and about 150 supplemental sites, all operated by state and local monitoring
12                  agencies. Higher spatial and temporal resolution of the CSN facilitates increased utility in
13                  the scientific community, and the data from the CSN also assists states in formulating
14                  their emission control strategies, even if the network is not compliance-oriented. Pb is
15                  one of 33 elements in PM2 5 collected on Teflon filters every third day and analyzed by
16                  energy dispersive XRF spectrometry.
      November 2012                             3-81                  Draft - Do Not Cite or Quote

-------
 1                  In the IMPROVE networks, PM2 5 monitors, operated by the National Park Service,
 2                  largely with funding by EPA, are placed in "Class I" areas (including National Parks and
 3                  wilderness areas) and are mostly in rural locations. IMPROVE monitoring is intended to
 4                  establish current visibility conditions, track changes in visibility and determine causal
 5                  mechanisms of visibility impairment in 156 national parks and wilderness areas. There
 6                  are  110 formally designated IMPROVE sites and approximately 80 additional sites at
 7                  various urban and rural areas, informally treated as part of this network and operating
 8                  under IMPROVE protocols. At these sites, Pb in PM2 5 is determined by XRF (UC.
 9                  1995).

10                  The NATTS network is designed to monitor concentrations  of hazardous  air pollutants
11                  (HAPs). The NATTS is intended to provide model input, to  observe long-term trends in
12                  HAP concentrations, and to examine emission control strategies. The NATTS network
13                  measures several inorganic HAPs in PMi0, along with several volatile organic
14                  compounds (VOCs), carbonyls, and polycyclic aromatic hydrocarbons (PAHs). It is
15                  operated by state and local agencies and has less extensive national coverage than the
16                  other Pb monitoring networks. PMi0 is collected either by high volume sampling with a
17                  quartz fiber filter or low volume sampling with a PTFE filter following EPA
18                  Compendium Method IO-3.5 (U.S. EPA. 1999).  Pb is one of seven core inorganic HAPs
19                  collected on Teflon filters and analyzed by ICPMS. As of December 2009, the network
20                  consisted of 27 monitoring stations, including 20 urban and  7 rural stations operating on a
21                  one in six day sampling frequency.

22                  Pb monitoring is also conducted at NCore monitoring sites.  Monitoring for Pb-PM2 5 is
23                  currently being conducted at NCore sites as part of the larger CSN (described above). As
24                  described in Section 3.4.2. monitoring for Pb-PMi0 was required to be operational at
25                  NCore sites by December 27, 2011. Methods for Pb in PMi0-2.5 are being developed as
26                  part of the PM10.2.5 speciation pilot project and may be implemented at some NCore sites
27                  in the future.
      November 2012                            3-82                  Draft - Do Not Cite or Quote

-------
     Alaska
     0  250  500    1,000 Miles   0 70 140  280 Miles   0     245    490
                                                                         A CSN
                                                                         . IMPROVE
                                                                        Puerto Rico &
                                                                        Virgin Islands
                                                                               £*
                                                                        0 25 50 100 Miles
Figure 3-16     Pb-PM2.s monitoring sites for CSN and IMPROVE networks.
                                                              NATTS Monitors
                                                              Counties
                                                              United States
Figure 3-17     Pb-PMi0 monitoring sites for NATTS network.
November 2012
3-83
Draft - Do Not Cite or Quote

-------
      3.5       Ambient Air Pb Concentrations

 1                  The following section synthesizes ambient air Pb concentration data obtained during the
 2                  years 2008-2010 with data from studies in the literature presenting Pb concentrations
 3                  under varied source influences. The 3-month average ambient air Pb concentrations
 4                  presented here were created using a simplified approach of the procedures detailed in
 5                  40 CFR part 50 Appendix R and, as such, cannot be directly compared to the Pb NAAQS
 6                  for determination of compliance with the Pb NAAQS. For the purpose of analyses within
 7                  this ISA, monitors were initially designated to be source-oriented if either (1) they were
 8                  designated in AQS as source-oriented, (2) they were located within one mile of a
 9                  0.5 ton/year or greater source as noted in the 2005 NEI (U.S. EPA. 2008a). or (3) they
10                  were located within one mile of a 0.5 tons/year or greater source as noted in the 2008 NEI
11                  (U.S. EPA. 201 la). The remainder of Pb-TSP FRM monitors reporting to the AQS were
12                  classified as non-source-oriented.: For this analysis, 120 Pb-TSP FRM monitors were
13                  considered source-oriented, while 184 were considered to be non-source-oriented.
14                  However, the number of source-oriented and non-source-oriented monitors differed for
15                  each analysis year because there were changes in monitor siting. Summary information is
16                  presented within this section, and detailed data are included the Chapter 3 Appendix
17                  (Section 3.8. Supplemental Material) to this chapter.
      3.5.1        Spatial  Distribution of Air Pb
                    3.5.1.1     Variability across the U.S.

18                  This section presents nationwide Pb concentration data measured using source-oriented
19                  and non-source-oriented Pb-TSP FRM monitors from 2008-2010 and PM10 and PM25
20                  monitors for 2007-2009. The source and non-source-oriented Pb-TSP FRM monitors
21                  present data pertaining to compliance with the current level of the NAAQS. Pb-PMi0 data
22                  obtained from the NATTS network and Pb-PM2 5  data from the CSN are presented in the
23                  Chaper 3 Appendix (Section 3.8) to illustrate the nationwide distribution of Pb
24                  concentration in different classes of particle size. This information is useful to develop a
25                  sense of variability in Pb concentrations at a national scale.
      1 Following this initial classification, staff from the EPA Regional offices tasked with acting as liaisons to the states
      reviewed all monitors listed to fall within their Regions and reported any discrepancies between the initial
      classification and ground observations of the sites made by EPA Regional or state staff. The source and non-source
      monitor listing was edited accordingly. The definition of source-oriented monitoring is applied flexibly with input
      from regions in this ISA because 2008 data were obtained before the latest monitor designation requirements were
      implemented.
      November 2012                             3-84                  Draft - Do Not Cite or Quote

-------
                    Concentrations of Pb Measured using Pb-TSP Monitors (Source-Oriented
                    and Non-Source-Oriented Monitors)

 1                  Maximum 3-month average Pb concentrations1 were calculated for source-oriented
 2                  Pb-TSP monitors for 50 counties across the U.S. (1.6% of U.S. counties) during the
 3                  period 2008-2010. Figure 3-18 illustrates that the level of the NAAQS was exceeded in
 4                  twenty counties where source-oriented monitoring was performed. The mean exceeded
 5                  the level of the NAAQS and was skewed toward the 75th percentile of the distribution,
 6                  indicating that highest ambient air Pb concentrations are  near a small subset of the
 7                  monitors. Upper 90th percentile ambient air Pb concentrations for 2008-2010 occurred in
 8                  Pike Co., AL, Los Angeles Co., CA, Iron Co., MO, Jefferson Co., MO, and Collin Co.,
 9                  TX. Summary statistics for the monitor-specific one-month and three-month averages are
10                  presented in Table 3-5. and detailed statistics for the one-month and three-month
11                  averages are provided in Table 3-12. Table 3-14. Table 3-16. and Table 3-18 in the
12                  Chapter 3 Appendix (Section 3.8).
     1 Maximum 3-month average Pb concentrations are calculated as the maximum 3-month average of 3 consecutive
     monthly averages within the 2008-2010 time period.
     November 2012                             3-85                  Draft - Do Not Cite or Quote

-------
Table 3-5    Summary data for source-oriented Pb monitors across the U.S.,
              2008-2010.
                 Mean, |jg/m
Median, [jg/m3     95th%, [jg/m3   99th%, [jg/m3    Max,
Monthly
                 0.20
                                  0.063
                                                   0.86
                                                                 1.6
                                                                                4.4
3-mo rolling avg
                 0.21
                                  0.079
                                                   0.88
                                                                 1.6
                                                                                2.9
      Concentration:
      • >= 1.55 u.g/m3 (5 counties)
      • 0.76 -1.54 u.g/m3 (2 counties)
        0.16 - 0.75 u.g/m3 (13 counties)
      • 0.06-0.15 u.g/m3 (13 counties)
        <= .05 u.g/m3 (17 counties)
      U no data
Figure 3-18      Highest county-level source-oriented Pb-TSP concentrations
                  (ug/m3), maximum 3-month average, 2008-2010.

              Maximum 3-month average Pb concentrations were calculated for non-source-oriented
              Pb-TSP monitors for 47 counties across the U.S. (1.5% of U.S. counties) during the
              period 2008-2010.  Figure 3-19 illustrates that the level of the NAAQS was never
              exceeded at non-source-oriented monitors. Summary statistics are presented below in
              Table 3-6. and detailed statistics for the one-month and three-month average and maxima
November 2012
      3-86
Draft - Do Not Cite or Quote

-------
             non-source-oriented Pb-TSP concentrations are provided in Table 3-13. Table 3-15.
             Table 3-17. and Table 3-19 in the Chapter 3 Appendix (Section 3.8).
Table 3-6    Summary data for non-source-oriented Pb monitors across the U.S.,
             2008-2010.

Monthly
3-mo rolling avg
Mean, [jg/m3
0.012
0.012
Median, [jg/m3
0.010
0.010
95th%, Hg/rr
0.040
0.037
i3 99th%, (jg/rr
0.052
0.048
i3 Max, [jg/m3
0.14
0.073
     Concentration:
     • 0.06 - 0.15 u.g/m3 (4 counties)
       <= .05 u.g/m3 (43 counties)
     Q no data
Figure 3-19     Highest county-level non-source-oriented Pb-TSP concentrations
                (ug/m3), maximum 3-month average, 2008-2010.
November 2012
3-87
Draft - Do Not Cite or Quote

-------
                     3.5.1.2      Intra-urban Variability

 1                   Intra-urban variability is defined as the variation in Pb concentration across an urban
 2                   area. Because the source characteristics and size distribution of particle-bound Pb can
 3                   vary considerably in urban areas, spatial variability of Pb concentrations in urban areas
 4                   may also be high. Moreover, larger Pb-PM tends to settle quickly over short distances
 5                   after becoming airborne; short settling distances also contribute to high spatio-temporal
 6                   variability in ambient air Pb concentrations. Such variability may not be detected if one
 7                   or a small number of central site monitors is in use, so cities with multiple monitors are
 8                   used to characterize intra-urban variability.

 9                   Data for intraurban variability in six U.S. counties are presented in Section 3.8.2 of the
10                   Chapter 3 Appendix. When collectively reviewing the data from the six counties, it
11                   became apparent that spatial and temporal variability of Pb-PM concentrations were
12                   commonly high compared for example with PM2 5, which tends to have fairly
13                   homogenous concentrations over urban areas because it is subject to secondary
14                   formation. Variability was high for areas that included a Pb source, with high
15                   concentrations downwind of the sources and low concentrations at areas far from sources.
16                   When no large sources of Pb were present, variability of Pb concentrations were lower,
17                   and more  data were observed  to lie below the MDL. For example, Los Angeles County,
18                   CA data illustrated very high  concentrations adjacent to a Pb recycling facility, but
19                   non-source-oriented concentrations were well below the level of the NAAQS at all times,
20                   including  at sites near roads. As described in Section 3.3. PM size distribution influences
21                   how far the particle will travel upon initial emission or resuspension before being
22                   deposited. Meteorology, nature of the sources, distance from sources, and positioning of
23                   sources with respect to the monitors all appeared to influence the level of concentration
24                   variability across time and space.


                     Airborne Pb near Roads

25                   Five monitors, described in Table 3-7. were selected from the TSP network to examine
26                   Pb concentrations in the near  road environment. These monitors were selected because
27                   they are located in the vicinity of major roadways in urban areas with different
28                   characteristics and because they each have long-term data available. Further, based on
29                   reviews of emissions inventory information as well as satellite image searches, these  sites
30                   are not known to be near metals-related industrial sites. Time series of Pb-TSP monthly
31                   concentration for all five monitors are shown in Figure  3-20. The annual average over the
32                   two sites that were reporting data in 1980 was 0.90 (ig/m3. This Pb-TSP concentration
33                   from 1980 likely reflected the influence of Pb emissions from leaded automobile gasoline
34                   (see Figure 3-7 for annual national consumption of leaded motor vehicle gasoline). By

      November 2012                             3-88                  Draft - Do Not Cite or Quote

-------
 1                  1986, when all five monitors were reporting data, the annual average of Pb-TSP
 2                  concentration over all five monitors dropped to 0.18 (ig/m3. Over 2001-2010, the annual
 3                  average Pb-TSP concentration over all sites was 0.02 (ig/m3 with a standard deviation of
 4                  0.01 (ig/m3. The highest 2008-2010 design value was 0.04 (ig/m3, which occurred at the
 5                  Chicago site (17-031-6003) located less than 10 meters to Interstate 1-290 at a monitor
 6                  height of 2 meters AGL. The multi-site average was not substantially larger than the
 7                  maximum three-month rolling average of 0.012 (ig/m3 for non-source-oriented monitors
 8                  for the period 2008-2010, and the Pb-TSP concentration varied little over the period
 9                  2001-2010. Note that the monitor heights were 2-6 meters AGL, which may be higher
10                  than the vertical distance likely traveled by some particles (depending on particle size)
11                  following initial resuspension (see Section 3.3.1.3).
     November 2012                             3-89                   Draft - Do Not Cite or Quote

-------
Table 3-7     Sample of U.S. near road Pb TSP monitors
2008-2010
Design
County, Value
State Site ID Latitude Longitude (ug/m3)*"
Los 06-037-4002 33.82376 -118.18921 0.01
Angeles,
CA
Riverside 06-065-1003 33.94603 -117.40063 0.01
, CA
Cook, IL 17-031-0052 41.96548 -87.749928 0.02°
Cook, IL 17-031-6003 41.872202 -87.826165 0.04°
Suffolk, 25-025-0002 42.348873 -71.097163 0.02°
MA
Monitor
Height
(m AGL) Distance from Roads
6 500 m to Interstate
I-405 (San Diego
Freeway), 10 m to
Long Beach Blvd
4 Within 20 m of
intersection of
Magnolia and
Arlington Ave.
5 Near to intersection
of Interstates I-90
and I-94, 80 m to
Interstate I-90,
200 m to Interstate
I-94, 70 m to
railroad
2 Less than 10 m to
Interstate I-290
(Dwight D.
Eisenhower
Expressway)
5 95 m to Intgerstate
I-90, inside median
of Commonwealth
Ave.
Surrounding Area
High intensity
residential, urban
High intensity
residential, mixed
use urban
Located at public
utilities water
pumping station,
high density
residential urban
Parking lot of
Circuit Court of
Cook County,
% surrounded by
Concordia
Cemetary
High intensity
urban, mixed use
residential and
commercial
"The level of the 2008 NAAQS for Pb is 0.15 ug/m not to be exceeded in any 3-month period. The design value for the 2008 Pb
NAAQS is the maximum rolling 3-month Pb-TSP average within the 3-year design period.
bThe design values shown here are computed for the latest design value period using Federal Reference Method (FRM) or
equivalent data reported by States, Tribes, and local agencies to EPA's AQS as of 7/12/2011. Concentrations flagged by States,
Tribes, and local agencies as exceptional events (e.g., high winds, wildfires, volcanic eruptions, construction) and concurred by the
associated EPA Regional Office are not included in the calculation of these design values. Although the indicator for the 2008 Pb
NAAQS is Pb-TSP at "local conditions" (i.e., actual temperature and pressure; parameter 14129), 2008 Pb-TSP data reported in
"standard temperature and pressure" (i.e., 25 ° C, 760 mmHg; parameter 12128)  are also considered valid for NAAQS comparisons
and related attainment/nonattainment determinations if the sampling and analysis methods that were utilized to collect that data
were consistent with previous or newly designated FRMs or FEMs and quality assurance requirements were met.
°Fewer than 36 rolling 3-month  Pb-TSP average data are available at this site for this 3-year period; the value  shown here is the
highest valid 3-month mean.
November 2012
3-90
Draft - Do Not Cite or Quote

-------
                                                                                            	060374002
                                                                                            	060651003
                                                                                            	170310052
                                                                                            	170316003
                                                                                            	250250002
                                                                                9 •€ '5
                                                                                        9 •€ '5
      Note: Monitor \Ds from Table 3-7: Los Angeles, CA: 06-037-4002 (dark red); Riverside, CA: 06-065-1003 (dark green);
      Two from Cook, IL: [17-031-0052 (light red), and 17-031-6003 (light green)]; and Suffolk, MA: 25-025-0002 (blue).
      Figure 3-20     Time series of monthly average Pb-TSP concentration at five
                        near-road monitors.
 i
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
Airborne Pb near Airports

There have been only a few studies of air Pb concentrations near airports, but they have
generally demonstrated consistent results. Levin et al. (2008) summarized findings from
measurements at Buttonville Airport near Toronto, Canada that median air Pb-PM10
levels were not substantially higher than average reported background levels (0.01 (ig/m3
versus 0.007 (ig/m3) (Conor Pacific Environmental Technologies Inc. 2000). although the
Buttonville analysis averaged upwind and downwind data. The maximum 24-hour
concentration measured in this 10-day study was 0.13 ug/m3. The Illinois report noted
that air Pb concentrations were elevated downwind of O'Hare airport compared with
upwind levels (Illinois Environmental Protection Agency. 2002). Carr et al. (2011) also
noted elevated downwind Pb concentrations when studying Pb concentrations at the
Santa Monica Airport and surrounding neighborhood in Santa Monica, CA in 2009.  The
downwind location was higher than upwind (winter:  0.040 (ig/m3 versus 0.0075 (ig/m3;
summer: 0.049 (ig/m3 versus 0.0040 (ig/m3). Summer measurements also included a
residential neighborhood  100 meters further downwind, which were still higher than
upwind (0.033 (ig/m3 versus 0.004 (ig/m3) Modeling results from Carr et al. (2011) also
suggest that three-month average Pb concentrations above local background extended
beyond the airport property and that the preflight runup check, taxi, and takeoff emissions
were the most important contributors to Pb concentrations. This airport had a Pb
     November 2012
                            3-91
Draft - Do Not Cite or Quote

-------
 1                  emissions inventory of 0.3 tons/year (U.S. EPA. 201 la), which is below the threshold for
 2                  airports for requiring consideration of Pb NAAQS compliance monitoring (see
 3                  Section 3.4.2).
                    Airborne Pb at Urban and Rural Sites

 4                  A number of studies characterizing airborne Pb-bearing PM distribution at the
 5                  neighborhood scale suggest that spatial variability in Pb concentrations is related to local
 6                  sources. Yu et al. (2011) measured Pb-PM10 concentration using a four-channel PM
 7                  sampler (Thermo Scientific) at four roof-top sites (10-13 meters AGL) within Paterson,
 8                  NJ: (1) background, (2) near-road (within 0.8 km of two major roads), (3) industrial
 9                  (within 1 km of three metal processing facilities), and (4) commercial. Coefficient of
10                  variation (CV), defined as the standard deviation of site measurements divided by the
11                  average) was 31.3%, with concentrations ranging of 5.61 ng/m3 (near road), 6.48 ng/m3
12                  (industrial), and 6.58 ng/m3 (commercial), compared with 2.95 ng/m3 at the background
13                  site. Harrison and Yin (2010) also noted that median urban background Pb concentrations
14                  were elevated compared with rural background (urban: 13.9 ng/m3; rural: 8.0 ng/m3).
15                  Martuzevicius et al. (2004) examined the spatial variability of Pb-PM2 5 samples obtained
16                  in the greater Cincinnati,  OH area at 6 urban, 4 suburban, and 1 rural site using Harvard
17                  PM25 Impactors. They found that Pb-PM25 had a CV of 33.8%, compared with a CV for
18                  PM2 5 of 11.3% over all sites. Average Pb-PM2 5 concentration among the sites  varied
19                  from 1.79-28.4 ng/m3. Martuzevicius et al. (2004) suggested that differences between
20                  mass and Pb spatial variability implied that Pb originated primarily from local sources.
21                  Sabin et al. (2006a) measured Pb-PM with a Noll Rotary Impactor having an upper
22                  cutpoint of 29 urn and found that urban concentrations ranged from 2.2 to 7.4 ng/m3  with
23                  a CV of 40%. In contrast, a rural location had a concentration of 0.62 ng/m3. Li et al.
24                  (2009a) collected PM2 5 with a Harvard Impactor and observed that Pb concentration in
25                  PM2 5 samples was 2.2-3.0 times higher near a bus depot than next to a rural-suburban
26                  road; in this study, the authors provided ratios but not actual concentrations. Ondov et al.
27                  (2006) measured Pb-PM2 5 concentration at three Baltimore sites using an FRM. Average
28                  Pb-PM25 concentrations at the different sites were 8.3 ng/m3, 7.2 ng/m3, and 1.9 ng/m3,
29                  with the two higher concentration sites located within two miles of industrial facilities.
30                  The industrial sites include a major steel plant; several chemical manufacturing plants;
31                  and incinerators for municipal waste, medical waste, and sludge. Although these
32                  concentrations are low, they agree with the body of literature to suggest that intra-urban
33                  variability is most strongly related to source type, strength, and location.
      November 2012                             3-92                   Draft - Do Not Cite or Quote

-------
      3.5.2       Temporal Variability

 1                  The following sections present data for multi-year trends and seasonal variability of Pb
 2                  concentrations on a nationwide basis. The data presented here provide information on the
 3                  success of Pb reduction efforts over past decades as well as on areas for continued
 4                  attention with respect to Pb monitoring. The multi-year trends illustrate changes in air Pb
 5                  concentrations resulting from the phase-out of leaded gasoline for automobiles and
 6                  smaller reductions of industrial Pb usage. The seasonal variability plots demonstrate
 7                  changes in concentration within a given year, potentially related to climate or source
 8                  variation.
                    3.5.2.1      Multi-year Trends

 9                  Pb-TSP concentrations have declined substantially during the years 1980-2010. For
10                  source and non-source monitors combined in the trends network, the annual average of
11                  the maximum 3-month averages across 74 Pb-TSP monitors reporting air Pb
12                  concentrations over the period from 1980-2010 has dropped by 89% from 1.3 ug/m3 in
13                  1980 to 0.14 ug/m3 in 2010 (n = 31); see Figure 3-21. The median maximum 3-month
14                  average concentration has declined by 97% from 0.87 ug/m3 in 1980 to 0.03  ug/m3 in
15                  2010. The decline can be attributed to the phase-out of Pb antiknock agents in on-road
16                  fuel and reductions in industrial use and processing of Pb, as described in Section 3.2.1.
17                  Average concentrations in these calculations are heavily influenced by the source-
18                  oriented monitors in the network.
     November 2012                             3-93                  Draft - Do Not Cite or Quote

-------
          3,5
          2.5
        j 1.5
           1
          0.5
                                      x
                                       *
1978-2008 Level of the NAAQS
                   l\
                   I \
           1990     1992      1994     1996      1998     2000     2002     2004      200S     2008      2010
     Note: Annual average of maximum 3-month average Pb-TSP concentrations is shown by the solid black line, annual median of
     maximum 3-month average concentrations is shown by the solid blue line, and the 10th and 90th percentiles are shown by the
     dashed lines.

     Figure 3-21     National trends in Pb concentration (ug/m3), 74 trends sites,
                       1990-2010.
i
2
3
4
5
6
7
     3.5.2.2      Seasonal Variations

     This section outlines seasonal variability among Pb monitors. Seasonal variation may
     provide insight related to differential influences of sources and climate throughout a year.
     Figure 3-22 illustrates average monthly trends in Pb-PM2 5 at four IMPROVE sites: Lake
     County, CA (060333010), Bronx County, NY (360050110), Monroe County, NY
     (360551007), Chittenden, VT (500070007). In each plot, some month-to-month
     variability is evident. Seasonal trends, with peaks in median,75th, and 95th percentiles in
     the spring and fall, are apparent in the Bronx and Chittenden sites. Variability in the
     median values is less pronounced. These sites  do not illustrate national trends, but they do
     collectively suggest that there can be seasonal variation in ambient air Pb concentration
     November 2012
                                 3-94
Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
within sites consistent with findings from the 1970s and 1980s (U.S. EPA. 1986a).
National trends in monthly concentrations, provided in the Chapter 3 Appendix (in Figure
3-58. Figure 3-59. Figure 3-60. and Figure 3-61). do not illustrate variability because
inter-site variability is averaged out.
                                        nnn
                                                    ficr  MM  DK
                                                                                          s*f  (XI  MM  DM
      Notes: Data were not available for all years at all sites. IMPROVE sites were chosen where at least three years of data were
      available during 2001 -2011. Boxplots are clipped at all but the Bronx site to improve illustration of the variability among the monthly
      interquartile ranges.
      Legend: Top left panel: Lake County, CA 060333010, Top right panel: Bronx County, NY 360050110,
      Bottom left panel: Monroe County, NY 360551007, and Bottom right panel: Chittenden, VT 500070007.

      Figure 3-22     Boxplots of average monthly Pb-PM2.s concentrations  measured
                        at four IMPROVE sites, 2001-2010.
10
n
Other data regarding seasonal variability of ambient air Pb concentrations have been
limited. Laidlaw et al. (2012) also explored the seasonal variability of Pb-PM2 5 at four
cities (Birmingham, AL, Chicago, IL, Detroit, MI, and Pittsburgh, PA) using data from
the Interagency Monitoring of Protected Visual Environments (IMPROVE) network.
They observed a strong seasonal pattern with elevated Pb-PM2 5 levels in the summer
compared with the winter for all four cities. Likewise, Harrison and Yin (2010) observed
      November 2012
                            3-95
Draft - Do Not Cite or Quote

-------
 1                  that winter background concentrations of Pb were 88% and 81% of summer background
 2                  concentrations for urban and rural settings, respectively. In addition to the data presented
 3                  above, monthly average Pb concentrations averaged across sites from the TSP, NATTS,
 4                  and CSN networks are provided in the Chapter 3 Appendix (Section 3.8.3).
      3.5.3        Size Distribution of Pb-Bearing PM

 5                   The diverse nature of the main source types of ambient air Pb contributes to variations in
 6                   Pb-PM size distribution. Such variation in the size distribution, along with size-dependent
 7                   biases in Pb-TSP collection efficiency (Section 3.4.1.1). can lead to uncertainties in the
 8                   interpretation of results from Pb-PM measurements. Accordingly, depending on the
 9                   locations and magnitudes of nearby sources, ambient air Pb may be 1) mainly Pb in PMi0
10                   and PM2 5, for which good sampler performance is well established, 2) Pb-PM with a size
11                   distribution that ranges up to slightly larger than 10 (im, in which case the existing
12                   Pb-TSP FRM could potentially be subject to wind related bias, or 3) a Pb-PM size range
13                   that extends well above 10 (im, or too large to be efficiently collected even by an
14                   improved Pb-TSP method. In the latter case, air sampling is likely to be less
15                   representative of actual concentrations of Pb. The role of ambient air Pb size distribution
16                   on human exposure, along with the role of the size distribution of Pb in soil and dust, is
17                   described in Section 4.1.1.1.

18                   Because atmospheric lifetime is dependent on particle size, as described in
19                   Section 3.3.1.3 and in the U.S. EPA 2009 PM ISA (2QQ9a), TSP sampling is  likely to be
20                   representative only on a very small spatial scale. Ultra-coarse particles have a sharp
21                   concentration gradient with distance from source, because coarser particles have greater
22                   settling velocities. Hence, concentrations of particles larger than 10 (im are likely to be
23                   very spatially and temporally heterogeneous compared with finer particles (U.S. EPA.
24                   2009a; Hinds. 1999). As a consequence, in locations near sources of ultra-course
25                   particles, measurements may reflect true concentrations only in small areas in close
26                   proximity to the monitor. This issue has been thoroughly discussed in the 2006 Pb AQCD
27                   (U.S. EPA. 2006b). as well as in the 1977 Pb AQCD (U.S. EPA. 1977).

28                   Size-selective monitoring data from AQS and the literature are examined in this section.
29                   Size distribution data enhances understanding of the relationship between sources and
30                   characteristics of airborne Pb-bearing PM and hence informs monitoring strategies.
31                   Several studies in the literature since the last review have been designed to characterize
32                   the size distribution of Pb concentrations in the vicinity of sources. In the following
33                   subsections, the currently available information is presented for locations in the vicinity
      November 2012                             3-96                  Draft - Do Not Cite or Quote

-------
 1                   of industrial sources (active and closed), near roadways, and in other urban and rural
 2                   environments.
                     3.5.3.1      Co-located Monitoring Data Analysis

 3                   This section employs AQS data for Pb concentrations from co-located TSP, PMi0, and/or
 4                   PM2 5 monitors to analyze correlations and ratios of concentrations obtained from the
 5                   different monitors. These data were used because relationships among the monitors
 6                   provide information about the nature of Pb-bearing PM at different locations (e.g.,
 7                   whether the mode is in the fine or coarse fraction). Correlations indicate the extent to
 8                   which the size fractions vary together in time, and ratios signify the average proportion of
 9                   the smaller fraction to the larger fraction (e.g., the ratio of PM25 to PMi0 concentrations).

10                   Estimation of the size distribution of Pb-bearing PM is possible at a limited number of
11                   monitoring sites where monitors having different size-selective cut-points are co-located.
12                   Data for correlations between concentrations at co-located monitors having different size
13                   cuts and average ratios of concentrations from these co-located monitors are available per
14                   co-location site in Table 3-26 in the Chapter 3 Appendix (Section 3.8). and a summary of
15                   these data are provided in Table 3-8. To ensure quality of the comparisons, data from
16                   co-located monitoring pairs are presented if there were at least 30 co-located data points,
17                   the data from both monitors were above the MDL, and data from both monitors were
18                   presented at standard temperature and pressure.

19                   The collective size cut comparison data illustrate that the correlations and concentration
20                   ratios of Pb-TSP with the Pb-PMi0 and Pb-PM25 fractions are moderate, with less
21                   correspondence of Pb-PM25 with Pb-TSP compared with Pb-PMi0. The findings indicate
22                   that, on average, 81% of Pb-TSP is in the Pb-PM10 fraction, and 50% is in the Pb-PM25
23                   fraction, and 74% of the Pb-PMi0 was  in the Pb-PM2 5 fraction (assuming no bias in the
24                   Pb-PM measurements, which may not be a reasonable assumption based on Section
25                   3.4.1). However, for co-located pairs of Pb-TSP with Pb-PMi0 or Pb-PM2 5, the ranges of
26                   correlations and ratios were large, indicating  substantial spatiotemporal variability. There
27                   appeared to be little difference between urban and suburban correlations and
28                   concentration ratios. For three co-located Pb-PMi0 :Pb-TSP pairs in Wichita, KS, the
29                   average concentration ratios greater than one were observed. This suggests that some
30                   portion of the particles captured by the PMi0  sampler were not collected by the TSP
31                   sampler, perhaps as a result of instrument biases, as discussed in  Section 3.4.1. Likewise,
32                   if such a bias is consistent across monitors, it is possible that, even for average ratios less
33                   than one, the average ratios would be lower if the particles were sampled more efficiently
34                   with the TSP monitor.
      November 2012                             3-97                  Draft - Do Not Cite or Quote

-------
     Table 3-8    Summary of comparison data for co-located ambient air Pb monitors.
Monitors*

Correlation
N
Average
Standard
Deviation
Range
Average
Average Ratio
Standard
Deviation
Range
All Sites
Pb-PM10 vs. Pb-TSP
Pb-PM2.5 vs. Pb-TSP
Pb-PM2.5 vs. Pb-PM10
36
20
28
0.74
0.62
0.91
0.23
0.28
0.12
0.13-0.99
0.11-0.96
0.50-0.99
0.81
0.50
0.74
0.19
0.12
0.06
0.38-1.28
0.30-0.73
0.59-0.90
Urban and City Center
Pb-PM10 vs. Pb-TSP
Pb-PM2.5 vs. Pb-TSP
Pb-PM2.5 vs. Pb-PM10
18
12
14
0.76
0.58
0.89
0.20
0.31
0.15
0.40-0.99
0.11-0.96
0.50-0.99
0.80
0.50
0.74
0.12
0.09
0.06
0.57-0.99
0.36-0.62
0.69-0.90
Suburban
Pb-PM10 vs. Pb-TSP
Pb-PM2.5 vs. Pb-TSP
Pb-PM2.5 vs. Pb-PM10
*Note: For comparability
samples were obtained,
18
8
12
0.73
0.69
0.92
0.26
0.22
0.08
0.13-0.96
0.31-0.91
0.74-0.99
, comparisons are limited to monitors where all samples were
and both monitors reported data at standard temperature and
0.81
0.51
0.73
0.25
0.15
0.07
0.38-1.28
0.30-0.73
0.59-0.82
above the MDL, at least 30 co-located
pressure. N: sample size, PM: particulate
     matter, TSP: total suspended particulate matter.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
3.5.3.2      Studies of Pb-bearing PM Size Distribution in the
             Literature

The size distribution of Pb-bearing PM has changed over time and by site. Table 3-9 is
reproduced from Cho et al. (2011). which reviewed studies of the size distribution of
Pb-bearing PM. Studies included in Cho et al. (2011) from the late 1960s to the early
1980s reported substantially higher Pb concentrations compared with current levels.
Traffic-related emissions produced higher contributions from Pb-PM2 5 compared with
industrial emissions. More recent studies from the 1990s and 2000s illustrated variability
in the size distribution regardless of whether the source was traffic or industrial. Cho et
al. (2011) concluded that the size distribution appears to have shifted after the 1980s,
with the mode appearing to fall somewhere between 2.5 \\m and 10 pirn, compared with
previous estimations of a primary mode smaller than 2.5 nm; however, Cho et al. (2011)
maintained that additional data are needed to improve characterization of the Pb-PM size
distribution. Metadata and size distribution data from cited studies are provided in Table
3-27 and Table 3-28. respectively, of the Chapter 3 Appendix (Section 3.8).
     November 2012
                           3-98
Draft - Do Not Cite or Quote

-------
Table 3-9
Study
Lee et al.
(1968)


Lee et al.
(1972)





Dorn et al.
(1976)a'b



Alpert and
Hopke (1981)
Holsen et al.
(1993)
Sweet et al.
(1998)h
Singh et al.
(2002)
Harrison et al.
(2003)
Lough et al.
(2005)
Zereini et al.
(2005)
Goforth and
Christoforou
(2006)
Sabin et al.
(2006b)
Wang et al.
(2006d)a
Dall'Osto et al.
(2008)a'g
Summary of studies reporting Pb size distribution
reviewed literature.
Location or Pb-TSP
Site Type (M9/m3)
Urban -
Suburban
Chicago, IL
Cincinnati, OH -
Denver, CO
Philadelphia, PA
St. Louis, MO
Washington, DC
Near a Pb smelter
Year 1 .04
Winter 1 .76
Summer 0.78
Control site
Year 0.11
Winter 0.10
Summer 0.08
Urban 0.913
Urban
Rural
Lake Michigan
Lake Erie -
Lake Michigan
Lake Superior -
Traffic + Industrial
Receptor -
9 m from a
highway
Traffic Tunnel
Main street 0.0326 f
Side street 0.01 26 f
Rural 0.0116'
Rural 0.0150
10m from hwy 0.0200
Urban bkg 0.0110
Traffic + Industrial 0.0045
Near a large
steelwork site + 0.0306
major motorway
Pb-PM10
(ug/m3)
_
-
-
:
-
-

0.91
1.69
0.54
0.09
0.10
0.08
-
0.0257
0.0052
0.0112
0.0009
0.0014
0.0013
0.0069
0.0039
0.0274
_
-
-
0.0132
0.0091
0.0044
0.0290
Pb-PM2.5 Pb-PM,
(ug/m3) (ug/m3)
_ _
-
-
: :
-
-

0.47 0.27
0.84 0.46
0.32 0.18
0.06 0.04
0.07 0.04
0.04 0.03
0.720
0.0189
0.0043
0.0091
0.0070
0.0021
0.0031
0.0059 0.0051
0.0021 0.0017
-
_ _
-
0.0061
: :
0.0031 0.0017
0.0245 0.0140
Pb-PM10
Pb-TSP
_
0.93
1.00
0.99
0.98
0.98
0.99

0.88
0.96
0.69
0.83
0.93
0.94
-
-
-
_
0.98
0.85d
-
-
0.66
0.83
0.99
0.95
in the peer-
Pb-PM^
Pb-TSP
_
0.83
0.88
0.88
0.87
0.81
0.90

0.45
0.48
0.41
0.52
0.70
0.51
0.79
-
-
_
0.89
0.39d
-
0.41
:
0.69
0.80
Pb-PM,
Pb-TSP
0.75
0.65
0.59
0.72
0.70
0.70
0.62
0.74

0.26
0.26
0.23
0.32
0.36
0.34
-
-
-
0.67°
0.41°
0.80
0.20d
0.45°
0.60°
0.64°
-
:
0.37
0.46
Pb-PM^
Pb-PM,0
_
0.89
0.88
0.89
0.89
0.83
0.91

0.51
0.50
0.59
0.63
0.75
0.54
-
0.69
0.92
0.81
-
0.86
0.60
0.90
0.46d
0.1 7e
0.59
0.78
0.82
-
:
0.69
0.84
November 2012
3-99
Draft - Do Not Cite or Quote

-------
Study
Bruggemann et
al. (2009)
Zota et al.
(2009)
Makkonen et
al. (201 0)
aTSP calculated
bPM >0.43 urn.
Location or Pb-TSP
Site Type (ug/m3)
Curbside of a busy
street
Near mine waste
Traffic + Mine
waste
Upwind
Rural bkg
No wildfire -
Wildfire
as a sum of all size fractions.
Pb-PM10
(ug/m3)
0.0169
0.0114
0.0052
0.0030
0.0099
0.0153

Pb-PM25
(ug/m3)
0.0154
0.0035
0.0022
0.0019
0.0055

Pb-PM,
(ug/m3)
0.0120
-
0.0035
0.0097

Pb-PM10 Pb-PM^ Pb-PM,
Pb-TSP Pb-TSP Pb-TSP
0.71°
- - -
0.35°
0.64°

Pb-PM^
Pb-PM,0
0.86
0.31
0.42
0.63
0.55

      dEstimated from mass emissions distribution measured using MOUDIs.
      eUniversity of Wisconsin samplers.
      fPM >0.22 urn.
      90.10um 
-------
 1                  Recent studies have indicated temporal variation in the size distribution given differences
 2                  among wind direction and industrial production. Bein et al. (2006) measured the size
 3                  distribution of PM containing Pb from the Pittsburgh Supersite using rapid single particle
 4                  mass spectrometry and a MOUDI. Source apportionment illustrated that Pb was
 5                  contained in a sub-population of particles of almost every major particle-containing class
 6                  in this study, emanating from point sources including fuel combustion, steel processing,
 7                  incinerators, foundries, battery manufacturing, and glass manufacturing (Pekney et al..
 8                  2006). Bein et al.'s (2006) measurements yielded different results on different days, with
 9                  a bimodal distribution with modes around 140 nm and 750 nm during an October, 2001
10                  measurement and a single dominant mode around 800 nm during a March, 2002
11                  measurement. Differences in the size distributions could have been related to differences
12                  among wind speed, wind direction, and source contributions on the respective dates.
13                  Weitkamp et al. (2005) used a HI-VOL sampler to measure Pb-bearing PM25
14                  concentrations across the river from a coke plant in the Pittsburgh, PA area and analyzed
15                  the data with ICP-MS. Pb comprised 0.088% of the PM2 5 mass, and the mode of the size
16                  distribution (measured overall but not specifically for Pb) was observed to shift between
17                  50 nm to 1 nm. Dall'Osto et al. (2008) used a MOUDI (cut points range: 0.196-18 (im) to
18                  measure the  size distribution of Pb emissions from a steel works facility in a coastal town
19                  within the United Kingdom (U.K.). The size distribution was multimodal with a primary
20                  mode around 1 urn, a secondary mode around 300 nm, and a very small additional mode
21                  around 5 um. This multimodal distribution was thought to be associated with sintering
22                  and steel working processes, from which Pb was emitted. Pekey et al. (2010) measured
23                  PM2 5 and PMi0 concentrations in a heavily industrialized area of Kocaeli City, Turkey
24                  using a low-volume PM10 stack filter unit. They observed PM2 5/PM10 ratios of 0.60
25                  during summer and 0.45 during winter.


                    Airborne Pb Near Roadways

26                  Traffic-induced turbulence may be a cause of resuspension of Pb-bearing particles from
27                  deposited contemporaneous wheel weights,  industrial emissions, or historic sources. Pb
28                  mass in near-road PM is predominantly associated with the coarse mode (U.S. EPA.
29                  2006b). The Pb fraction in resuspended dust generally ranges from 0.002 to 0.3%, with
30                  the highest fractions observed for paved road dust and lowest for agricultural soil. Sabin
31                  et al. (2006b) compared the size distribution of coarse Pb-PM measured using a Noll
32                  Rotary Impactor at an urban background site and  at a location 10 meters from the 1-405
33                  Freeway in the southern California air basin; data from Sabin et al. (2006b) are displayed
34                  in Figure 3-23. For both the urban background and near-road sites, the largest fraction
35                  was from PM sampled below the 6 um cut point,  but the near-road Pb-PM distribution
36                  appeared bimodal with a mode in the largest size  fraction. Sabin et al. (2006b) point out
      November 2012                            3-101                  Draft - Do Not Cite or Quote

-------
 1                  that the freeway tends to be a source of very large particles that are dispersed via the
 2                  turbulent motion of the vehicular traffic. Song et al. (2011) used an eight-stage MOUDI
 3                  (cut point range: 0.18-18 (im) to measure roadside PM, 5 meters from the New Jersey
 4                  Turnpike in Carlstadt, NJ and speciated the samples. They observed a bimodal
 5                  distribution of the Pb concentration in summer and a trimodal distribution in winter. 85%
 6                  of the Pb-PM mass was measured as PM2 5 during the summer, and 68% was measured as
 7                  PM2 5 in the winter. Similarly, Zereini et al. (2005) observed that roughly 80% of particle-
 8                  bound Pb measured with a MOUDI was smaller than 5.8 (im for an urban main street,
 9                  and more than 90% were smaller than 5.8 (im for a rural area included in that study.
10                  However, in a study of automotive emissions in a traffic tunnel, Lough et al. (2005)
11                  observed that 85% of Pb measured with a MOUDI was in the PM10, with just 39% in the
12                  PM2 5 fraction and 20% in the PMi fraction. In  a near-road study conducted in Raleigh,
13                  NC with a 13-stage low-pressure impactor, Hays et al. (2011) note that the proportion of
14                  Pb  within particles in ultrafine, fine, and coarse size ranges was the same at 50 mg/kg;
15                  similar to Lough et al. (2005). mass concentrations were measured by Hays et al. (2011)
16                  to be 0.4 ± 0.4 ng/m3, 1.4 ± 0.6 ng/m3, and 0.1 ± 0.02 ng/m3 for PM10-2.5, PM25.01, and
17                  PM0.i, respectively. The Pb-PMi0 samples from Hays et al. (2011) were highly correlated
18                  with As samples (p = 0.7, p <0.0001); both Pb and As are found in wheel weights (see
19                  Section 3.2.2.6). Hays et al. (2011) did not report correlations between Pb and As for
20                  smaller size fractions, but they did state that the correlations for other size fractions were
21                  lower compared with Pb-PMi0 and As. Likewise, the Pb samples were not well correlated
22                  with crustal elements in the coarse size distribution, so it is more likely that resuspended
23                  Pb  originated from contemporary roadway sources rather than historic Pb on-road
24                  gasoline emissions. Chen et al.  (201 Ob) measured Pb in PMi0.2 5, PM2 5.01, and PM0.i
25                  using a MOUDI at a roadside location and in a tunnel in Taipei, Taiwan in 2008. While
26                  roadside and tunnel concentrations of PM10 and PM2 5 were roughly equivalent around
27                  20-30 ng/m3, Pb in PM01 was approximately 15 times higher in the tunnel (during the
28                  hours 9:00 a.m. - 9:00 p.m.) than by the roadside (tunnel: 20 ng/m3; roadside: 1 ng/m3).
29                  The authors suggest that particle-bound Pb was emitted from on-road gasoline and diesel
30                  engines. This could possibly be attributed to trace levels of Pb in diesel fuel and
31                  lubricating oil. Birmili et al. (2006) compared concentrations of Pb in PM measured with
32                  a Sierra-Anderson high volume cascade impactor at various traffic and background sites
33                  in Birmingham, U.K. Birmili et al. (2006) captured PM at the stage below a 0.5 um
34                  cutpoint and on the 1.5-3.0 um  stage for near-road, in a traffic tunnel, and remote and
35                  urban background sites. The highest concentrations were measured in the tunnel, at
36                  3.3 ng/m3 for Pb-PM05 and 10 ng/m3 for Pb-PMj 5.3 0. In contrast, urban background was
37                  more enriched in the finer size fraction, with concentrations of 5.4 ng/m3 for Pb-PM0 5
38                  and 0.84 ng/m3 for Pb-PM] 5.3 0. Remote background concentrations were on 0.16 ng/m3
39                  for Pb-PMo 5 and 0.03 ng/m3 for Pb-PMj 5.3 o Bruggemann et al. (2009) measured
      November 2012                            3-102                 Draft - Do Not Cite or Quote

-------
1
2
3
4
5
6
7
roadside distribution of Pb in PM in Dresden, Germany using a 5-stage Berner-type
low-pressure impactor to analyze the effect of season and direction of the air mass. For
all data combined as well as for data broken down by season or by wind direction, it was
found that the data followed a unimodal distribution with a peak at the 0.42-1.0 um size
bin with roadside measurements averaging 13-22 ng/m3, depending on wind direction.
Evidence of Pb in road dust related to near road ambient air Pb concentrations is
described in Section 3.6.1.
                                                                          Urban Background
                                                                          Near-Road
                    <6 u.m
                6-11 urn
11-20 urn       20-29 u.m
>29 u.m
     Source: Adapted, with permission of Elsevier Publishing, Sabin et al. (2006b).
     Figure 3-23     Comparison of urban background and near-road size fractions of
                      Pb-bearing PM.
    November 2012
                          3-103
               Draft - Do Not Cite or Quote

-------
 1                  Several studies have suggested that near-road ambient air Pb samples are derived from
 2                  sources other than from the road. Harrison et al. (2003) measured the distribution of Pb in
 3                  PM10 at a roadside sampler in Birmingham, U.K. using a MOUDI fitted with only
 4                  stages  1, 2, 4, and 8 with outpoints of 10 pirn, 2 um, 1 um, and 0.2 um. The size
 5                  distribution was unimodal with approximately 2% of the Pb mass (totaling 26.5 ng/m3)
 6                  above the 10 um cut point, 12% of the mass in the 2-10  pirn bin, 8% in the 1-2 um bin,
 7                  53% of the Pb mass in the 0.2-1 um bin, and 25% collected below the 0.2 um cut point.
 8                  Regression analysis against NOX concentration in the Harrison et al.  (2003) paper
 9                  provided a weak indication that Pb-PM0 2 was associated with NOX ((3 = 0.067, R2 = 0.38)
10                  as well as PM10 ((3 = 0.26, R2 = 0.35). Bruggemann et al. (2009) observed a unimodal Pb
11                  size distribution with 51% of the mass in the 0.42-1.2 um size bin. Observed Pb-PMi0
12                  concentration was 17 ng/m3. During winter, Pb concentrations were more than twice as
13                  high as during the summer (winter: 24 ng/m3; summer: 10 ng/m3), and they were also
14                  higher when winds blew from the east (0.42-1.2 um mode, east: 60 ng/m3; west: 25
15                  ng/m3). Bruggemann et al. (2009) suggested that this finding reflected coal burning
16                  sources dominating Pb emissions rather than road dust resuspension  during winter. Wang
17                  et al. (2006d) used a nine-stage cascade impactor (cut point range: 0.43-11 um) to
18                  measure the Pb-PM size distribution in a heavily trafficked area of Kanazawa, Japan with
19                  incineration and generation facilities nearby. They observed a bimodal distribution with
20                  modes at the 0.65-1.1 um and the 3.3-4.7 um size bins. Average concentration in the
21                  coarse mode was 2.1 ng/m3, while fine-mode average concentration was 3.7 ng/m3. Wang
22                  et al.'s (2006d) source apportionment work in this study suggested that the fine  mode
23                  derives from incineration and combustion of oil and coal.


                    Airborne Pb at Other Urban and Rural Sites

24                  Spatial and temporal concentration variability is also reflected in varying Pb-PM size
25                  distributions within and between cities. Martuzevicius et al. (2004) measured the size
26                  distribution of Pb in Cincinnati, OH at the city center site using a MOUDI and showed it
27                  to be bimodal with a primary peak at 0.56 um and a slightly smaller secondary peak at
28                  5.6 um. Using high volume samplers, Moreno et al. (2008) measured Pb concentrations
29                  in PM2 5 and PMi0 at urban, suburban, and rural sites around Mexico City, Mexico to
30                  illustrate differences among the land use categories. At the urban site, PM2 5/PMi0 ratios
31                  were 0.51 during the day and 0.57 at night (Pb-PM!0 was 59 ng/m3 and 162 ng/m3,
32                  respectively). At the suburban site, Pb-PM2 5/Pb-PMi0 ratios were 0.63 during the  day and
33                  0.81 at night (Pb-PMi0 was 24 ng/m3 and 42 ng/m3, respectively). Goforth and
34                  Christoforou (2006) measured Pb-TSP  and Pb-PM2 5 with a high volume cyclone
35                  separator in rural Georgia and observed a Pb-PM2 5 concentration of 6 ng/m3 and a
36                  Pb-TSP concentration of 15 ng/m3. Makkonen et al. (2010) measured concentrations of
      November 2012                            3-104                 Draft - Do Not Cite or Quote

-------
 1                  Pb-PMj, Pb-PM25, and Pb-PM10 during a spate of wildfires in rural southeastern Finland
 2                  with high volume size-selective samplers. They found that the ratio of Pb-PMi/Pb-PMi0
 3                  varied substantially from day to day (examples provided of 64% on 8/14/07 and 35% on
 4                  8/25/07, with Pb-PM2 5/Pb-PM10 ratio of 51% on  8/25/07), and they attributed the highest
 5                  concentrations to long-range transport of wildfire emissions via southerly winds;
 6                  variability in concentration and ratios was related to shifting wind conditions.
      3.5.4        Pb Concentrations in a Multipollutant Context

 7                  The correlations between Pb and copollutant concentrations were investigated because
 8                  correlation may indicate commonality of sources among the pollutants. For example,
 9                  correlation between Pb and SO2 may suggest common industrial sources. Correlation
10                  between Pb and NO2 or CO may suggest roadway sources, such as trace Pb in unleaded
11                  on-road gasoline or resuspension of material from pulverized wheel weights or
12                  contaminated soil. Additionally, seasonality can influence correlations, potentially from
13                  differences among sources or the contaminants' responses to climate differences.

14                  Pb concentrations exhibit varying degrees of association with other criteria pollutant
15                  concentrations. At most sites, Pb monitors are co-located with monitors for other criteria
16                  pollutants, but monitoring the full suite of criteria pollutants at a single monitoring site is
17                  rare. As a result, the number of observations for each copollutant varies. Pearson
18                  correlations of monitored non-source Pb-TSP concentrations with concentrations of other
19                  criteria pollutants are summarized in Figure 3-24 for 2008-2010 data for 46 Pb-TSP
20                  monitors at which data were above MDL and more than 30 data pairs were measured at
21                  each point. Seasonal co-pollutant measurement data from the literature are also provided
22                  in the Chapter 3 Appendix (see Figure 3-63 through Figure 3-67).
      November 2012                            3-105                 Draft - Do Not Cite or Quote

-------
CO
PM10
NO2
PM2.5
S02
03
0
o oo
o
o oo o
O O O CTEBD
VlOGm-VKXMSiOOa
OQB9GQDO O O

OCD
-------
 1                  Associations between Pb-PM25 and other species were generally low-to-moderate. The
 2                  strongest association was with zinc (Zn) (median R = 0.38). Elemental carbon (EC),
 3                  PM2 5 mass, copper (Cu), organic carbon (OC) mass, and bromine (Br) also exhibited
 4                  low-to-moderate associations with Pb-PM2 5 concentrations (median R = 0.28 to 0.33).
 5                  Such correlations may suggest some common sources affecting the pollutants, as
 6                  described in Section 3.2.2. For example, correlation with EC and OC mass may be
 7                  diagnostic indicators of some crustal, general combustion, industrial emission, and coal
 8                  combustion processes. Piston-engine aircraft emit Pb  as PbBr2 so this source may explain
 9                  the weak covariation in Pb and Br concentrations at the CSN sites. At the same time,
10                  these species must have other disparate sources that drive the Pearson correlations down.

11                  A few recent studies have used speciation techniques  to characterize Pb and other
12                  components of PMi0, PM25, and PMi. Pingitore et al.  (2009) used XAFS to speciate air
13                  samples obtained near a defunct smelter in El Paso, TX, in 1999 and 2005 and found that
14                  air Pb-TSP concentrations of 0.10 to 0.50 (ig/m3 could largely be attributed to Pb-humate.
15                  Similarly, Laidlaw et al. (2012) observed statistically  significant associations between
16                  ambient air Pb-PM2 5 and ambient air soil in Pittsburgh, PA, Chicago, IL, Detroit, MI,
17                  and Birmingham, AL (R2 = 0.31-0.49, p <0.01). Together, these results suggest a link
18                  between soil resuspension  and Pb-TSP levels

19                  Murphy et al. (2008) studied weekly patterns of metals and other aerosol components
20                  using data collected from 2000 to 2006 at IMPROVE sites. The authors concluded that
21                  Pb concentrations were impacted by piston aircraft emissions. They reached this
22                  conclusion because, in contrast to other species, Pb was elevated on weekends when there
23                  is typically a peak in general aviation flights. The authors also note that Zn and Pb were
24                  highly correlated in atmospheric samples, and they suggest that this is due to similar
25                  sources (i.e., electric utility and industrial sources). Murphy et al. (2007) also carried out
26                  a detailed study of the distribution of Pb in single atmospheric particles. During the fifth
27                  Cloud and Aerosol Characterization Experiment in the Free Troposphere (CLACE 5)
28                  campaign conducted  at the Jungfraujoch High Altitude Research Station, Switzerland,
29                  about 5% of analyzed aerosol particles in PMi contained Pb. Of these, 35% had a relative
30                  signal for Pb greater than 5% of the total mass spectrum measured by an aerosol time of
31                  flight mass spectrometer (ATOFMS). These "high Pb" particles also contained one or
32                  more positive ions (e.g., of Na+, Mg2+, A13+, K+, Fe3+,  Zn2+, Mo6+, Ag+, Ba2+). Sulfate
33                  fragments were present in 99% of the negative ion spectra associated with high Pb
34                  particles, and 50% also contained nitrite and nitrate. About 80% contained positive and/or
35                  negative polarity organic fragments. The average aerodynamic diameter of the Pb-rich
36                  particles (500 nm)  was larger than the background aerosol (350 nm) but none had a
37                  diameter less than 300 nm. Murphy et al. (2007) suggest that this mixture can be
      November 2012                            3-107                 Draft - Do Not Cite or Quote

-------
1                  attributed to combined emissions from combustion (e.g., Pb and organics) and industry

2                  (e.g., Pb sulfates).
                 EC
            Pr.l2.3.1a: :
                 Cn
              OC ma;:
               OCadl
                 Br
                  h
                 h +
                 Ft
                 Su
               HHt*
                 r.n
              NH+NOJ
            H H +_i» i
                 Ca
              c .iui lal


                 1
                 Co
                 a
                 a
                 Tl
                 LJ
                  V
                  ¥
                 Ha
                 El
                 PI.
                 r.fc
                 d:
                 Tl.
                 C.I
                 se  =
                Na*
                 Ta
                 HD
                 C;
                 an
                 D

                 Hr





















1
q
i 	 ^^^j^^ | 	 iro o



i 	 r r r- 	 -aooio o


"~ i- ^™T^™T~r _i n rvi


n i 	 t^^^3^ 	 ' ark
l 	 -i 1 i 	 CO Oi O O O O
i 	 1 ¥ 1 	 1 o o


, 	 1 I =p^~
i 	 1 i r — i o


i 	 -| • I 	 1 oo flu
i — i 1 — i
" — I I i 	 1
i 	 • 1 i I — — i
1 1 1 1
900 —
                                                  Pearson R
     Note: Correlations were calculated from available data when data were above MDL and there were at least 30 data pairs available
     for comparison; organic carbon (OC) samples were blank-adjusted.


     Figure 3-25     Pearson correlations of monitored Pb-PM2.s concentration with

                       copollutant concentrations, 2008-2010.
     November 2012
3-108
Draft - Do Not Cite or Quote

-------
      3.5.5       Background Pb Concentrations

 1                  The 2006 and 1986 Pb AQCDs evaluated evidence on Pb emissions from natural sources,
 2                  such as volcanoes, sea-salt spray, biogenic sources, wild forest fires and wind-borne soil
 3                  particles in rural areas without elevated Pb soil concentration. The 1986 Pb AQCD
 4                  concluded that the natural particulate Pb concentration was probably lower than the
 5                  concentration of 0.076 ng/m3 reported at the South Pole (U.S. EPA. 1986a). A 1980
 6                  National Academy of Sciences (NAS) report estimated that average natural background
 7                  levels of airborne Pb might range from 0.02 to 0.5 ng/m3 (NAS. 1980).

 8                  Global transport can carry airborne Pb to remote areas with no industrial activity, thus it
 9                  is difficult to estimate a natural background  concentration of Pb. Hong et al. (1994) found
10                  that Pb concentrations in Greenland ice cores remained nearly constant (at about 0.55 pg
11                  Pb/g ice) from about 7,760 years ago to about 3,000 years ago. Ratios of Pb to major
12                  crustal elements were not enriched compared with natural background levels in this
13                  section of the ice core suggesting that Pb was natural in origin, produced by rock and soil
14                  dust. At about 2,500 years ago, Pb concentrations started to increase (to about 100
15                  pg Pb/g snow averaged from 1930 to 1990) (Boutron et al.. 1991) corresponding to  an
16                  enrichment of-200 times natural background levels. McConnell and Edwards (2008)
17                  also noted elevated Pb levels in Greenland ice cores, with high correlation to black
18                  carbon (BC), cadmium (Cd), sulfur (S), and thallium (Tl) during the period 1860-1940,
19                  suggesting coal  combustion  sources from North America. Osterberg et al. (2008)
20                  observed elevated Pb levels  in a  1970-1998 ice core from Mt. Logan, Canada, indicating
21                  elevated Pb levels corresponding to increased industrial activity in Asia during this time
22                  period.

23                  Measurements of Pb from IMPROVE sites and source apportionment modeling have
24                  been used to assess the potential  input from  intercontinental transport. Liu et al. (2003)
25                  used positive matrix factorization to attribute sources of Asian dust to the measurements
26                  at two western IMPROVE sites at high elevations, Crater Lake (Oregon) and Lassen
27                  Volcanic Park (California) from  1988 to 2000. Geometric mean Pb concentrations of 0.34
28                  and 0.48 ng/m3 were found in the samples with only a few percent of these values
29                  attributable to transport from Asia. No enrichment in Pb and  other metals (As, Cr, Cu, Ni,
30                  Pb, V and Zn) above reference Asian-dust material was found. Their results suggest
31                  either that arriving air masses did not entrain contributions from Asian pollution sources
32                  or that these contributions were preferentially scrubbed out during transport.  Large
33                  enrichments in sulfur (S) were found, however, which might  have been due to pollution
34                  sources but also due to model artifacts. However, other studies have found some evidence
35                  of trans-Pacific transport. Murphy et al. (2007) measured single Pb particles off the coast
36                  of California (using a National Oceanic and  Atmospheric Administration [NOAA]
      November 2012                            3-109                  Draft - Do Not Cite or Quote

-------
 1                  aircraft elevated more than 2 km above ground level). Given the elevation of the
 2                  measurement and the timing of trans-Pacific plume events, the authors concluded that
 3                  these Pb-bearing PM2 5 originated in Asia. They also noted Pb/Zn ratios in PM2 5 at the
 4                  Mount Zirkel, CO IMPROVE site of 0.6 corresponding to measurements at Mauna Loa,
 5                  HI in spring, when measurements at other times of year produced Pb/Zn ratios of 0.3-0.4.
 6                  Ewing et al. (2010) used time series analysis of Pb isotope measurements to estimate
 7                  Asian and local contributions to Pb-PM2 5 concentrations measured at two observatories
 8                  near San Francisco, CA. They estimated a springtime contribution of Asian dust to
 9                  Pb-PM2 5 measurements. In both the Murphy et al. (2007) and Ewing et al. (2010) studies,
10                  the authors conclude that the Asian contribution is still generally less than 1 ng/m3.

11                  The use of data for PM25, PMi0.2.5, and PMi0 from monitoring sites in the East will
12                  generally result in gross overestimates of background concentrations because
13                  anthropogenic sources will cause extensive contamination. Intercontinental transport of
14                  African dust contributes to PM and is observed mainly in the Southeast but is apparent on
15                  an episodic basis elsewhere in the eastern U.S. [see e.g., 2004 PM AQCD (U.S. EPA.
16                  2004) and 2009 PM ISA (U.S. EPA. 2009a')1. Data obtained at four eastern IMPROVE
17                  sites ([1] Moosehorn National Wildlife Refuge, ME; [2] Acadia National Park, ME;
18                  [3] Swanquarter, NC; and [4] Cape Remain National Wildlife Refuge, SC) from 2007 to
19                  2009 indicate a median Pb-PM2 5 concentration of 1.0 ng/m3 with a 95th percentile value
20                  of 2.5 ng/m3. As noted above, these sites are likely to be affected by upwind
21                  anthropogenic sources within the U.S.

22                  Rough estimates  for the natural source of Pb in different size fractions of Pb-PM can be
23                  made by multiplying the abundance of Pb in soils by the crustal component of PM in the
24                  different size fractions. The mean abundance of Pb in surface rocks is ~20 mg/kg (Potts
25                  and Webb. 1992): the 2006 Pb AQCD (U.S. EPA. 2006^) reported Pb concentrations in
26                  different types of rocks to range from 3.5 to 32 mg/kg (Reuer and Weiss. 2002). There is
27                  substantial variation with location depending on composition, in particular on the
28                  abundances of uranium  (U) and thorium (Th), since Pb is produced mainly by radioactive
29                  decay of these elements. The mean Pb concentration of 863 soil samples taken across the
30                  U.S. at 2 meters depth is -16 mg/kg; this value was derived by  sampling residual Pb of
31                  the weathered rocks on which they formed [Wedepohl (1978) and references therein].

32                  Concentrations of the Pb content of soils can be used with estimates of the crustal
33                  component of PM25, PM10_25 (which is mainly crustal), PM10, and TSP produced by wind
34                  erosion of natural surfaces to estimate contributions to Pb concentrations in these size
35                  fractions. U.S. annual average PM10 concentrations in some arid counties most affected
36                  by windblown dust in the western U.S. are ~20 ug/m3. If it is assumed that these levels of
37                  PM10 are entirely due to natural wind erosion without any anthropogenic contribution and
      November 2012                            3-110                  Draft - Do Not Cite or Quote

-------
 1                  that the Pb concentration in all airborne size fractions if the same as the Pb concentration
 2                  in bulk soil or surface rock, an estimate of ~0.3 ng/m3 for the contribution of wind
 3                  erosion on natural surfaces to Pb in PM10 is obtained; however, it must be observed that
 4                  the natural contribution is probably lower than this estimate. An assumed ratio 3.5 for
 5                  TSP to PM10 in dust storms, derived by Bacon et al. (2011). indicates a contribution of
 6                  ~1 ng/m3 for Pb from natural sources in TSP. The more recent estimate indicates that
 7                  background airborne Pb concentrations are well below current ambient concentrations.
 8                  These estimates exceed estimates of natural background presented in the  1986 AQCD
 9                  (U.S. EPA. 1986a) and the National Academy of Sciences Report (NAS.  1980) by a
10                  factor of 2 to 50. Hence, a plausible range of natural background airborne Pb is 0.02 to
11                  1 ng/m3.
      3.6       Ambient Pb Concentrations in Non-Air Media and Biota

12                  There have been some major recent research efforts to characterize geographic and
13                  temporal trends in Pb concentrations across a variety of environmental media and biota.
14                  In general these concentrations reflect the decreases observed in atmospheric Pb
15                  concentrations due to reduced on-road Pb emissions.

16                  The 2006 Pb AQCD (U.S. EPA. 2006b) describes several studies showing higher Pb
17                  concentrations in plants grown in Pb contaminated soil related to mine spoils, smelting
18                  operations, sludge amendment, contaminated irrigation water, and Pb containing  agro-
19                  chemicals. In general, metal accumulation occurs more readily for Pb salts applied to
20                  soils than for the same quantity of metal in sewage sludge or fly ash. Root uptake is the
21                  dominant means of accumulation, and it is strongly influenced by pH. Root vegetables
22                  are the most strongly affected, and fruits and grains are the least susceptible. More Pb is
23                  also generally found in roots than in other parts of the plant.

24                  The 2006 Pb AQCD (U.S. EPA. 2006b) identified ingestion and water intake as major
25                  routes of Pb exposure for aquatic organisms, and it identified food, drinking water, and
26                  inhalation as major routes of exposure for livestock and terrestrial wildlife. The
27                  2006 Pb AQCD (U.S. EPA. 2006b) reports data from the U.S. Geologic Service National
28                  Water-Quality Assessment (NAWQA), which are updated every ten years. In the
29                  NAWQA survey, maxima concentrations in surface waters, sediments, and fish tissues
30                  were 30 ug/L, 12,000 mg/kg, and 23 mg/kg, respectively, compared with median values
31                  of 0.50 ug/L, 28 mg/kg, and 0.59 mg/kg. Some of the highest levels of Pb contamination
32                  occur near major  sources, like smelters, and fatal doses have been measured in tissue
33                  from sheep and horses near sources. High levels in cattle have also been observed.
34                  Wildlife in urban areas tend to  contain higher Pb concentrations than in rural areas, and
      November 2012                            3-111                 Draft - Do Not Cite or Quote

-------
 1                  higher Pb accumulations have been observed for aquatic organisms living in polluted
 2                  coastal zones than in the open sea. Ingestion of deposited Pb-PM on plant surfaces was
 3                  consistently observed to be more important than Pb accumulated from soil. Some
 4                  important variations between animals have been observed, and ruminants appear to be
 5                  less susceptible to Pb uptake than other animals. Uptake of Pb by lowest trophic levels,
 6                  including invertebrates, phytoplankton, krill, were described as the most important means
 7                  of introduction into food chains. Elevated Pb levels have been observed in aquatic
 8                  organisms that feed from sediments when the sediments contain appreciable Pb. In
 9                  shrimp, a substantial fraction of Pb can be absorbed from prey, and considerably more
10                  accumulated Pb from food has been observed to be irreversibly retained than is the case
11                  for dissolved Pb from water. These examples all illustrated that substantial Pb uptake by
12                  livestock and wildlife readily occurs in Pb contaminated environments.
      3.6.1       Soils

13                  Several studies suggest that soil can act as a reservoir for contemporary and historical Pb
14                  emissions. The importance of soil Pb to human exposure is described in Section 4.1. At
15                  the same time, soils in remote or rural areas tend to have lower Pb concentrations. The
16                  most extensive survey of background soil Pb concentration in the conterminous U.S. was
17                  conducted between 1961 and 1976 and comprised 1,319 non-urban, undisturbed sample
18                  locations, where 250 cm3 of soil was collected at a depth of 20  cm (Shacklette and
19                  Boerngen, 1984). The lower detection limit was 10 mg/kg, and 14% of the 1,319 samples
20                  were below it. The mean Pb concentration was 19.3 mg/kg, the median 15 mg/kg, and the
21                  95th percentile was 50 mg/kg. Sixteen locations had Pb concentrations between 100 and
22                  700 mg/kg. These results were in agreement with 3 previous  surveys. When creating the
23                  Ecological Soil Screening Level (Eco-SSL) Guidance and Documents, the U.S. EPA
24                  (U.S. EPA. 2007d. 2003b) augmented these data with observations from an additional 13
25                  studies conducted between 1982 and 1997, most of them limited to one state. The
26                  resulting data were summarized using state means for each of the fifty  states. Those
27                  means ranged between 5 and 38.6 mg/kg, with an overall national mean of 18.9 mg/kg.
28                  This is reasonably close to the values reported by Wedepohl  (1978) and references
29                  therein with a mean soil Pb concentration of roughly 16  mg/kg when samples were taken
30                  at 2-meter depths. Biasioli et al. (2006) contrasted urban and rural soils (tested at soil
31                  depths of 0-20 cm) of the same alluvial composition near Torino, Italy  to assess the
32                  influence of anthropogenic inputs. The urban soils had a median Pb concentration of
33                  117 mg/kg, while the median Pb concentration for rural soil was 19 mg/kg. Table 3-10
34                  presents data from seven metropolitan areas (Cobb et al.. 2006). Differences among the
35                  intraurban concentration ranges illustrate a high level of spatial variability within
      November 2012                            3-112                 Draft - Do Not Cite or Quote

-------
 1                   individual cities as well as high inter-urban variability. The rural New Orleans site
 2                   reported relatively low Pb soil concentrations, and the highest average Pb soil
 3                   concentrations were reported for the city of New Orleans.
      Table 3-10   Soil concentrations in various cities, prior to 2005.

      City                              Avg Pb Concentration (mg/kg)a      Pb Concentration Range (mg/kg)a
      Baltimore, MD                                                        1-10,900
      Miami, FL                          275                              25-1,612
      Mt. Pleasant, Ml                     320                              100-840
      New Orleans, LA                     784                              31.7-5,195
      New Orleans, LA (rural outskirts)         11                                4.8-17.3
      St. Louis, MO                        427                              35-1,860
      Syracuse, NY                        80                                20-800
      "Dry weight basis.
      Source: Reprinted with permission of the American Chemical Society, (Cobb et al.. 2006).

 4                   In North American forest soils, Pb concentrations have decreased substantially since the
 5                   phase out of leaded motor vehicle gasoline. When sampling from the O horizon (often at
 6                   0-2 cm), Evans et al. (2005) observed Pb concentrations ranging from 60 to 200 mg/kg in
 7                   Vermont, Maine, and Quebec, with lower concentrations in Quebec than in southern
 8                   Vermont in 1979, but in 1996 concentrations had decreased to between 32  and 66 mg/kg
 9                   with no spatial trend. Johnson and Richter (2010) also observed a substantial decrease in
10                   O-horizon (depth not specified) Pb concentrations in soil between 1978 and 2004 in West
11                   Virginia, Maryland, Pennsylvania, New Jersey, New York, and Connecticut,  with a
12                   median change of-65%. However, elevation also appears to be an important  factor in
13                   determining whether appreciable decreases in Pb concentration have occurred since the
14                   phase out of leaded gasoline (Kaste et al.. 2006). At sites above 800 meters in the
15                   northeastern U.S., O-horizon concentrations (depths not specified) ranged from 11 to
16                   29 kg Pb/ha, and little  change in Pb concentration was observed between 1980 and 2000.
17                   In contrast, concentrations ranged  from 10 to 20  kg Pb/ha at low elevation  sites and
18                   decreased to 2  to 10 kg Pb/ha by 2000. This difference was likely due to greater organic
19                   turnover increasing Pb mobility at the lower elevations (Kaste et al.. 2006).

20                   Soil Pb variability depends on the  strength and prevalence of nearby sources. Joshi et al.
21                   (2009) observed Pb dust concentrations to be highest at industrial sites (260 mg/kg)
22                   followed by commercial sites (120 mg/kg) and residential sites (60 mg/kg) in Singapore.
23                   Griffith et al. (2002) investigated spatial autocorrelation of soil Pb concentration at three
24                   sites: urban Syracuse, NY (0-10 cm), rural Geul River, The Netherlands (0-5  cm), and an
      November 2012                             3-113                  Draft - Do Not Cite or Quote

-------
 1                  abandoned Pb Superfund site in Murray, UT (0-5 cm). In both Syracuse and Geul River,
 2                  the soil Pb concentrations were not strongly correlated in space, with the exception of soil
 3                  obtained near roads, which exhibited less variability. The smelting and shooting areas of
 4                  the Superfund site were both demonstrated to have spatial clusters that were well
 5                  correlated. Later work on the spatial distribution of metals in Syracuse (sampling depth
 6                  not specified) produced similar results for that city (Griffith et al.. 2009). These studies
 7                  did not adjust for age of housing, although Griffith et al. (2009) did find that housing age
 8                  and Pb co-vary. An association between housing age and soil Pb would likely be
 9                  enhanced by such co-variation.

10                  Emissions trends have shown that industrial activities are now one of the largest sources
11                  of soil Pb following phase out of Pb in on-road gasoline. Pruvot et al.  (2006) compared
12                  urban and agricultural soils at depths of 0-25 cm near a closed Pb smelter with soils in
13                  similar environments not exposed to smelter emissions  in northern France. For samples
14                  near the smelter, Pruvot et al. (2006) observed that median soil Pb levels in lawns were
15                  roughly 2 times higher, while kitchen garden soil Pb concentrations were 10 times higher
16                  and agricultural soil Pb was almost 15 times higher than soil not exposed to smelter
17                  emissions. Bonnard and McKone (2009) reported surface soil Pb concentrations at depths
18                  of 0-20 cm of 66-493 mg/kg outside homes of children  living within  1 km of a Pb smelter
19                  in France; air Pb levels reported by Bonnard and McKone (2009) for this town ranged
20                  from 0.025-0.20 (ig/m3. The air samples Pingitore et al. (2009) obtained near a defunct
21                  El Paso, TX smelter (described in Section 3.5.4) found that the air Pb-TSP concentrations
22                  could largely be attributed to Pb-humate, which is created by sorption of Pb onto humic
23                  substances in  soil and can be resuspended. Spalinger et al. (2007) compared soil Pb
24                  samples at depths of 0-2.5 cm from surrounding towns with those from the Bunker Hill
25                  Superfund remediation site in Idaho. Median background soil-Pb concentration was
26                  48 mg/kg, while the median soil-Pb concentration at Bunker Hill was 245 mg/kg.

27                  Recent studies of brownfield soils have shown variable Pb concentrations. Van
28                  Herwijnen et al. (2007) measured soils at depths of 0-2  cm near a defunct Zn smelter in
29                  Avonmouth, U.K. in areas termed low and high contamination by the authors. Total soil
30                  Pb concentration in the low contamination area was 315 mg/kg, while soil Pb
31                  concentration in the high contamination area was 1,688 mg/kg. Deng and Jennings (2006)
32                  tested various Pb extraction methods on soils obtained from over 50 brownfield sites in
33                  the greater Cleveland, OH area at depths of 0-5 cm. Comparison of twelve extraction
34                  methods for three samples produced a range of 1,780-2,636 mg/kg for one sample,
35                  283-491 mg/kg  for a second sample, and 273-499 mg/kg for a third sample. Verstraete
36                  and Van Meirvenne (2008) measured Pb in soils at a remediated brownfield site at depths
37                  of 0-5 meters  in Belgium and reported average Pb concentrations to be 188  mg/kg and
38                  224 mg/kg in two sampling campaigns. Dermont et al. (2010) fractionated soil sampled at
      November 2012                            3-114                  Draft - Do Not Cite or Quote

-------
 1                   depths of 0-150 cm by particle size class and measured the Pb concentration in each. Pb
 2                   concentrations by size bin were as follows: 125-250 (im: 1,132 mg/kg; 63-125 (im:
 3                   1,786 mg/kg; 38-63 (im: 1,712 mg/kg; 20-38 (im: 2,465 mg/kg; 0-20 (im: 3,596 mg/kg.
 4                   Hence, the highest concentration was in the smallest soil particle fraction. Bulk Pb
 5                   concentration over 0-250 (im particle sizes was 2,168 mg/kg.

 6                   Several studies explore the relationship between  soil Pb concentration and land use.
 7                   Laidlaw and Filippelli (2008)  displayed data for Indianapolis, IN showing the Pb
 8                   concentration at the soil surface  (depths not specified) had a smoothed "bull's eye"
 9                   pattern. Cities generally have a similar pattern consisting of larger quantities of Pb
10                   accumulated within the inner city and smaller quantities of Pb in outer cities (i.e., near the
11                   outskirts or suburban areas) (Filippelli and Laidlaw. 2010).  Similarly, Filippelli et al.
12                   (2005) reported surface (depths not specified) soil Pb concentration distribution to have a
13                   maximum at the center of Indianapolis, IN, around the location where two interstate
14                   highways intersect, and to decrease with distance away from the center. However, the
15                   spatial distribution of Pb was presumed to reflect contributions from historic sources of
16                   on-road gasoline (Section 3.2.2.6) and Pb paint (Section 3.2.2.7). In this paper, soil Pb
17                   concentrations were also shown  to decrease with distance from roadways, but the levels
18                   were roughly four times higher in urban areas compared with suburban areas. This is also
19                   illustrated for urban scale Pb accumulation in New Orleans, LA in Figure 3-26. Brown et
20                   al. (2008) also measured soil Pb concentration along three transects of Lubbock, TX at
21                   depths of 0-2 cm and observed that soil Pb decreased with increasing distance from the
22                   city center, which was the oldest part of the city.
      November 2012                             3-115                  Draft - Do Not Cite or Quote

-------
                                                                                        Median Soil Pb (mg/kg)
                                                                                            3-99
                                                                                            100-199
                                                                                            200-299
                                                                                            300-399
                                                                                            400 - 499
                                                                                            500-599
                                                                                            600 - 699
                                                                                            700 - 799
                                                                                            800-899
                                                                                            900-999
                                                                                            1000-1768
      Note: At the urban scale, Pb quantities are largest within the inner-city residential communities that surround the Central Business
      District where pavement and concrete cover the soil. Note the several orders of magnitude difference between the interior and the
      exterior areas of the city. Note that the  number on each census tract indicates the number of blood Pb samples taken from that tract
      during the six years from which the study data were obtained.
      Source: Reprinted with permission of Elsevier Publishing, Mielke et al. (2007a)

      Figure  3-26     Map of median Pb content in soil in New Orleans.

 1                   Mielke et al. (2008) compared soil Pb concentrations at depths of 0-2.5 cm for public and
 2                   private housing at the center and outer sections of New Orleans and found that median
 3                   and maximum  soil Pb concentrations were substantially higher in the city center
 4                   compared with the outer portions of the city. This study also found that private residences
 5                   had higher soil Pb compared with public housing. In a separate study to examine surface
 6                   soil Pb loading and concentration  on 25 properties at depths of 0-2.5 cm in 25
 7                   New Orleans properties that were  previously identified as having median soil Pb
 8                   concentrations of at least 1,000 mg/kg,  Mielke et al. (2007b) reported median and
 9                   maxima soil loading rates of roughly 25,000 and 265,000 (ig/m2, respectively. Median
10                   and maxima surface soil Pb concentrations were  observed to be 1,000 and 20,000 mg/kg,
11                   respectively. Clark et al. (2006) performed isotopic analysis on urban garden soils at
12                   depths of 0-10  cm and 30-40 cm in an area of Boston, MA with no large industrial
13                   sources of Pb and estimated that 40-80% of the soil Pb  could be attributed to Pb-based
14                   paint while the remainder was attributed to historic Pb on-road gasoline emissions.
15                   Additional discussion of historic sources of Pb is provided in Section 3.2.2.7. Isotope
      November 2012
3-116
Draft - Do Not Cite or Quote

-------
 1                   ratios for paint and gasoline references used in the Clark et al. (2006) study were obtained
 2                   from Rabinowitz (1986).

 3                   Several studies have examined the effects of roadway attributes on Pb content in roadside
 4                   dust. Yesilonis et al. (2008) measured metal content in surface soil samples (0-10 cm) at
 5                   selected land parcels throughout Baltimore, MD, based on a stratified random sampling
 6                   design that accounted for land use factors. They compared soil metals within 100 meters
 7                   buffers of roadways and outside those buffers and found that median soil Pb
 8                   concentration inside the buffer was significantly higher than median soil Pb concentration
 9                   outside the buffers (outside: 38.7 kg/ha; inside: 134 kg/ha; p <0.0001). In an analysis of
10                   the relationship between land use parameters and  Pb concentration in soil in Los Angeles,
11                   CA, Wu et al. (2010) observed that soil Pb concentration at depths of 0-2.5 cm was
12                   higher near freeways and major traffic arteries compared with other locations. The
13                   (square-root transformed) age of the building on a sampled land parcel, length of
14                   highway within a 1,000-meter buffer, and length of local road within a 20-meter buffer in
15                   which the sample was obtained were significant predictors of Pb. Home age within 30
16                   meters of a soil sample and road length within 3,000 meters of a road sample were also
17                   shown to be significant predictors of soil Pb concentration in areas not designated to be
18                   near a freeway or major traffic artery. Wu et al. (2010) concluded that both historical
19                   traffic and leaded paint contributed to Pb contamination in soils. However, Wu et al.
20                   (2010) acknowledged uncertainty in historical roadway and traffic count data, which
21                   introduces uncertainty into that conclusion. Study areas were classified as residential,
22                   commercial, park,  and industrial (not specific to Pb emissions), although the authors were
23                   not able to distinguish the relative effects of each  area on Pb content in roadside dust. Wu
24                   et al. (2010) reported that the  highest median measured concentrations of Pb content in
25                   roadside dust were in residential freeway samples (112 mg/kg), followed by residential
26                   arterial samples (98 mg/kg), and industrial freeway samples (90 mg/kg). Additional
27                   sources of Pb to soil near roadways, such as traces of Pb in unleaded gasoline and
28                   Pb-containing wheel weights  (described in Section 3.2.2.6) were not considered in this
29                   study. Amato et al. (2009) observed that deposited PM onto roadways, measured as dust
30                   samples, in Barcelona, Spain  was differentially enriched with Pb compared with dust
31                   collected at a harbor area. Pb  concentration in PM10 was highest at ring roads
32                   (229 mg/kg) and in the city center (225 mg/kg), followed by demolition and construction
33                   sites (177 mg/kg) and near a harbor (100 mg/kg).  Roadside dust Pb concentration was
34                   also found to vary  with roadway activity by Preciado and Li (2006);  average Pb dust
35                   concentrations at a busy road  were 90 mg/kg, compared with 56 mg/kg at a less busy
36                   road. Preciado and Li (2006)  also examined soil Pb depth to ascertain availability of soil
37                   Pb for exposure. They observed peak soil Pb  concentrations of 250-800 mg/kg at depths
38                   of 0.12-0.23 meters, depending on the soil measurement location and roadway traffic.
      November 2012                             3-117                  Draft - Do Not Cite or Quote

-------
 1
 2

 3
 4
 5
 6
 7
This finding may suggest that over time, historic emissions of Pb deposited to soil are
being covered by fresh soil and hence moving further down within the soil horizons.

Size distributions of Pb-containing dust settled on the ground contain larger particles than
the size distribution of ambient air Pb, described in Section 3.5.3. Preciado and Li (2006)
measured the size distribution of Pb-containing dust near the roadside, as seen in Figure
3-27. For the busy highway, the mass median dust diameter estimated from the figure
ranged from 34- 42  pirn, depending on distance from the road. For the lower traffic
highway, the mass median dust diameter estimated from the figure ranged from
64-99 urn.
JD
"CO
E
U)
73
ca
0.
-^0

iuu-
80-

60-
40-

20-
0^
                                    Particle Size for HWY 1
                                    dustfall
                          Particle Size for H_WY
                          17 dustfall
                                           10                 100
                                           Particle Size (urn)
                                                               1000
     Source: Reprinted with permission of Springer, Preciado and Li (2006)

     Figure 3-27     Size distribution of Pb-containing dust collected near busy
                       (HWY 1) and low traffic (HWY 17) highways.

10                 Two recent studies focused on Pb from paint degradation by examining Pb dust loading
11                 to hard surfaces located along transects of each of the five boroughs of New York City
12                 (Caravanos et al., 2006a; Weiss et al., 2006). Caravanos et al. (2006a) used GIS to
13                 examined Pb dust loadings on top of pedestrian traffic signals and observed "hot spots,"
14                 defined by the authors as at least twice the Pb dust loading at adjacent samples near major
15                 elevated bridges in upper Manhattan, the Bronx, and Queens. In Brooklyn and Staten
16                 Island, areas with high dust loading were not clearly attributed to a source. "Low spots,"
17                 defined by the authors as at least two times lower Pb dust loading compared with
     November 2012
                          3-118
Draft - Do Not Cite or Quote

-------
 1                  adjacent samples were observed in lower Manhattan, were thought to correspond with
 2                  intensive cleaning efforts that followed the September 11, 2001 World Trade Center
 3                  attack. Weiss et al. (2006) studied Pb concentrations of grit (granules of mixed
 4                  composition found to accumulate alongside street curbs) along the transects and found
 5                  that median Pb concentrations in grit under the elevated steel structures were 2.5-11.5
 6                  times higher than those obtained away from steel structures; 90th percentile values were
 7                  up to 30 times higher near steel structures compared with those further from these
 8                  structures.

 9                  Outdoor Pb dust has been also associated with demolition activities. Farfel et al. (2005.
10                  2003) measured  Pb dust within 100 meters of a demolition site before, immediately after,
11                  and 1 month following the demolition. They found that the rate of Pb dust fall increased
12                  by a factor of more than 40 during demolition (Farfel et al.. 2003). Immediately after
13                  demolition, one demolition site had dust loadings increase by a factor of 200% for streets
14                  (87,000 (ig/m2),  138% for alleys (65,000 (ig/m2), and 26% for sidewalks (23,000 (ig/m2)
15                  compared with pre-demolition Pb dust levels. One month following demolition, Pb dust
16                  levels dropped by a factor of 45% for the street (48,000 (ig/m2), compared with post-
17                  demolition concentrations, 67% for alleys (21,000 (ig/m2), and 41% for sidewalks
18                  (14,000 (ig/m2). At another demolition site, smaller increases were observed: 29% for
19                  streets (29,000 (ig/m2), 18% for alleys (19,000 (ig/m2) and 18% for sidewalks
20                  (22,000 (ig/m2). No values were reported for the 1-month follow-up for the second site
21                  (Farfel et al.. 2005).

22                  Pb can be elevated in soils located where ammunition is used for military or hunting
23                  purposes. In a study of Pb content in sand used to cover a firing range, Lewis et al. (2010)
24                  found that 93% of bullet mass was recovered in the top 0.3 meters of the sand,  and 6.4%
25                  was recovered at a depth of 0.3-0.45 meter. Pb oxides were observed to be the dominant
26                  species in the contaminated sand. Berthelot et al. (2008) studied soil Pb concentrations in
27                  grounds (0-15 cm) used for testing military tanks and munitions and measured  soil Pb
28                  levels to range from 250 to 2,000 mg/kg dry basis.
      3.6.2        Sediments

29                  The recently completed Western Airborne Contaminants Assessment Project (WACAP)
30                  is the most comprehensive database, to date, on contaminant transport and depositional
31                  effects on sensitive ecosystems in the U.S. (Landers et al.. 2010). The transport, fate, and
32                  ecological impacts of semi-volatile compounds and metals from atmospheric sources
33                  were assessed on ecosystem components collected from 2002-2007 in watersheds of
34                  eight core national parks (Landers et al.. 2008). The goals of the study were to assess
      November 2012                            3-119                  Draft - Do Not Cite or Quote

-------
 1                  where these contaminants were accumulating in remote ecosystems in the western U.S.,
 2                  identify ecological receptors for the pollutants, and to determine the source of the air
 3                  masses most likely to have transported the contaminants to the parks. Pb was measured in
 4                  sediments, as well as snow, water, lichen, fish, and moose during the multiyear project,
 5                  and although Pb was not measured in air as a part of this study, routine monitoring find
 6                  particle Pb was monitored at IMPROVE sites in the majority of national parks included
 7                  in the study.

 8                  Pb concentrations in sediments from all  lakes in which Pb was measured in the
 9                  conterminous 48 states exhibited higher Pb concentrations near the surface relative to
10                  pre-industrial Pb levels measured at greater depth. This was not the case for other metals
11                  measured, except for cadmium (Cd) and mercury (Hg). Sediments in most lakes exhibited
12                  maximum concentrations between 1960 and 1980, followed by a decrease, as shown in
13                  Figure 3-28. A clear decline in Pb concentrations in sediments after the  discontinued use
14                  of leaded on-road gasoline was observed at almost all WACAP locations, of nearly all
15                  WACAP sites in the western U.S. Sediment Pb concentrations averaged over the year in
16                  which they were obtained correlated moderately well with annual average Pb-TSP
17                  concentrations from the AQS with R = 0.63 for 1980-2004, in which WACAP data were
18                  available (NPS. 2011). Pb concentrations in sediments were much lower in Alaska, and
19                  no such decline was observed. Pb in sediments was mainly attributed to on-road gasoline
20                  use, but for some lakes a strong influence from other local sources of Pb to lake
21                  sediments was shown to be important, including Pb mining, smelting, logging, and other
22                  industrial activities. The reduction in sediment Pb concentrations shown in Figure 3-28
23                  for recent years coincides with declines  in air Pb concentrations following the phase-out
24                  of Pb anti-knock agents in gasoline and reductions of air Pb emissions from  industrial
25                  activities. Elevated Pb deposition at the Glacier, Rocky Mountain, and Sequoia and Kings
26                  Canyon National Park and Preserve sites was thought by Landers et al. (2008) to reflect
27                  regional scale bioaccumulation of airborne contaminants in remote ecosystems in the
28                  western U.S.  Accumulation of contaminants was shown to vary geographically; Landers
29                  et al. (2008) lists potentially influential factors causing variation in Pb deposition
30                  including proximity to individual sources or source areas, primarily agriculture, mining,
31                  and smelting  operations. This finding was counter to the original working hypothesis that
32                  most of the contaminants found in western parks would originate from eastern Europe
33                  and Asia.
      November 2012                            3-120                 Draft - Do Not Cite or Quote

-------
                      CD
                      CD
                      a
                      CO
                  g
                  ^—•
                  (13
                  i=
                  OJ
                  o
                  O
                  "cr
                  a:
                  ^
                  (Li
                  CO
                      o
                      CN
Park
* DENA
* GAAR
• GLAC
* MORA
NO AT
n OLYM
ROMO
SEKI



• I
A
i
n D
1
1800










v

4 n
18










*'-\'
A A A A
A^-1 ^*
%* r* G] U CH n
*D* B ^
* .
1 	
1—
50 19
D;






• •
,,
•
i« 
-------
 1                  trends, Balogh et al. (2010) observed that the mean Pb concentration peaked in the 1970s
 2                  then declined, with levels from the 1990s below 1930s levels.

 3                  Data from select regions of the U.S. illustrate that Pb concentrations in surface waters and
 4                  sediment are likely to be higher in urbanized areas compared with rural locations. Figure
 5                  3-29 illustrates such variability within a single watershed for the Apalachicola,
 6                  Chattahoochee, and Flint River Basin, which runs south from north of the greater Atlanta,
 7                  GA metropolitan area and drains into the Gulf of Mexico at the Apalachicola Bay in the
 8                  Florida panhandle. Sediment concentrations peaked near the Atlanta area and diminished
 9                  as distance from the Apalachicola Bay decreased. This observation suggests that rural
10                  areas have lower Pb sediment levels compared with urban areas. Consistent with the
11                  WACAP trends shown in Figure 3-28. the data also illustrated that Pb concentrations in
12                  sediment have declined in the U.S. since 1975 (Figure 3-30). Note that Figure 3-30 does
13                  not include data near Atlanta, so the urban peak cannot be seen here as in Figure 3-29.
      November 2012                             3-122                  Draft - Do Not Cite or Quote

-------
                 100


                 90


                 80


              •3  70


              j?  60

              1

              *  50


              I  40


              £  30


                 20


                 10


                  0

                    0
                            100      200       300      400      500

                                           River km above Apalachteola Bay, FL
                        Downstream
                                                                        600
                                                                                 700
                                                                                         soo
                          —*—Pb in streambed-sediment and reservoir-core samples

                           X Pb background in streambed-sediment and baseline reservoir-core samples
Note: The background refers to concentrations from undeveloped geographic regions and baseline samples are obtained from the
bottom of the sediment core to minimize anthropogenic effects on the sample. Pb concentrations reported on a dry basis.
The lakes and reservoirs along the Apalachicola, Chattahoochee, and Flint River Basin (ACF) feed from north of the Atlanta, GA
metropolitan area into the Gulf of Mexico at Apalachicola Bay in the Florida panhandle.
Source: Reprinted with permission of the American Chemical Society, Callenderand Rice (2000).


Figure 3-29      Sediment core data (1992-1994) for the lakes and reservoirs along

                    the Apalachicola, Chattahoochee, and Flint River Basin (ACF).
November 2012
3-123
Draft - Do Not Cite or Quote

-------
                  160
                  140
                  120
                I
                "w 10°
                ¥
                £  so
                I
                3  60
                fi
                   40
                   20
                              —•—1975-1980
                              -••--1980-1985
                              -A-1985-1990
                              —X- 1990-1995
                                                                       Downstream
                              100       200       300      400      500
                                             River km above Apalachicola Bay, FL
                                                                          600
                                                                                   700
                                                                                            800
      Note: The background refers to concentrations from undeveloped geographic regions and baseline samples are obtained from the
      bottom of the sediment core to minimize anthropogenic effects on the sample. Pb concentrations reported on a dry basis. Sediment
      samples were not obtained for various time periods in Atlanta, so the graph does not indicate a lack of elevated sediment Pb in
      Atlanta.
      Lakes and reservoirs along the Apalachicola, Chattahoochee, and Flint River Basin (ACF) feed from north of the Atlanta, GA
      metropolitan area into the Gulf of Mexico at Apalachicola Bay in the Florida panhandle.
      Source: Reprinted with permission of the American Chemical Society, Callenderand Rice (2000).

      Figure 3-30     Sediment core data (1975-1995) for the lakes and  reservoirs along
                        the Apalachicola,  Chattahoochee, and Flint River  Basin (ACF).

 1                   Many recent studies have illustrated the effects of natural disasters on Pb concentrations
 2                   in surface water and sediment in the wake of Hurricane Katrina, which made landfall on
 3                   August 29,  2005 in New Orleans, LA, and Hurricane Rita, which made landfall west of
 4                   New Orleans on September 23, 2005. Pardue et al. (2005) sampled floodwaters on
 5                   September 3 and September 7, 2005 following the hurricanes and observed that elevated
 6                   concentrations of Pb along with other trace elements and contaminants were not irregular
 7                   for stormwater but were important because human exposure to the stormwater was more
 8                   substantial for Hurricane Katrina than for a typical storm. Floodwater samples obtained
 9                   throughout  the city on September 18, 2005 and analyzed for Pb by Presley et al. (2006)
10                   were below the limit of detection (0.04 (ig/mL). Likewise, Hou et al. (2006) measured
11                   trace metal  concentration in the water column of Lake Pontchartrain and at various
12                   locations within New Orleans during the period September 19 through October 9, 2005
13                   and found that almost all Pb concentrations were below the limit of detection
      November 2012
3-124
Draft - Do Not Cite or Quote

-------
 1                  (0.0020 mg/kg). However, several studies noted no appreciable increase in Pb
 2                  concentration within Lake Pontchartrain soils and sediments (Abel et al.. 2010; Abel et
 3                  al.. 2007; Schwab et al.. 2007; Cobb et al.. 2006; Presley et al.. 2006). Shi et al. (2010)
 4                  analyzed Lake Pontchartrain sediment samples using a factored approach and found that
 5                  most Pb was sequestered in carbonate-rich, iron oxide-rich, and magnesium oxide-rich
 6                  sediments in which it can be more readily mobilized and potentially more bioaccessible.
 7                  Zahran et al. (2010) and Presley et al. (2010) noted that soil Pb samples obtained outside
 8                  schools also tended to decrease in the wake of Hurricanes Katrina and Rita, with some
 9                  sites observing substantial increases and others noting dramatic reductions. These studies
10                  suggest that floodwaters can change the spatial distribution of Pb in soil and sediments to
11                  result in increased or reduced concentrations.
      3.6.3        Rain

12                  There are currently no routine measurements of Pb in precipitation in the U.S. Recent
13                  results from locations outside the U.S. were consistent with decreasing rain water
14                  concentrations described in the 2006 Pb AQCD, reflecting the elimination of Pb from
15                  on-road gasoline in most countries. From the 2006 Pb AQCD (U.S. EPA. 2006b). volume
16                  weighted Pb concentrations in precipitation collected in 1993-94 from Lake Superior,
17                  Lake Michigan and Lake Erie ranged from -0.7 to ~1.1 (ig/L (Sweet etal.. 1998). These
18                  values fit well with the temporal trend reported in Watmough and Dillon (2007). who
19                  calculated annual volume-weighted Pb concentrations to be 2.12, 1.17 and 0.58 (ig/L for
20                  1989-1990,  1990-1991 and 2002-2003, respectively, in precipitation from  a central
21                  Ontario, Canada, forested watershed. A similar value of 0.41  (ig/L for 2002-03 for Plastic
22                  Lake, Ontario, was reported in Landre et al. (2009). For the nearby Kawagama Lake,
23                  Shotyk and Krachler (2010) gave Pb concentrations in unfiltered rainwater collected in
24                  2008. For August and September 2008, the values were 0.45 and 0.22 (ig/L, respectively,
25                  and so there had been little discernible change  over the post-2000 period. In support, Pb
26                  concentrations in  snow pit samples collected in 2005 and 2009 collected 45 km northeast
27                  of Kawagama Lake had not changed to any noticeable extent(0.13, 0.17, and 0.28 (ig/L
28                  in 2005; 0.15 and 0.26 (ig/L in 2009) (Shotvk and Krachler. 2010).

29                  There have also been a few recently published, long-term European studies of Pb
30                  concentration in precipitation including Berg et al. (2008) and Farmer et al. (2010).  Berg
31                  et al. (2008) compared the trends in Pb concentration in precipitation at Norwegian
32                  background sites in relation to the decreasing European emissions of Pb over the period
33                  1980-2005. The Birkenes site at the southern tip of Norway is most affected by long-
34                  range transport of Pb from mainland Europe but there had been a 97% reduction in the
35                  concentration of Pb in precipitation over the 26-year time period. This was similar to the


      November 2012                            3-125                  Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
reductions of 95% and 92% found for the more northerly sites, Karvatn and
Jergul/Karasjok, respectively (Figure 3-31). A decline of-95% in Pb concentrations in
moss (often used as a biomonitor of Pb pollution) from the southernmost part of Norway,
collected every 5 years over the period 1977-2005, agreed well with the  Birkenes
precipitation results (Berg et al., 2008). The reductions in Pb concentration in both
precipitation and moss appear to agree well with the reductions in emissions in Europe
(-85%) and Norway (-99%). Similar to the situation in the U.S., the greatest reductions
occurred prior to the late 1990s, and relatively minor reductions have occurred thereafter;
see Figure 3-31.
                    G)
                    -0
                    Q-
                                                                   Birkenes
                                                                   Karvatn
                                                                   Jergul/Karasjok
                          1980
                 1985
1990
1995
2000
2005
      Source: Reprinted with permission of Pergamon Press, Berg et al. (2008)
      Figure 3-31      Trends in Pb concentration in precipitation from various sites in
                        Norway over the period 1980-2005.

10                  Farmer et al. (2010) showed the trends in concentration of Pb in precipitation collected in
11                  a remote part of northeastern Scotland over the period 1989-2007. The 2.6- and 3.0-fold
12                  decline in mean concentration from 4.92 (ig/L (1989-1991) to  1.88 (ig/L (1999) and then
13                  to 0.63 (ig/L (2006-2007) is qualitatively but not quantitatively in line with the sixfold
14                  decline in annual total U.K. emissions of Pb to the atmosphere over each of these time
15                  periods. After leaded on-road gasoline was banned in the U.K. in 2000, the ratio of
16                  rainwater Pb concentrations to Pb emissions (metric tons) appears to have stabilized to a
17                  near-constant value of 0.009 (ig/L per metric ton. The concentrations in precipitation
18                  reported in these studies are all at the lower end of the range reported in the
     November 2012
                           3-126
                    Draft - Do Not Cite or Quote

-------
 1                  2006 Pb AQCD (U.S. EPA. 2006b). and similar to concentrations reported for those
 2                  studies conducted after the removal of Pb from on-road gasoline. Overall, recent studies
 3                  of wet deposition tended to confirm the conclusions of the 2006 Pb AQCD (U.S. EPA,
 4                  2006b) that wet deposition fluxes have greatly decreased since the removal of Pb from
 5                  on-road gasoline.
      3.6.4       Snowpack

 6                  The location of Pb deposition impacts its further environmental transport. For example,
 7                  Pb deposited to some types of soil may be relatively immobile, while Pb deposited to
 8                  snow is likely to undergo further transport more easily when snow melts. Deposition to
 9                  snow was investigated in several studies. Measurements of Pb in snowmelt during the
10                  WACAP study, showed that median Pb concentration ranged form 20-60 ng/L, with 95th
11                  percentile values ranging from 30-130 ng/L; see Figure 3-32 (NPS. 2011). Measurements
12                  in WACAP of Hg and  particulate carbon deposition onto snow were thought to reflect
13                  coal combustion, and Pb was not significantly correlated with Hg in terms of either
14                  concentration or of calculated enrichment factors normalized to Al concentrations.
15                  Shotyk and Krachler (2009) reported considerably higher concentrations at two North
16                  American sites, Johnson and Parnell, in Ontario, Canada. Mean Pb concentration for
17                  contemporary snow was 672 (Johnson, n = 6; Parnell, n = 3) ng/L. Shotyk et al. (2010)
18                  presented additional values for Pb in contemporary snow samples in Simcoe County,
19                  Ontario, and these were higher than for ground and surface waters. Luther Bog and Sifton
20                  Bog snow had mean Pb concentrations of 747 and 798 ng/L, respectively. The relatively
21                  high concentrations in  snow were attributed to contamination with predominantly
22                  anthropogenic Pb, although it was noted that the extent of contamination was
23                  considerably lower than in past decades.

24                  Seasonal patterns of heavy metal deposition to snow on Lambert Glacier basin, east
25                  Antarctica, were determined by Hur et al. (2007). The snow pit samples covered the
26                  period from austral spring 1998 to summer 2002 and Pb concentrations ranged from
27                  1.29-9.6 pg/g with a mean value of 4.0 pg/g. This was similar to a mean value of 4.7 pg/g
28                  (1965-1986) obtained by Planchon et al. (2003) for Coats Land, northwest Antarctica.
29                  Estimated contributions to the Pb in Lambert Glacier basin snow were ~1 % from rock
30                  and soil dust (based on Al concentrations) and -4.6% from volcanoes (based on the
31                  concentrations of nss-sulfate). There was almost negligible contribution from seaspray
32                  (based on Na concentrations), and so it was suggested that a substantial part of the
33                  measured Pb concentration must originate from anthropogenic sources, ffighest Pb
34                  concentrations were  generally observed in spring/summer with an occasional peak in
35                  winter. This contrasts with data for the Antarctic Peninsula, where highest concentrations


      November 2012                             3-127                  Draft - Do Not Cite or Quote

-------
1                  occurred during autumn/winter, and again with Coats Land, where high concentrations
2                  were observed throughout the winter. These differences were attributed to spatial changes
3                  in input mechanism of Pb aerosols arriving at different sites over Antarctica, which could
4                  be due to their different source areas and transport pathways. Hur et al. (2007). however,
5                  suggested that the good correlation between Pb and crustal metals in snow samples shows
6                  that Pb pollutants and crustal PM are transported and deposited in Lambert Glacier basin
7                  snow in a similar manner.
                     o
                     "
                     o
                     r-j
                     d
                     O
                     r-j
                 o
                 "(5
                 1=
                 111
                 o
                 O
                 o
                 d
                 CO
                     o
                     d
                     P
                     d
                     g
                     d
                             I      I       I      I      I     I      I      I       I
                           DENA  NOAT  MORA NOCA OLYM  SEKI  GAAR ROMO  GLAC
     Source: WACAP Database (NFS. 2011)
     (DENA = Denali, GAAR = Gates of the Arctic, GLAC = Glacier, MORA = Mount Ranier, NOCA = North Cascades, NOAT = Noatak,
     OLYM = Olympic, ROMO = Rocky Mountain, SEKI = Sequoia and Kings Canyon)

     Figure 3-32     Box plots illustrating Pb concentration in snow melt at nine
                      National Parks and Preserves.

                   Lee et al. (2008b) collected 42 snow samples during the period autumn 2004-summer
                   2005 from a 2.1-meter snow pit at a high-altitude site on the northeast slope of Mount
     November 2012
3-128
Draft - Do Not Cite or Quote

-------
 1                  Everest, Himalayas. Pb concentrations ranged from 5-530 pg/g with a mean value of
 2                  77 pg/g. The mean value is clearly higher than the Hur et al. (2007) value for Antarctica
 3                  but is substantially lower than a mean concentration of 573 pg/g for snow from Mont
 4                  Blanc, France [ 1990-1991; Lee et al. (2008b)]. The mean Pb concentration for Mount
 5                  Everest snow was lower during the monsoon (28 pg/g) compared with the non-monsoon
 6                  periods (137 pg/g). From calculated enrichment factors (Pb/Alsnow:Pb/Alcmst),
 7                  anthropogenic inputs of Pb were partly important but soil and rock dust also contributed.
 8                  The low Pb concentrations during monsoon periods are thought to be attributable to low
 9                  levels of atmospheric loadings of crustal dusts. Lee et al. (2008b) noted that their
10                  conclusions differ from those in Kang et al. (2007). who stated that anthropogenic
11                  contributions of Pb to Mount Everest snow were negligible because the Everest
12                  concentrations were similar to those in Antarctica. Kang et al.  (2007) did not take account
13                  of the difference in accumulation rates at the two sites and had also used Pb
14                  concentrations for Antarctic snow from a study by Ikegawa et  al. (1999). Lee et al.
15                  (2008b) suggested that these Pb concentrations were much higher than expected and that
16                  their snow samples may have suffered  from contamination during sampling and analysis.
      3.6.5       Natural Waters

17                  Monitoring data for streams, rivers, and lakes are summarized in periodic national
18                  assessments of surface waters that are carried out periodically by EPA, and they include
19                  measurement of major biological and chemical stressors. Human exposure to Pb in
20                  drinking water is described in Section 4.1.3.3. Pb concentrations in natural waters also
21                  may reflect deposition of Pb even in remote locations. WACAP data at five National
22                  Parks and Preserves show median Pb concentrations in surface waters to range from 6 to
23                  75 ng/L (NPS. 2011); see Figure 3-33. Four sites (Denali, Mt. Ranier, Glacier, and
24                  Olympic National Parks) were in the lower range of 6 to 20 ng/L. One site (Noatak)
25                  reported a single value of 75 ng/L. With the exception of the Noatak site, the WACAP
26                  values were in line with measurements by Shotyk and Krachler (2007) of Pb
27                  concentrations in six artesian flows in Simcoe County, near Elmvale, Ontario, Canada.
28                  The values  ranged from 0.9 to  18 ng/L with a median (n = 18) of 5.1 ng/L. These are
29                  comparable with reports of a range of 0.3-8 ng/L for Lake Superior water samples (Field
30                  and Sherrell. 2003). Shotyk and Krachler (2007) also commented that such low
31                  concentrations for ground and surface waters are not significantly different from those
32                  (5.1 ± 1.4 ng/L) reported for Arctic ice from Devon Island, Canada, dating from
33                  4,000-6,000 years ago. In a separate study, Shotyk and Krachler (2009) reported
34                  concentrations of Pb in groundwater (from two locations,  Johnson and Parnell), surface
35                  water (Kawagama Lake [Ontario, Canada]) and contemporary snow (Johnson and
      November 2012                             3-129                  Draft - Do Not Cite or Quote

-------
 1                  Parnell, as described in Section 3.6.4). The lowest mean dissolved Pb concentrations
 2                  were found for groundwater: 5.9 (Johnson, n = 11) and 3.4 (Parnell, n = 12) ng/L. For
 3                  lake water the mean Pb concentration was 57 (Kawagama Lake, n = 12) ng/L. The
 4                  extremely low concentrations of Pb in the groundwaters were attributed to natural
 5                  removal processes. Specifically, at the sampling location in Canada, there is an
 6                  abundance of clay minerals with high surface area and high cation exchange capacity and
 7                  these, combined with the elevated pH values (pH=8.0) resulting from flow through a
 8                  terrain rich in limestone and dolostone, provide optimal circumstances for the removal of
 9                  trace elements such as Pb from groundwater. Although such removal mechanisms have
10                  not been demonstrated, the vast difference between Pb concentration in snow and that in
11                  the groundwaters, indicate that the removal process is very effective. Shotyk and
12                  Krachler (2010) speculate that even at these very low Pb  concentrations, much if not
13                  most of the Pb is likely to be colloidal, as suggested by the 2006 Pb AQCD (U.S. EPA.
14                  2006b). Finally, Shotyk et al.  (2010) suggest that the pristine groundwaters from Simcoe
15                  County, Canada, provide a useful reference level against which other water samples can
16                  be compared.
     November 2012                            3-130                 Draft - Do Not Cite or Quote

-------
                 1=
                 o
                 o
                 O
                     i--
                     CD
                     to
                     CD
                     CD
                     CD
                     Q
                     CD
  CO
  CD
  CD
                     CD
                     CD
                     CD
                     CD
                              DENA
                     MORA
GU\C
OLYM
NO AT
     Source: WACAP Database (NFS. 2011)
     Note: (DENA = Denali, GLAC = Glacier, MORA = Mount Ranier, NOAT = Noatak, OLYM = Olympic)

     Figure 3-33     Boxplots of Pb concentration in surface waters measured at five
                       National Parks and Preserves.
 i
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11
Although Pb concentrations in Kawagama Lake (Ontario, Canada) water were
approaching "natural values," the 206Pb/207Pb ratios for the samples that had the lowest
dissolved Pb concentrations of 10, 10 and 6 ng/L were 1.16, 1.15 and 1.16, respectively.
These values are inconsistent with those expected for natural Pb (the clay fraction from
the lake sediments dating from the pre-industrial period had values of 1.19-1.21) and it
was concluded that most of the dissolved Pb in the lake water was of industrial origin.
Shotyk and Krachler (2010) found that the full range of isotope ratios for Kawagama
Lake water samples (Ontario, Canada) was 1.09 to 1.15; this was not only much lower
than the stream water values entering the lake but also lower than the values attributed to
leaded on-road gasoline in Canada (-1.15). The streamwater ratio values were ~1.16 to
1.17, while those for rainwater were as low as 1.09; in good agreement with the lower
     November 2012
                          3-131
                Draft - Do Not Cite or Quote

-------
 1                  lake water values. This means that there must be an additional atmospheric source of Pb,
 2                  which has a lower 206Pb/207Pb ratio than leaded on-road gasoline. Supporting evidence
 3                  came from contemporary samples such as near surface peat, rainwater and snow, all of
 4                  which confirmed a trend away from natural Pb (1.191 to 1.201) to lower 206Pb/207Pb
 5                  ratios. The local smelting activities (Sudbury) were unlikely to be the source of
 6                  anthropogenic Pb as Sudbury-derived emissions exhibit a typical 206Pb/207Pb ratio of
 7                  -1.15, similar to leaded on-road gasoline. Instead, it was suggested that long-range
 8                  transport of Pb from the smelter at Rouyn-Noranda (known as the "Capital of Metal,"
 9                  NW Quebec) may still be impacting on Kawagama Lake but no  Pb isotope data was
10                  quoted to support this supposition. Several studies, summarized  in Mager (2012).
11                  reported Pb concentrations in matched reference and mining-disturbed  streams in
12                  Missouri and the western U.S. They are summarized in Table 3-11.

13                  The range of Pb levels in various saltwater environments are available from several
14                  studies although the values are not specific to the U.S. A range of 0.005-0.4 (ig Pb/L for
15                  seawater was reported by Leland and Kuwabara (1985) to reflect localized anthropogenic
16                  inputs in marine environments based on references from prior to 1980 and 0.01 to
17                  27 (ig Pb/L by Sadiq (1992). In general, Pb in seawater is higher in coastal areas and
18                  estuaries since these locations are closer to sources of Pb contamination and loading from
19                  terrestrial systems (U.S. EPA. 2008b).
     November 2012                             3-132                  Draft - Do Not Cite or Quote

-------
     Table 3-11   Pb concentrations from stream food-webs; in mining-disturbed areas
                   of Missouri and the western U.S.
Area
Total Pb in water (H9/L)
Dissolved Pb (\igl\-)
Animas River, CO (Besser et al., 2001):
Reference Streams
Mining-disturbed areas
Boulder River, MT (Faraa et al., 2007):

Reference Streams
Mining-disturbed areas
<1.8
0.9-8.6

0.4 (colloidal)
0.1-44
<0.2
<0. 1-6.9

0.3-0.4
0.1-2
Coeur d'Alene River, ID (Clark, 2003; Farag et al., 1998):
Reference Streams
Mining-disturbed areas
New Lead Belt, MO (Besser et al., 2007; Brumbaugh et al.
Reference Streams
Mining-disturbed areas
2-20
6-2,000
, 2007):
NR
NR
0.01-2
2-50

O.01-1.6
0.02-1 .7
Adapted with permission of Elsevier: Table 4-4 in Mager (2012)
     3.6.6       Vegetation

 1                 The 2006 Pb AQCD (U.S. EPA. 2006b) presented data on Pb in vegetation. The main
 2                 conclusions were that Pb uptake was strongly affected by pH, and acidic soils are most
 3                 likely to have Pb in solution for absorption by plants. Additionally, the 2006 Pb AQCD
 4                 (U.S. EPA. 2006b) states that most Pb stored within vegetation is stored in roots rather
 5                 than fruits or shoots. Recent measurements from the WACAP study (NPS. 2011) have
 6                 shown some Pb storage in lichens. Median Pb concentrations ranged from 0.3 mg/kg in
 7                 Noatak National Park (Alaska) to 5 mg/kg in Glacier National Park (Montana), with
 8                 substantial variation in the Glacier and Olympic National Park (Washington State)
 9                 samples; Figure 3-34. Landers et al. (2008) state that lichen Pb concentrations have
10                 decreased substantially from the 1980s and that metal concentrations were within
11                 background levels for these remote Western  sites.
     November 2012
3-133
Draft - Do Not Cite or Quote

-------
                  o
                 O
                             NOAT    DENA   GAAR   MORA   OLYM    SEKI    GLAC
     Source: WACAP Database (NFS. 2011)
     Note: (DENA = Denali, GAAR = Gates of the Arctic, GLAC = Glacier, MORA = Mount Ranier, NOAT = Noatak, OLYM = Olympic,
     SEKI = Sequoia and Kings Canyon)

     Figure 3-34     Boxplots of Pb concentration in lichen measured at seven
                       National Parks and Preserves.
 i
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11
Mosses can be used effectively for monitoring trends in Pb deposition as demonstrated in
many studies (Harmens et al., 2010; Harmens et al., 2008). For example, Harmens et al.
(2008) showed that a 52% decrease in deposited Pb concentrations corresponded to a
57% decrease in Pb concentrations in moss. Farmer et al. (2010) showed that there was
good agreement between the 206Pb/207Pb ratio for precipitation and mosses collected in
northeast Scotland. A study in the Vosges Mountains (France) also found a ratio value of
1.158 for a moss sample and a surface soil litter value of 1.167 and concluded that 1.158
to 1.167 represented the current atmospheric baseline (Geagea et al.. 2008). For rural
northeast Scotland, a combination of sources is giving rise to a 206Pb/207Pb ratio of-1.15
in recent precipitation and mosses (Farmer et al.. 2010). Clearly, sources with a lower
ratio than coal (-1.20) must be contributing substantially to the overall emissions. Pb
     November 2012
                          3-134
Draft - Do Not Cite or Quote

-------
 1                   from waste incineration has been implicated as a possible current source (cf. typical
 2                   206Pb/207Pb ratios for Pb from European incineration plants are -1.14 to 1.15 [de la Cruz
 3                   et al. (2009) and references therein].

 4                   Pb has been measured on vegetation near roads in recent years. Hasselbach et al. (2005)
 5                   measured Pb and other metals in mosses to assess deposition of metals along a haul road
 6                   leading from a port to the Red Dog Zn-Pb mine in Northwest Alaska. They observed that
 7                   moss concentrations of Pb decreased with increasing distance from the road, while
 8                   subsurface soils (average depth = 62 meters) did not vary with distance from the road.
 9                   The strong moss Pb gradient and constant subsurface soil Pb concentrations imply that Pb
10                   concentrations in mosses were primarily attributed to deposition and did not have
11                   appreciable contributions from soil. Throughout the study area, median moss Pb
12                   concentration was  16.2 mg/kg (dry basis), with a range of 1.1-912.5 mg/kg.
13                   Concentrations along the port road also diminished with increasing distance from the
14                   port, where ore loading operations take place. Hasselbach et al. (2005) attributed the
15                   concentrations to ore dust generated during loading operations at the port and mine along
16                   with fugitive dust escaping during truck transport. Maher et al. (2008) measured average
17                   Pb loading onto tree leaves near highways to be 29 (ig/m2 (max: 81 (ig/m2) at elevations
18                   ranging from 0.30 to 2.1 meters.

19                   Trends in Pb concentration among flora have decreased in recent years. For example,
20                   Franzaring et al. (2010) evaluated data from a 20-year biological monitoring study of Pb
21                   concentration in permanent forest and grassland plots in Baden-Wurttemberg, southwest
22                   Germany. Grassland and tree foliage samples were collected from 1985-2006. The
23                   samples were not washed and so atmospheric deposition rather than uptake from the soil
24                   probably predominates. For all foliage (beech and spruce), Pb concentrations have shown
25                   large reductions over time, particularly in the early 1990s. The Pb concentrations  in the
26                   grassland vegetation also decreased from the late 1980s to the early 1990s but the trend
27                   thereafter was found to be statistically non-significant. The reduction corresponded to the
28                   phase-out of leaded on-road gasoline in Germany. Similarly, Aznar et al. (2008a)
29                   observed that the decline in Pb concentrations in the outer level of tree rings
30                   corresponded with the decline in Cu smelter emissions in Gaspe Peninsula in Canada;
31                   Figure 3-35. Both Pb concentrations and Pb isotope ratios declined with distance from the
32                   smelter (Aznar et al.. 2008b; Aznar etal., 2008a).
      November 2012                             3-135                  Draft - Do Not Cite or Quote

-------
                                                    Trends
                                              1

                                             0.8

                                             0.6-

                                             0.4

                                             0.2

                                              0
                         Regional
                         pollution
800 •
700 -
600 •
500 •
400 •
300 •
200 -
100 •
Humus
•
ii (11)
.


• (41)
i (29)
o (21)
                                                 1950  1960  1970  1980  1990  2000

                                                    Tree rings
                                  2002            1950  1960  1970  1980  1990  2000
                                                Sapwood-heartwood
                                                   boundary
     Source: Reprinted with permission of Elsevier Publishing, Aznar et al. (2008a)
     Notes: Humus Pb concentration reported in units of mg/kg dry basis, and tree ring Pb concentration reported in units of ug/kg dry
     basis.

     Figure 3-35     Trends in regional pollution near a copper (Cu) smelter in Canada
                       and Pb concentrations at the boundary of heartwood trees within
                       roughly 75 km of the smelter.
     3.6.7       Aquatic Bivalves

1                  Data from invertebrate waterborne populations can serve as in indicator of Pb
2                  contamination because animals such as mussels and oysters take in contaminants during
3                  filter feeding. Kimbrough et al. (2008) surveyed Pb concentrations in mussels, zebra
                              O         O       V    /     J                             '
4                  mussels, and oysters along the coastlines of the continental U.S. In general, they observed
5                  the highest concentrations of Pb in the vicinity of urban and industrial areas. Company et
6                  al. (2008) measured Pb concentrations and Pb isotope ratios in bivalves along the
7                  Guadiana River separating Spain and Portugal. Analysis of Pb isotope ratio data
8                  suggested that high Pb concentrations were related to historical mining activities in the
     November 2012
3-136
Draft - Do Not Cite or Quote

-------
 1                  region. Elevated Pb concentrations were also observed by Company et al. (2008) in the
 2                  vicinity of more populated areas. Couture et al. (2010) report data from a survey of the
 3                  isotopic ratios of Pb mMytilus edulis blue mussel, collected off the coast of France from
 4                  1985-2005. The results indicated that the likely source of Pb in mussel tissue is from
 5                  resuspension of contaminated sediments enriched with Pb runoff from wastewater
 6                  treatment plants, municipal waste incinerators, smelters and refineries rather than from
 7                  atmospheric deposition (Couture et al.. 2010).
      3.6.8       Vertebrate Populations

 8                  Pb concentrations in fish fillet and liver were measured through the WACAP study in
 9                  eight National Parks and Preserves (NPS. 2011). For fish fillet, Pb concentrations ranged
10                  from 0.0033-0.30 mg/kg dry basis, with a median of 0.016 mg/kg dry basis. Liver stores
11                  were several times higher, with Pb concentrations ranging from 0.011-0.97 mg/kg dry
12                  basis and a median of 0.096 mg/kg dry basis. Pb concentrations in moose meat and liver
13                  were also measured at the Denali National Park and Preserve (Alaska) as part of WACAP
14                  (NPS. 2011). Moose meat Pb concentrations ranged from 0.021-0.23 mg/kg dry basis
15                  with a median of 0.037 mg/kg dry basis. Pb concentrations in moose liver ranged from
16                  0.025-0.11 mg/kg dry basis with a median of 0.053 mg/kg dry basis. Boxplots of
17                  measured Pb concentrations in fish fillet and liver are shown in Figure 3-36.  and boxplots
18                  of measured Pb concentrations for moose meat and liver are shown in Figure 3-37.  For
19                  fish and meat tissues, median and maximum Pb concentrations were substantially lower
20                  than values reported in the 2006 Pb AQCD (U.S. EPA. 2006b). Similarly, in a study of
21                  Pb levels in moose teeth from Isle Royale, MI, (Vucetich et al.. 2009) median and mean
22                  Pb levels underwent a statistically significant decrease from the period 1952-1982 to
23                  1983-2002 in both calves and adult moose. For 1952-1982, Pb concentrations were
24                  relatively constant, and a linear decline (R2 = 0.86) was observed for 1983-2002. These
25                  findings suggest an overall decline but still some Pb accumulation in fish and moose in
26                  these remote locations occurring recently.
      November 2012                            3-137                 Draft - Do Not Cite or Quote

-------
                1.0 -
                o.s -
                0.6 -
            o
            "(5
            c
            o
            O
                0.2 -
                o.o -
                                           ^J  -L  Q
                        1    1   1   1    1   1    1   1    1   1    1   1    1
                                                                     O  01
                                                                     CC  -i-i
                                                                  ill   ,11  :=
                                                                            i
                                                                            "2
                                                                            UJ
                                                                            tn
                        =  .s;=i   =  .s=  S  =  .s=   i^Sil
                        L^  _l   U-  Ll   Ll-  -I   U_   I  ^-  -I  Ll-   Lj  i7   ' I
(DENA = Denali, GAAR = Gates of the Arctic, GLAC = Glacier, MORA = Mount Ranier, NOAT = Noatak, OLYM = Olympic,
ROMO = Rocky Mountain, SEKI = Sequoia and Kings Canyon)
Note: Tissue concentration reported on a dry basis.
Source: WACAP Database (NFS. 2011)

Figure 3-36     Boxplots of Pb concentration in fish fillet and fish liver, measured
                  at eight National Parks and/or Preserves.
November 2012
3-138
Draft - Do Not Cite or Quote

-------
               CD

               CD
           o
           "oo
           O
           O
           m
           z;
           i.U
           01
               CD
               CD
                               Liver
                  Meat
Note: Tissue concentration reported on a dry basis.
Source: WACAP Database (NFS. 2011)

Figure 3-37      Boxplots of Pb concentration in moose meat and moose liver
                 measured at Denali National  Park and Preserve.
November 2012
3-139
Draft - Do Not Cite or Quote

-------
      3.7       Summary and Conclusions
      3.7.1       Sources of Atmospheric Pb

 1                  The 2006 Pb AQCD (U.S. EPA. 2006b) documented the decline in ambient air Pb
 2                  emissions following the ban on alkyl-Pb additives for on-road gasoline. Pb emissions
 3                  declined by 98% from 1970 to 1995 and then by an additional 76% from 1995 to 2008, at
 4                  which time national Pb emissions were 964 tons/year. As was the case for the 2008
 5                  NAAQS review, piston-engine aircraft emissions currently comprise the largest share
 6                  (57%) of total atmospheric Pb emissions nationally (U.S. EPA. 201 la). Other sources of
 7                  ambient air Pb, in approximate order of importance with regard to national totals, include
 8                  metal working and mining, fuel combustion, other industrial sources, roadway related
 9                  sources, and historic Pb. Although piston-engine aircraft collectively comprise the largest
10                  emissions source, the highest emitting individual industrial sites produce more ambient
11                  air Pb emissions than individual airports.
      3.7.2       Fate and Transport of Pb

12                  The atmosphere is the main environmental transport pathway for Pb, and on a global
13                  scale atmospheric Pb is primarily associated with fine PM. Pb in fine PM is transported
14                  long distances and found in remote areas. Atmospheric Pb deposition peaked in the
15                  1970s, followed by a decline. Both wet and dry deposition are important removal
16                  mechanisms for atmospheric Pb. Wet deposition is more important for fine Pb, and Pb
17                  associated with coarse PM is usually removed by dry deposition. Local deposition fluxes
18                  are much higher near industrial sources, and a substantial amount of emitted Pb is
19                  deposited near sources, leading to increased soil Pb concentrations. Deposition does not
20                  cause  an ultimate sink for Pb because particles are potentially resuspended and
21                  redeposited many times before reaching a site where further transport is unlikely,
22                  especially for dry deposition.

23                  In water,  Pb is transported as free ions, soluble chelates, or on surfaces of iron and
24                  organic rich colloids. In most surface waters, atmospheric deposition is the largest source
25                  of Pb, but urban runoff and industrial discharge are  also considerable. A substantial
26                  portion of Pb in runoff ultimately originates from atmospheric deposition, but substantial
27                  amounts of Pb from vehicle wear and building materials can also be transported by runoff
28                  waters without becoming airborne. Often the majority of Pb transport by runoff occurs at
29                  the beginning of a rainfall event. Pb is rapidly dispersed in water, and highest
30                  concentrations of Pb are observed near sources where Pb is deposited.
      November 2012                            3-140                 Draft - Do Not Cite or Quote

-------
 1                   Transport in surface waters is largely controlled by exchange with sediments. The cycling
 2                   of Pb between water and sediments is governed by chemical, biological, and mechanical
 3                   processes, which are affected by many factors.  Organic matter in sediments has a high
 4                   capacity for accumulating trace elements like Pb. Binding of anoxic sediments to sulfides
 5                   is a particularly important process that affects Pb bioavailability. Pb is relatively stable in
 6                   sediments, with long residence times and limited mobility. However, Pb-containing
 7                   sediment particles can be remobilized into the water column. Resuspended Pb is largely
 8                   associated with OM or Fe and Mn particles. This resuspension of contaminated sediments
 9                   strongly influences the lifetime  of Pb in water bodies and can be a more important Pb
10                   source to the water column than atmospheric deposition. Resuspension of sediments
11                   largely occurs during discrete events related to  storms.

12                   A complex variety of factors influence Pb retention in soil, including hydraulic
13                   conductivity, solid composition, OM content, clay mineral content, microbial activity,
14                   plant root channels, animal holes, geochemical  reactions, colloid amounts, colloidal
15                   surface charge, and pH. Leaf litter can be an important temporary sink for metals from
16                   the soil around and below leaves, and decomposition of leaf litter can reintroduce
17                   substantial amounts of Pb into soil "hot spots,"  where re-adsorption of Pb is favored. A
18                   small fraction of Pb in  soil is present as the free Pb2+ ion. The fraction of Pb in this form
19                   is strongly dependent on soil pH.

20                   In summary, environmental distribution of Pb occurs mainly through the atmosphere,
21                   from where  it is deposited into surface waters and soil. Pb associated with coarse PM
22                   deposits to a great extent near sources, while fine Pb-PM can be transported long
23                   distances. Surface waters act as an important reservoir, with half-lives of Pb in the water
24                   column largely controlled by rates of deposition to and resuspension from bottom
25                   sediments. Pb retention in soil depends on Pb speciation and a variety of factors intrinsic
26                   to the soil.
      3.7.3        Ambient Pb Monitoring

27                   Since the publication of the 2006 Pb AQCD for Pb (U.S. EPA. 2006b) there has been
28                   little progress in the state of the science regarding monitoring technology and monitor
29                   siting criteria for representation of population exposures to airborne Pb and Pb of
30                   atmospheric origin. Our understanding of sampling errors in the existing FRM, of
31                   possible alternatives to existing Pb-TSP sampling technology, and of particle size ranges
32                   of Pb particles occurring in different types of locations have changed little in that time. In
33                   addition to monitors used historically for sampling Pb-PM, several single stage and
34                   multi-stage impactors and inlets used for sampling PM are also potential options for
      November 2012                             3-141                  Draft - Do Not Cite or Quote

-------
 1                  monitoring Pb particles smaller than 15 pirn. Ambient air Pb deposits onto soil or dust. As
 2                  described in Section 4.1, the size distribution of dust and soil Pb particles is larger than
 3                  the size distribution of ambient air Pb particles. The existing samplers reasonably capture
 4                  the airborne fraction of ambient Pb that is available for human exposure.

 5                  The current Pb monitoring network design requirements include two types of monitoring
 6                  sites: source-oriented and non-source-oriented. Source-oriented monitoring sites are
 7                  required near sources of air Pb emissions which are expected to or have been shown to
 8                  contribute to ambient air Pb concentrations in excess of the NAAQS. Non-source-
 9                  oriented monitoring of Pb-TSP or Pb-PMi0 is also required at NCore sites in CBSAs with
10                  a population of at least 500,000.

11                  In addition to Pb-TSP monitoring for the purposes  of judging attainment with the
12                  NAAQS, Pb is also routinely measured in smaller PM fractions in the CSN, IMPROVE,
13                  and the NATTS networks. While monitoring in multiple networks provides extensive
14                  geographic coverage, measurements between networks are not directly comparable in all
15                  cases because different PM size ranges are sampled in different networks. Depending on
16                  monitoring network, Pb is monitored in TSP, PM10, or PM2 5 using high-volume or
17                  low-volume samplers.
      3.7.4       Ambient Air Pb Concentrations

18                  Ambient air Pb concentrations have declined drastically over the period 1980-2010. The
19                  median annual maximum 3-month average concentration of Pb-TSP has dropped by 97%
20                  from 0.87 ug/m3 in 1980 to 0.03 ug/m3 in 2010. The decline can be attributed to the
21                  phase-out of Pb antiknock agents in on-road fuel and reductions in industrial use and
22                  processing of Pb, as described in Section 3.2.1. The mean of maximum 3-month average
23                  concentrations for source-oriented monitors was skewed toward the 75th percentile of the
24                  data distribution and exceeded the level of the NAAQS, indicating that highest ambient
25                  air Pb concentrations occur near a subset of source-oriented monitors. Studies in the peer-
26                  reviewed literature have shown slightly elevated Pb concentrations downwind of
27                  industrial sources and airports.

28                  Spatial variability was observed in ratios and correlations of Pb within different size
29                  fractions. Urban or suburban land types did not appear to affect sampled size
30                  distributions. Studies in the peer-reviewed literature suggest that proximity to industrial
31                  sources or some roadways can affect the Pb-PM  size distribution. Pb concentrations
32                  exhibit varying degrees of association with other criteria pollutant concentrations.
33                  Overall, non-source Pb-TSP was moderately associated with CO, PM2 5, and PMi0, which
34                  may indicate some role of traffic in Pb exposure. Among trace metals speciated from

      November 2012                             3-142                  Draft - Do Not Cite or Quote

-------
 1                  PM2 5, Pb was not associated with most pollutants; Pb did associate with Zn, although
 2                  that association was low-to-moderate, suggesting mobile source emissions contributing to
 3                  the Pb. EC, Cu, OC, and Br concentrations also exhibited low-to-moderate associations
 4                  with Pb concentrations. Such correlations may suggest some common sources affecting
 5                  the pollutants. Finally, the evidence on natural background Pb suggests a plausible
 6                  background airborne Pb range of 0.02 to 1 ng/m3.
      3.7.5        Ambient Pb Concentrations in Non-Air Media and Biota

 7                  Atmospheric deposition has led to measurable Pb concentrations observed in rain,
 8                  snowpack, soil, surface waters, sediments, agricultural plants, livestock, and wildlife
 9                  across the world, with highest concentrations near Pb sources, such as metal smelters.
10                  Since the phase-out of Pb from on-road gasoline, concentrations in these media have
11                  decreased to varying degrees. In rain, snowpack, and surface waters, Pb concentrations
12                  have decreased considerably. Declining Pb concentrations in tree foliage, trunk sections,
13                  and grasses have also been observed. In contrast, Pb is retained in soils and sediments,
14                  where it provides a historical record of deposition and associated ambient concentrations.
15                  In remote lakes, sediment profiles indicate higher Pb concentrations in near surface
16                  sediment as compared to pre-industrial era sediment from greater depth and indicate peak
17                  concentrations between  1960 and 1980, when leaded on-road gasoline was at peak use.
18                  Concentrations of Pb in  moss, lichens, peat, and aquatic bivalves have been used to
19                  understand spatial and temporal distribution patterns of air Pb concentrations. Ingestion
20                  and water intake are the  major routes of Pb exposure for aquatic organisms, and food,
21                  drinking water, and inhalation are major routes of exposure for livestock and terrestrial
22                  wildlife. Overall, Pb concentrations have decreased substantially in media through which
23                  Pb is rapidly transported, such as air and water. Substantial Pb remains in soil and
24                  sediment sinks. In areas  less affected by major local sources, the highest concentrations
25                  are below the surface layers and reflect the previous use of Pb in on-road gasoline and
26                  emissions reductions from other sources.
      November 2012                             3-143                  Draft - Do Not Cite or Quote

-------
3.8      Chapter 3 Appendix (Supplemental Material)
3.8.1
Variability across the U.S.
Table 3-12  Distribution of 1-month average Pb-TSP concentrations (ug/m3) nationwide, source-oriented
           monitors, 2008-2010.
State/ County
Year Season County State name
Site N: mo
ID means
N
sites Mean
Min
1
5
10
25
50
75
90
95
99
max
Nationwide statistics
2008-2010
2008
2009
2010
Winter
Spring
Summer
Fall
2,318
548
629
1141
554
579
601
584
0.202
0.318
0.212
0.141
0.202
0.239
0.186
0.184
0.000
0.004
0.002
0.000
0.000
0.000
0.001
0.000
0.003
0.004
0.004
0.002
0.002
0.003
0.003
0.004
0.006
0.013
0.008
0.005
0.006
0.007
0.006
0.007
0.010
0.024
0.013
0.008
0.008
0.012
0.010
0.011
0.029
0.050
0.038
0.018
0.026
0.034
0.030
0.026
0.063
0.110
0.084
0.045
0.055
0.070
0.066
0.064
0.217
0.348
0.256
0.136
0.184
0.272
0.212
0.206
0.578
0.841
0.611
0.408
0.502
0.738
0.559
0.505
0.856
1.240
0.856
0.625
0.883
0.977
0.755
0.758
1.576
2.557
1.357
1.233
2.438
1.905
1.233
1.178
4.440
4.440
2.438
1.828
3.103
3.123
4.440
4.225
November 2012
                      3-144
Draft - Do Not Cite or Quote

-------
Year Season
State/
County
State
County Site
name ID
N: mo
means
N
sites
Mean
Min
1
5
10 25
50
75 90
95
99 max
Nationwide statistics, pooled by site
2008-2010
2008
2009
2010
Winter
Spring
Summer
Fall
















Statistics for individual counties








(2008-2010)









111
47
54
101
108
110
111
110

0.161
0.323
0.214
0.140
0.156
0.185
0.148
0.152

0.002
0.007
0.007
0.002
0.000
0.002
0.002
0.002

0.003
0.007
0.007
0.003
0.003
0.002
0.003
0.004

0.008
0.022
0.013
0.005
0.006
0.010
0.006
0.009

0.013 0.031
0.028 0.055
0.018 0.043
0.013 0.030
0.009 0.021
0.015 0.027
0.012 0.025
0.013 0.034

0.056
0.148
0.090
0.052
0.048
0.057
0.050
0.062

0.177 0.441
0.419 0.890
0.343 0.669
0.165 0.392
0.160 0.475
0.210 0.568
0.153 0.430
0.168 0.421

0.687
1.205
0.849
0.586
0.879
0.921
0.696
0.616

0.997 1 .275
1 .540 1 .540
0.921 0.921
0.888 1.185
1.130 1.488
1.189 1.548
0.882 1 .031
1.081 1.189


















01109
06037
12057
13015
13215
17031
17115
17119
17143
17195
17201
18035
18089
18097
18127
19155
AL
CA
FL
GA
GA
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IA
Pike
Los Angeles
Hillsborough
Bartow
Muscogee
Cook
Macon
Madison
Peoria
Whiteside
Winnebago
Delaware
Lake
Marion
Porter
Pottawattamie
32
131
81
12
12
11
12
36
24
12
11
59
57
70
12
12
1
4
3
1
1
1
1
1
2
1
1
2
3
2
1
1
0.5252
0.2380
0.1755
0.0128
0.0361
0.1515
0.0800
0.1367
0.0119
0.0194
0.0339
0.2746
0.0309
0.0195
0.0125
0.1536
0.054
0.018
0.007
0.007
0.004
0.028
0.018
0.018
0.010
0.010
0.010
0.034
0.004
0.003
0.004
0.025
0.054
0.019
0.007
0.007
0.004
0.028
0.018
0.018
0.010
0.010
0.010
0.034
0.004
0.003
0.004
0.025
0.083
0.026
0.017
0.007
0.004
0.028
0.018
0.022
0.010
0.010
0.010
0.040
0.007
0.005
0.004
0.025
0.164 0.252
0.034 0.047
0.020 0.053
0.008 0.008
0.010 0.013
0.028 0.050
0.025 0.035
0.024 0.037
0.010 0.010
0.012 0.012
0.014 0.020
0.049 0.080
0.008 0.012
0.005 0.008
0.005 0.007
0.026 0.063
0.402
0.085
0.104
0.014
0.027
0.074
0.074
0.068
0.010
0.015
0.024
0.128
0.020
0.012
0.009
0.164
0.798 1 .053
0.246 0.602
0.187 0.530
0.016 0.017
0.043 0.058
0.196 0.304
0.118 0.144
0.175 0.304
0.012 0.016
0.024 0.036
0.032 0.050
0.241 0.427
0.035 0.052
0.025 0.046
0.021 0.024
0.257 0.276
1.117
0.905
0.567
0.019
0.140
0.580
0.168
0.363
0.023
0.040
0.118
1.011
0.079
0.050
0.026
0.282
1.277 1.277
2.501 2.880
1 .007 1 .007
0.019 0.019
0.140 0.140
0.580 0.580
0.168 0.168
0.836 0.836
0.024 0.024
0.040 0.040
0.118 0.118
4.440 4.440
0.298 0.298
0.125 0.125
0.026 0.026
0.282 0.282
November 2012
3-145
Draft - Do Not Cite or Quote

-------
State/
Year Season County
20169
21019
21151
26067
27003
27037
27145
29093
29099
29179
31053
31127
36071
39035
39051
39091
39101
39151
39155
40121
41071
42003
42007
42011
42045
42055
42063
State
KS
KY
KY
Ml
MN
MN
MN
MO
MO
MO
NE
NE
NY
OH
OH
OH
OH
OH
OH
OK
OR
PA
PA
PA
PA
PA
PA
County
name
Saline
Boyd
Madison
Ionia
Anoka
Dakota
Stearns
Iron
Jefferson
Reynolds
Dodge
Nemaha
Orange
Cuyahoga
Fulton
Logan
Marion
Stark
Trumbull
Pittsburg
Yamhill
Allegheny
Beaver
Berks
Delaware
Franklin
Indiana
Site N: mo
ID means
11
7
12
12
12
36
12
171
453
48
9
8
105
72
34
102
10
11
8
11
12
24
54
117
12
11
12
N
sites
1
1
1
1
1
1
1
7
19
4
1
1
3
3
1
4
1
1
1
1
1
2
3
6
1
1
1
Mean
0.2020
0.0042
0.0255
0.1781
0.0157
0.1966
0.0028
0.3388
0.4795
0.0428
0.0515
0.0476
0.0281
0.0941
0.1462
0.0480
0.0358
0.0175
0.0075
0.0023
0.0157
0.0369
0.1130
0.0989
0.0452
0.0449
0.0454
Min
0.043
0.002
0.004
0.016
0.003
0.037
0.000
0.007
0.011
0.007
0.005
0.008
0.001
0.004
0.009
0.003
0.025
0.008
0.004
0.002
0.006
0.006
0.042
0.034
0.043
0.042
0.042
1
0.043
0.002
0.004
0.016
0.003
0.037
0.000
0.008
0.015
0.007
0.005
0.008
0.001
0.004
0.009
0.003
0.025
0.008
0.004
0.002
0.006
0.006
0.042
0.035
0.043
0.042
0.042
5
0.043
0.002
0.004
0.016
0.003
0.048
0.000
0.014
0.033
0.008
0.005
0.008
0.003
0.007
0.009
0.004
0.025
0.008
0.004
0.002
0.006
0.006
0.044
0.038
0.043
0.042
0.042
10
0.044
0.002
0.008
0.023
0.005
0.058
0.000
0.018
0.048
0.011
0.005
0.008
0.004
0.008
0.026
0.005
0.026
0.009
0.004
0.002
0.007
0.006
0.047
0.042
0.043
0.043
0.043
25
0.083
0.004
0.013
0.054
0.007
0.084
0.000
0.033
0.141
0.017
0.021
0.010
0.006
0.014
0.057
0.020
0.027
0.010
0.005
0.002
0.008
0.010
0.068
0.048
0.043
0.043
0.043
50
0.133
0.004
0.017
0.169
0.011
0.137
0.003
0.093
0.336
0.027
0.031
0.024
0.018
0.038
0.091
0.042
0.033
0.018
0.007
0.002
0.016
0.017
0.096
0.066
0.045
0.045
0.044
75
0.320
0.004
0.022
0.279
0.021
0.259
0.005
0.518
0.659
0.060
0.053
0.049
0.044
0.121
0.170
0.070
0.041
0.024
0.008
0.003
0.020
0.040
0.128
0.119
0.047
0.047
0.046
90
0.457
0.007
0.032
0.361
0.022
0.424
0.006
0.850
1.118
0.087
0.149
0.206
0.063
0.210
0.420
0.090
0.054
0.025
0.017
0.003
0.025
0.121
0.198
0.200
0.048
0.047
0.047
95
0.488
0.007
0.121
0.414
0.054
0.572
0.008
1.110
1.451
0.099
0.149
0.206
0.081
0.400
0.490
0.100
0.066
0.028
0.017
0.003
0.037
0.144
0.272
0.295
0.048
0.047
0.058
99
0.488
0.007
0.121
0.414
0.054
0.738
0.008
2.557
2.220
0.268
0.149
0.206
0.101
0.719
0.510
0.120
0.066
0.028
0.017
0.003
0.037
0.149
0.286
0.347
0.048
0.047
0.058
max
0.488
0.007
0.121
0.414
0.054
0.738
0.008
4.225
3.123
0.268
0.149
0.206
0.134
0.719
0.510
0.170
0.066
0.028
0.017
0.003
0.037
0.149
0.286
0.348
0.048
0.047
0.058
November 2012
3-146
Draft - Do Not Cite or Quote

-------
                  State/
                                    County
Year   Season   County   State   name
Site
ID
N: mo
means
N
sites
Mean    Min    1
10
25
50
75
90
95
99
                                                                                                                  max
                  42073
                            PA
                                    Lawrence
                                                                                  0.0438   0.042   0.042   0.042   0.042   0.043  0.044   0.045   0.046   0.046   0.046  0.046
                  42079
                            PA
                                    Luzerne
                                                            10
                                                                                  0.0953   0.043   0.043   0.043   0.044   0.045  0.071   0.102   0.215   0.268   0.268  0.268
                  42129
                            PA
                                    Westmoreland
                                                            12
                                                                                  0.0439   0.041   0.041    0.041   0.041   0.043  0.044   0.045   0.046   0.047   0.047  0.047
                  47093
                            TN
                                    Knox
                                                            48
                                                                                  0.0165   0.002   0.002   0.005   0.006   0.008  0.012   0.019   0.032   0.038   0.063  0.063
                  47163
                            TN
                                    Sullivan
                                                            120
                                                                                  0.0534   0.021   0.023   0.030   0.032   0.037  0.045   0.059   0.083   0.124   0.145  0.156
                  48085
                            TX
                                    Collin
                                                            108
                                                                                  0.3062   0.007   0.028   0.040   0.052   0.104  0.189   0.438   0.717   0.904   1.178   1.564
                  48375
                            TX
                                    Potter
                                                                                  0.0044   0.004   0.004   0.004   0.004   0.004  0.004   0.005   0.006   0.006   0.006  0.006
                  51770
                            VA
                                    Roanoke City
                                                            12
                                                                                  0.0412   0.005   0.005   0.005   0.008   0.010  0.015   0.035   0.054   0.272   0.272   0.272
                  55117
                            Wl
                                    Sheboygan
                                                            12
                                                                                  0.0802   0.001   0.001    0.001   0.003   0.007  0.054   0.136   0.182   0.279   0.279  0.279
                  72013
                            PR
                                    Arecibo
                                    (Puerto Rico)
                                                            12
                                                                                  0.1774   0.038   0.038   0.038   0.064   0.102   0.178   0.264   0.290   0.310   0.310  0.310
Statistics for individual sites where overall average monthly mean > national 90th percentile (2008-2010)
                                    11090003
                                                            32
                                                                                  0.525    0.054   0.054   0.083   0.164   0.252   0.402   0.798   1.053   1.117   1.277   1.277
                                    060371405
                                                            36
                                                                                  0.671    0.100   0.100   0.188   0.235   0.285  0.359   0.771    2.086   2.501   2.880  2.880
                                    290930016
                                                            36
                                                                                  0.670    0.166   0.166   0.186   0.219   0.330  0.466   0.726   0.974   2.435   4.225  4.225
                                    290930021
                                                            36
                                                                                  0.681    0.082   0.082   0.084   0.095   0.194  0.650   0.879   1.437   2.438   2.557  2.557
                                    290990004
                                                            36
                                                                                  0.997    0.256   0.256   0.307   0.408   0.598  0.918   1.236   1.690   1.905   2.416  2.416
                                    290990015
                                                            21
                                                                                  1.275    0.340   0.340   0.421   0.646   0.756   1.118   1.349   2.440   3.103   3.123  3.123
                                    290990020a
                                                            31
                                                                                  0.687    0.191   0.191    0.195   0.297   0.368  0.620   0.808   1.111   1.280   2.220  2.220
                                    290990021a
                                                            21
                                                                                  0.719    0.084   0.084   0.141   0.359   0.572   0.666   0.876   1.164   1.168   1.553   1.553
                                    290990022a
                                                            31
                                                                                  0.441
                                                                                          0.140   0.140   0.171   0.208   0.303  0.409   0.599   0.683   0.754   0.861   0.861
                                    290999001"
                                                            24
                                                                                  0.850    0.186   0.186   0.208   0.319   0.449  0.845   1.071    1.382   1.558   1.623   1.623
                                    290999005a
                                                            24
                                                                                  0.986    0.155   0.155   0.250   0.330   0.558  0.864   1.487   1.802   1.828   1.985   1.985
                                    480850009a
                                                            36
                                                                                  0.601
                                                                                          0.137   0.137   0.138   0.185   0.420  0.579   0.757   1.101   1.178   1.564   1.564
"Sites listed in the bottom six rows of this table fall in the upper 90th percentile of the data pooled by site.
November 2012
  3-147
                       Draft - Do Not Cite or Quote

-------
Table 3-13 Distribution of 1 -month average Pb-TSP concentrations
monitors, 2008-2010.
Year Season ^^ State C°^ S,'Je me™
N sites Mean
»
Min
1
(ug/m3) nationwide,
5
10
25
50
non-source-oriented
75
90
95
99
max
Nationwide statistics
2008-2010 2290
2008 685
2009 768
2010 837
Winter 556
Spring 574
Summer 584
Fall 576
0.0120
0.0126
0.0114
0.0120
0.0109
0.0122
0.0119
0.0129
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.002
0.002
0.002
0.000
0.001
0.002
0.002
0.002
0.004
0.005
0.004
0.004
0.004
0.004
0.005
0.005
0.010
0.010
0.010
0.009
0.008
0.009
0.010
0.010
0.015
0.015
0.014
0.016
0.013
0.015
0.016
0.016
0.026
0.029
0.023
0.026
0.022
0.028
0.026
0.026
0.040
0.040
0.040
0.036
0.038
0.040
0.040
0.040
0.052
0.052
0.048
0.054
0.056
0.052
0.050
0.053
0.136
0.066
0.128
0.136
0.087
0.128
0.057
0.136
Nationwide statistics, pooled by site
2008-2010
2008
2009
2010
Winter
Spring
Summer
Fall
88 0.0120
59 0.0125
66 0.0116
73 0.0119
88 0.0115
86 0.0119
88 0.0117
88 0.0130
0.000
0.001
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.001
0.000
0.000
0.000
0.000
0.000
0.000
0.001
0.002
0.001
0.001
0.001
0.001
0.000
0.001
0.002
0.003
0.002
0.001
0.001
0.002
0.001
0.003
0.005
0.006
0.004
0.005
0.004
0.004
0.005
0.005
0.011
0.010
0.010
0.010
0.009
0.009
0.010
0.011
0.016
0.016
0.014
0.018
0.016
0.016
0.016
0.017
0.024
0.024
0.024
0.023
0.025
0.027
0.026
0.028
0.033
0.043
0.032
0.028
0.038
0.032
0.034
0.031
0.046
0.051
0.050
0.046
0.048
0.059
0.043
0.054
0.046
0.051
0.050
0.046
0.048
0.059
0.043
0.054
Statistics for individual counties (2008-2010)
04013 AZ Maricopa 6
06025 CA Imperial 33
06037 CA Los Angeles 224
06065 CA Riverside 72
1 0.0218
1 0.0162
8 0.0098
2 0.0077
0.009
0.004
0.000
0.000
0.009
0.004
0.000
0.000
0.009
0.006
0.000
0.003
0.009
0.009
0.002
0.004
0.014
0.011
0.006
0.006
0.021
0.015
0.010
0.008
0.028
0.019
0.012
0.010
0.038
0.025
0.017
0.010
0.038
0.032
0.020
0.012
0.038
0.035
0.038
0.014
0.038
0.035
0.044
0.014
November 2012
3-148
Draft - Do Not Cite or Quote

-------
State/
Year Season
County
06071
08005
08031
13089
17031
17117
17119
17143
17163
18089
18097
18163
25025
26081
26163
27017
27037
27053
27075
27123
27137
27163
29097
29187
29189
36047
State
CA
CO
CO
GA
IL
IL
IL
IL
IL
IN
IN
IN
MA
Ml
Ml
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
NY
County
name
San
Bernardino
Arapahoe
Denver
DeKalb
Cook
Macoupin
Madison
Peoria
Saint Clair
Lake
Marion
Vanderburgh
Suffolk
Kent
Wayne
Carlton
Dakota
Hennepin
Lake
Ramsey
Saint Louis
Washington
Jasper
Saint Francois
Saint Louis
Kings
Site N: mo
ID means
71
9
12
10
288
24
36
36
36
36
35
33
31
12
36
12
118
126
10
71
72
72
12
24
33
24
N sites
2
1
1
1
8
1
1
1
1
1
1
2
2
1
2
1
5
4
1
3
2
3
1
2
1
1
Mean
0.0091
0.0120
0.0056
0.0033
0.0195
0.0101
0.0188
0.0105
0.0206
0.0150
0.0058
0.0045
0.0087
0.0053
0.0112
0.0000
0.0035
0.0032
0.0000
0.0062
0.0015
0.0016
0.0125
0.0327
0.0230
0.0131
Min
0.001
0.004
0.003
0.002
0.010
0.010
0.010
0.010
0.010
0.005
0.002
0.001
0.004
0.003
0.003
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.007
0.008
0.005
0.010
1
0.001
0.004
0.003
0.002
0.010
0.010
0.010
0.010
0.010
0.005
0.002
0.001
0.004
0.003
0.003
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.007
0.008
0.005
0.010
5
0.003
0.004
0.003
0.002
0.010
0.010
0.010
0.010
0.010
0.005
0.002
0.001
0.004
0.003
0.003
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.007
0.009
0.005
0.010
10
0.004
0.004
0.004
0.002
0.010
0.010
0.010
0.010
0.012
0.005
0.003
0.002
0.005
0.003
0.004
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.007
0.009
0.005
0.010
25
0.007
0.007
0.005
0.003
0.012
0.010
0.012
0.010
0.014
0.008
0.004
0.003
0.007
0.005
0.005
0.000
0.000
0.000
0.000
0.002
0.000
0.000
0.009
0.018
0.006
0.011
50
0.010
0.012
0.005
0.003
0.016
0.010
0.016
0.010
0.018
0.014
0.005
0.004
0.008
0.005
0.009
0.000
0.002
0.002
0.000
0.004
0.000
0.000
0.012
0.032
0.008
0.012
75
0.012
0.016
0.006
0.004
0.025
0.010
0.020
0.010
0.026
0.019
0.008
0.005
0.010
0.006
0.015
0.000
0.005
0.005
0.000
0.008
0.002
0.003
0.017
0.039
0.050
0.014
90
0.014
0.018
0.008
0.005
0.034
0.010
0.032
0.013
0.032
0.030
0.010
0.006
0.013
0.008
0.021
0.000
0.008
0.006
0.000
0.013
0.004
0.004
0.018
0.054
0.050
0.018
95
0.014
0.018
0.008
0.006
0.040
0.010
0.053
0.013
0.038
0.033
0.012
0.010
0.016
0.008
0.023
0.000
0.010
0.008
0.000
0.020
0.006
0.005
0.019
0.080
0.050
0.020
99
0.022
0.018
0.008
0.006
0.060
0.012
0.066
0.014
0.054
0.049
0.013
0.010
0.020
0.008
0.032
0.000
0.017
0.010
0.000
0.028
0.010
0.006
0.019
0.089
0.066
0.020
max
0.022
0.018
0.008
0.006
0.070
0.012
0.066
0.014
0.054
0.049
0.013
0.010
0.020
0.008
0.032
0.000
0.036
0.044
0.000
0.028
0.010
0.006
0.019
0.089
0.066
0.020
November 2012
3-149
Draft - Do Not Cite or Quote

-------
State/
Year Season
County
39017
39029
39035
39049
39143
39167
40115
42003
42021
42045
42101
42129
48061
48141
48201
48479
49035
51087
State
OH
OH
OH
OH
OH
OH
OK
PA
PA
PA
PA
PA
TX
TX
TX
TX
UT
VA
County
name
Butler
Columbiana
Cuyahoga
Franklin
Sandusky
Washington
Ottawa
Allegheny
Cambria
Delaware
Philadelphia
Westmoreland
Cameron
El Paso
Harris
Webb
Salt Lake
Henrico
Site N: mo
ID means
34
107
107
36
12
54
16
36
23
20
24
24
35
68
32
29
12
7
N sites
1
3
3
1
1
2
2
1
1
1
1
1
1
3
1
1
1
1
Mean
0.0055
0.0155
0.0143
0.0092
0.0048
0.0048
0.0124
0.0105
0.0463
0.0432
0.0210
0.0419
0.0041
0.0206
0.0053
0.0134
0.0173
0.0066
Min
0.002
0.004
0.004
0.004
0.003
0.002
0.003
0.000
0.040
0.040
0.011
0.037
0.002
0.014
0.003
0.004
0.003
0.003
1
0.002
0.004
0.004
0.004
0.003
0.002
0.003
0.000
0.040
0.040
0.011
0.037
0.002
0.014
0.003
0.004
0.003
0.003
5
0.003
0.006
0.006
0.005
0.003
0.002
0.003
0.000
0.040
0.040
0.011
0.040
0.003
0.014
0.003
0.005
0.003
0.003
10
0.004
0.007
0.007
0.005
0.003
0.003
0.005
0.000
0.040
0.040
0.012
0.040
0.003
0.014
0.004
0.006
0.006
0.003
25
0.004
0.008
0.009
0.007
0.004
0.003
0.006
0.004
0.040
0.040
0.014
0.040
0.003
0.015
0.004
0.008
0.009
0.003
50
0.005
0.011
0.012
0.009
0.005
0.005
0.013
0.009
0.040
0.043
0.020
0.040
0.004
0.017
0.005
0.011
0.011
0.004
75
0.007
0.018
0.017
0.011
0.006
0.006
0.017
0.015
0.044
0.046
0.027
0.042
0.005
0.019
0.006
0.018
0.024
0.005
90
0.008
0.027
0.024
0.013
0.006
0.007
0.021
0.019
0.054
0.047
0.033
0.050
0.006
0.029
0.007
0.026
0.040
0.024
95
0.009
0.034
0.030
0.014
0.007
0.008
0.025
0.024
0.058
0.048
0.033
0.050
0.007
0.056
0.008
0.028
0.043
0.024
99
0.009
0.065
0.041
0.016
0.007
0.010
0.025
0.053
0.128
0.048
0.039
0.053
0.009
0.087
0.010
0.035
0.043
0.024
max
0.009
0.136
0.041
0.016
0.007
0.010
0.025
0.053
0.128
0.048
0.039
0.053
0.009
0.087
0.010
0.035
0.043
0.024
November 2012
3-150
Draft - Do Not Cite or Quote

-------
                          "*
                                                        10
                                    25
50
75
90
95
99    max
Statistics for individual sites where overall average monthly mean > national 90th percentile (2008-2010)
170310022
170310026
170316003
291870006"
291870007"
42021 0808a
420450002"
421290007"
481410002"
36
36
36
12
12
23
20
24
23
0.0330
0.0282
0.0249
0.0383
0.0271
0.0463
0.0432
0.0419
0.0236
0.012
0.014
0.012
0.009
0.008
0.040
0.040
0.037
0.016
0.012
0.014
0.012
0.009
0.008
0.040
0.040
0.037
0.016
0.014
0.014
0.014
0.009
0.008
0.040
0.040
0.040
0.016
0.016
0.018
0.018
0.015
0.009
0.040
0.040
0.040
0.016
0.020
0.020
0.020
0.024
0.013
0.040
0.040
0.040
0.017
0.033
0.028
0.026
0.035
0.026
0.040
0.043
0.040
0.018
0.040
0.034
0.031
0.042
0.035
0.044
0.046
0.042
0.021
0.056
0.044
0.033
0.080
0.052
0.054
0.047
0.050
0.033
0.062
0.048
0.038
0.089
0.054
0.058
0.048
0.050
0.056
0.070
0.052
0.040
0.089
0.054
0.128
0.048
0.053
0.087
0.070
0.052
0.040
0.089
0.054
0.128
0.048
0.053
0.087
'Sites listed in the bottom six rows of this table fall in the upper 90th percentile of the data pooled by site.
November 2012
3-151
Draft - Do Not Cite or Quote

-------
Table 3-14 Distribution of 3-month moving average Pb-TSP concentrations (ug/m3) nationwide, source-oriented
monitors, 2008-2010.
Year
Nationwide
j O*«*A/ ! ! LI
Season i State JCountyname Site ID I ° N sites Mean Min 1 5 10 25 50 75 90 95 99 max
jCounty | jmeans
statistics3
2008-2010
2008




2,112

0.2134
0.000
0.004
0.010
0.014
0.035
0.079
0.250
0.600
0.881
1.555
2.889
III I J537 I I 0.3225 I 0.005 I 0.006 I 0.016 I 0.028 I 0.056 I 0.129 I 0.385 I 0.900 I 1.197 I 2.452 I 2.889
2009
2010



Winter
jSpring


Nationwide
Summer
Fall
statistics, pooled b
2008-2010 |
2008
2009
2010











ysite


























600
975
443
535
572
562












106
47
54
96
0.2177
0.1507
0.2366
0.2376
0.2022
0.1835

0.1671
0.3309
0.2203
0.1415
0.004
0.000
0.003
0.000
0.002
0.002

0.002
0.007
0.007
0.002
0.005
0.002
0.004
0.004
0.003
0.004

0.003
0.007
0.007
0.002
0.011
0.008
0.011
0.011
0.009
0.009

0.012
0.024
0.013
0.008
0.016
0.012
0.014
0.014
0.015
0.013

0.015
0.029
0.019
0.013
0.040
0.024
0.040
0.035
0.034
0.033

0.030
0.056
0.042
0.027
0.090
0.052
0.083
0.078
0.077
0.078

0.059
0.154
0.086
0.053
0.292
0.173
0.272
0.323
0.240
0.220

0.173
0.461
0.311
0.163
0.622
0.436
0.647
0.642
0.580
0.521

0.577
0.814
0.632
0.407
0.799
0.694
0.963
0.999
0.869
0.714

0.717
1.284
0.840
0.619
1.217
1.055
2.070
2.017
1.261
1.186

1.009
1.639
0.886
1.110
2.070
1.375
2.621
2.889
2.163
2.456

1.316
1.639
0.886
1.110
          iWinter
                         M04
        0.1700 i 0.003 i 0.004 i 0.011 i 0.013 i 0.025 i 0.055 i 0.171 i 0.522 i 0.827 i 1.097 i 1.324



Statistics fc





Spring
Summer






Fall | | |
>r individual counties (2008-2010)
1 01 109 JAL I Pike
1 06037 |CA | Los Angeles

12057
FL
Hillsbo rough
|13015 |GA |Bartow
J13215 JGA jMuscogee













25
131
79
11
12
101
106
105

1
4
3
1
1
0.1904
0.1597
0.1538

0.5771
0.2521
0.1940
0.0125
0.0367
0.001
0.002
0.002

0.223
0.023
0.011
0.009
0.014
0.002
0.004
0.004

0.223
0.023
0.011
0.009
0.014
0.013
0.010
0.009

0.247
0.036
0.015
0.009
0.014
0.016
0.015
0.012

0.256
0.041
0.037
0.009
0.020
0.028
0.028
0.032

0.302
0.055
0.063
0.011
0.022
0.060
0.058
0.066

0.574
0.078
0.110
0.013
0.031
0.186
0.174
0.170

0.719
0.237
0.249
0.014
0.052
0.502
0.520
0.462

1.088
0.543
0.423
0.015
0.066
0.874
0.788
0.630

1.178
0.832
0.582
0.016
0.070
1.231
0.989
0.960

1.210
2.452
1.770
0.016
0.070
1.740
1.104
1.161

1.210
2.489
1.770
0.016
0.070
November 2012
3-152
Draft - Do Not Cite or Quote

-------
Year





i State/
Season i
| County
1 17031
J17115
17119
J17143
J17195
State
IL
IL
IL
IL
IL
County name
Cook
Macon
Madison
Peoria
Whiteside
Site ID !me^n°s
i9
J10
36
J20
ho
Hs«es
1
1
1
2
1
Mean
0.1364
0.0806
0.1346
0.0121
0.0191
Min
0.068
0.048
0.027
0.010
0.012
1
0.068
0.048
0.027
0.010
0.012
5
0.068
0.048
0.035
0.010
0.012
10
0.068
0.052
0.036
0.010
0.014
25
0.109
0.067
0.063
0.011
0.016
50
0.135
0.080
0.113
0.012
0.019
75
0.150
0.088
0.207
0.014
0.022
90
0.241
0.117
0.283
0.015
0.025
95
0.241
0.123
0.341
0.016
0.025
99
0.241
0.123
0.416
0.016
0.025
max
0.241
0.123
0.416
0.016
0.025


























17201
18035
18089
18097
18127
19155
20169
21151
26067
27003
27037
27145
29093
IL
IN
IN
IN
IN
IA
KS
KY
Ml
MN
MN
MN
MO
Winnebago
Delaware
Lake
Marion
Porter
Pottawattamie
Saline
Madison
Ionia
Anoka
Dakota
Stearns
Iron













9
57
46
66
10
12
9
10
10
10
36
10
158
1
2
2
2
1
1
1
1
1
1
1
1
6
0.0356
0.2866
0.0305
0.0198
0.0131
0.1581
0.2286
0.0212
0.1980
0.0161
0.2026
0.0032
0.3465
0.019
0.053
0.007
0.005
0.007
0.034
0.096
0.013
0.106
0.006
0.068
0.000
0.010
0.019
0.053
0.007
0.005
0.007
0.034
0.096
0.013
0.106
0.006
0.068
0.000
0.011
0.019
0.059
0.011
0.006
0.007
0.034
0.096
0.013
0.106
0.006
0.072
0.000
0.019
0.019
0.073
0.012
0.007
0.007
0.067
0.096
0.014
0.110
0.008
0.088
0.001
0.022
0.021
0.090
0.016
0.011
0.007
0.113
0.107
0.015
0.128
0.010
0.104
0.002
0.033
0.027
0.159
0.027
0.014
0.013
0.153
0.231
0.017
0.212
0.013
0.216
0.004
0.142
0.057
0.246
0.036
0.025
0.017
0.220
0.324
0.024
0.259
0.022
0.248
0.004
0.549
0.063
0.495
0.040
0.036
0.020
0.246
0.421
0.037
0.273
0.029
0.357
0.005
0.901
0.063
1.867
0.057
0.043
0.022
0.263
0.421
0.049
0.284
0.031
0.415
0.005
1.167
0.063
2.163
0.129
0.079
0.022
0.263
0.421
0.049
0.284
0.031
0.429
0.005
2.076
0.063
2.163
0.129
0.079
0.022
0.263
0.421
0.049
0.284
0.031
0.429
0.005
2.456








29099
29179
31053
31127
MO
MO
NE
NE
Jefferson
Reynolds
Dodge
Nemaha




423
40
7
6
19
4
1
1
0.4925
0.0397
0.0474
0.0447
0.023
0.012
0.019
0.019
0.033
0.012
0.019
0.019
0.050
0.014
0.019
0.019
0.071
0.015
0.019
0.019
0.187
0.017
0.020
0.024
0.385
0.031
0.060
0.032
0.723
0.057
0.067
0.075
0.989
0.087
0.072
0.087
1.186
0.089
0.072
0.087
2.017
0.100
0.072
0.087
2.889
0.100
0.072
0.087




36071
39035
NY
OH
Orange
Cuyahoga


99
70
3
3
0.0271
0.0905
0.003
0.006
0.003
0.006
0.004
0.010
0.005
0.011
0.007
0.021
0.027
0.050
0.037
0.122
0.068
0.221
0.075
0.287
0.086
0.531
0.086
0.531
                    139051
iOH    i Fulton
i30
0.1609 i 0.025 i 0.025 i 0.027 i 0.046 i 0.054 i 0.092 i 0.254 i 0.354 i 0.453 i 0.567 i 0.567




39091
39101
OH
OH
Logan
Marion


100
8
4
1
0.0499
0.0379
0.004
0.032
0.004
0.032
0.004
0.032
0.006
0.032
0.033
0.034
0.047
0.037
0.072
0.042
0.090
0.047
0.095
0.047
0.100
0.047
0.100
0.047
November 2012
                     3-153
             Draft - Do Not Cite or Quote

-------
Year





i State/
Season i
| County
1 391 51
| 391 55
40121
J41071
I 42003
State
OH
OH
OK
OR
PA
County name
Stark
Trumbull
Pittsburg
Yamhill
Allegheny
Site ID !me^n°s
i9
J6
9
|10
J20
Hs«es
1
1
1
1
2
Mean
0.0180
0.0080
0.0021
0.0166
0.0414
Min
0.015
0.005
0.002
0.009
0.009
1
0.015
0.005
0.002
0.009
0.009
5
0.015
0.005
0.002
0.009
0.011
10
0.015
0.005
0.002
0.011
0.012
25
0.016
0.006
0.002
0.013
0.017
50
0.018
0.008
0.002
0.016
0.030
75
0.019
0.010
0.002
0.019
0.054
90
0.023
0.011
0.003
0.026
0.099
95
0.023
0.011
0.003
0.027
0.120
99
0.023
0.011
0.003
0.027
0.138
max
0.023
0.011
0.003
0.027
0.138


























42007
42011
42045
42055
42063
42079
42129
47093
47163
48085
51770
55117
72013
PA
PA
PA
PA
PA
PA
PA
TN
TN
TX
VA
Wl
PR
Beaver
Berks
Delaware
Franklin
Indiana
Luzerne
Westmoreland
Knox
Sullivan
Collin
Roanoke City
Sheboygan
Arecibo
(Puerto Rico)













41
105
10
7
10
6
10
44
118
108
10
10
10
3
6
1
1
1
1
1
2
4
3
1
1
1
0.1160
0.0995
0.0447
0.0447
0.0447
0.1078
0.0434
0.0165
0.0554
0.3101
0.0466
0.0897
0.1725
0.043
0.038
0.043
0.043
0.043
0.084
0.041
0.007
0.030
0.048
0.013
0.012
0.059
0.043
0.039
0.043
0.043
0.043
0.084
0.041
0.007
0.030
0.051
0.013
0.012
0.059
0.052
0.041
0.043
0.043
0.043
0.084
0.041
0.009
0.033
0.070
0.013
0.012
0.059
0.056
0.045
0.043
0.043
0.043
0.084
0.042
0.009
0.035
0.085
0.016
0.034
0.068
0.083
0.051
0.043
0.043
0.043
0.085
0.042
0.012
0.039
0.120
0.019
0.058
0.129
0.114
0.078
0.045
0.045
0.044
0.103
0.044
0.016
0.045
0.217
0.026
0.076
0.194
0.159
0.145
0.046
0.046
0.046
0.135
0.044
0.020
0.060
0.469
0.097
0.126
0.213
0.170
0.183
0.047
0.046
0.049
0.137
0.046
0.023
0.100
0.682
0.108
0.164
0.241
0.187
0.197
0.047
0.046
0.049
0.137
0.046
0.027
0.125
0.753
0.109
0.170
0.245
0.206
0.242
0.047
0.046
0.049
0.137
0.046
0.035
0.134
1.189
0.109
0.170
0.245
0.206
0.251
0.047
0.046
0.049
0.137
0.046
0.035
0.168
1.262
0.109
0.170
0.245
November 2012
3-154
Draft - Do Not Cite or Quote

-------
Year
! State/
Season i
| County
State
County name
ojtpin |N:mo
Slte ID imeans
N sites
Mean
Min
1
5
10
25
50
75
90
95
99
max
Statistics for individual sites where overall average monthly mean > national 90th percentile (2008-2010)
                                                          011090003
                                                                         25
                                                                                          0.5771   0.223  0.223  0.247  0.256  0.302   0.574  0.719  1.088  1.178  1.210   1.210
                                                          060371405!
                    36
       i 0.7174 i 0.188 i  0.188 i 0.234 i 0.237 i 0.309 i 0.476 i 0.791 i 2.178 i  2.452 i 2.489 i 2.489













































290930016
290930021
290990004
290990015"
290990020"
290990021"
290999001"
290999005"
480850009"
36
36
36
21
29
21
22
22
36









0.6682
0.6950
1 .0090
1.3162
0.6680
0.7317
0.8413
0.9875
0.6068
0.207
0.173
0.640
0.612
0.452
0.429
0.587
0.612
0.196
0.207
0.173
0.640
0.612
0.452
0.429
0.587
0.612
0.196
0.258
0.192
0.655
0.632
0.471
0.435
0.592
0.630
0.268
0.313
0.218
0.699
0.743
0.482
0.507
0.600
0.644
0.335
0.418
0.346
0.775
0.921
0.555
0.547
0.699
0.783
0.469
0.543
0.689
0.913
1.074
0.651
0.685
0.845
0.995
0.585
0.634
0.954
1.081
1.258
0.754
0.900
0.963
1.220
0.704
1.167
1.214
1.555
2.621
0.891
0.999
1.061
1.271
0.965
2.076
1.275
2.011
2.634
0.943
1.013
1.100
1.278
1.189
2.456
1.937
2.017
2.889
0.989
1.141
1.204
1.375
1.262
2.456
1.937
2.017
2.889
0.989
1.141
1.204
1.375
1.262
aThe 3-month averages presented here were created using a simplified approach of the procedures detailed in 40 CFR part 50 appendix R and as such cannot be directly compared to the Pb NAAQS for
determination of compliance with the Pb NAAQS.
"Sites listed in the bottom six rows of this table fall in the upper 90th percentile of the data pooled by site.
November 2012
3-155
Draft - Do Not Cite or Quote

-------
Table 3-15 Distribution of 3-month moving average Pb-TSP concentrations (ug/m3) nationwide, non-source-
oriented monitors, 2008-2010.
Year
JO*«*»I I I
Season! 1ST JCountyname Site ID N: mo N Mean Min 1 5 10 25 50 75 90 95 99 max
County | means sites
Nationwide statistics
2008-2010
2008




2,164

0.0120
0.000
0.000
0.001
0.002
0.005
0.010
0.015
0.025
0.037
0.048
0.073
III I I 663 I I 0.0130 I 0.000 I 0.000 I 0.001 I 0.002 I 0.005 I 0.011 I 0.016 I 0.027 I 0.040 I 0.050 I 0.055
2009
2010






Winter
Spring
Summer
Fall
Nationwide statistics, p
2008-2010
2008
2009
2010








>oolec










i by site















727
774
494
548
565
557












86
59
65
71
0.0114
0.0118
0.0113
0.0119
0.0121
0.0126

0.0120
0.0127
0.0117
0.0118
0.000
0.000
0.000
0.000
0.000
0.000

0.000
0.001
0.001
0.000
0.000
0.000
0.001
0.000
0.000
0.000

0.000
0.001
0.001
0.000
0.001
0.001
0.001
0.001
0.001
0.001

0.001
0.002
0.001
0.001
0.002
0.001
0.002
0.002
0.002
0.002

0.002
0.003
0.003
0.001
0.004
0.005
0.005
0.005
0.005
0.005

0.005
0.005
0.004
0.005
0.009
0.010
0.009
0.009
0.010
0.011

0.010
0.011
0.010
0.010
0.014
0.016
0.014
0.015
0.016
0.017

0.016
0.016
0.014
0.017
0.024
0.025
0.023
0.025
0.026
0.027

0.024
0.024
0.026
0.022
0.038
0.035
0.037
0.036
0.037
0.037

0.034
0.043
0.031
0.028
0.043
0.047
0.050
0.050
0.046
0.048

0.046
0.050
0.049
0.045
0.073
0.057
0.055
0.073
0.053
0.057

0.046
0.050
0.049
0.045
      iWinter
                84  i  0.0118  i 0.000 i  0.000 i  0.001 i  0.002 i  0.005  i 0.010  i 0.015  i 0.025 i 0.036 i  0.048 i  0.048



Statist





Spring
Summer




Fall | |
ics for individual counties (2008-2010
06025JCA jlmperial
06037 1 CA | Los Angeles
06065
CA
Riverside
06071 |CA | San Bernardino
08005JCO jArapahoe



)









31
218
72
69
7
83
86
86

1
8
2
2
1
0.0118
0.0118
0.0126

0.0165
0.0100
0.0078
0.0091
0.0126
0.000
0.000
0.000

0.007
0.000
0.002
0.003
0.011
0.000
0.000
0.000

0.007
0.000
0.002
0.003
0.011
0.001
0.001
0.001

0.008
0.002
0.004
0.005
0.011
0.002
0.002
0.002

0.011
0.004
0.005
0.006
0.011
0.004
0.005
0.005

0.013
0.006
0.007
0.007
0.011
0.010
0.009
0.011

0.017
0.009
0.008
0.009
0.013
0.015
0.016
0.016

0.021
0.013
0.010
0.011
0.014
0.025
0.023
0.026

0.023
0.016
0.011
0.013
0.014
0.034
0.037
0.030

0.023
0.020
0.011
0.014
0.014
0.059
0.043
0.046

0.023
0.028
0.011
0.017
0.014
0.059
0.043
0.046

0.023
0.035
0.011
0.017
0.014
November 2012
3-156
Draft - Do Not Cite or Quote

-------
Year





istate/
Season i
| County
08031
13089
17031
17117
17119
ST
CO
GA
IL
IL
IL
County name
Denver
DeKalb
Cook
Macoupin
Madison
Site ID





N:mo
means
10
8
287
24
36
N
sites
1
1
8
1
1
Mean
0.0054
0.0035
0.0196
0.0101
0.0188
Min
0.004
0.003
0.010
0.010
0.010
1
0.004
0.003
0.010
0.010
0.010
5
0.004
0.003
0.010
0.010
0.010
10
0.004
0.003
0.010
0.010
0.011
25
0.005
0.003
0.012
0.010
0.014
50
0.006
0.004
0.017
0.010
0.016
75
0.006
0.004
0.025
0.010
0.022
90
0.006
0.004
0.033
0.011
0.036
95
0.006
0.004
0.038
0.011
0.036
99
0.006
0.004
0.047
0.011
0.039
max
0.006
0.004
0.051
0.011
0.039

















17143
17163
18089
18097
18163
25025
26081
26163
27017
27037
27053
27075
27123
27137
27163
29097
29187
IL
IL
IN
IN
IN
MA
Ml
Ml
MN
MN
MN
MN
MN
MN
MN
MO
MO
Peoria
Saint Clair
Lake
Marion
Vanderburgh
Suffolk
Kent
Wayne
Carlton
Dakota
Hennepin
Lake
Ramsey
Saint Louis
Washington
Jasper
Saint Francois

















36
36
36
33
31
24
10
32
10
112
124
8
65
70
70
10
21
1
1
1
1
2
2
1
2
1
5
4
1
3
2
3
1
2
0.0105
0.0204
0.0149
0.0056
0.0047
0.0093
0.0055
0.0119
0.0000
0.0036
0.0033
0.0000
0.0061
0.0016
0.0017
0.0135
0.0337
0.010
0.012
0.007
0.003
0.002
0.005
0.004
0.004
0.000
0.000
0.000
0.000
0.001
0.000
0.000
0.009
0.011
0.010
0.012
0.007
0.003
0.002
0.005
0.004
0.004
0.000
0.000
0.001
0.000
0.001
0.000
0.000
0.009
0.011
0.010
0.012
0.007
0.003
0.003
0.006
0.004
0.004
0.000
0.001
0.001
0.000
0.001
0.000
0.000
0.009
0.012
0.010
0.014
0.007
0.003
0.003
0.006
0.005
0.005
0.000
0.001
0.001
0.000
0.001
0.000
0.000
0.011
0.012
0.010
0.016
0.010
0.004
0.004
0.008
0.005
0.005
0.000
0.001
0.002
0.000
0.002
0.001
0.001
0.012
0.027
0.010
0.020
0.014
0.005
0.005
0.009
0.006
0.012
0.000
0.003
0.003
0.000
0.005
0.001
0.001
0.014
0.035
0.011
0.024
0.018
0.007
0.005
0.011
0.006
0.017
0.000
0.005
0.004
0.000
0.008
0.002
0.003
0.015
0.042
0.012
0.029
0.024
0.009
0.006
0.013
0.006
0.021
0.000
0.007
0.006
0.000
0.014
0.004
0.004
0.016
0.048
0.012
0.033
0.032
0.010
0.007
0.015
0.006
0.023
0.000
0.012
0.006
0.000
0.016
0.004
0.004
0.017
0.053
0.013
0.036
0.037
0.011
0.007
0.016
0.006
0.024
0.000
0.013
0.015
0.000
0.017
0.005
0.005
0.017
0.054
0.013
0.036
0.037
0.011
0.007
0.016
0.006
0.024
0.000
0.015
0.016
0.000
0.017
0.005
0.005
0.017
0.054


29189
36047
MO
NY
Saint Louis
Kings


33
24
1
1
0.0243
0.0131
0.005
0.011
0.005
0.011
0.005
0.011
0.006
0.011
0.007
0.012
0.008
0.013
0.050
0.014
0.050
0.016
0.050
0.018
0.055
0.019
0.055
0.019
                39017iOH  iButler
       30
0.0055  i 0.003 i 0.003 i  0.004 i  0.004 i  0.005 i  0.006  i 0.006  i 0.007  i 0.007 i 0.008 i 0.008


39029
39035
OH
OH
Columbiana
Cuyahoga


105
105
3
3
0.0148
0.0144
0.005
0.005
0.005
0.006
0.007
0.006
0.008
0.008
0.010
0.010
0.013
0.013
0.017
0.018
0.021
0.023
0.028
0.027
0.054
0.033
0.057
0.035
November 2012
3-157
    Draft - Do Not Cite or Quote

-------
Year





istate/
Season i
| County
39049
39143
39167
40115
42003
ST
OH
OH
OH
OK
PA
County name
Franklin
Sandusky
Washington
Ottawa
Allegheny
Site ID





N:mo
means
36
10
48
12
36
N
sites
1
1
2
2
1
Mean
0.0092
0.0052
0.0047
0.0128
0.0101
Min
0.005
0.004
0.002
0.005
0.000
1
0.005
0.004
0.002
0.005
0.000
5
0.005
0.004
0.002
0.005
0.000
10
0.005
0.004
0.003
0.006
0.000
25
0.008
0.005
0.004
0.010
0.007
50
0.010
0.005
0.004
0.014
0.012
75
0.011
0.006
0.006
0.016
0.014
90
0.011
0.006
0.007
0.018
0.016
95
0.012
0.006
0.007
0.019
0.018
99
0.012
0.006
0.008
0.019
0.025
max
0.012
0.006
0.008
0.019
0.025









42021
42045
42101
42129
48061
48141
48201
48479
49035
PA
PA
PA
PA
TX
TX
TX
TX
UT
Cambria
Delaware
Philadelphia
Westmoreland
Cameron
El Paso
Harris
Webb
Salt Lake









23
14
22
24
33
56
30
23
10
1
1
1
1
1
3
1
1
1
0.0459
0.0427
0.0214
0.0417
0.0042
0.0212
0.0051
0.0121
0.0145
0.040
0.040
0.013
0.037
0.002
0.014
0.004
0.006
0.007
0.040
0.040
0.013
0.037
0.002
0.014
0.004
0.006
0.007
0.040
0.040
0.014
0.040
0.003
0.014
0.004
0.007
0.007
0.040
0.040
0.014
0.040
0.003
0.015
0.004
0.007
0.007
0.040
0.040
0.018
0.040
0.004
0.016
0.005
0.008
0.008
0.041
0.042
0.022
0.041
0.004
0.018
0.005
0.010
0.011
0.046
0.045
0.025
0.043
0.005
0.023
0.006
0.016
0.016
0.069
0.046
0.029
0.046
0.005
0.038
0.006
0.021
0.032
0.070
0.047
0.029
0.047
0.006
0.040
0.007
0.022
0.036
0.073
0.047
0.030
0.048
0.006
0.040
0.007
0.026
0.036
0.073
0.047
0.030
0.048
0.006
0.040
0.007
0.026
0.036
Statistics for individual sites where overall average monthly mean > national 90th percentile (2008-2010)




































170310022
170310026
170316003
291870006"
291870007"
291892003"
420210808"
420450002"
421290007"
36
36
36
10
11
33
23
14
24









0.0335
0.0281
0.0245
0.0412
0.0268
0.0243
0.0459
0.0427
0.0417
0.016
0.018
0.015
0.017
0.011
0.005
0.040
0.040
0.037
0.016
0.018
0.015
0.017
0.011
0.005
0.040
0.040
0.037
0.018
0.019
0.015
0.017
0.011
0.005
0.040
0.040
0.040
0.026
0.022
0.017
0.026
0.012
0.006
0.040
0.040
0.040
0.028
0.023
0.020
0.035
0.012
0.007
0.040
0.040
0.040
0.032
0.026
0.025
0.043
0.028
0.008
0.041
0.042
0.041
0.038
0.032
0.028
0.048
0.035
0.050
0.046
0.045
0.043
0.047
0.038
0.031
0.054
0.036
0.050
0.069
0.046
0.046
0.048
0.043
0.035
0.054
0.041
0.050
0.070
0.047
0.047
0.051
0.046
0.036
0.054
0.041
0.055
0.073
0.047
0.048
0.051
0.046
0.036
0.054
0.041
0.055
0.073
0.047
0.048
aThe 3-month averages presented here were created using a simplified approach of the procedures detailed in 40 CFR part 50 appendix R and as such cannot be directly compared to the Pb NAAQS for
determination of compliance with the Pb NAAQS.
"Sites listed in the bottom six rows of this table fall in the upper 90th percentile of the data pooled by site.
November 2012
3-158
Draft - Do Not Cite or Quote

-------
Table 3-16   Distribution of annual 1-month site maxima TSP Pb concentrations (ug/m3) nationwide, source-
              oriented monitors, 2008-2010.
Year
jSitelD-year
N (sites)
Mean
Min
10
25
50
75
90
95
99
                                                                                                                      max
Nationwide statistics
2008-2010
2008
2009



111
47
54
0.5003
0.8138
0.4486
0.003
0.012
0.016
0.006
0.012
0.016
0.016
0.052
0.022
0.032
0.057
0.050
0.066
0.096
0.090
0.156
0.320
0.170
0.575
0.850
0.618
1.530
2.557
1.280
2.416
3.123
1.623
4.225
4.440
2.438
4.440
4.440
2.438
2010

101
0.3105
0.003
0.006
0.008
0.024
0.054
0.142
0.347
0.854
1.117
1.576
1.828
Annual site max 1 -month means >= national 90th percentile (2008-20010)








060371405-2008
180350009-2008
290930016-2008
290930021-2008
290930021-2009
290990004-2008
290990004-2009
290990004-2010








2.8800
4.4400
4.2252
2.5566
2.4380
2.4156
1.5599
1.5762
























































































1 29099001 1-2008 1.5295
1 29099001 5a-2008 3.1228


290990020a-2008
29099002 1a-2008


2.2204
1 .5528






















i 290999001 a-2009 1.6228
1 290999001 a-2010



290999005a-2009
290999005a-2010
480850009a-2008



1 .5576
1 .9850
1 .8278
1 .5640












































aSites listed in the bottom eight rows of this table fall in the upper 90th percentile of the data pooled by site.
November 2012
                           3-159
                          Draft - Do Not Cite or Quote

-------
Table 3-17   Distribution of annual 1-month site maxima TSP Pb concentrations (ug/m3) nationwide, non-source-
              oriented monitors, 2008-2010.
Year
jSitelD-year
N (sites)
Mean
Min
10
25
50
75
90
95
99
                                                                                                                          max
Nationwide statistics
2008-2010
2008
2009



88
59
66
0.0284
0.0232
0.0210
0.000
0.004
0.003
0.000
0.004
0.003
0.004
0.005
0.005
0.006
0.006
0.006
0.010
0.010
0.008
0.020
0.016
0.014
0.041
0.033
0.026
0.057
0.053
0.040
0.070
0.058
0.056
0.136
0.066
0.128
0.136
0.066
0.128
2010

73
0.0233
0.000
0.000
0.002
0.004
0.008
0.015
0.029
0.049
0.065
0.136
0.136
Annual site max 1-month means >= national 90th percentile (2008-2010)


170310022-2009
170310022-2010


0.0700
0.0620






















            1171193007-2008
                                 0.0660








291870006a-2010
291892003a-2008
39029001 9a-2010
390290022a-2010
42021 0808a-2008
42021 0808a-2009
481410002a-2010
481410033a-2009








0.0894
0.0660
0.1360
0.0652
0.0583
0.1280
0.0870
0.0570
























































































aSites listed in the bottom eight rows of this table fall in the upper 90th percentile of the data pooled by site.
November 2012
                            3-160
                           Draft - Do Not Cite or Quote

-------
Table 3-18   Distribution of annual 3-month site maxima Pb-TSP concentrations (ug/m3) nationwide, source-
               oriented monitors, 2008-2010.
Year
jSitelD-year
N (sites)
Mean
Min
10
25
50
75
90
95
99
                                                                                                                                  max
Nationwide statistics3
2008-2010
2007
2008
2009
Annual site max







3-month means >= nation
011090003-2008
060371405-2008
120571066-2008
106
47
54
96
al 90th percent



0.3605
0.5831
0.3611
0.2112
ile (2008-2C
1.2100
2.4890
1 .7700
0.003
0.009
0.012
0.003
110)



0.005
0.009
0.012
0.003




0.016
0.038
0.017
0.011




0.023
0.043
0.035
0.021




0.047
0.085
0.060
0.046




0.109
0.242
0.121
0.091




0.378
0.815
0.467
0.262




1.204
2.017
1.079
0.630




1.937
2.456
1.258
0.865




2.489
2.889
2.070
1.375




2.889
2.889
2.070
1.375








180350009-2008
29093001 6b-2008
29093001 6b-2009
29093002 1b-2009




2.1630
2.4560
2.0700
1 .9370
















































290990004b-2008
29099001 5b-2008
290999001 b-2009
290999005b-2009




2.0170
2.8890
1 .2040
1 .2580












































J290999005b-2010
|480850009b-2008


1 .3750
1 .2620






















"The 3-month averages presented here were created using a simplified approach of the procedures detailed in 40 CFR part 50 appendix R and as such cannot be directly compared to the Pb NAAQS for
determination of compliance with the Pb NAAQS.
bSites listed in the bottom nine rows of this table fall in the upper 90th percentile of the data pooled by site.
November 2012
                              3-161
                             Draft - Do Not Cite or Quote

-------
Table 3-19    Distribution of annual 3-month site maxima Pb-TSP concentrations (ug/m3) nationwide, non-source-
                oriented monitors, 2008-2010.
Year
jSitelD-year
|  N (sites)  |  Mean   j  Min   j   1
10
25
50
75
90
95
99
                                                                                                                             max
Nationwide statistics3
2008-2010
                                     86
                                              0.0198    0.000   0.000   0.002   0.004   0.007    0.015    0.028    0.044   0.051    0.073   0.073
2008
2009
2010



59
65
71
0.0176
0.0162
0.0171
0.002
0.002
0.000
0.002
0.002
0.000
0.004
0.003
0.001
0.005
0.004
0.002
0.007
0.006
0.006
0.014
0.013
0.013
0.024
0.021
0.024
0.039
0.038
0.037
0.048
0.041
0.047
0.055
0.073
0.057
0.055
0.073
0.057
Annual site max 3-month means >= national 90th percentile (2008-2010)


170310022-2008
170310022-2009


0.0480
0.0470





























170310022-2010
170310026b-2008
291870006b-2010
291892003b-2008
39029001 9b-2010
390290022b-2010
420210808b-2008







0.0510
0.0460
0.0540
0.0550
0.0570
0.0440
0.0490
















































































420210808b-2009
420450002b-2010
421290007b-2008



0.0730
0.0470
0.0480

































"The 3-month averages presented here were created using a simplified approach of the procedures detailed in 40 CFR part 50 appendix R and as such cannot be directly compared to the Pb
NAAQS for determination of compliance with the Pb NAAQS.
b Sites listed in the bottom nine rows of this table fall in the upper 90th percentile of the data pooled by site.
November 2012
                                3-162
                                   Draft - Do Not Cite or Quote

-------
3
4
5
Concentrations of Pb Measured using PM10 Monitors (for Concentrations
and Trends)

Figure 3-38 displays maximum 3-month averages for Pb-PMi0 concentrations for 36
counties in which measurements were obtained. Among the 36 counties in which PM10
monitoring was conducted, only one county, Gila County, AZ, reported concentrations
above 0.076 ug/m3. Three other counties reported concentrations greater than
0.016 ug/m3: Wayne County, MI, Boyd County, KY, and the county of St. Louis City,
MO.
                        2007-2009 Pb-PM10 County Maximum 3-Month Mean
             Concentration:
             * >= 0.076 ng/m3 (1 county)
             * 0.016 -0.075 n£/m (3 counties)
             • 0.006 - 0.015 ug/m- (17 counties)
              <= .005 ug/m- (15 counties)
             _1 no data
     Figure 3-38     Highest county-level Pb-PMi0 concentrations (ug/m ), maximum
                      3-month average, 2007-2009.
    November 2012
                         3-163
Draft - Do Not Cite or Quote

-------
1
2
3
4
5
Concentrations of Pb Measured using PM2.s Monitors (for Speciation
Concentrations and Trends)

Figure 3-39 displays maximum 3-month average county-level data for Pb in PM2 5
concentrations for 323 counties in which PM2 5 measurements were obtained for
speciation in the CSN and IMPROVE networks. The data presented here are not
compared to the NAAQS because PM2 5 monitors are not deployed for the purpose of
evaluating compliance for the NAAQS. Among the 323 counties in which PM25
monitoring was conducted, only eleven counties reported concentrations greater than
0.016 ug/m3: Jefferson, AL, San Bernardino, CA, Imperial, CA, Wayne, MI, Jefferson,
MO, Erie, NY, Lorain, OH, Allegheny, PA, Berks, PA, Davidson, TN, and El Paso, TX.
                         2007-2009 Pb-PM2 5 County Maximum 3-Month Mean
                                                         .v -?       -V " __--
                                                           V *    	-- - -^~~ M
                                                  • 0.016 -0.075 pg/m-ill counties)
                                                  • 0.006 -0.015 yg/m!(71counties)
                                                  <= .005 H£/m: (241 counties)
    Figure 3-39      Highest county-level Pb-PM2.s concentrations (ug/m ), maximum
                      3-month average, 2007-2009.
    November 2012
                         3-164
Draft - Do Not Cite or Quote

-------
      3.8.2        Intra-urban Variability

 1                  Intra-urban variability in Pb concentrations reported to AQS was described in detail for
 2                  Los Angeles County, CA (Los Angeles), Hillsborough and Pinellas Counties, FL
 3                  (Tampa), Cook County, IL (Chicago), Jefferson County, MO (Herculaneum), Cuyahoga
 4                  County, OH (Cleveland), and Sullivan County, TN (Bristol) were selected for this
 5                  assessment to illustrate the variability in Pb concentrations measured across different
 6                  metropolitan regions with varying Pb source characteristics. Four of the counties
 7                  encompass large cities (Los Angeles, Tampa, Chicago, and Cleveland). All six counties
 8                  contain source-oriented monitors. Maps and wind roses (graphs representing wind
 9                  direction and wind speed at a location) are presented in this Chapter 3 Appendix for each
10                  of the six urban areas. Additionally, annual and seasonal box plots of the Pb
11                  concentration distributions and intra-monitor correlation tables are presented to illustrate
12                  the level of variability throughout each urban area.

13                  Maps of six areas (Los Angeles County, CA; Hillsborough/Pinellas Counties, FL; Cook
14                  County, IL; Jefferson County, MO; Cuyahoga County, OH; and Sullivan County, TN) are
15                  shown to illustrate the location of all Pb monitors meeting the inclusion criteria. Wind
16                  roses for each season are also provided to help put the source concentration data in
17                  context. Letters on the maps identify the individual monitor locations and correspond
18                  with the letters provided in the accompanying concentration box plots and pair-wise
19                  monitor comparison tables. The box plots for each monitor include the annual and
20                  seasonal concentration  median and interquartile range with whiskers extending from the
21                  5th to the 95th percentile. Data from 2008-2010 were used to generate the box plots,
22                  which are stratified  by season as follows:  1 = winter (December-February), 2 = spring
23                  (March-May), 3 = summer (June-August), and 4 = fall (September-November). The
24                  comparison tables include the Pearson correlation coefficient (R), Spearman rank-ordered
25                  correlation coefficient (p), the 90th percentile of the absolute difference in concentrations
26                  (P90) in  (ig/m3, the  coefficient of divergence (COD) and the straight-line distance
27                  between monitor pairs (d) in km. The COD provides an indication of the variability
28                  across the monitoring sites within each county and is defined as follows:
                                             CODJk =
                                                          p /_, \Xt] + Xtk
                                                            '-1
                                                                                         Equation 3A-1
29                  where Xtj and Xik represent the observed hourly concentrations for time period / at sites j
30                  and k, and/? is the number of paired hourly observations. A COD of 0 indicates there are


      November 2012                            3-165                  Draft - Do Not Cite or Quote

-------
 1                  no differences between concentrations at paired sites (spatial homogeneity), while a COD
 1                  approaching 1 indicates extreme spatial heterogeneity.

 3                  In certain cases, the information contained in these figures and tables should be used with
 4                  some caution since many of the reported concentrations forthe years 2008-2010 are near
 5                  or below the analysis method's stated method detection limit (MDL). The MDL is
 6                  generally taken as 0.01 because it is the upper value of the range of MDLs reported for
 7                  atomic absorption (AA) and Emissions Spectra ICAP methods, which were the two
 8                  methods reported in the AQS to have been used for analysis of FRM samples (Rice.
 9                  2007). Generally, data are reported to the hundredth place, so this assumption is
10                  reasonable. The approximate percentage of data below the MDL (to the nearest 5%) is
11                  provided for each site along with box plots of seasonal Pb concentration at monitors
12                  within each urban area studied.

13                  Figure 3-40 illustrates Pb monitor locations within Los Angeles County, CA. Ten
14                  monitors are located within Los Angeles County, five of which were source-oriented and
15                  the other five were non-source-oriented monitors. Monitor A was located immediately
16                  downwind of the Quemetco battery recycling facility in the City of Industry, CA. This
17                  source was estimated to produce 0.32 tons of Pb/yr (U.S. EPA. 2008c). Monitor C was
18                  sited in a street canyon just upwind of the Exide Pb recycling facility, which was
19                  estimated to produce  2.0 tons of Pb/yr (U.S. EPA. 2008c). Monitor D was situated
20                  slightly northwest of the same Pb recycling facility. It is still in relatively close proximity
21                  but not downwind on most occasions. Monitor B was located 12 km downwind of the
22                  Exide facility. Monitor E was located nearby the Trojan Battery recycling facility, which
23                  emitted 0.79 tons Pb/yr (U.S. EPA. 2008c). Location of the non-source-oriented monitors
24                  varied. Monitor F was positioned on a rooftop 60 meters away from a 4-lane arterial road
25                  and 100 meters from  of a railroad. Monitor G was located on a rooftop approximately
26                  20 meters from an 8-lane arterial road, and monitor H was positioned at the curbside of a
27                  four-lane road roughly  650 meters north of that road's junction with Interstate  1-405.
28                  Monitor I was sited in a parking lot roughly 80 meters from a four-lane road, and monitor
29                  J was located approximately 130 meters south of a 4-lane highway. Figure 3-41 displays
30                  seasonal wind roses for Los Angeles County. In spring, summer, and fall, the
31                  predominant winds come from the west-southwest. During winter, wind direction varies
32                  with a portion from the west-southwest and the remainder from the east. The highest
33                  winds during winter come more frequently from the west-southwest.

34                  The maps shown in Figure 3-40 for source-oriented monitors A-E illustrate the different
35                  conditions captured by the monitors; this informs analysis of the seasonal and year-round
36                  concentrations reported in Figure 3-42. The average annual concentration  at monitor A
37                  was 0.074 ug/m3. The 95th percentile exceeded the level of the NAAQS in the spring
      November 2012                            3-166                 Draft - Do Not Cite or Quote

-------
 1                   (0.16 ug/m3) and summer (0.18 ug/m3). Monitor C reported the highest concentrations in
 2                   Los Angeles County, with a year-round mean of 0.68 ug/m3. Given the position of this
 3                   monitor with respect to the Exide facility, there is the potential for recirculation of
 4                   fugitive Pb emissions in the air sampled by that monitor. The average annual Pb
 5                   concentration at monitor D was 0.12 ug/m3, and the 75th percentile of year-round data
 6                   exceeded the level of the NAAQS; in spring, the 70th percentile exceeded 0.15  ug/m3.
 7                   Monitor B reported the lowest values among the source-oriented monitors with  an
 8                   average annual concentration of 0.013 ug/m3. Note that 75% of reported values  were
 9                   below the MDL for this site, and no data from this site exceeded the level of the NAAQS.
10                   The annual average concentration at monitor E was 0.068 ug/m3, and the 95th percentile
11                   of concentration was 0.17 ug/m3.

12                   The non-source-oriented monitors located at sites F-J all recorded low concentrations,
13                   with average values ranging from 0.004 to 0.018 ug/m3 (Figure 3-42). The highest
14                   average year-round concentrations were recorded at site F. The 95th percentiles at these
15                   sites ranged from 0.01 to 0.04 ug/m3.  There is much less certainty in the data recorded at
16                   the non-source-oriented sites, because 45-95% of the data from these monitors were
17                   below the MDL. Additionally, only one of the non-source-oriented monitors (monitor H)
18                   was positioned at roadside, and none of the non-source-oriented monitors were located at
19                   the side of a major highway.

20                   Intersampler correlations (Table 3-20). illustrate that Pb has high intra-urban spatial
21                   variability. For the source-oriented monitors, the highest correlation (R = 0.59, p = 0.57)
22                   occurred for monitors C and D, which covered the same site. Because monitor D was
23                   slightly farther from the Exide source and slightly upstream of the predominant wind
24                   direction, the signal it received from the source  site was correspondingly lower. Hence,
25                   the correlation between these sites was moderate despite their relatively close proximity.
26                   In general, low or even negative correlations were observed between the source-oriented
27                   and non-source-oriented monitors. The  exception to this was the Spearman-ranked
28                   correlation between source-oriented monitor B and non-source-oriented monitor F, with
29                   p = 0.74. Pearson correlation was much lower for this pair (R = 0.33). Monitors B and F
30                   are roughly  16 km apart, whereas monitor B is only 12 km from monitors D and C, 8 km
31                   from monitor E, and 6 km from monitor A. It is possible that monitors B  and F both
32                   captured a source that was either longer in range or more ubiquitous and so would have
33                   been obscured by the stronger source  signals at  sites A, C,  D, and E. Comparisons
34                   between the non-source-oriented monitors revealed moderate correlation between sites
35                   (G to J  [R = 0.29 to 0.71, p = 0.37 to 0.65]). Sites G, H, I and J are all located in the
36                   southwestern quadrant of Los Angeles.  It is possible that they are also exposed to a
37                   ubiquitous source that produces a common signal at these four sites.
      November 2012                             3-167                  Draft - Do Not Cite or Quote

-------
                                 Legend
                                 0  TSP Source Monitors
                                 •  TSP Non-source Monitors
                                 •  City-based Population Center
                                 •  County-based Population Center
                                 	Inters) ate s
                                 	 Mitjof Highways
                                    Bodies o( Waler
                                    Urban Areas
                                           V.CA
       0   10  20     40 Kilometers

Note: Monitor locations are denoted by green markers, and source locations are denoted by red markers. Top: view of all Pb FRM
monitors in Los Angeles County. Bottom left: Close up of the industrial site near monitors C and D. Bottom right: Close up of the
populated area captured by monitor F.
Figure 3-40      Pb TSP monitor and source locations within  Los Angeles County,
                    CA (06-037), 2007-2009.
November 2012
3-168
Draft - Do Not Cite or Quote

-------
       I
       i
Note: Clockwise from top left: January, April, July, and October. Note that the wind percentages vary from month to month.
Source: NRCS (2011).

Figure 3-41     Wind roses for Los Angeles County, CA, from meteorological data
                 at the Los Angeles  International Airport, 1961-1990.
November 2012
3-169
Draft - Do Not Cite or Quote

-------
Site
SITE ID
MEAN
SD
OBS
% BELOW
MDL
Source
orientation
A
06-037-
1404
0.074
0.040
66
0
Source
B
06-037-
1602
0.013
0.017
112
75
Source
C
06-037-
1405
0.68
1.0
617
0
Source
D
06-037-
1406
0.12
0.092
242
0
Source
E
06-037-
1403
0.068
0.052
128
0
Source
F
06-037-
1103
0.018
0.011
121
45
Non-
source
G
06-037-
1301
0.015
0.012
108
65
Non-
source
H
06-037-
4002
0.0083
0.0068
120
85
Non-
source
1
06-037-
4004
0.0087
0.0069
117
85
Non-
source
J
06-037-
5005
0.0040
0.0064
109
95
Non-
source
 E
15
 01

 C.
 o
J.U -
2.9-
2.8 -
2.7
2.6-
2.5
2.4-
2.3 -
2.2-
2.1 -
2.0-
1.9-
1.7-
1.6
1.4-
1.3 -
1.1 -
1.0-
0.9-
0.8 -
0.7
0.6-
0.5 -
0.4-
0.3 -
0.2-
0.1 -
0.0 -
A






















*lt*+
B






















= , 4
























C















































D






















m
E






















iii = i
F























G






















*i
H























I























J























       Y1
         234 Y1234  Y1234 Y1234  Y1234 Y1234  Y1234 Y1234  Y1234 Y1234
                                     season
Figure 3-42     Box plots of annual and seasonal 24-h Pb TSP concentrations
               (ug/m3) from source-oriented and non-source-oriented monitors
               within Los Angeles County, CA (06-037), 2007-2009.
November 2012
3-170
Draft - Do Not Cite or Quote

-------
Table 3-20 Comparisons between Pb TSP concentrations from source-oriented and non-source-oriented
monitors within Los Angeles County, CA (06-037), 2007-2009.


A Source R
P
P90
COD
B Source R
P
P90
COD
C Source R
P
P90
COD
D Source R
P
P90
COD
E Source R
P
P90
COD
ABC
Source Source Source
1.00 -0.04 0.14
1.00 0.16 0.10
0.00 0.08 0.49
0.00 0.63 0.64
1.00 0.06
1.00 0.05
0.00 3.59
0.00 0.96
1.00
1.00
0.00
0.00








D
Source
0.10
0.08
0.10
0.31
0.17
0.05
0.25
0.84
0.59
0.57
1.76
0.68
1.00
1.00
0.00
0.00




E
Source
0.17
0.27
0.10
0.34
-0.06
0.07
0.10
0.71
0.08
0.03
2.14
0.77
0.18
0.12
0.17
0.42
1.00
1.00
0.00
0.00
F
Non-
Source
0.03
-0.15
0.08
0.57
0.33
0.74
0.02
0.46
0.12
-0.08
3.59
0.95
0.33
0.17
0.24
0.78
0.05
0.13
0.10
0.61
G
Non-
Source
0.00
0.00
0.06
0.57
0.29
0.12
0.02
0.48
0.24
0.26
4.22
0.96
0.09
0.11
0.25
0.80
0.07
0.06
0.10
0.64
H
Non-
Source
-0.08
0.14
0.08
0.79
0.40
0.28
0.01
0.61
0.28
0.28
3.59
0.98
0.32
0.24
0.25
0.89
0.00
0.24
0.11
0.78
1
Non-
Source
-0.07
-0.02
0.08
0.77
0.22
0.11
0.02
0.60
0.18
0.20
3.59
0.98
0.20
0.21
0.25
0.89
0.09
0.07
0.11
0.79
J
Non-
Source
-0.27
-0.09
0.08
0.85
0.20
0.10
0.02
0.81
0.08
0.13
3.92
0.99
0.03
0.07
0.25
0.95
-0.07
0.18
0.11
0.90
November 2012
3-171
Draft - Do Not Cite or Quote

-------


F Non-Source R
P
P90
COD
G Non-Source R
P
P90
COD
H Non-Source R
P
P90
COD
1 Non-Source R
P
P90
COD
J Non-Source R
P
P90
COD
ABCDEFGH
Non- Non- Non-
Source Source Source Source Source
Source Source Source
1.00 0.10 0.43
1.00 0.02 0.19
0.00 0.02 0.02
0.00 0.39 0.61
1.00 0.71
1 .00 0.65
0.00 0.01
0.00 0.54
1.00
1.00
0.00
0.00








1
Non-
Source
0.34
0.09
0.02
0.58
0.55
0.39
0.02
0.61
0.60
0.51
0.01
0.55
1.00
1.00
0.00
0.00




J
Non-
Source
0.21
0.09
0.02
0.82
0.54
0.38
0.02
0.85
0.51
0.40
0.01
0.77
0.29
0.37
0.01
0.78
1.00
1.00
0.00
0.00
Each comparison contains (in order): Pearson rank-order correlation (R), Spearman rank-order correlation (p), the difference between the 90th and 10th percentile data (P90), and the
coefficient of divergence (COD).
November 2012                                  3-172                     Draft - Do Not Cite or Quote

-------
 1                  Figure 3-43 illustrates Pb monitor locations within Hillsborough and Pinellas Counties in
 2                  FL, which comprise the greater Tampa-St. Petersburg metropolitan area. Two source-
 3                  oriented monitors (A and B) were located within Hillsborough County, and one non-
 4                  source-oriented monitor (C) was located in Pinellas County. Monitor A was located
 5                  360 meters north-northeast of the EnviroFocus Technologies battery recycling facility,
 6                  which produced 1.3 tons/year (U.S. EPA. 2008d). and monitor B was located 320 meters
 7                  southwest of the same facility. Monitor C was located next to a two-lane road in Pinellas
 8                  Park, FL.

 9                  Figure 3-44 displays seasonal wind roses for the Tampa-St. Petersburg metropolitan area.
10                  These wind roses suggest shifting wind directions throughout the winter, spring, and
11                  summer. During the winter, the highest winds came from the north and northeast with
12                  little influence from the west and southwest. During spring and summer, easterly and
13                  westerly winds were evident from the wind rose, with winds from the west being slightly
14                  higher in wind speed. During autumn, winds came predominantly from the northeast with
15                  little signal from the west or south.

16                  Seasonal and year-round concentrations are reported for Hillsborough and Pinellas
17                  Counties in Figure 3-45. The average annual concentration at monitor A was 0.15 ug/m3,
18                  and the 95th percentile was 0.70 ug/m3. During winter, the 60th percentile of the  data met
19                  the level of the NAAQS. At this site, the highest concentrations occurred during summer,
20                  which corresponded to the time when westerly winds were stronger.  Concentration data
21                  at monitor B were much higher, with an annual average of 0.45 ug/m3 and a 95th
22                  percentile of 1.9 ug/m3. Annually, the 55th percentile exceeded the level of the NAAQS,
23                  and in autumn the 45th percentile exceeded the NAAQS. The highest concentrations
24                  occurred in autumn, coinciding with the time when winds blew from the northeast,  when
25                  monitor B was most often downwind of the battery recycling facility. The  non-source-
26                  oriented monitor C always reported concentrations of 0.0 ug/m3. This is likely related to
27                  its location next to a quiet road in a small city.

28                  Intersampler correlations, shown in Table 3-21. illustrate that Pb has high intra-urban
29                  spatial variability. The source-oriented monitors were anticorrelated  (R = -0.09,
30                  p = -0.08). This was likely related to the fact that they were designated to monitor the
31                  same source and were downwind of the source at different times.
      November 2012                            3-173                  Draft - Do Not Cite or Quote

-------
Top: view of all Pb FRM monitors in Hillsborough and Pinellas Counties.
Bottom: Close up of industrial site around monitors A and B.

Figure 3-43     Pb TSP monitor locations within Hillsborough and Pinellas
                 Counties, FL (12-057 and 12-103), 2007-2009.
November 2012
3-174
Draft - Do Not Cite or Quote

-------
       I
          •
       I
          I
Note: Clockwise from top left: January, April, July, and October. Note that wind percentages vary from month to month.
Source: NRCS (2011).

Figure 3-44    Wind roses for Hillsborough/Pinellas Counties, FL, obtained from
                meteorological data at Tampa International Airport, 1961-1990.
November 2012
3-175
Draft - Do Not Cite or Quote

-------
      Site

      SITE ID
                               12-057-1073
                                            12-057-1066
                                                        12-103-3005
      MEAN
                                  0.15
                                              0.45
                                                          0.00
      SD
                                  0.27
                                              1.08
                                                          0.00
      DBS
                                  154
                                              155
                                                           58
       , BELOW MDL
                                  20
                                                           95
      Source orientation
                     E

                     1
                     c
                     O
                     c
                     (U
                     u
                     c
                     O
                     u
                                 Source
                                             Source
                                                        Non-source
3.0-
2.9 -
2.8 -
2.7-
2.6 -
2.5 -
2.4-
2.3 -
2.2 -
2.1 -
2.0 -
1.9 -
1.8 -
1.7 -
1.6 -
1.5 -
1.4 -
1.3 -
1.2 -
1.1 -
1.0 -
0.9 -
0.8 -
0.7-
0.6 -
0.5 -
0.4-
0.3 -
0.2 -
0.1 -
o.o -
































/
















































































































































B














































































































































C
u



























	
                              Y1234    Y1234    Y1234

                                             season
Figure 3-45     Box plots of annual and seasonal 24-h Pb TSP concentrations
                (ug/m3) from source-oriented and non-source-oriented monitors
                within Hillsborough and Pinellas Counties, FL (12-057 and
                12-103), 2007-2009.
November 2012
3-176
Draft - Do Not Cite or Quote

-------
      Table 3-21   Correlations between Pb TSP concentrations from source-oriented
                    and non-source-oriented monitors within Hillsborough and Pinellas
                    Counties, FL (12-057 and 12-103), 2007-2009.


A Source R
P
P90
COD
g Source R
P
P90
COD
Q Non-source R
P
P90
COD
A B
Source Source
1 .00 -0.09
1 .00 -0.08
0.00 1.20
0.00 0.71
1.00
1.00
0.00
0.00




C
Non-source


0.50
1.00


2.20
1.00
1.00
1.00
0.00
0.00
      Each comparison contains (in order): Pearson rank-order correlation (R), Spearman rank-order correlation (p), the difference
      between the 90th and 10th percentile data (P90), and the coefficient of divergence (COD).
 1
 2
 o
 5
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
Figure 3-46 illustrates Pb monitor locations within Cook County, IL. Eight monitors were
located within Cook County, four of which were designated by the Illinois Environmental
Protection Agency (IEPA) in data reported to the AQS as source-oriented and the other
four were non-source-oriented monitors. Monitor A was situated within 10 km of 6
sources ranging in emissions from 0.14 to 1.08 tons/year (U.S. EPA. 2008a). Monitor A
was also sited in the median of Interstate I-90/I-94. Monitor B was located on the
northern roadside of Interstate 1-290, 5 meters from the closest lane of traffic and was
within 10 km of 2 Pb sources (0.41 and 1.08 tons/year) (U.S. EPA. 2008a). Monitor C
was also located within 10 km of 6 sources in Cook County and Lake County, IN; the
largest of those sources was 2.99 tons/year and was located 8 km southeast of monitor C
(U.S. EPA. 2008a). Monitor C was placed on the roof of a high school. Monitor D was
located roughly 60 meters west of Interstate 1-294 and adjacent to O'Hare International
Airport. Monitor E was located on the rooftop of a building rented for government offices
in Alsip, IL,  a suburb south of Chicago. This location was roughly 1 km north of
Interstate 1-294 but not located on an arterial road; it was 9 km southeast of a
0.56 tons/year source (U.S. EPA. 2008a). Monitor F was sited in the parking lot of a
water pumping station, 100 meters north of Interstate 1-90 and 300 meters northwest of
the junction between Interstates 1-90 and 1-94. This site was 2 km  north-northwest of a
0.10 tons/year source (U.S. EPA. 2008a). Monitor G was situated  atop an elementary
school in a residential neighborhood on the south side of Chicago, roughly  100 meters
      November 2012
                           3-177
Draft - Do Not Cite or Quote

-------
 1                  south of a rail line and over 300 meters west of the closest arterial road. Although not
 2                  designated as a source monitor, monitor G was located 2 km southwest of facilities
 3                  emitting 0.30 and 0.41 tons/year (U.S. EPA. 2008a). Monitor H was sited on the grounds
 4                  of the Northbrook Water Plant. Interstate 1-94 curves around this site and was
 5                  approximately 700 meters from the monitor to the east and around to the north. Figure
 6                  3-47 displays seasonal wind roses for Cook County. Wind patterns were quite variable
 7                  during each season for this area. During the winter, winds mostly came from the west,
 8                  with smaller contributions from the northwest, southwest, and south. In spring,
 9                  measurable winds were omni-directional, with the highest winds coming from the south
10                  and northeast. Winds originated predominantly from the southwest and south during the
11                  summer, with measurable contributions from the northeast as well. In autumn, wind flow
12                  was predominantly from the south, but smaller contributions also came from the
13                  southwest, west, and northwest.

14                  Figure 3-48 presents seasonal box plots of Pb concentration at the eight monitors located
15                  within Cook County. The maximum 95th percentile concentration on this plot was
16                  0.14 ug/m3, so the scale of this box plot makes the variability in these data appear wider
17                  than the data presented for Los Angeles County and Hillsborough/Pinellas Counties.

18                  Monitor C was in closest proximity to the industrial steel facilities located in Lake
19                  County, IN. The average of concentrations measured at monitor C was 0.031 ug/m3, with
20                  a median of 0.02 ug/m3 and a maximum concentration of 0.31 ug/m3. In winter, the 95th
21                  percentile of data was 0.14 ug/m3. The higher values could potentially be attributed to
22                  transport of emissions; winds blow from the southeast roughly 10-15% of the time
23                  throughout the year. No other monitors in Cook County reported values above the level
24                  oftheNAAQS.

25                  Three "near-road" monitors, A, B, and D can be compared with the other monitors to
26                  consider the possibility of roadside resuspension of Pb dust from contemporaneous
27                  sources, as discussed in Section 3.2.2.6. It would be expected that resuspension would
28                  diminish with distance from the road. The 2 roadside monitors, A and B, reported
29                  average concentrations of 0.030 ug/m3 and 0.024 ug/m3, respectively. The median
30                  concentrations for monitors A and B  were 0.02 ug/m3. Fifteen percent of data were below
31                  the MDL for monitor A, and  25% were below the MDL for monitor B. Note that data
32                  obtained from monitor A may reflect industrial emissions as well. Monitor D was located
33                  roughly 60 meters from the closest interstate and 570 meters from the closest runway at
34                  O'Hare International Airport. The average concentration at this site was 0.012  ug/m3, and
35                  85% of data were below the MDL. Non-source monitors, E, F, G, and H had average
36                  concentrations of 0.011-0.017 ug/m3. It is possible that the difference between Pb
37                  concentrations at monitors A and B and Pb concentrations at the other monitors was
      November 2012                            3-178                 Draft - Do Not Cite or Quote

-------
 1                  related to proximity to the roadway, although this cannot be stated with certainty without
 2                  source apportionment data to confirm or refute the influence of industrial plumes from
 3                  Lake County, IN or local sources at each of the monitors.

 4                  Comparison among the monitor data demonstrates a high degree of spatial variability
 5                  (Table 3-22). None of the source-oriented monitors were well correlated with each other.
 6                  The highest correlation between source-oriented monitors occurred for monitors (A and
 7                  B [R = 0.32, p = 0.26]). This might have reflected more substantial differences related to
 8                  the additional influence of industrial sources nearby monitor A. Monitors (C and D) were
 9                  uncorrelated with each other and with monitors (A and B), likely because their exposure
10                  to sources was substantially different. The source-oriented and non-source-oriented
11                  monitors were generally not well correlated. The highest Spearman correlation occurred
12                  between monitors D and H (p = 0.53), but Pearson correlation was much lower for this
13                  pair (R = 0.19). Both were located on the north side of Cook County, but monitor H was
14                  roughly 20 km northeast of monitor D. Winds blew from the southwest roughly 20-30%
15                  of the time throughout the year and from the northeast 20-25% of the time between the
16                  months of March and July, so the correlation may have been related to a common signal
17                  transported across both sites. Monitors B and F (R = 0.52, p = 0.46) were also moderately
18                  correlated.  Monitor F is roughly 12 km northeast of monitor B, so the same common
19                  wind influence for monitors D and H may have also caused the moderate correlation
20                  between monitors (B and F). Monitor F was also moderately correlated with the  other 3
21                  non-source monitors (R = 0.42 to 0.54, p = 0.36 to 0.45), and the correlation between
22                  monitors (E and G) was moderate (R = 0.65, p = 0.40). The data from monitor H did not
23                  correlate well with those from monitors E and G. The non-source monitors were oriented
24                  from north to south over a distance of roughly  50 km in the following order: monitor H,
25                  monitor F,  monitor G, and monitor E. The correlation pattern may have been related to
26                  distance between samplers. Monitor H was located in the suburb of Northbrook, monitors
27                  F and G were sited within the Chicago city limits, and monitor E was situated in a town
28                  near the south side of Chicago. Differences among land use may have been related to the
29                  lack of correlation of the monitor H data with those from monitors E and G. It is likely
30                  that data from monitor F was at times belter correlated with monitors E and G and at
31                  other times with monitor H, since it had moderate correlation with all three other
32                  non-source monitors.
      November 2012                            3-179                  Draft - Do Not Cite or Quote

-------
                                  Legend
                                  O  TSP Source Monitors
                                  •  TSP Non-source Monitors
                                  •  City-based Population Cenlei
                                  •  County-based Population Cenlei
                                  — I nlor states
                                  	Major Highways
                                     Bodies of Water
                                     Urban Aieas
                                     Cook County. IL
                   "-«**HE
                                                      0  S  10
Top: view of all Pb FRM monitors in Cook County.
Bottom left: Close up of the high traffic site around monitor A.
Bottom right: Close up of O'Hare International Airport adjacent to monitor D.
Figure 3-46       Pb TSP Monitor locations within Cook County,  IL (17-031),
                     2007-2009.
November 2012
3-180
Draft - Do Not Cite or Quote

-------
       I
       I
          I
          I
Note: Clockwise from the top left: January, April, July, and October. Note that the wind percentages vary from month to month.
Source: NRCS (2011)

Figure 3-47     Wind roses for Cook County, IL, obtained from meteorological
                 data at O'Hare International Airport, 1961-1990.
November 2012
3-181
Draft - Do Not Cite or Quote

-------
Site
A
B
C
D


E F

G H
SITE ID 17-031-0026 17-031-6003 17-031-0022 17-031-3103 17-031-0001 17-031-0052 17-031-3301 17-031-4201
MEAN
SD
DBS
% BELOW
MDL
Source
orientation
0.15 -
0.14 -
0.13 -
0.12 -
0.11 -
0.10 -
- 0.09 -
E
^ 0.08 -
° 0.07 -
4-1
c 0.06 -
01
o
§ 0.05 -
0.04 -
0.03 -
0.02 -
0.01 -
0.00 -
0.030
0.020
179
15

0.024
0.013
175
25


Source



Source
0.031
0.036
177
25


0.012
0.0062
168
85




Source Source



0.013 0.017
0.0078 0.0098
177 175
75 55




Non-source Non-
source










|
I
A









I
I





























B




















1
f









||
11










1
1











C
i









|







II









1
It









|
1










||









'I
D










LI




































E










.nil
F










III























0.017 0.011
0.0097 0.0031
171 168
50 95

Non- Non-source
source
G









I
III!
H










.i.l.
         Y1234  Y1234  Y1234   Y1234  Y1234  Y1234  Y1234  Y1234
                                        season

Figure 3-48     Box plots of annual and seasonal 24-h Pb TSP concentrations
               (ug/m3) from source-oriented and non-source-oriented monitors
               within Cook County, IL (17-031), 2007-2009.
November 2012
3-182
Draft - Do Not Cite or Quote

-------
Table 3-22 Correlations between Pb TSP concentrations from source-oriented
and non-source-oriented monitors within Cook County, IL (17-031),
2007-2009.


A



B



C



D



E



F



G



H





Source R
P
P90
COD
Source R
P
P90
COD
Source R
P
P90
COD
Source R
P
P90
COD
Non-Source R
P
P90
COD
Non-Source R
P
P90
COD
Non-Source R
P
P90
COD
Non-Source R
P
P90
COD
A B C D
Source Source Source Source
1.00 0.32 0.00 0.05
1.00 0.26 -0.01 0.08
0.00 0.03 0.06 0.04
0.00 0.29 0.38 0.43
1.00 0.14 0.07
1.00 0.05 0.10
0.00 0.04 0.03
0.00 0.33 0.36
1.00 0.01
1.00 0.04
0.00 0.05
0.00 0.40
1.00
1.00
0.00
0.00
















Each comparison contains (in order): Pearson rank-order correlation (R), Spearman
90th and 10th percentile data (P90), and the coefficient of divergence (COD).
Figure 3-49 illustrates Pb monitor locations
E
Non-
Source
0.17
0.06
0.04
0.41
0.54
0.32
0.03
0.34
0.24
0.16
0.05
0.39
0.18
0.21
0.01
0.19
1.00
1.00
0.00
0.00












F
Non-
Source
0.39
0.32
0.03
0.36
0.52
0.46
0.02
0.29
0.05
0.10
0.04
0.35
0.12
0.37
0.01
0.24
0.42
0.36
0.02
0.24
1.00
1.00
0.00
0.00








G
Non-
Source
0.34
0.18
0.03
0.36
0.60
0.35
0.02
0.30
0.19
0.17
0.05
0.35
0.08
0.07
0.02
0.28
0.65
0.40
0.01
0.24
0.54
0.41
0.01
0.24
1.00
1.00
0.00
0.00




H
Non-
Source
0.06
0.06
0.04
0.45
0.06
-0.01
0.03
0.40
-0.04
0.06
0.05
0.42
0.19
0.53
0.01
0.15
-0.01
0.07
0.01
0.20
0.42
0.45
0.02
0.26
0.01
0.05
0.02
0.27
1.00
1.00
0.00
0.00
rank-order correlation (p), the difference between the
with Jefferson County, MO. Ten source-
              oriented monitors surrounded the Doe Run primary Pb smelter in Herculaneum, MO on
              the west and northwestern sides. The largest distance between these monitors was
November 2012
3-183
Draft - Do Not Cite or Quote

-------
 1                  approximately 1.5 km. Monitor E located on the Doe Run facility roughly 20 meters west
 2                  of the nearest building. Monitors A, B, C, D, F, G, and H were all located approximately
 3                  200 meters west of the facility. Monitors D, E, and H were situated alongside service
 4                  roads to the facility. Monitor I was  sited  100 meters north of the smelter, and monitor J
 5                  was located approximately 600 meters northwest of the facility. The Doe Run smelter
 6                  was the only active primary smelter in the U.S. at the time of this review, and the facility
 7                  was estimated to have emitted 41.1  tons Pb/yr (U.S. EPA. 2008f). Figure 3-50 displays
 8                  seasonal wind roses for Jefferson County. During winter, predominant winds originated
 9                  from the northwest, with a smaller fraction  of calmer winds originating in the south-
10                  southeast. During the spring, the south-southeasterly winds became more prevalent with a
11                  measurable fraction of stronger winds still originating in the north-northwest. In the
12                  summer, winds were omni-directional and generally calmer. A slightly larger percentage
13                  came from the south compared with other wind directions. Autumn winds were most
14                  predominantly south-southeastern, with a smaller fraction from the west and northwest.

15                  Figure 3-51 illustrates the seasonal  distribution of concentrations at monitors A-J in
16                  Jefferson County. The annual average concentrations ranged from  0.18 to 1.36 ug/m3
17                  across the monitors. The maximum concentration was measured at monitor C to be
18                  21.6 ug/m3, which was 144 times higher than the level of the standard. For this monitor,
19                  the 25th percentile of the data was at the level of the standard. In general, median and
20                  75th percentile concentrations were highest during the springtime and second highest
21                  during the fall. These seasons coincide with periods when the  southeastern winds were
22                  stronger and more prevalent. Because the Doe Run facility had two 30-meter stacks
23                  (Bennett, 2007). it is possible that the Pb measured at the closer monitors were due to
24                  either fugitive emissions from the plant; or, if vechiles and ground equipment were
25                  operated nearby, the previously-deposited emissions from the  plant were resuspended.

26                  Spatial variability among the monitors is lower than at many sites, because the monitors
27                  are relatively close together and are located on one side of the same source (Table 3-23).
28                  Correlations range substantially (R  = -0.03  to 0.96, p = -0.04 to 0.96). High correlations
29                  (R > 0.75, p > 0.75) occurred for monitors (A and C), (A and D), (C and D), (D and F),
30                  (E and F), (G and H), and (I and J). Monitors (A and C), (A and D), (C and D), (D and F),
31                  (E and F), and (G and H) are all within 250 meters of each other. For the highest
32                  correlation (R = 0.96,  p = 0.96, [for monitors E and  F]), monitor F is 250 meters directly
33                  east of monitor E. Low correlation (R < 0.25, p < 0.25) generally occurred when monitors
34                  B, I, and J were compared with monitors A, C, D, E, F, G, and H. Monitors B, I, and J
35                  were on the outskirts of the measurement area and so were likely oriented such that the
36                  southeasterly winds did not carry pollutants to these sites concurrently with the signal
37                  recorded by the other monitors.
      November 2012                             3-184                  Draft - Do Not Cite or Quote

-------
                             1-MIJfll.l
                              »  T&PSoulcs Mwllor*
                              •  c-ty wsea Population C»nl«r
                              •  Coui%-tkl«*d Populalion C»nw
Note: All monitors surround the Doe Run industrial facility. Top: Map view of all monitors in Jefferson County. Bottom: Satellite view
of the monitors and the Doe Run facility.
Figure 3-49      Pb TSP Monitor locations within  Jefferson County, MO (29-099),
                    2007-2009.
November 2012
3-185
Draft - Do Not Cite or Quote

-------
       i
          I
       i
          •
Note: Clockwise from top left: January, April, July, and October. Note wind percentages vary from month to month.
Source: NRCS (2011)

Figure 3-50    Wind roses for Jefferson County, MO, obtained from
                meteorological data at St. Louis/Lambert International Airport,
                1961-1990.
November 2012
3-186
Draft - Do Not Cite or Quote

-------
Site
SITE ID
MEAN
SD
OBS
% BELOW
MDL
Source
orientation
7.0-
A
29-099-00
22
0.43
0.54
622
0
Source

B
C
29-099-002 29-099-00
4 15
0.36
0.49
209
5
Source

1.36
1.97
1E3
0
Source

D E F G
29-099-002 29-099-00 29-099-00 29-099-002
3 04 20 1
0.39 1.12 0.69 0.75
0.54 1.67 1.01 1.25
632 1E3 575 953
0505
Source Source Source Source
I I I
H
29-099-00
05
0.29
0.59
351
25
Source

1
29-099-001
1
0.34
0.85
366
5
Source

J
29-099-00
13
0.18
0.33
177
15
Source

7.0 -
6.5-
6.0 -
5.5 -
5.0 -
4.5 -
4.0 -
3.5 -
3.0
2.5 -
2.0-
1.5-
1.0-
0.5 -

A










I.

II











!
i













B





































ll
III



























C


















D
































E














































I



F












































|
!





G



































































H












,
ll













I



























!
I












,
[I













ii














j












i.i.i
        Y1234 Y1234  Y1234  Y1234  Y1234  Y1234  Y1234  Y1234  Y1234  Y1234
                                         season

Figure 3-51     Box plots of annual and seasonal 24-h Pb TSP concentrations
               (ug/m3) from source-oriented and non-source-oriented monitors
               within Jefferson County, MO (29-099), 2007-2009.
November 2012
3-187
Draft - Do Not Cite or Quote

-------
Table 3-23 Correlations between Pb TSP concentrations from source-oriented
and non-source-oriented monitors within Jefferson County, MO
(29-099), 2007-2009.


A



B



C



D



E



F



G



H





Source R
P
P90
COD
Source R
P
P90
COD
Source R
P
P90
COD
Source R
P
P90
COD
Source R
P
P90
COD
Source R
P
P90
COD
Source R
P
P90
COD
Source R
P
P90
COD
A B C D
Source Source Source Source
1.00 0.66 0.80 0.84
1.00 0.59 0.80 0.83
0.00 0.71 1.55 0.42
0.00 0.46 0.48 0.30
1 .00 0.54 0.40
1 .00 0.53 0.43
0.00 1 .86 0.87
0.00 0.58 0.51
1 .00 0.86
1 .00 0.86
0.00 1.56
0.00 0.50
1.00
1.00
0.00
0.00
















E
Source
0.60
0.57
1.93
0.55
0.15
0.10
2.77
0.69
0.56
0.59
2.26
0.50
0.70
0.71
1.83
0.50
1.00
1.00
0.00
0.00












F
Source
0.65
0.64
1.14
0.45
0.15
0.14
1.96
0.62
0.72
0.72
1.26
0.46
0.80
0.80
1.02
0.36
0.96
0.96
0.86
0.35
1.00
1.00
0.00
0.00








G
Source
0.33
0.33
1.41
0.57
0.08
0.07
2.08
0.68
0.28
0.26
2.94
0.60
0.41
0.41
1.38
0.53
0.57
0.54
2.16
0.49
0.56
0.56
1.13
0.47
1.00
1.00
0.00
0.00




H
Source
0.32
0.35
0.74
0.64
0.16
0.22
0.94
0.68
0.32
0.27
2.65
0.74
0.48
0.56
0.76
0.61
0.53
0.46
2.50
0.66
0.56
0.54
1.51
0.63
0.85
0.87
1.53
0.61
1.00
1.00
0.00
0.00
1
Source
0.07
0.07
0.92
0.67
0.11
0.10
1.04
0.65
-0.03
-0.04
3.18
0.73
0.17
0.14
0.88
0.63
0.09
0.06
3.09
0.70
0.12
0.10
1.74
0.65
0.36
0.28
2.10
0.63
0.24
0.20
0.89
0.67
J
Source
0.05
0.05
0.78
0.69
0.01
0.09
0.91
0.65
-0.03
0.04
2.60
0.73
0.10
0.18
0.70
0.66
0.14
0.16
2.57
0.72
0.20
0.19
1.40
0.70
0.34
0.38
2.08
0.66
0.33
0.30
0.56
0.65
November 2012
3-188
Draft - Do Not Cite or Quote

-------

1 Source R
P
P90
COD
J Source R
P
P90
COD
ABODE FGH 1
1.00
1.00
0.00
0.00




J
0.87
0.79
0.62
0.48
1.00
1.00
0.00
0.00
      Each comparison contains (in order): Pearson rank-order correlation (R), Spearman rank-order correlation (p), the difference
      between the 90th and 10th percentile data (P90), and the coefficient of divergence (COD).

 1                   Figure 3-52 illustrates Pb monitor locations in Cuyahoga County, OH. Five monitors are
 2                   located within Cuyahoga County, three of which were designated by the Ohio EPA
 3                   (OEPA) as source-oriented and the other two were non-source-oriented monitors.
 4                   Monitors A, B, and C were all located within 1-10 km of six 0.1 tons/year source
 5                   facilities and one 0.2 tons/year source (U.S. EPA. 2008g). Additionally, monitor B was
 6                   located 30 meters north of the Ferro Corporation headquarters. This facility was stated in
 7                   the 2005 NEI to have no emissions, but it was thought by the OEPA to be the source of
 8                   exceedances at this monitor (U.S. EPA. 2008g). Monitor A was  sited roughly 300 meters
 9                   south of the Ferro Corporation facility. Monitor C was located 2.2 km west-northwest of
10                   the 0.5 tons/year Victory White Metal Co. facility. Monitor C was also roughly 250
11                   meters southeast of Interstate 1-490. Monitors D and E were designated as non-source-
12                   oriented monitors, although monitor D was just 600 meters further from the Victory
13                   White Metal facility than was monitor C. Monitor D was sited on a residential street
14                   located 50 meters north of Interstate 1-490. Monitor E was located on the rooftop of a
15                   building within 20 meters of a four-lane arterial road. Figure 3-53 displays seasonal wind
16                   roses for Cuyahoga County. During winter, summer, and autumn, the predominant winds
17                   were from the southwest, with stronger winds recorded during the winter. In the spring,
18                   the strongest winds still emanated from the south-southwest, but measurable winds were
19                   also scattered from the northeast to the northwest.

20                   Figure 3-54 illustrates the seasonal distribution of Pb concentration data at the five
21                   monitoring sites. The influence of southern winds, along with close proximity to a
22                   potentially-emitting facility, could have caused the elevated concentrations observed at
23                   monitor B (average: 0.10 ug/m3). The 80th percentile of data was at the level of the
24                   NAAQS at this monitor, and during autumn the 60th percentile of data met the level of
25                   the NAAQS. The maximum concentration during fall and for the monitor year-round was
26                   0.22  ug/m3. Concentration data from all other monitors were below the level of the
27                   NAAQS. For monitor A, the average concentration was 0.025 ug/m3, and the median
      November 2012                             3-189                  Draft - Do Not Cite or Quote

-------
 1                   reached 0.04 ug/m3 during the summer. Maximum concentration at this monitor was
 2                   0.07 ug/m3. Concentrations at monitor C averaged 0.017 ug/m3, and those at monitors D
 3                   and E averaged 0.014 ug/m3 and 0.013 ug/m3, respectively. Maximum concentrations
 4                   reached 0.04 ug/m3 at all three monitors.

 5                   The level of spatial variability is illustrated by the intersampler correlations presented in
 6                   Table 3-24. Monitors A and B appear to be anticorrelated (R = -0.06, p = -0.13). If the
 7                   Ferro site was the dominant source in this area, then the anticorrelation was likely caused
 8                   by the positioning of monitors A and B on opposite sides of that facility. At any given
 9                   time, potential emissions from the Ferro plant may have affected monitors A and B at
10                   distinct times. Monitors C, D, and E correlated moderately to well with each other
11                   (R = 0.37 to 0.74, p = 0.67 to 0.77). Given that all 3 monitors are separated by roughly
12                   2.8 km, it is possible that the relatively high correlations related to common sources, as
13                   suggested in the previous paragraph.  Little correlation was observed between the source-
14                   oriented and non-source-oriented monitors.
      November 2012                             3-190                  Draft - Do Not Cite or Quote

-------
                                 Legend
                                    TSP Source Monrtots
                                    TSP Non-source Monitors
                                    City-based Population Center
                                    County-based Population Center
                                   - Interslates
                                   - Majw Highways
                                    Bodies of Wfater
                                    Urban Areas
                                    Cuyahoga County, OH
Note: Top: view of all Pb FRM monitors in Cuyahoga County. Bottom left: Close up of industrial site around monitors A and B.
Bottom right: Close up of monitor D north of Interstate 1-490.
Figure 3-52     Pb TSP Monitor locations within Cuyahoga County, OH (39-035),
                    2007-2009.
November 2012
3-191
Draft - Do Not Cite or Quote

-------
       I
       I
          I
          I
Note: Clockwise from top left: Jan, April, July, and October. Note wind percentages vary from month to month.
Source: NRCS (2011)

Figure 3-53    Wind roses for Cuyahoga County, OH, obtained from
                meteorological data at Cleveland/Hopkins International Airport,
                1961-90.
November 2012
3-192
Draft - Do Not Cite or Quote

-------
Site
SITE ID
MEAN
SD
DBS
% BELOW MDL
Source orientation
0.25 -
0.24 -
0.23 -
0.22 -
0.21 ;
0.20 -
0.19 -
0.18 -
0.17 -
0.16 -
"E 0.15 -
"M 0.14 -
~ 0.13 -
•B 0.12 :
(D
£ 0.11 -
g 0.10 -
I °-09:
0.08 -
0.07 -
0.06 -
0.05 -
0.04 -
0.03 -
0.02 -
0.01 -
o.oo -


A

B
39-035-0050 39-035-0049
0.025
0.018




36
20


Source
0.10
0.060
36
0
Source
A

































|
















i


















































I
B























































































C
D E
39-035-0061 39-035-0038 39-035-0042
0.017
0.010
36
30
Source
0.014 0.013
0.0072 0.0076
35 36
45 45
Non-source Non-source
C

































^

































I
D
















i(
















i,
E
















Lit
Figure 3-54
 Y1234   Y1234   Y1234   Y1234    Y1234
                         season

Box plots of annual and seasonal 24-h Pb TSP concentrations
(ug/m3) from source-oriented and non-source-oriented monitors
within Cuyahoga County, OH (39-035), 2007-2009.
November 2012
                   3-193
Draft - Do Not Cite or Quote

-------
 1
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11
Table 3-24


A Source



B Source



C Source



D Non-Source


E Non-Source



Each comparison
between the 90th
Correlations between Pb TSP concentrations from source-oriented
and non-source-oriented monitors within Cuyahoga County, OH
(39-035), 2007-2009.
A
Source
R 1.00
p 1.00
P90 0.00
COD 0.00
R
P
P90
COD
R
P
P90
COD
R
P
P90
COD
R
P
P90
COD
B C
Source Source
-0.06 0.21
-0.13 0.24
0.18 0.05
0.64 0.33
1.00 0.26
1.00 0.31
0.00 0.18
0.00 0.69
1.00
1.00
0.00
0.00







D
Non-Source
0.17
0.19
0.04
0.35
0.43
0.24
0.19
0.71
0.74
0.77
0.01
0.17
1.00
1.00
0.00
0.00




contains (in order): Pearson rank-order correlation (R), Spearman rank-order correlation (p),
and 10th percentile data (P90), and the coefficient of divergence (COD).
Figure 3-55 illustrates Pb monitor locations within Sullivan County, TN
E
Non-Source
0.24
0.21
0.05
0.37
0.11
0.34
0.19
0.73
0.51
0.67
0.01
0.18
0.37
0.67
0.01
0.17
1.00
1.00
0.00
0.00
the difference
. Three source-
oriented monitors were situated around an Exide Pb recycling facility emitting
0.78 tons/year (U.S. EPA. 200810. Monitors A and C are positioned along the facility's
service road and are approximately 100 meters and 200 meters away from the facility,
respectively. Monitor A is directly next to the road, and monitor C is roughly 15 meters
from the road. Monitor B is located in the facility's parking lot roughly 50 meters from
the closest building. The facility and all three monitors are approximately 1.5 km
northwest of the Bristol Motor Speedway and Dragway racetracks, which hosts a variety
of auto races each year, including NASCAR, KART, and drag racing. Although the
NASCAR circuit no longer uses tetraethyl Pb as an anti-knock agent in its fuel, some of
the smaller racing circuits continue to do so. However, the speedway is rarely upwind of
     November 2012
                           3-194
Draft - Do Not Cite or Quote

-------
 1                  the monitoring sites and so likely had minimal influence on the reported concentrations.
 2                  Figure 3-56 displays seasonal wind roses for Sullivan County. During winter and spring,
 3                  the predominant winds come from the southwest and west. In the summer, the percentage
 4                  of wind coming from the west and southwest is roughly equal to that for wind coming
 5                  from the east and northeast, although the easterly winds are calmer. During autumn,
 6                  winds come predominantly from the northeast and east, although these winds tend to be
 7                  calmer than those originating from the southwest and west.

 8                  The data presented in Figure 3-57 illustrates that concentrations above the level of the
 9                  NAAQS occurred frequently at the monitors. The average concentrations at monitors A,
10                  B, and C were 0.11 ug/m3, 0.051 ug/m3, and  0.059 ug/m3, respectively.  Median
11                  concentrations were 0.08 ug/m3, 0.03 ug/m3, and 0.04 ug/m3, respectively. The 75th
12                  percentile of year-round data at monitor A was at the level of the NAAQS, while the 95th
13                  percentile of data were below the NAAQS level for monitors B and C. The maxima at
14                  each monitor were 0.76 ug/m3, 0.26 ug/m3, and 0.43 ug/m3 for monitors A, B, and C. The
15                  concentrations measured at monitor A tended to be higher because the predominant and
16                  stronger winds came from the southwest, so in many cases monitor A was upwind of the
17                  facility. It is possible that Pb that had either deposited or was stored in waste piles
18                  became readily resuspended by traffic-related turbulence and was measured at monitor A
19                  since that monitor was closest to the road. The slightly higher concentrations at monitor
20                  A compared with those from monitor C are consistent with the southwestern winds.

21                  Not surprisingly, the correlations of monitor  A with monitors B and C (R = 0.06 to 0.14,
22                  p = -0.04 to 0.13) were quite low (Table 3-25). The correlation between monitors B and
23                  C was moderate (R = 0.31, p = 0.45). It makes sense that the correlation for these
24                  monitors would be somewhat higher because they are both oriented to the east of the Pb
25                  recycling facility, although monitor C is to the northeast and monitor B  to the east-
26                  southeast.
      November 2012                            3-195                 Draft - Do Not Cite or Quote

-------
                      Legend
                       • TSPSource Monitors
                       * City-based Population Canter
                       • County-based Population Center
                      	 Interstate*
                         Major Highways
                         Bodies of Water
                         Urban Areas
                         Sullivan County, TN
   0    S    10       20 Kilometers
Note: Top: Map, bottom: Satellite image. Monitors A, B, and C surround the Exide Pb recycling facility. Just to the southeast is the
Bristol motor speedway.
Figure 3-55      Pb TSP Monitor locations within Sullivan  County, TN (47-163),
                     2007-2009.
November 2012
3-196
Draft - Do Not Cite or Quote

-------
       i
       •
          S:
Source: NRCS (2011)
Note: Clockwise from top left: January, April, July, and October. Note that the wind percentages vary from month to month.

Figure 3-56     Wind roses for Sullivan County, TN, obtained from meteorological
                 data at Bristol/Tri City Airport, 1961-90.
November 2012
3-197
Draft - Do Not Cite or Quote

-------
Site
SITE ID
MEAN
SD
OBS
% BELOW MDL
Source orientation
0.44 -
0.42 -
0.40 -
0.38 -
0.36 -
0.34 -
0.32 -
0.30 -
0.28 -
mE 0.26 -
"M 0.24 -
"c 0.22 -
0
'•£ °-20 "
c 0.18 -
0)
c 0.16 -
0
u o.H -
0.12 -
0.10 -
0.08 -
0.06 -
0.04 -
0.02 -
o.oo -
A
47-163-3001
0.11
0.11
334
0
Source
A
































































B
47-163-3002
0.051
0.036
362
0
Source
B


























































































c
47-1 63-3003
0.059
0.047
345
0
Source
C















Figure 3-57
                      Y1234   Y1234   Y1234

                                  season
Box plots of annual and seasonal 24-h Pb TSP concentrations
(ug/m3) from source-oriented monitors within Sullivan County, TN
(47-163), 2007-2009.
November 2012
                   3-198
Draft - Do Not Cite or Quote

-------
     Table 3-25   Correlations between Pb TSP concentrations from source-oriented
                    monitors within Sullivan County, TN (47-163), 2007-2009.


A Source R
P
P90
COD
B Source R
P
P90
COD
C Source R
P
P90
COD
A B
Source Source
1.00 0.06
1 .00 -0.04
0.00 0.21
0.00 0.47
1.00
1.00
0.00
0.00




C
Source
0.14
0.13
0.19
0.43
0.31
0.45
0.06
0.23
1.00
1.00
0.00
0.00
     Each comparison contains (in order): Pearson rank-order correlation (R), Spearman rank-order correlation (p), the difference
     between the 90th and 10th percentile data (P90), and the coefficient of divergence (COD).
     3.8.3       Seasonal Variation in Pb Concentrations

 1                  Monthly average Pb concentrations averaged over multiple sites and over 3 years from
 2                  2008-2010 are shown for Pb-TSP from source-oriented monitors (Figure 3-58). Pb-TSP
 3                  from non-source-oriented monitors (Figure 3-59). Pb-PMi0 (Figure 3-60). and Pb-PM2 5
 4                  (Figure 3-61). For source-oriented Pb-TSP (Figure 3-58). monthly average concentrations
 5                  were determined from between 146 and 154 samples in each month.  For non-source-
 6                  oriented TSP (Figure 3-59). monthly average concentrations were determined from
 7                  between 141 and 151 samples in each month. A winter minimum was observed with
 8                  December, January, and February exhibiting the three lowest monthly averages. In both
 9                  cases, there is little seasonal variation. Minor variations in monthly averages are probably
10                  driven by exceptional events. Monthly median concentrations are very similar for all
11                  months.
     November 2012
3-199
Draft - Do Not Cite or Quote

-------
  CO
  2
  
-------
         0.05
         0.04-
         0.03 -
       0  0.02
         0.01
         0.00-

A
1

Ja

K
r

n






j|


F(

I


;b









M




ar









A




nr









M




ay









Ji




n




vlontr




J




j|









A




jg









S<




5P









O




ct









N




DV






^


D

K


3C
Note: Box and whisker plots are used for each month, with the box comprising the interquartile range of the data and the whiskers
comprising the range within the 5th to 95th percentiles. The median is noted by the red line, and the blue star denotes the mean.


Figure 3-59      Monthly non-source-oriented Pb-TSP average (ug/m3) over 12

                   months of the year, 2008-2010.
0.028 -
0.026 -
0.024 •
0.022 -
0.020 -
CO
j= 0.018-
D)
-3 0.016-
•B 0.014-
•E 0.012-
0)
^ 0.010-
o
0 0.008 -
0.006 •
0.004 -
0.002 -
0.000 -

















































T





-L
I
Jan









I
i
Feb





























_3





r

=>=
1
Mar





























3






f
	
->-
I
Apr





























3




)c

-L-
1
May












































^


















Jun






— . —





r^





{-,

-L-
i
Jul





























3






C

-L-
i
Aug





























3





C

-1-
I
Sep





























•}




c

1
I
Oct


















__^_










3










f


J-
I
Nov





























)













I
-L-

































Dec
Month
Note: Box and whisker plots are used for each month, with the box comprising the interquartile range of data and the whiskers
comprising the range from 5th to 95th percentiles. The median is noted by the red line, and the blue star denotes the mean.


Figure 3-60      Monthly Pb-PMi0 average (ug/m3) over 12 months of the year,
                   2007-2009.
November 2012
3-201
Draft - Do Not Cite or Quote

-------
0.010-
0.009 -
0.008 -
ff 0.007-
E
g 0.006 -
•2 0.005 -
1 0.004 •
|
° 0.003 -
0.002 -
0.001 •
0.000 -




















—
_l_
i
Jan


















*

i
i
Feb



















*
_L
Mar















3


C

J_
Apr















3




C

J_
i
May














-j







L.
Jr-
_l_







i
Jun







3






K

_l_
i
Jul











__^



3



C

_l_
i
Aug














p


t

_L
i
Sep





















J_
i
Oct















J






C




i
Nov















3




C

_L
Dec
Month
     Note: Box and whisker plots are used for each month, with the box comprising the interquartile range of the data and whiskers
     comprising the range from 5th to 95th percentiles. The median is noted by the red line, and the blue star denotes the mean.
     Figure 3-61      Monthly Pb-PM2 5 average (ug/m3) over 12 months of the year,
                       2007-2009.
i
2
3
4
5
For both Pb-PMi0 (Figure 3-60) and Pb-PM2 5, (Figure 3-61) there is also little seasonal
variation, with minor fluctuations in monthly averages probably driven by exceptional
events, and similar monthly median concentrations for all months. Pb-PMi0 monthly
average concentrations were determined from between 100 and 109 samples and
Pb-PM2.5 from between 866 and 1,034 samples each month.
     3.8.4        Size Distribution of Pb-bearing PM

6                  Table 3-26 presents data for co-located Pb-TSP, Pb-PMi0, and/or Pb-PM2 5 monitors.
7                  Table 3-27 contains metadata for studies in Section 3.5.3 involving size distribution data,
8                  and Table 3-28 contains the size distribution data for those studies. At times, the data
9                  were extracted from figures in the original references.
     November 2012
                           3-202
Draft - Do Not Cite or Quote

-------
Table 3-26   Correlations and average of the concentration ratios for co-located monitors, TSP versus PMio, TSP
            versus PM2.s, and PM™ versus PM2.s.
Site ID*
060190008
060190008
060190008
060190008
060250004
060250005
060250005
060251003
060290004
060290004
060290014
060290014
060290014
060290014
060292004
060310003
060310004
060370002
060374002
CBSA
Fresno, CA
Fresno, CA
Fresno, CA
Fresno, CA
El Centra, CA
El Centra, CA
El Centra, CA
El Centra, CA
Bakersfield, CA
Bakersfield, CA
Bakersfield, CA
Bakersfield, CA
Bakersfield, CA
Bakersfield, CA
Bakersfield, CA
Hanford-Corcoran, CA
Hanford-Corcoran, CA
Los Angeles-Long Beach-
Santa Ana, CA
Los Angeles-Long Beach-
Santa Ana, CA
Land Type
Suburban
Suburban
Suburban
Suburban
Suburban
Suburban
Suburban
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Suburban
Unknown
Suburban
Suburban
Suburban
Avg
Years Corr Ratio
PMio: TSP
1995-2001 0.93 0.82
1995-2001 0.93 0.83



1996-2001 0.79 0.98
1996-2001 0.92 0.89



1995-2000 0.94 0.75
1995-2000 0.92 0.78
1995-2000 0.47 0.80
1995-2000 0.43 0.81




1995-2000 0.76 0.38
Years Corr
Avg
Ratio
PM2.5: TSP
1992-2001 0.82
1992-2001 0.80
0.59
0.56



1996-2001 0.77
1996-2001 0.91
0.73
0.64

1992-1994 0.90
1992-1994 0.94
1994-2000 0.96
1994-2000 0.92
1995-2000 0.28
1995-2000 0.27
0.55
0.43
0.51
0.53
0.60
0.62




1992-2000 0.62
0.30
Years
Corr
Avg
Ratio
PM2.s: PMio
1995-2001
1995-2001
1995-2001
1995-2001
1995-1996
1996-2001
0.99
0.96
0.99
0.98
0.96
0.99
0.77
0.82
0.79
0.77
0.80
0.71

1995-1995
0.81
0.72


1995-2000
1995-2000
1995-2000
1995-2000
1995-2000
1995-1998
1996-2000
1995-2000
1995-2000
0.96
0.91
0.84
0.98
0.74
0.97
0.95
0.89
0.91
0.71
0.77
0.80
0.72
0.74
0.83
0.77
0.59
0.62
November 2012
3-203
Draft - Do Not Cite or Quote

-------

060374002
060390001
060631008
060658001
060658001
060670010
060710014
060771002
060771002
060850004
060850004
060850004
060990002
060990002
060990005
060990005
061072002
170310022
170310052
171190010
Los Angeles-Long Beach-
Santa Ana, CA
Madera-Chowchilla, CA
NONE (Plumas Co., CA)
Riverside-San Bernardino-
Ontario, CA
Riverside-San Bernardino-
Ontario, CA
Sacramento-Arden-Arcade-
-Roseville, CA
Riverside-San Bernardino-
Ontario, CA
Stockton, CA
Stockton, CA
San Jose-Sunnyvale-
Santa Clara, CA
San Jose-Sunnyvale-
Santa Clara, CA
San Jose-Sunnyvale-
Santa Clara, CA
Modesto, CA
Modesto, CA
Modesto, CA
Modesto, CA
Visalia-Porterville, CA
Chicago-Nape rville-Joliet,
IL-IN-WI
Chicago-Nape rville-Joliet,
IL-IN-WI
St. Louis, MO-IL
Suburban
Urban and Center City
Unknown
Suburban
Suburban
Urban and Center City
Suburban
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Suburban
Suburban
Urban and Center City
Avg
Years Corr Ratio
1995-2000 0.87 0.72


1995-1996 0.13 0.39
1995-1997 0.93 0.72


1995-2000 0.70 0.84
1995-2000 0.91 0.74
1995-2000 0.87 0.63


1995-1998 0.96 0.79




1992 0.81 0.94
1992 0.84 0.86
1992 0.40 0.96
Avg
Years Corr Ratio
1992-2000 0.43 0.44


1992-1996 0.31 0.33
1992-1997 0.86 0.46


1992-2000 0.59 0.56
1992-2000 0.76 0.48
1994-1997 0.11 0.36
1992-1993 0.42 0.37
1992-2000 0.54 0.39
1992-1998 0.24 0.61







Years
Avg
Corr Ratio

1995-1996
1997-1999
1995-1997
0.98 0.90
0.95 0.72
0.94 0.67

1995-2001
1996-2000
1995-2000
0.99 0.75
0.78 0.73
0.94 0.71

1995-2000
0.95 0.69


1995-1998
1997-1998
1998-2000
1998-2000
1995-2000
0.64 0.80
0.50 0.73
0.99 0.71
0.97 0.71
0.99 0.70



November 2012
3-204
Draft - Do Not Cite or Quote

-------

171191007
171630010
180890023
201730007
201730008
201730009
201731012
201770007
202090015
202090020
270530053
300490719
300490719
450430001
450790014
450791003
450791003
St. Louis, MO-IL
St. Louis, MO-IL
Chicago-Nape rville-Joliet,
IL-IN-WI
Wichita, KS
Wichita, KS
Wichita, KS
Wichita, KS
Topeka, KS
Kansas City, MO-KS
Kansas City, MO-KS
Minneapolis-St. Paul-
Bloomington, MN-WI
Helena, MT
Helena, MT
Georgetown, SC
Columbia, SC
Columbia, SC
Columbia, SC
Urban and Center City
Suburban
Urban and Center City
Suburban
Suburban
Suburban
Suburban
Urban and Center City
Urban and Center City
Urban and Center City
Urban and Center City
Suburban
Suburban
Urban and Center City
Suburban
Urban and Center City
Urban and Center City
Years Corr
1992 0.92
1992 0.96
2007 - 2008 0.73
1990-1997 0.18
1990-1997 0.34
1990-1997 0.54
1990-1997 0.85
1990-1997 0.56
1990-1997 0.75
1990-1997 0.99
1996-2001 0.54
1990-1991 0.81
1990 0.79
1990-1991 0.71
1990 0.82
1990 0.90
1990 0.90
Avg
Ratio
0.91
0.91
0.86
1.28
1.12
1.05
0.89
0.99
0.80
0.81
0.57
0.48
0.49
0.60
0.89
0.94
0.83
Avg
Years Corr Ratio

















Avg
Years Corr Ratio

















*Note: For comparability, comparisons are limited to monitors where all samples were above the MDL, at least 30 co-located samples were obtained, and both monitors reported data at standard
temperature and pressure.
November 2012
3-205
Draft - Do Not Cite or Quote

-------
Table 3-27    Metadata for studies of Pb-PM size distribution.
Reference
Location
Nearest source
Proximity to source
Sampling dates
                           Jersey City, NJ- an
                           urban/industrial area
                           New Brunswick, NJ- a suburban
                           area
                                 Jersey City- near Manhattan, NJ Turnpike,
                                 Hudson River- high gas/oil consumption for
                                 industry/domestic heating, heavy gasoline &
                                 diesel powered vehicles and ship traffic from
                                 harbor.
                                 New Brunswick- near NJ parkway and Garden
                                 State Parkway
                                                                                                       Close
                                                                            ASD measurements:
                                                                            June 10-20,2002
Bein et al. (2006)
Pittsburg, PA
Article does not describe sources; other articles
also use data from Pittsburgh Air Quality Study
(PAQS) and may have more info on sources
                                 July 2001-September
                                 2002
Pekey et al. (2010)
Kocaeli, Turkey
Kocaeli is a very industrialized and urbanized
region in Turkey; sources include a large
refinery, a petrochemical complex, a hazardous
waste incinerator and industry operations for
textile, machine, mine, metal, food, automotive,
paper, chemistry, wood, petroleum, tanning,
and coal sectors, plus heavy traffic
Singh etal. (2002)
Downey, CA- a city in Los Angeles
County along "Alameda corridor"
joining coastal area to downtown
LA
Riverside, CA- an inland county
east of LA
Downey- a "source" site affected by fresh PM
emissions from nearby oil refineries, industry,
and heavy diesel emissions
Riverside- a "receptor" site affected by aged
PM emissions including high vehicle emissions
in LA
Downey: 10 km downwind of
refineries; 2-4 km from Interstates
1-710 and I-605
Riverside- 70 km east of
downtown LA
Downey: September
2000-January 2001
Riverside: February
2001-June 2001
Dall'Osto et al. (2008)
U.K. national air quality monitoring
station in Port Talbot, U.K.
One of the U.K.'s largest integrated steelworks;
near major roadways; (Table 1 of this study
describes the plants, operations, emission
types, and emission components for steelwork
sources)
Monitoring site next to steelwork    April 24-May 5, 2006
Weitkamp et al. (2005)
Coking facility near Pittsburgh, PA
Large coking facility that converts 6 million tons
of coal to 4 million tons of metallurgic coke
every year
Sampling site downwind, directly
across river from coke
facility(~400m)
August 22-September 5,
2002
Pekey et al. (2010)
Indoor/outdoor sample points for
15 homes in Kocaeli, Turkey
Kocaeli is a very industrialized and urbanized
region in Turkey; sources include a large
refinery, a petrochemical complex, a hazardous
waste incinerator and industry operations for
textile, machine, mine, metal, food, automotive,
paper, chemistry, wood, petroleum, tanning,
and coal sectors, plus heavy traffic
15 Kocaeli homes chosen as
representative sample; 10 close to
high traffic roads, 5 near
low/moderate traffic roads
Summer: May 31-June
29, 2006
Winter: December 16-
January 20, 2007
November 2012
                        3-206
                 Draft - Do Not Cite or Quote

-------
Reference
Location
Nearest source
Proximity to source
Sampling dates
                          Near Interstate 1-405 between
                          Sunset Blvd and Wilshire Blvd,
                          Los Angeles, CA
                                 Heavy traffic on Interstate 1-405 Freeway
                                            UP: 150m upwind of Interstate
                                            I-405 (background)
                                            DW1: 10m downwind
                                            DW2: 150m downwind
                                            DW3: 450m downwind
                                April 13-May 1, 2004
Song and Gao (2011)
Carlstadt, NJ
Heavy traffic on NJ Turnpike
~5m from roadside of the highway
Winter: December 2007-
February 2008
Summer: July 2008
Zereini et al. (2005)
3 Sites in Frankfurt, Germany
with different traffic densities
                                                           Vehicle emissions
                                            Site 1: next to main street with
                                            32500 cars/day
                                            Site 2: next to side street with
                                            <1000 cars/day
                                            Site 3: large garden 8km NWof
                                            city
                                August 2001 to July 2002
Lough et al. (2005)
Two road traffic tunnels in
Milwaukee, Wl
                                                           Traffic emissions
                                           5 m upwind from entrance (inside
                                           tunnel); 15 m upwind from tunnel
                                           exit
                                Summer: July 31-August
                                28, 2000
                                Winter: December 13-
                                January 17, 2001
Haysetal. (2011)
20m downwind of Interstate I-440
highway in Raleigh, NC
Traffic emissions from highway
                                                                                                      20 m downwind
                                July 26-31  and August
                                3-10,2006
                          Roadside site and site inside
                          highway tunnel in Taipei, Taiwan
                                                           Vehicle emissions
                                                                            Roadside: sidewalk 4m from road
                                                                            Tunnel: relay station in tunnel
                                                                            1.4km from outlet; 2 m from traffic
                                                                            lane
                                                                            January to December
                                                                            2008
Birmili et al. (2006)
Remote background: Mace Head
atmospheric research station in
Connemara, Ireland
Urban background: University of
Birmingham campus, U.K.
Roadside: A38 Bristol Road,
Birmingham, U.K.
Road Tunnel: Queensway
underpass in Birmingham, U.K.
                                                           Traffic emissions
                                            Remote background:
                                            Urban background: at least 100m
                                            from road traffic
                                            Roadside: 4 m from traffic
                                            Road Tunnel: 30 m from tunnel
                                            exit
                                Remote background:
                                August 8-28, 2002
                                Urban background:
                                April 23-October 7, 2002
                                Roadside: July 8-12,
                                2002
                                Road Tunnel: July 2,
                                2002
Bruggemann et al. (2009)
Harrison et al. (2003)
Roadside in
Dresden, Germany
Roadside in
Birmingham, U.K.
Traffic emissions from busy street (traffic
density-55,000 per day; 8% trucks), tramline,
railway station
Traffic emissions from A38
Next to road, near tramline
crossing, 200 m to railway station
9m from road
September 2003-August
2004
October 26, 2000 to
January 17, 2001
November 2012
                       3-207
                 Draft - Do Not Cite or Quote

-------
Reference
Location
                                                             Nearest source
Proximity to source
Sampling dates
                           Suburb of Kanazawa, Japan- a
                           western coastal city; the largest in
                           Hokuriku region of Japan
                                  Emissions from road traffic, nearby incinerators
                                  and electricity generation plants, and sea salt
                                                                                                         Next to road; ~5km from
                                                                                                         incinerators and electricity
                                                                                                         generation plants; situated on
                                                                                                         west coast
                                 May-June 2003
Martuzevicius et al. (2004)
9 Locations in Cincinnati, Ohio
metropolitan area
                                                            Vehicle emissions from Cincinnati highway
                                                            network; emission from industry (233 facilities
                                                            within municipal area limits)
11 Sites- varying distance to linear
and point sources; distance to
major highways ranges from
21 Om to 4,590m
December 2001-
November2002
Moreno et al. (2008)
3 Sampling sites in Mexico City
(Mexico) Metropolitan Area:
1 Site in the industrial center (TO),
1 Site NE outside city limits (T1),
and
1 Rural site north of city (T2)
                                                             Urban pollution sources- traffic emissions,
                                                             industry
3 Sites with varying relation to
"Mexico urban plume" by
distance, wind direction
                                                                                                                                           March 2006
Goforth and Christoforou
(2006)
                           Lake Hartwell, GA
                           (rural southeast U.S.)
                                                                                                                February-March 2003
Makkonen et al. (2010)
Virolahti EMEP station, Finland
                                                             European Route E18 (3,000 vehicles/day)
                                                                                                         5 km
                                 August 2007
Wojas and Almquist (2007)
Oxford, OH and other towns in
Greater Cincinnati region
                                                            Transportation, manufacturing processes, and
                                                            coal-fired power plants
                                                                                                          -80 km to northwest Cincinnati
                                 January to December
                                 2005
Lin et al. (2005)
Roadside in city in southern
Taiwan
                                                            Traffic emissions (avg traffic load = 72,000
                                                            vehicles/day), Pingtung Industrial Park
                                                            (146 factories), Nearby crematory
10 m from road, 2 km from
industrial park (146 factories,
i.e., electron apparatus, metal,
and food manufacturing), 1 km
from crematory
February to April 2004
Csavina et al. (2011)
Main sampling site in Winkelman,
Arizona at Hayden High School
                                                             Nearby an active mining and smelting site in
                                                             Hayden, Arizona
2km from mine tailings pile; 1km
from smelting operations, main
smoke stack, and slag pile
December 2008-
November2009
November 2012
                        3-208
                                                                              Draft - Do Not Cite or Quote

-------
Reference                Location                       Nearest source                            Proximity to source            Sampling dates

Fang and Huang (2011)      3 Sites in central Taiwan:           Emissions from vehicle traffic and industry        Hung-kuang (HK): in residential     November 2010-
                          A school                                                                    area 2 km from maJ°r expressway   December 2010
                          (Hung-kuang),                                                               Gaomei in Taichung (GM): 300
                          A wetland                                                                   hectare wetland with coal
                          (Gaomei in Taichung),                                                         combustion-based Taichung
                                                                                                     Thermal Power Plant (located
                          An industrial site                                                             a,    the coast of the west side of
                          (Quan-xing)                                                                 the sampNng site)

                                                                                                     Quan-xing (QX): town with lots of
                                                                                                     industry including metal
                                                                                                     manufacturing, textiles, petroleum
                                                                                                     and coal products

                          7 Sites around Los Angeles, CA,     Not stated                                   -                               August, 2002 - June,
                          including 6 urban watershed sites                                                                              2003
                          and one non-urban coastal
                          watershed.
November 2012                                  3-209                     Draft - Do Not Cite or Quote

-------
Table 3-28 Size distribution data for various studies described in Table 3-27.
Reference Location Size Bin
Yi et al. (2QQ6) Jersey City, NJ 0.18-0.32
0.32-0.56
0.56-1
1-1.8
1 .8-3.2
3.2-5.6
5.6-10
10-14.4
14.4-19.9
19.9-26.1
26.1-36.1
36.1-100
New Brunswick, NJ 0.18-0.32
0.32-0.56
0.56-1
1-1.8
1 .8-3.2
3.2-5.6
5.6-10
10-14.4
14.4-19.9
19.9-26.1
26.1-36.1
36.1-100
Bein et al. (2006) Pittsburgh, PA 1 -1 .8
1 .8-3.2
3.2-5.6
5.6-10
Concentration
0.001054
0.000668
0.000952
0.000852
0.000609
0.001229
0.001591
0.000948
0.000171
0.000693
0.000333
0.000447
0.001146
0.00078
0.001733
0.001083
0.000373
0.000446
0.000347
0.000182
2.02E-09
1 .35E-05
1 .62E-05
0.000152
0.096608
0.314846
0.187393
0.239094
November 2012
3-210
Draft - Do Not Cite or Quote

-------
Reference Location Size Bin
Singh et al. (2002) Downey, CA <0.1 |jm
0.1-0.35
0.35-1.0
1 .0-2.5
2.5-10
Riverside, CA <0.1 |jm
0.1-0.35
0.35-1.0
1 .0-2.5
2.5-10
Dall'Osto et al. (2008) Port Talbot, U.K. 0.1-0.196
0.196-0.356
0.356-0.57
0.57-1
1-1.8
1.8-3.1
3.1-6.2
6.2-9.9
9.9-18
Sabin et al. (2006b) Los Angeles, CA, 10 m downwind of road <6
6-11
11-20
20-29
>29
Concentration
0.00133
0.00419
0.00334
0.00189
0.00175
0.0004
0.00089
0.0018
0.001
0.00297
0.00021 1
0.001871
0.005424
0.004935
0.010229
0.002216
0.001847
0.000807
0.000141
0.007953
0.004172
0.00013
0.000522
0.004563
November 2012
3-211
Draft - Do Not Cite or Quote

-------
Reference Location Size Bin
Zereini et al. (2005) Frankfurt, Germany <0.43
Main street 0.43-0.63
0.63-1.1
1.1-2.1
2.1-3.3
3.3-4.7
4.7-5.8
5.8-9.0
>9.0
Side street <0.43
0.43-0.63
0.63-1.1
1.1-2.1
2.1-3.3
3.3-4.7
4.7-5.8
5.8-9.0
>9.0
Rural background <0.43
0.43-0.63
0.63-1.1
1.1-2.1
2.1-3.3
3.3-4.7
4.7-5.8
5.8-9.0
>9.0
Concentration
0.005904
0.005332
0.004285
0.002857
0.002666
0.002857
0.001809
0.002476
0.004285
0.00332
0.002818
0.002239
0.001544
0.000849
0.000772
0.000386
0.00054
0.000733
0.003312
0.002442
0.002208
0.001405
0.000602
0.000502
0.000201
0.000201
0.000502
November 2012
3-212
Draft - Do Not Cite or Quote

-------
Reference Location Size Bin
Hays etal. (2011) Raleigh, NC 0.03-0.06
0.06-0.108
0.108-0.17
0.17-0.26
0.26-0.4
0.4-0.65
0.65-1
1-1.6
1 .6-2.5
2.5-4.4
4.4-6.8
6.8-10
10-18
Chen et al. (201 Ob) Taipei, Taiwan tunnel <0.1
0.1-2.5
2.5-10
Concentration
0.000186
0.000395
0.000732
0.001486
0.003593
0.007315
0.00423
0.002719
0.002701
0.003346
0.00123
0.000934
0.001265
0.018409
0.019773
0.030682
<0.1 0.00125
0.1-2.5
2.5-10
Bruggemann et al. (2009) Dresden, Germany 0.05-0.14
Summer 0.14-0.42
0.42-1.2
1 .2-3.5
3.5-10
Winter 0.05-0.14
0.14-0.42
0.42-1.2
1 .2-3.5
3.5-10
0.020625
0.024375
0.001078
0.002874
0.004671
0.001617
0.000539
0.002335
0.00521
0.013293
0.003054
0.000539
November 2012
3-213
Draft - Do Not Cite or Quote

-------
Reference Location Size Bin
Harrison et al. (2003) Birmingham, U.K. <0.2
0.2-1
1-2
2-10
Concentration
0.00685
0.014923
0.002446
0.00318
>10 0.000489
Wang et al. (2006d) Kanazawa, Japan 0.1-0.43
0.43-0.65
0.65-1.1
1.1-2.1
2.1-3.3
3.3-4.7
4.7-7
7-11
11-18
Martuzevicius et al. (2004) Cincinnati, OH ~0.1-0.18
Cycle VIII 0.18-0.32
0.32-0.56
0.56-1
1-1.8
1 .8-3.2
3.2-5.6
5.6-10
Cycle IX -0.1 -0.1 8
0.18-0.32
0.32-0.56
0.56-1
1-1.8
1 .8-3.2
3.2-5.6
5.6-10
0.000792
0.000748
0.00118
0.00103
0.000393
0.000678
0.000375
0.000229
0.000125
0.000758
0.002045
0.003258
0.00447
0.005758
0.00697
0.008333
0.009545
0.000455
0.001591
0.002879
0.004091
0.005379
0.006667
0.007879
0.009091
November 2012
3-214
Draft - Do Not Cite or Quote

-------
Reference
Location
  Size Bin
Concentration
Lim et al. (2006)
'Summer and Fall data not provided
because there is more uncertainty in
the sites for those  datasets.
Los Angeles, CA
                                    Los Angeles River Watershed #1
  Winter
                                      Spring
                                    Los Angeles River Watershed #2
                                     Winter
                                      Spring
                                    Los Angeles River Watershed #3
                                     Winter
                                      Spring
                                    Santa Ana River Watershed
                                     Winter
                                                                            6-11
                                                                            11-20
                                        20-29
                                                                            >29
                                                                            6-11
                                                                            11-20
                                                                            20-29
                                                                            >29
                                                                            6-11
                                                                            11-20
                                                                            20-29
                                                                            >29
                                                                            6-11
                                                                            11-20
                                                                            20-29
                                                                            >29
                                                                            6-11
                                                                            11-20
                                                                            20-29
                                                                            >29
                                                                            6-11
                                                                            11-20
                                                                            20-29
                                                                            >29
                                                                            6-11
                                                                            11-20
                                                                            20-29
                                                                            >29
                                                                                                                    1.315
                                                                                                                    0.743
                                                                                                                    0.821
                                                                                                                    2.302
                                                                                                                    1.212
                                                                                                                    1.485
                                                                                                                    3.025
                                                                                                                    1.251
                                                                                                                    0.547
                                                                                                                    1.212
                                                                                                                    1.042
                                                                                                                    1.212
                                                                                                                    0.312
                                                                                                                    0.782
                                                                                                                    1.116
                                                                                                                    1.055
                                                                                                                    0.547
                                                                                                                    1.016
                                                                                                                    2.299
                                                                                                                    1.055
                                                                                                                    0.508
                                                                                                                    1.524
                                                                                                                    0.097
                                                                                                                    0.235
                                                                                                                    0.195
November 2012
            3-215
Draft - Do Not Cite or Quote

-------
Reference Location Size Bin
Spring 6-11
11-20
20-29
>29
Ballona Creek Watershed 6-1 1
Winter 11-20
20-29
>29
Spring 6-11
11-20
20-29
>29
Dominguez Creek Watershed 6-1 1
Winter 11-20
20-29
>29
Spring 6-11
11-20
20-29
>29
Malibu Creek (non-urban) 6-1 1
Winter 11-20
Concentration
0.185
0.235
0.039
0.156
1.263
1.016
0.313
0.664
5.064
1.29
0.312
2.58
2.315
1.368
0.547
0.625
0.683
0.469
0.078
0.508
0.201
0.391
20-29
>29
Spring 6-11
11-20
0.039
0.211
0.156
20-29
>29
0.117
November 2012
3-216
Draft - Do Not Cite or Quote

-------
3.8.5
Pb Concentration in a Multipollutant Context
CO
PM10
N02
PM2.5
SO2
03
0
O 00
0
O 0® C
OCO ODGDD
> OOOarOTDOOO 0 0
 ooo
fjTlTi OCX IHQQDKS
O OCOBOOQD
CO 0 OOO «
3SDDO O OO O
                -1
                   -0.5
                                                                0.5
Note: Correlations were calculated from available data when data were above MDL and there were at least 30 data pairs available
for comparison.
Correlations for individual sites are shown with black open circles, while median correlations are illustrated with a red square.

Figure 3-62     Spearman correlations of monitored non-source Pb-TSP
                 concentration  with daily averages of copollutant concentrations,
                 2008-2010.
November 2012
                           3-217
Draft - Do Not Cite or Quote

-------
            Source, SO2
        Non-Source, SO2
           Source, PM2s
       Non-Source, PM25
           Source, PMio
       Non-Source, PMio
             Source, O3
         Non-Source, O3
            Source, N02
        Non-Source, NO2
            Source, CO
        Non-Source, CO
            Source, SO2
        Non-Source, SO2
          Source, PM2 5
       Non-Source, PM25
           Source, PMio
       Non-Source, PM10
             Source, O3
         Non-Source, O3
            Source, NO2
        Non-Source, NO2
            Source, CO
        Non-Source, CO
         US Summer
             o      oo
OOO  GOKHDGD  QDO O  O    O
      O    O    
                         OOO OCD O O OBDODDODO
                                  O O        O    O
         O OOGD CHD GOOD   OOO   QtJOO O dD CH8DCXBD  00   O      (
                                               OO
                             O  O       OODOOOOIBEMDCDO C5DCHWZID <3D
                                           O     O
                       O          O     O OOCDDSD O O dHOSDCJ)  OO (3D
                       1.0
                    -0.5            0.0             0.5
                    Spearman Correlation Coefficient
                                  1.0
Note: Top panel: Summer; Bottom panel: Fall.
Figure 3-63
Seasonal correlations of monitored Pb-TSP concentration with
copollutant concentrations, 2007-2008.
November 2012
                      3-218
            Draft - Do Not Cite or Quote

-------
        Non-Source, SO2 ' "


          Source, PM2 5


       Non-Source, PM2s


       Non-Source, PM,0


         Non-Source, O3


        Non-Source, N02 • -


        Non-Source, CO . -
US Winter
o
0 0
000 0
0
o o o o o
O 000 O (
 0
0 0
0 0 00 0
                      -1.0
  -0.5            0.0             0.5

   Spearman Correlation Coefficient
                       1.0
        Non-Source, S02 -


           Source, PM2.s -


       Non-Source, PM25 -


       Non-Source, PMio


         Non-Source, 03 .


        Non-Source, NO2 .


         Non-Source, CO .
                         US Spring
o      o

    QD O
                      -1.0
Note: Top panel: Winter; Bottom panel: Spring.
                                  oo  o
                            o oo
QD
  -0.5            0.0            0.5

   Spearman Correlation Coefficient
                      1.0
Figure 3-64     Seasonal correlations of monitored Pb-TSP concentration with
                 copollutant concentrations, 2009.
November 2012
     3-219
  Draft - Do Not Cite or Quote

-------
              Non-Source, SO2
                Source. PM25 "

             Non-Source, PM2 5 -
             Non-Source, PM,0 -
               Non-Source, O3 -
              Non-Source, NO2
              Non-Source, CO -
                             US Summer
             ooo          o
                    o
          O      00         O
             o              o  o
          O  ODD    O
                          O  O
     OO  O                O
                           -1.0
                                        -0.5            0.0           0.5
                                        Spearman Correlation Coefficient
                                                                                1.0
              Non-Source, SO2
                Source, PM2s
             Non-Source, PM2 5
             Non-Source, PM,0
               Non-Source, Os
              Non-Source, NO2
              Non-Source. CO
                              US Fall
                             O  00
                          00   O  O
                     000
             O  00 O O
                       o  o
                           -1.0
Note: Top panel: Summer; Bottom panel: Fall.
                                        -0.5           0.0           0.5
                                         Spearman Correlation Coefficient
                                                                                 1.0
Figure 3-65     Seasonal correlations of monitored Pb-TSP concentration with
                  copollutant concentrations, 2009.
November 2012
3-220
Draft - Do Not Cite or Quote

-------
Top panel: Winter; Bottom panel: Spring.
Note: "nvol" = non-volatile, "vol" = volatile, and organic carbon (OC) samples were blank-adjusted.


Figure 3-66     Seasonal correlations of monitored Pb-PM2.s concentration with

                 copollutant concentrations, 2007-2009.
November 2012
3-221
Draft - Do Not Cite or Quote

-------
Zn —
Br -
Cu -
K -
s -
$04-2 —
Ca -
luh -
Cmstal —
Fe —
Se -
N03- -
11 -
EC -
^ -
Si —
Mg -
K+ —
V —
NH4t -
Ha -

Hg
Cd —
Na+2 -
vol N03- -
Ni —
fs —
nvol H03- —
0 | 	 1 |
o TO o i 	 H |
0, 	 1 ,
0 | 	 1 |
0 0 1 	 1 I
0 00 1 	 < I
0 0 1 	 1 1
o o  cm 	 1 | |-
0 | 	 ( , |_
0 0 1 	 H 1 H-
0 1 	 1 1 1
0 01 	 1 • h 	
0 0 1 	 1 | | 	
1 	 1 1 1 	
1 _J 1 	 J
OO O 1 	 1 1 1 	 1
0 | 	 1 , | 	
1 	 1 1 1 	
(. 	 1
-] 	 1
( 	 ,
	 1
]::::::::::
	 1 o
	 1 0
	 1
	 1 0
	 10 0
	 100 00

o o
-HO 0

Zn —
K —
Cu -
Br -
oc -
Fe -
vol H03- -
Cmstal —
Ca -
K+ -


NH4+ -
Si -


Cr -
V -

Mg
Na+2 -
Ni -
Hg -
Cl —
Cd -
cool 	 1 | h 	 1
0 , 	 H , | 	 ,
O O OTC O 1 	 ^ 1 1 	 1
i-CD
o  	 1 0 0
o mi- n__J__^ — r H


1 • — i H

000 1 	 1 1 1 	 1 OJ 0
1 	 | | | 	 1010

Top panel: Summer; Bottom panel: Fall.
Note: "nvol" = non-volatile, "vol" = volatile, and organic carbon (OC) samples were blank-adjusted.


Figure 3-67     Seasonal correlations of monitored Pb-PM2.s concentration with

                 copollutant concentrations, 2007-2009.
November 2012
3-222
Draft - Do Not Cite or Quote

-------
Table 3-29 Copollutant exposures for various trace metal studies.
Adgate et al. (2007)

Location
PM2.5
Pb
S
Ca
Al
Na
Fe
Mg
K
Ti
Zn
Cu
Ni
Mn
Sb
Cd
V
La
Cs
Th
Sc
Ag
Co
Cr
Si
Cl
Se
Rb
Sr
As
l-R(med)a'b
Personal
(median)0
Minnesota, U.S.

1.5
272.1
85.0
23.3
20.6
43.1
16.3
38.4
0.8
6.5
1.-0.15
2.4
0.21
0.12
0.12
0.05
0.00
0.00
0.00
0.00
0.07
0.02
1.2







3.2
351.6
174.1
58.6
31.9
78.6
27.5
47.5
1.4
9.6
4.9
1.8
2.3
0.30
0.14
0.16
0.11
0.00
0.00
0.01
0.08
0.07
2.6






Riedikeretal. (2003)
Vehicle
(range)0
Roadside
(range)0
New Jersey, U.S.
24,000
2-3
905-1 ,592
31-44


307-332

6-75
9-10
5-10
18-32
0
3-4

4-6
1






2
198-464
7-32
1
1
5-28
1
31 ,579
4-6
1,416-2,231
18-40


82-163

23-57
6-10
14-17
8-16
0
3

4-7
1






1
338-672
3-9
1-2
1
1
1
Pekey et al.
(2010) Molnar et al. (2007)
I -near industry I-R l-School l-Pre-School
(range)3 (median)a'b (median)3 (median)3
Kocaeli, Turkey Stockholm, Sweden
24,400-29,800
34-85 2.8 2.5 1.7
435-489 330 290 220
309-452 70 110 58
53-60

44-58 57 100 71

160-215 120 96 67
29-39 8.0 13 8.7
51-88 14 17 11
21-58 9.3 1.7 2.1
2-3 0.99 1.0 0.72
28-32 2.2 2.5 2.1


3-5 2.5 2.7 1.8






3-8 <1.1 1.3 1.1
387-401




1-2
Mo
Br
al: Indoor; Units
bR: Residential;
°Units: ng/m3

: ng/m3
Units: ng/m3






2.1 1.3 1.3

November 2012
3-223
Draft - Do Not Cite or Quote

-------
References for Chapter 3
     Abel. MT; Cobb. GP; Presley. SM; Ray. GL; Rainwater. TR; Austin. GP; Cox. SB; Anderson. TA;
        Leftwich. BD; Kendall. RJ: Suedel. BC. (2010). Lead distributions and risks in New Orleans
        following hurricanes Katrina and Rita. Environ Toxicol Chem 29: 1429-1437.
        http://dx.doi.org/10.1002/Etc.205

     Abel. MT; Presley. SM; Rainwater. TR; Austin. GP: Cox. SB; Mcdaniel. LN; Marsland. EJ; Leftwich.
        BD; Anderson. TA;  Kendall. RJ: Cobb. GP. (2007). Spatial and temporal evaluation of metal
        concentrations in soils and sediments from New Orleans, Louisiana, USA, following hurricanes
        Katrina and Rita. Environ Toxicol Chem 26: 2108-2114. http://dx.doi.Org/10.1897/06-595R.l

     Adgate. JL; Mongin. SJ: Pratt. GC: Zhang. J; Field. MP; Ramachandran. G: Sexton. K. (2007).
        Relationships between personal, indoor, and outdoor exposures to trace elements in PM2.5. Sci Total
        Environ 386: 21-32. http://dx.doi.Org/10.1016/i.scitotenv.2007.07.007

     Aelion. CM; Davis. HT; Mcdermott. S: Lawson. AB. (2009). Soil metal concentrations and toxicity:
        Associations with distances to industrial facilities and implications for human health. Sci Total
        Environ 407: 2216-2223. http://dx.doi.Org/10.1016/i.scitotenv.2008.ll.033

     Al-Malack. MH. (2001). Migration of lead from unplasticized polyvinyl chloride pipes. J Hazard Mater
        82: 263-274.  http://dx.doi.org/10.1016/S0304-3894(00)00366-6

     Al-Masri. MS: Al-Kharfan. K; Al-Shamali. K. (2006). Speciation of Pb, Cu and Zn determined by
        sequential extraction for identification of air pollution sources in Syria. Atmos Environ 40: 753-761.
        http://dx.doi.0rg/10.1016/i.atmosenv.2005.10.008

     Allott. RW: Hewitt. CN; Kelly. MR. (1989). The environmental half-lives and mean residence times of
        contaminants in dust for an urban environment: Barrow-in-Furness. Sci Total Environ 93: 403-410.
        http://dx.doi.org/10.1016/0048-9697(90)90131-0

     Alpert. DJ: Hopke. PK. (1981). A determination of the sources of airborne particles collected during the
        regional air pollution study. Atmos Environ 15: 675-687.
        http://dx.doi.org/10.1016/0004-6981(81)90272-9

     ALRA. Airborne Lead Reduction Act of 1984. S. 2609. (1984).
        http://heinonline.org/HOL/Page?collection=congrec&handle=hein.cbhear/cbhearings2029&tvpe=Ima
        ge&id=l

     Amato. F; Pandolfi. M; Viana. M; Querol. X; Alastuev. A; Moreno. T. (2009). Spatial and chemical
        patterns of PM10 in road dust deposited in urban environment. Atmos Environ 43: 1650-1659.
        http://dx.doi.0rg/10.1016/i.atmosenv.2008.12.009

     Annibaldi. A; Truzzi. C: Illuminati. S: Bassotti. E; Scarponi. G. (2007). Determination of water-soluble
        and insoluble (dilute-HCl-extractable) fractions of Cd, Pb and Cu in Antarctic aerosol by square wave
        anodic stripping voltammetry: Distribution and summer seasonal evolution at Terra Nova Bay
        (Victoria Land). Anal Bioanal Chem 387: 977-998. http://dx.doi.org/10.1007/s00216-006-0994-0

     Anttila. P; Makkonen. U; Hellen. H; Kyllonen. K; Leppanen. S: Saari. H; Hakola. H. (2008). Impact of
        the open biomass fires in spring and summer of 2006 on the chemical composition of background air
        in south-eastern Finland. Atmos Environ 42: 6472-6486.
        http://dx.doi.0rg/10.1016/i.atmosenv.2008.04.020

     Ariola. V: D'Alessandro. A; Lucarelli. F; Marcazzan. G: Mazzei. F; Nava. S: Garcia-Orellana. I; Prati.  P;
        Valli. G: Vecchi. R;  Zucchiatti. A. (2006). Elemental characterization of PM10, PM2.5 and PM1 in  the
        town of Genoa (Italy). Chemosphere 62: 226-232.
        http://dx.doi.0rg/10.1016/i.chemosphere.2005.05.004
November 2012                              3-224                  Draft - Do Not Cite or Quote

-------
     ASTM (American Society for Testing and Materials). (2007). Standard specification for aviation
        gasolines. (ASTM D910-06). West Conshohocken, PA: American Society for Testing and Material.
        http://www.astm.org/Standards/D910.htm

     Atkinson. CA; Jollev. DF; Simpson. SL. (2007). Effect of overlying water pH, dissolved oxygen, salinity
        and sediment disturbances on metal release and sequestration from metal contaminated marine
        sediments. Chemosphere 69: 1428-1437. http://dx.doi.Org/10.1016/i.chemosphere.2007.04.068

     Aucott. M; Caldarelli. A. (2012). Quantity of Lead Released to the Environment in New Jersey in the
        Form of Motor Vehicle Wheel Weights. Water Air Soil Pollut 223:  1743 -175 2.
        httrj://dx.doi.org/10.1007/sll270-011-0979-2

     Aznar. JC: Richer-Lafleche. M; Begin. C: Rodrigue. R. (2008a). Spatiotemporal reconstruction of lead
        contamination using tree rings and organic soil layers. Sci Total Environ 407: 233-241.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.09.044

     Aznar. JC: Richer-Lafleche. M; Cluis. D. (2008b). Metal contamination in the lichen Alectoria
        sarmentosa near the copper smelter of Murdochville,  Quebec. Environ Pollut 156:  1.
        http://dx.doi.0rg/10.1016/i.envpol.2007.12.037

     Bacardit. M; Camarero. L. (2010a). Atmospherically deposited major and trace elements in the winter
        snowpack along a gradient of altitude in the Central Pyrenees: the seasonal record of long-range fluxes
        over SWEurope. Atmos Environ 44: 582-595. http://dx.doi.Org/10.1016/i.atmosenv.2009.06.022

     Bacardit. M; Camarero. L. (201 Ob). Modelling Pb, Zn and As transfer from terrestrial to aquatic
        ecosystems during the ice-free season in three Pyrenean catchments. Sci Total Environ 408:
        5854-5861. http://dx.doi.Org/10.1016/i.scitotenv.2010.07.088

     Bacon. SN; Mcdonald. EV; Amit. R; Enzel. Y; Crouvi. O. (2011). Total suspended paniculate matter
        emissions at high friction velocities from desert landforms. J Geophys Res 116: F03019.
        http://dx.doi.org/10.1029/2011JF001965

     Balogh. SJ: Triplett. LD; Engstrom. PR; Nollet. YH. (2010). Historical trace metal loading to a large river
        recorded in the sediments of Lake St.  Croix, USA.  J Paleolimnol 44: 517-530.
        http://dx.doi.org/10.1007/sl0933-010-9434-4

     Barrett. J: Taylor. K; Hudson-Edwards. K; Charnock. J. (2010). Solid-phase speciation of Pb in urban
        road dust sediment: A XANES and EXAFS study. Environ Sci Technol 44: 2940-2946.
        http://dx.doi.org/10.1021/es903737k

     Batonneau. Y; Bremard. C: Gengembre. L; Laureyns. J: Le Maguer. A; Le Maguer. D; Perdrix. E;
        Sobanska. S. (2004). Speciation of PM10 sources of airborne nonferrous metals within the 3-km zone
        of lead/zinc smelters. Environ Sci Technol 38: 5281-5289. http://dx.doi.org/10.1021/es0497173

     Bein. KJ; Zhao. Y; Peknev. NJ; Davidson. CI; Johnston. MV; Wexler. AS. (2006). Identification of
        sources of atmospheric PM at the Pittsburgh supersite, part II: Quantitative comparisons of single
        particle, particle number, and particle mass measurements. Atmos Environ 40: S424-S444.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.01.039

     Bein. KJ: Zhao. YJ; Johnston. MV: Wexler. AS. (2007). Identification of sources of atmospheric PM at
        the Pittsburgh supersite, part III: Source characterization. Atmos Environ 41:  3974-3992.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.01.039

     Bench. G: Grant. PG: Ueda. D: Cliff. SS: Perry. KD; Cahill. TA. (2002). The use of STIM and PESA to
        measure profiles of aerosol mass and hydrogen content, respectively, across Mylar rotating drums
        impactor samples. Aerosol Sci Technol 36:  642-651. http://dx.doi.org/10.1080/02786820252883874

     Bennett. JD. (2007). Memorandum regarding Doe Run-Herculaneum State Implementation Plan (SIP)
        dispersion modeling review. Jefferson City, MO: Missouri Department of Natural Resources.
        http://www.dnr.mo.gov/env/apcp/docs/drhsipmodmemol.pdf
November 2012                             3-225                   Draft - Do Not Cite or Quote

-------
     Berg. T; Aas. W: Pacytia. J: Uggerud HT; Vadset M. (2008). Atmospheric trace metal concentrations at
        Norwegian background sites during 25 years and its relation to European emissions. Atmos Environ
        42: 7494-7501. http://dx.doi.Org/10.1016/i.atmosenv.2008.05.020

     Berthelot. Y; Valton. E; Auroy. A; Trottier. B; Robidoux. PY. (2008). Integration of lexicological and
        chemical tools to assess the bioavailability of metals and energetic compounds in contaminated soils.
        Chemosphere 74: 166-177. http://dx.doi.Org/10.1016/i.chemosphere.2008.07.056

     Besser. JM; Brumbaugh. WG: May. TW: Church. SE; Kimball. BA. (2001). Bioavailability of metals in
        stream food webs and hazards to brook trout (Salvelinus fontinalis) in the upper Animas River
        watershed, Colorado. Arch Environ Contam Toxicol 40: 48-59.
        http://dx.doi.org/10.1007/s002440010147

     Besser. JM; Brumbaugh. WG: May. TW: Schmitt CJ. (2007). Biomonitoring of lead, zinc, and cadmium
        in streams draining lead-mining and non-mining areas, Southeast Missouri, USA. Environ Monit
        Assess 129: 227-241. http://dx.doi.org/10.1007/sl0661-006-9356-9

     Biasioli. M; Barberis. R; Ajmone-Marsan. F. (2006). The influence of a large city on some soil properties
        and metals content. Sci Total Environ 356: 154-164. http://dx.doi.Org/10.1016/i.scitotenv.2005.04.033

     Biggins. PDE: Harrison. RM. (1979). Atmospheric chemistry of automotive lead. Environ Sci Technol 13:
        558-565. http://dx.doi.org/10.1021/es60153a017

     Biggins. PDE: Harrison. RM. (1980). Chemical speciation of lead compounds in street dusts. Environ Sci
        Technol 14: 336-339. http://dx.doi.org/10.1021/es60163a005

     Bindler. R; Renberg. I; Klaminder. J. (2008). Bridging the gap between ancient metal pollution and
        contemporary biogeochemistry.  J Paleolimnol 40: 755-770.
        http://dx.doi.org/10.1007/sl0933-008-9208-4

     Birch. G: McCreadv. S.  (2009). Catchment condition as a major control on the quality of receiving basin
        sediments (Sydney Harbour, Australia). Sci Total Environ 407: 2820-2835.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.12.051

     Birch. G: O'Hea. L. (2007). The chemistry of suspended paniculate material in a highly contaminated
        embayment of Port Jackson (Australia) under quiescent, high-wind and heavy-rainfall conditions.
        Environ Geol 53: 501-516. http://dx.doi.org/10.1007/s00254-007-0662-5

     Birmili. W: Allen. AG: Barv. F; Harrison. RM. (2006). Trace metal concentrations and water solubility in
        size-fractionated atmospheric particles and influence of road traffic. Environ Sci Technol 40:
        1144-1153. http://dx.doi.org/10.1021/es0486925

     Bonnard. R: Mckone. TE. (2009). Integration of the predictions of two models with dose measurements in
        a case study of children exposed to the emissions from a lead smelter. Hum Ecol Risk Assess 15:
        1203-1226. http://dx.doi.org/10.1080/10807030903304849

     Boonfueng. T; Axe. L; Xu. Y; Tyson. TA. (2006). Nickel and lead sequestration in manganese oxide-
        coated montmorillonite. J  Colloid Interface Sci 303: 87-98.
        http://dx.doi.0rg/10.1016/i.jcis.2006.07.056

     Boutron. CF; Goerlach.  U; Candelone. JP; Bolshov. MA; Delmas. RJ. (1991). Decrease in anthropogenic
        lead, cadmium and zinc in Greenland snows since the late 1960s. Nature 353:  153-156.
        http://dx.doi.org/10.1038/353153aO

     Bouvet. M; Francois. D; Schwartz.  C. (2007). Road soil retention Pb leached from MSWI bottom  ash.
        Waste Manag 27: 840-849. http://dx.doi.Org/10.1016/j.wasman.2006.06.003

     Britter. RE; Hanna. SR.  (2003). Flow and dispersion in urban areas. Annu Rev Fluid Mech 35: 469-496.
        http://dx.doi.org/10.1146/annurev.fluid.35.101101.161147

     Brown. RW: Gonzales. C: Hooper. MJ; Bavat. AC: Fornerette. AM; McBride. TJ; Longoria. T; Mielke.
        HW. (2008). Soil lead (Pb) in residential transects through Lubbock, Texas: A preliminary assessment.
        Environ Geochem Health 30: 541-547. http://dx.doi.org/10.1007/sl0653-008-9180-y
November 2012                             3-226                   Draft - Do Not Cite or Quote

-------
     Bruggemann. E; Gerwig. H; Gnauk. T; Muller. K; Herrmann. H. (2009). Influence of seasons, air mass
        origin and day of the week on size-segregated chemical composition of aerosol particles at a kerbside.
        Atmos Environ 43: 2456-2463. http://dx.doi.Org/10.1016/i.atmosenv.2009.01.054

     Brumbaugh. WG: May. TW: Besser. JM; Allert. AL; Schmitt. CJ. (2007). Assessment of elemental
        concentrations in streams of the new lead belt in southeastern Missouri, 2002-05.
        (USGSSIR20075057). Reston, VA: U.S. Geological Survey, http://pubs.usgs.gov/sir/2007/5057

     Bukowiecki. N; Lienemann. P; Hill. M; Furger. M; Richard. A; Amato. F; Prevot. ASH; Baltensperger. U;
        Buchmann. B; Gehrig. R. (2010). PM10 emission factors for non-exhaust particles generated by road
        traffic in an urban street canyon and along a freeway in Switzerland. Atmos Environ 44: 2330-2340.
        http://dx.doi.0rg/10.1016/i.atmosenv.2010.03.039

     Buonanno. G: Fuoco. FC: Stabile. L. (2011). Influential parameters on particle exposure of pedestrians in
        urban microenvironments. Atmos Environ 45: 1434-1443.
        http://dx.doi.0rg/10.1016/i.atmosenv.2010.12.015

     Bur. T; Probst. JL; N'Guessan. M; Probst. A. (2009). Distribution and origin of lead in stream sediments
        from small agricultural catchments draining Miocene molassic deposits (SW France). Appl Geochem
        24: 1324-1338. http://dx.doi.Org/10.1016/i.apgeochem.2009.04.004

     Cahill. CF. (2003). Asian aerosol transport to Alaska during ACE-Asia. J Geophys Res  108: 8664.
        http://dx.doi.org/10.1029/2002JD003271

     Cahill. TA. (1981). Innovative aerosol sampling devices based upon PIXE capabilities.  Nuclear
        Instruments and Methods 181: 473-480. http://dx.doi.org/10.1016/0029-554X(81)90652-2

     Cahill. TA; Feenev. PJ: Eldred. RA. (1987). Size-time composition profile of aerosols using the drum
        sampler. Nucl lustrum Methods  Phys Res B 22: 344-348.
        http://dx.doi.org/10.1016/0168-583X(87)90355-7

     Callender. E; Rice. KG. (2000). The urban environmental gradient: Anthropogenic influences on the
        spatial and temporal distributions of lead and zinc in sediments. Environ Sci Technol 34: 232-238.
        http://dx.doi.org/10.1021/es990380s

     Canepari. S: Astolfi. ML; Moretti. S: Curini. R. (2010). Comparison of extracting solutions for elemental
        fractionation in airborne particulate matter. Talanta 82: 834-844.
        http://dx.doi.0rg/10.1016/i.talanta.2010.05.068

     Canepari. S: Cardarelli. E; Perrino.  C: Catrambone. M; Pietrodangelo. A; Strincone. M. (2006). Two-
        stage chemical fractionation method for the analysis of elements and non-volatile inorganic ions in
        PM10 samples: Application to ambient samples collected in Rome (Italy). Atmos Environ 40:
        7908-7923. http://dx.doi.Org/10.1016/i.atmosenv.2006.07.005

     Canepari. S: Perrino. C: Olivieri. F; Astolfi. ML. (2008). Characterisation of the traffic  sources of PM
        through size-segregated sampling, sequential leaching and ICP analysis. Atmos Environ 42:
        8161-8175. http://dx.doi.0rg/10.1016/i.atmosenv.2008.07.052

     Canovas. CR; Olias. M; Nieto. JM; Galvan. L. (2010).  Wash-out processes of evaporitic sulfate salts in
        the Tinto river: Hydrogeochemical evolution and environmental impact. Appl Geochem 25: 288-301.
        http://dx.doi.0rg/10.1016/i.apgeochem.2009.ll.014

     Cantwell. MG: Burgess. RM; King. JW. (2008). Resuspension of contaminated field and formulated
        reference sediments Part I: Evaluation of metal release under controlled laboratory conditions.
        Chemosphere73: 1824-1831. http://dx.doi.Org/10.1016/i.chemosphere.2008.08.007

     Caravanos. J: Weiss. AL; Blaise. MJ;  Jaeger. RJ. (2006a). A survey of spatially distributed exterior dust
        lead loadings in New York City. Environ Res 100: 165-172.
        http://dx.doi.0rg/10.1016/i.envres.2005.05.001
November 2012                             3-227                   Draft - Do Not Cite or Quote

-------
     Carr. E; Lee. M; Marin. K; Holder. C: Hover. M; Pedde. M; Cook. R; Touma. J. (2011). Development and
        evaluation of an air quality modeling approach to assess near-field impacts of lead emissions from
        piston-engine aircraft operating on leaded aviation gasoline. Atmos Environ 45: 5795-5804.
        http://dx.doi.0rg/10.1016/i.atmosenv.2011.07.017

     Chalmers. AT; Van Metre. PC: Callender. E. (2007). The chemical response of particle-associated
        contaminants in aquatic sediments to urbanization in New England, USA. J Contam Hydrol 91: 4-25.
        http://dx.doi.0rg/10.1016/i.jconhvd.2006.08.007

     Chan. YC: Cohen. DP; Hawas. O: Stelcer. E; Simpson. R; Denison. L; Wong. N; Hodge. M; Comino. E;
        Carswell. S. (2008). Apportionment of sources of fine and coarse particles in four major Australian
        cities by positive matrix factorisation. Atmos Environ 42: 374-389.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.09.030

     Chen. FL; Vanderpool. R; Williams. R:  Dimmick. F; Grover. BD; Long. R; Murdoch. R. (2011 a). Field
        evaluation of portable and central site PM samplers emphasizing additive and differential mass
        concentration estimates. Atmos Environ 45: 4522-4527.
        http://dx.doi.0rg/10.1016/i.atmosenv.2011.02.006

     Chen. S: Tsai. C: Huang. C: Chen. H; Chen. S: Lin. C: Tsai. J: Chou. C. (2010b). Chemical mass closure
        and chemical characteristics of ambient ultrafine particles and other PM fractions. Aerosol Sci Technol
        44: 713-723.

     Cheng. H; Hu. Y. (2010). Lead (Pb) isotopic fingerprinting and its applications in lead pollution studies in
        China: A review [Review]. Environ Pollut 158: 1134-1146.
        http://dx.doi.0rg/10.1016/i.envpol.2009.12.028

     Chester. R; Lin. FJ; Murphy. KJT. (1989). A three stage sequential leaching scheme for the
        characterisation of the sources and environmental mobility of trace metals in the marine aerosol.
        Environ Technol 10: 887 - 900. http://dx.doi.org/10.1080/09593338909384810

     Cho. SH; Richmond-Bryant. J: Thornburg. J: Portzer. J: Vanderpool. R; Cavender. K; Rice. J. (2011). A
        literature review of concentrations and size distributions of ambient airborne Pb-containing particulate
        matter. Atmos Environ 45: 5005-5015. http://dx.doi.Org/10.1016/j.atmosenv.2011.05.009

     Choel M; Deboudt. K; Flament.  P; Lecornet. G: Perdrix. E; Sobanska. S. (2006). Fast evolution of
        tropospheric Pb- and Zn-rich particles in the vicinity of a lead smelter. Atmos Environ 40: 4439-4449.
        http://dx.doi.0rg/10.1016/i.atmosenv.2006.04.027

     Chon. HS: Ohandja. DG: Voulvoulis. N. (2010). Implementation of E.U. water framework directive:
        Source assessment of metallic substances at catchment levels. J Environ Monit 12: 36-47.
        http://dx.doi.org/10.1039/b907851g

     Chow. TJ; Johnstone. MS. (1965). Lead isotopes in gasoline and aerosols of Los Angeles Basin,
        California. Science 147: 502-503. http://dx.doi.org/10.1126/science.147.3657.502

     Cizmecioglu. SC: Muezzinoglu. A. (2008). Solubility of deposited airborne heavy metals. Atmos Res 89:
        396-404. http://dx.doi.0rg/10.1016/i.atmosres.2008.03.012

     Clark. GM.  (2003). Occurrence and transport of cadmium, lead, and zinc in the spokane river basin, Idaho
        and Washington, water years  1999-2001. (Water-Resources Investigations Report 024183). Boise, ID:
        U.S. Geological Survey. http://id.water.usgs.gov/PDF/wri024183/clarkfinal2.pdf

     Clark. HF; Brabander. DJ; Erdil.  RM. (2006). Sources, sinks, and exposure pathways of lead in urban
        garden soil. J Environ Qual 35: 2066-2074. http://dx.doi.org/10.2134/ieq2005.0464

     Cloquet. C: Carignan. J: Libourel. G. (2006). Isotopic composition of Zn and Pb atmospheric depositions
        in an urban/periurban area of northeastern France. Environ Sci Technol 40: 6594-6600.
        http://dx.doi.org/10.1021/es0609654

     CNL (Crocker Nuclear Laboratory). (2001). Version II IMPROVE sampler:  Operating procedures manual
        for use in the IMPROVE monitoring network. (TI 201 A). Davis, CA.
        http://vista.cira.colostate.edu/improve/publications/SOPs/ucdavis sops/ti201a v2.pdf
November 2012                              3-228                  Draft - Do Not Cite or Quote

-------
     Cobb. GP: Abel MT; Rainwater. TR; Austin. GP; Cox. SB; Kendall. RJ: Marsland. EJ; Anderson. TA;
        Leftwich. BD; Zak. JC: Presley. SM. (2006). Metal distributions in New Orleans following hurricanes
        Katrina and Rita: A continuation study. Environ Sci Technol 40: 4571-4577.
        http://dx.doi.org/10.1021/es060041g

     Cohen. DP; Crawford. J: Stelcer. E; Bac. VT. (2010). Characterisation and source apportionment of fine
        paniculate sources at Hanoi from 2001 to 2008. Atmos Environ 44: 320-328.
        http://dx.doi.0rg/10.1016/i.atmosenv.2009.10.037

     Company. R; Serafim. A; Lopes. B; Cravo. A; Shepherd. TJ; Pearson. G: Bebianno. MJ. (2008). Using
        biochemical and isotope geochemistry to understand the environmental and public health implications
        of lead pollution in the lower Guadiana River, Iberia: A freshwater bivalve study. Sci Total Environ
        405: 109-119. http://dx.doi.0rg/10.1016/j.scitotenv.2008.07.016

     Cong. Z; Kang. S: Liu. X; Wang. G. (2007).  Elemental composition of aerosol in the Nam Co region,
        Tibetan Plateau, during summer monsoon season. Atmos Environ 41: 1180-1187.
        http://dx.doi.0rg/10.1016/i.atmosenv.2006.09.046

     Conor Pacific Environmental Technologies Inc. (2000). Airborne particulate matter, lead and manganese
        at Buttonville airport.  (041-6710). Toronto, Ontario:  Environment Canada.
     Courtin-Nomade. A; Soubrand-Colin. M; Marcus. MA; Fakra. SC. (2008). Evidence for the incorporation
        of lead into barite from waste rock pile materials. Environ Sci Technol 42: 2867-2872.
        http://dx.doi.org/10.1021/es702822k

     Couture. RM; J-F. C: Auger. D; Claisse. D; Gobeil. C: Cossa. D. (2010). Seasonal and decadal variations
        in lead sources to eastern North Atlantic mussels. Environ Sci Technol 44: 1211-1216.
        http://dx.doi.org/10.1021/es902352z

     Coynel. A; Schafer. J; Blanc. G: Bossy. C. (2007). Scenario of particulate trace metal and metalloid
        transport during a major flood event inferred from transient geochemical signals. Appl Geochem 22:
        821-836. http://dx.doi.0rg/10.1016/i.apgeochem.2006.10.004

     Csavina. JA; Landazuri. A; Wonaschutz. A; Rine. K; Rheinheimer. P; Barbaris. B; Conant. W: Saez. AE;
        Betterton. EA. (2011). Metal and metalloid contaminants in atmospheric aerosols from mining
        operations. Water Air  Soil Pollut 221: 145-157. http://dx.doi.org/10.1007/sll270-011-0777-x

     Curtis. CJ: Evans. CD; Helliwell. RC: Monteith.  DT. (2005). Nitrate leaching as a confounding factor in
        chemical recovery from acidification in U.K.  upland waters. Environ Pollut 137: 73-82.
        http://dx.doi.0rg/10.1016/i.envpol.2004.12.032
     Dahl. O; Poykio. R; Nurmesniemi. H. (2008). Concentrations of heavy metals in fly ash from a coal-fired
        power plant with respect to the new Finnish limit values. Journal of Material Cycles and Waste
        Management 10: 87-92. http://dx.doi.org/10.1007/sl0163-007-0189-6

     DaH'Osto. M; Booth. MJ: Smith. W: Fisher. R; Harrison. RM.  (2008). A study of the size distributions and
        the chemical characterization of airborne particles in the vicinity of a large integrated steelworks.
        Aerosol Sci Technol 42: 981-991. http://dx.doi.org/10.1080/02786820802339587

     Das. SK; Routh. J: Roychoudhury. AN; Klump. JV. (2008). Major and trace element geochemistry in
        Zeekoevlei, South Africa: A lacustrine record of present and past processes. Appl Geochem 23:
        2496-2511. http://dx.doi.0rg/10.1016/i.apgeochem.2008.02.011

     Dawson. JJC: Tetzlaff. D: Carey. AM; Raab. A; Soulsbv. C: Killham. K; Meharg. AA. (2010).
        Characterizing Pb mobilization from upland soils to streams using (206)Pb/(207)Pb isotopic ratios.
        Environ Sci Technol 44: 243-249. http://dx.doi.org/10.1021/es902664d

     de la Cruz. MT; Laborda. F; Callen. MS: Lopez.  JM; Mastral. AM. (2009). Study of Pb sources by Pb
        isotope ratios in the airborne PM10 of Zaragoza, Spain. J Environ Monit 11: 2052-2057.
        http://dx.doi.org/10.1039/b912274e
November 2012                             3-229                  Draft - Do Not Cite or Quote

-------
     Deng. X; Jennings. AA. (2006). Evaluating an eletrokinetically driven extraction method for measuring
        heavy metal soil contamination. J Environ Eng 132: 527-537.
        http://dx.doi.org/10.1061/(ASCE)0733-9372(2006)132:4(527)

     Dermont. G: Bergeron. M; Richer-Lafleche. M; Mercier. G. (2010). Remediation of metal-contaminated
        urban soil using flotation technique. Sci Total Environ 408:  1199-1211.
        http://dx.doi.0rg/10.1016/i.scitotenv.2009.ll.036

     Desta. MB; Bruen. M; Higgins. N; Johnston. P. (2007). Highway runoff quality in Ireland. J Environ
        Monit 9: 366-371. http://dx.doi.org/10.1039/b702327h

     Diaz-Somoano. M; Kylander. ME; Lopez-Anton. MA; Suarez-Ruiz. I; Martinez-Tarazona. MR; Ferrat.
        M; Kober. B; Weiss. DJ. (2009). Stable lead isotope compositions in selected coals from around the
        world and implications for present day aerosol source tracing. Environ Sci Technol 43: 1078-1085.
        http://dx.doi.org/10.1021/es801818r

     Porn. CR:  Pierce. JO. II; Philips. PE; Chase. GR. (1976). Airborne Pb, Cd, Zn, and Cu concentrations by
        particle size near a Pb smelter. Atmos Environ 10: 443-446.
        http://dx.doi.org/10.1016/0004-6981(76)90024-X

     Dos Santos. M; Gomez. D; Dawidowski. L; Gautier. E; Smichowski. P. (2009). Determination of water-
        soluble and insoluble compounds in size classified airborne paniculate matter. Microchem J 91:
        133-139. http://dx.doi.0rg/10.1016/i.microc.2008.09.001

     DRRC (Doe Run Company). (2010). DRRC multi-media Consent Decree. Herculaneum, MO: Doe Run
        Resources Corporation.

     Dunlap.  CE; Alpers.  CN; Bouse. R; Taylor. HE; Unruh. DM; Flegal. AR.  (2008). The persistence of lead
        from past gasoline emissions and mining drainage in a large riparian system: Evidence from lead
        isotopes in the  Sacramento River, California. Geochim Cosmo Act 72: 5935-5948.
        http://dx.doi.org/10.1016/i.gca.20Q8.10.006

     Duzgoren-Aydin. NS. (2007). Sources  and characteristics of lead pollution in the urban environment of
        Guangzhou. Sci Total Environ 385: 182-195. http://dx.doi.Org/10.1016/i.scitotenv.2007.06.047

     Duzgoren-Aydin. NS: Weiss. AL. (2008). Use and abuse of Pb-isotope fingerprinting technique and GIS
        mapping data to assess lead in environmental studies. Environ Geochem Health 30: 577-588.
        http://dx.doi.org/10.1007/sl0653-008-9179-4

     Edwards. RD; Lam. NL; Zhang. L; Johnson. MA; Kleinman. MT. (2009). Nitrogen dioxide and ozone as
        factors  in the availability of lead from lead-based paints. Environ Sci Technol 43: 8516-8521.
        http://dx.doi.org/10.1021/es901077m

     EIA (Energy Information Administration). (2012). Electric power monthly. Table 1.2. Net generation by
        energy  source:  Electric utilities, 2002-June 2012.
        http://www.eia.gov/electricitv/monthlv/epm  table  grapher.cfm?t=epmt  1  2

     Erel. Y. (1998). Mechanisms and velocities of anthropogenic Pb migration in Mediterranean soils.
        Environ Res 78: 112-117. http://dx.doi.org/10.1006/enrs.1997.3811

     Erel. Y; Davan. U; Rabi. R; Rudich. Y; Stein. M. (2006). Trans boundary transport of pollutants by
        atmospheric mineral dust. Environ  Sci Technol 40: 2996-3005. http://dx.doi.org/10.1021/es0515021

     ERG (Eastern Research Group Inc.). (2011). Documentation for aircraft component of the national
        emissions inventory methodology. (0245.03.402.001). Morrisville, NC: Eastern Research Group.

     Evans. GC: Norton. SA; Fernandez. IJ; Kahl. JS: Hanson. D. (2005). Changes in concentrations of major
        elements and trace metals in northeastern US-Canadian sub-alpine forest floors. Water Air Soil Pollut
        163: 245-267.

     Ewen. C: Anagnostopoulou. MA; Ward. NL (2009). Monitoring of heavy metal levels in roadside dusts of
        Thessaloniki, Greece in relation to motor vehicle traffic density and flow. Environ Monit Assess 157:
        483-498. http://dx.doi.org/10.1007/sl0661-008-0550-9
November 2012                              3-230                   Draft - Do Not Cite or Quote

-------
     Ewing. SA; Christensen. JN; Brown. ST; Vancuren. RA; Cliff. SS: Depaolo. DJ. (2010). Pb isotopes as an
        indicator of the Asian contribution to paniculate air pollution in urban California. Environ Sci Technol
        44: 8911-8916. http://dx.doi.org/10.1021/esl01450t
     FAA (U.S. Federal Aviation Administration). (2011). Emissions and Dispersion Modeling System
        (EDMS) (Version 5.1.3) [Computer Program]. Washington, D.C. Retrieved from
        http://www.faa.gov/about/office org/headquarters  offices/apl/research/models/edms  model/

     Falgayrac. G: Sobanska. S: Laureyns. J: Bremard. C. (2006). Heterogeneous chemistry between PbSO4
        and calcite microparticles using Raman microimaging. Spectrochim Acta A Mol Biomol Spectrosc 64:
        1095-1101. http://dx.doi.0rg/10.1016/j.saa.2005.ll.032

     Falta. T; Limbeck. A; Koellensperger. G: Hann. S. (2008). Bioaccessibility of selected trace metals in
        urban PM2.5 and PM10 samples: A model study. Anal Bioanal Chem 390: 1149-1157.
        http://dx.doi.org/10.1007/s00216-007-1762-5

     Fang. GC: Huang. CS.  (2011). Atmospheric Particulate and Metallic Elements (Zn, Ni, Cu, Cd and Pb)
        Size Distribution at Three  Characteristic Sampling Sites. Environ Forensics 12: 191-199.
        http://dx.doi.org/10.1080/15275922.2011.595052

     Fang. TH; Li. JY; Feng. HM;  Chen. HY. (2009). Distribution and contamination of trace metals in surface
        sediments of the East China Sea. Mar Environ Res 68: 178-187.
        http://dx.doi.0rg/10.1016/i.marenvres.2009.06.005

     Farag. AM; Nimick. DA; Kimball. BA; Church. SE; Harper. DP; Brumbaugh. WG. (2007).
        Concentrations of metals in water, sediment, biofilm, benthic macroinvertebrates, and fish in the
        Boulder River watershed, Montana, and the role of colloids in metal uptake. Arch Environ Contam
        Toxicol 52: 397-409. http://dx.doi.org/10.1007/s00244-005-0021-z

     Farag. AM; Woodward. DF; Goldstein. JN; Brumbaugh. W: Meyer. JS. (1998). Concentrations of metals
        associated with mining waste in sediments, biofilm, benthic macroinvertebrates, and fish from the
        Coeur d'Alene River Basin, Idaho. Arch Environ Contam Toxicol 34: 119-127.
        http://dx.doi.org/10.1007/s002449900295

     Farfel. MR; Orlova. AO; Lees. PSJ: Rohde. C: Ashley. PJ: Chisolm. JJ. Jr. (2003). A study of urban
        housing demolitions as sources of lead in ambient dust: Demolition practices and exterior dust fall.
        Environ Health Perspect 111: 1228-1234.

     Farfel. MR; Orlova. AO: Lees. PSJ: Rohde. C: Ashley. PJ: Chisolm. JJ. Jr. (2005). A study of urban
        housing demolition as a source of lead in ambient dust on sidewalks, streets, and alleys. Environ Res
        99: 204-213. http://dx.doi.Org/10.1016/i.envres.2004.10.005

     Farmer. JG: Eades. LJ; Graham. MC: Cloy. JM; Bacon. JR. (2010). A comparison of the isotopic
        composition of lead in rainwater, surface vegetation and tree bark at the long-term monitoring site,
        Glensaugh, Scotland, in 2007. Sci Total Environ 408: 3704-3710.
        http://dx.doi.0rg/10.1016/i.scitotenv.2010.03.050

     Farmer. JG: Eades. LJ: MacKenzie. AB; Kirika. A; Bailey-Watts. TE.  (1996). Stable lead isotope record
        of lead pollution in  Loch Lomond sediments since  1630 AD. Environ Sci Technol 30: 3080-3083.
        http://dx.doi.org/10.1021/es960162o

     Farmer. JG: Graham. MC: Bacon. JR; Dunn. SM; Vinogradoff. SI. (2005). Isotopic characterisation of the
        historical lead deposition record at Glensau, an organic-rich, upland catchment in rural NE Scotland.
        Sci Total Environ 346:  121-137. http://dx.doi.Org/10.1016/j.scitotenv.2004.ll.020

     Fauser. P. (1999) Particulate air pollution with emphasis  on traffic generated aerosols. (Doctoral
        Dissertation). Technical University of Denmark (DTU), Kgs. Lyngby, Denmark.
     Fernandez. C: Labanowski. J; Cambier. P; Jongmans. AG: Van Port. F. (2007). Fate of airborne metal
        pollution in soils as related to agricultural management. 1. Zn and Pb distributions in soil profiles. Eur
        J Soil Sci 58: 547-559. http://dx.doi.Org/10.llll/i.1365-2389.2006.00827.x
November 2012                              3-231                   Draft - Do Not Cite or Quote

-------
     Fernandez Espinosa. AJ; Ternero Rodriguez. M; Barragan De La Rosa. FJ; Jimenez Sanchez. JC. (2002).
        A chemical speciation of trace metals for fine urban particles. Atmos Environ 36: 773-780.
        http://dx.doi.org/10.1016/81352-2310(01)00534-9

     Fernandez Espinosa. AJ: Ternero Rodriguez. M; Fernandez Alvarez. F. (2004). Source characterisation of
        fine urban particles by multivariate analysis of trace metals speciation. Atmos Environ 38: 873-886.
        http://dx.doi.0rg/10.1016/i.atmosenv.2003.10.046

     Fernando. HJS. (2010). Fluid dynamics of urban atmospheres in complex terrain. Annu Rev Fluid Mech
        42: 365-389. http://dx.doi.org/10.1146/annurev-fluid-121108-145459

     Field. PM; Sherrell. RM.  (2003). Direct determination of ultra-trace levels  of metals in fresh water using
        desolvating micronebulization and HR-ICP-MS: Application to Lake Superior waters. J Anal At
        Spectrom 18: 254-259. http://dx.doi.org/10.1039/b210628k

     Filippelli. GM; Laidlaw. MAS. (2010). The elephant in the playground: Confronting lead-contaminated
        soils as an important source of lead burdens to urban populations. Perspect Biol Med 53: 31-45.
        http://dx.doi.0rg/10.1353/pbm.0.0136

     Filippelli. GM; Laidlaw. MAS: Latimer. JC: Raftis. R. (2005). Urban lead poisoning and medical
        geology: An unfinished story. Geological Society of America Today 15: 4-11.
        http://dx.doi.org/10.1130/1052-5173(2005)015<4:ULPAMG>2.0.CO:2

     Finlavson-Pitts. BJ; Pitts. JN. Jr. (2000). Chemistry of the upper and lower atmosphere: theory,
        experiments and applications. San Diego, CA: Academic Press.

     Flint. KR; Davis. AP (2007). Pollutant mass flushing characterization of highway storm water runoff from
        an ultra-urban area. J Environ Eng 133: 616-626.
        http://dx.doi.org/10.1061/(ASCE)0733-9372(2007)133:6(616)

     Fowler. D; McDonald. AG: Crosslev. A; Nemitz. E; Leaver. D;  Cape. JN; Smith. RI; Anderson. D;
        Rowland. P; Ainsworth. G: Lawlor. AJ: Guvatt H; Harmens. H. (2006). U.K. heavy metal monitoring
        network. (AS 06/07). Edinburgh, U.K.: NERC/Centre for Ecology and Hydrology.
        http://nora.nerc.ac.uk/3323

     Fowler. D; Smith. RI: Muller. JB; Havman. G: Vincent. KJ. (2005). Changes in the atmospheric
        deposition of acidifying compounds in the U.K. between 1986 and 2001. Environ Pollut 137:  15-25.
        http://dx.doi.0rg/10.1016/i.envpol.2004.12.028

     Franzaring. J: Holz. I; Zipperle. J: Fangmeier. A. (2010). Twenty years of biological monitoring of
        element concentrations in permanent forest and grassland plots in Baden-Wurttemberg (SW
        Germany). Environ Sci Pollut Res Int 17: 4-12. http://dx.doi.org/10.1007/sll356-009-0181-x

     Fraser. JL; Lum. KR. (1983). Availability of elements of environmental importance in incinerated sludge
        ash. Environ Sci Technol 17: 52-54. http://dx.doi.org/10.1021/es00107a013

     Freme. F. (2004). U.S. coal supply and demand: 2004 Review. Washington, DC: U.S. Energy Information
        Administration. http://tonto.eia.doe.gov/FTPROOT/features/feature04.pdf

     Fujiwara. F; Dos Santos. M; Marrero. J: Polla. G: Gomez. D; Dawidowskia. L; Smichowski. P. (2006).
        Fractionation of eleven elements by chemical bonding from  airborne paniculate matter collected in an
        industrial city in Argentina. J Environ Monit 8: 913-922. http://dx.doi.org/10.1039/b604307k

     Funasaka. K; Tojo. T; Katahira. K; Shinya. M; Miyazaki. T; Kamiura. T; Yamamoto. O: Moriwaki. H;
        Tanida. H; Takaoka. M. (2008). Detection of Pb-LIII edge XANES spectra of urban atmospheric
        particles combined with simple acid extraction. Sci Total Environ 403: 230-234.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.05.020

     Gallon. C: Tessier. A; Gobeil. C. (2006). Historical perspective  of industrial lead emissions to the
        atmosphere from a Canadian smelter.  Environ Sci Technol 40: 741-747.
        http://dx.doi.org/10.1021/es051326g
November 2012                             3-232                   Draft - Do Not Cite or Quote

-------
     Gao. Y; Lesven. L; Gillan. D; Sabbe. K; Billon. G: De Galan. S: Elskens. M; Baevens. W: Leermakers.
        M. (2009). Geochemical behavior of trace elements in sub-tidal marine sediments of the Belgian
        coast. Mar Chem 117:  88-96. http://dx.doi.Org/10.1016/j.marchem.2009.05.002

     Garland. JA; Nicholson. KW. (1991). A review of methods for sampling large airborne particles and
        associated radioactivity [Review]. J Aerosol Sci 22: 479-499.
        http://dx.doi.org/10.1016/0021-8502(91)90006-4

     Garrido. F; Serrano. S: Campbell. CG: Barrios. L; Garcia-Gonzalez. AT. (2008). Evidence of physical and
        chemical nonequilibrium in lead and cadmium transport and sorption in acid soils. Soil Sci Soc Am J
        72:  1434-1444. http://dx.doi.org/10.2136/sssaj2007.0411

     Geagea. ML; Stille. P; Gauthier-Lafave. F; Perrone. T; Aubert. D. (2008). Baseline determination of the
        atmospheric Pb, Sr and Nd isotopic compositions in the Rhine valley, Vosges mountains (France) and
        the Central  Swiss Alps. Appl Geochem 23: 1703-1714.
        http://dx.doi.0rg/10.1016/i.apgeochem.2008.02.004

     Gidnev. J: Twigg. M; Kittelson. D. (2010). Effect of organometallic fuel additives on nanoparticle
        emissions from a gasoline passenger car. Environ Sci Technol 44: 2562-2569.
        http://dx.doi.org/10.1021/es901868c

     Giere. R; Blackford. M; Smith. K. (2006). TEM study of PM2.5 emitted from coal and tire combustion in
        a thermal power station. Environ Sci Technol 40: 6235-6240. http://dx.doi.org/10.1021/es060423m

     Gillette. DA; Blifford. IH. Jr; Fryrear. DW. (1974). The influence of wind velocity on the size
        distributions of aerosols generated by the wind erosion of soils. J Geophys Res 79: 4068-4075.
        http://dx.doi.org/10.1029/JC079i027p04068

     Goforth. MR; Christoforou. CS.  (2006). Particle size distribution and atmospheric metals measurements
        in a rural area in the South Eastern USA.  Sci Total Environ 356: 217-227.
        http://dx.doi.0rg/10.1016/i.scitotenv.2005.03.017

     Goval. A; Small. MJ; Von Stackelberg. K; Burmistrov. D. (2005). Estimation of fugitive lead emission
        rates from secondary lead facilities using hierarchical Bayesian models. Environ Sci Technol 39:
        4929-4937.  http://dx.doi.org/10.1021/es035465e

     Graham. MC: Vinogradoff. SI; Chipchase. AJ; Dunn. SM; Bacon. JR; Farmer. JG. (2006). Using size
        fractionation and Pb isotopes to study Pb transport in the waters of an organic-rich upland catchment.
        Environ Sci Technol 40: 1250-1256. http://dx.doi.org/10.1021/es0517670

     Granev. JR: Landis. MS: Norris. GA. (2004). Concentrations and solubility of metals from indoor and
        personal exposure PM2.5 samples. Atmos Environ 38: 237-247.
        http://dx.doi.0rg/10.1016/i.atmosenv.2003.09.052

     Griffith. DA. (2002). The geographic distribution of soil lead concentration: Description and concerns.
        URISA Journal 14: 5-16.

     Griffith. DA; Johnson. PL; Hunt. A. (2009). The geographic distribution of metals in urban soils: The
        case of Syracuse, NY. GeoJournal 74: 275-291. http://dx.doi.org/10.1007/sl0708-008-9233-x

     Gulson. B; Korsch. M; Dickson. B; Cohen. D; Mizon. K; Davis. JM. (2007). Comparison of lead isotopes
        with source apportionment models, including SOM, for air particulates. Sci Total Environ 381:
        169-179. http://dx.doi.0rg/10.1016/i.scitotenv.2007.03.018

     Guo. T; Sansalone. J:  Piro. P. (2006a). The role of in situ unit operation/process infiltration treatment on
        partitioning and speciation of rainfall-runoff. Water Sci Technol 54: 255-261.
        http://dx.doi.org/10.2166/wst.2006.589

     Guo. XY: Zhang. SZ: Shan. XO: Luo. L: Pei. ZG: Zhu. YG: Liu. T: Xie. YN: Gault. A. (2006b).
        Characterization of Pb, Cu, and Cd adsorption on paniculate organic matter in soil. Environ Toxicol
        Chem 25: 2366-2373. http://dx.doi.Org/10.1897/05-636R.l
November 2012                              3-233                  Draft - Do Not Cite or Quote

-------
     Gutierrez-Castillo. ME; Olivos-Ortiz. M; De Vizcava-Ruiz. A; Cebrian. ME. (2005). Chemical
        characterization of extractable water soluble matter associated with PM10 from Mexico City during
        2000. Chemosphere 61: 701-710. http://dx.doi.Org/10.1016/i.chemosphere.2005.03.063

     H. K; M. K; W. S. (2010). Atmospheric Pb and Ti accumulation rates from sphagnum moss: Dependence
        upon plant productivity. Environ Sci Technol 44: 5509-5515. http://dx.doi.org/10.1021/esl00366d

     Halm. I; Brixev. LA; Wiener. RW: Henkle. SW: Baldauf. R. (2009). Characterization of traffic-related PM
        concentration distribution and fluctuation patterns in near-highway urban residential street canyons. J
        Environ Monit 11: 2136-2145. http://dx.doi.org/10.1039/b907130j

     Hallberg. M; Renman. G: Lundbom. T. (2007). Seasonal variations of ten metals in highway runoff and
        their partition between dissolved and paniculate matter. Water Air Soil Pollut 181: 183-191.
        http://dx.doi.org/10.1007/sll270-006-9289-5

     Harmens. H; Norris. DA; Koerber. GR; Buse. A;  Steinnes. E; Ruhling. A. (2008). Temporal trends
        (1990-2000) in the concentration of cadmium, lead and mercury in mosses across Europe. Environ
        Pollut 151: 368-376. http://dx.doi.Org/10.1016/i.envpol.2007.06.043

     Harmens. H; Norris. DA; Steinnes. E; Kubin. E; Piispanen. J; Alber. R; Aleksiavenak. Y; Blum. O;
        Coskun. M; Dam. M; De Temmerman. L; Fernandez. JA; Frolova. M; Frontasveva.  M; Gonzalez-
        Miqueo. L; Grodzinska. K; Jeran. Z; Korzekwa. S: Krmar. M; Kvietkus. K; Leblond. S: Liiv. S:
        Magnusson. SH; Mankovska. B; Pesch. R; Ruhling. A; Santamaria. JM; Schroder. W:  Spiric. Z;
        Suchara. I; Thoni. L; Urumov. V; Yurukova. L; Zechmeister. HG. (2010). Mosses as biomonitors of
        atmospheric heavy metal deposition: Spatial patterns and temporal trends in Europe. Environ Pollut
        158: 3144-3156. http://dx.doi.Org/10.1016/i.envpol.2010.06.039

     Harris. AR; Davidson. CI. (2005). The role of resuspended soil in lead flows in the California South Coast
        Air Basin. Environ Sci Technol 39: 7410-7415. http://dx.doi.org/10.1021/es050642s

     Harris. AR; Davidson. CI. (2008). Particle resuspension in turbulent flow: A stochastic model for
        individual soil grains. Aerosol Sci Technol 42: 613-628.
        http://dx.doi.org/10.1080/02786820802227337

     Harrison. RM; Tilling. R; Callen Romero. MS: Harrad. S: Jarvis. K. (2003). A study of trace metals and
        polycyclic aromatic hydrocarbons in the roadside environment. Atmos Environ 37: 2391-2402.
        http://dx.doi.org/10.1016/81352-2310(03)00122-5

     Harrison. RM; Yin. J. (2010).  Chemical speciation of PM2.5 particles at urban background and rural  sites
        in the U.K. atmosphere. J Environ Monit 12: 1404-1414. http://dx.doi.org/10.1039/c000329h

     Hartnett. M; Berry. A. (2010). Transport of lead in the Mersey Estuary: The development of a novel
        approach  to deriving partition coefficients. Adv Eng Software 41: 84-91.
        http://dx.doi.0rg/10.1016/i.advengsoft.2008.ll.007

     Hasselbach. L; Ver Hoef. JM; Ford. J: Neitlich. P; Crecelius. E; Berryman. S: Wolk. B;  Bohle. T. (2005).
        Spatial patterns of cadmium and lead deposition on and adjacent to National Park Service lands in the
        vicinity of Red Dog Mine, Alaska. Sci Total Environ 348:  211-230.
        http://dx.doi.0rg/10.1016/i.scitotenv.2004.12.084

     Hassellov. M; von der Kammer. F. (2008). Iron oxides as geochemical nanovectors for metal transport in
        soil-river systems. Elements 4: 401-406. http://dx.doi.Org/10.2113/gselements.4.6.401

     Hays. MD; Cho. SH; Baldauf. R: JJ. S: Shafer. M. (2011). Particle size distributions of metal and non-
        metal elements in an urban near-highway environment. Atmos Environ 45: 925-934.
        http://dx.doi.0rg/10.1016/i.atmosenv.2010.ll.010

     Heal. MR; Hibbs. LR; Agius.  RM: Beverland. IJ. (2005). Interpretation of variations in  fine, coarse and
        black smoke paniculate matter concentrations in a northern European city. Atmos Environ 39:
        3711-3718. http://dx.doi.0rg/10.1016/i.atmosenv.2005.03.007
November 2012                             3-234                  Draft - Do Not Cite or Quote

-------
     Helmreich. B; Hilliges. R; Schriewer. A; Horn. H. (2010). Runoff pollutants of a highly trafficked urban
        road: Correlation analysis and seasonal influences. Chemosphere 80: 991-997.
        http://dx.doi.0rg/10.1016/i.chemosphere.2010.05.037

     Hinds. WC. (1999). Aerosol technology: Properties, behavior, and measurement of airborne particles (2nd
        ed.). New York: John Wiley & Sons.

     Hirner. AV. (2006). Speciation of alkylated metals and metalloids in the environment. Anal Bioanal Chem
        385: 555-567. http://dx.doi.org/10.1007/s00216-006-0368-7

     Hitzenberger. R; Berner. A; Galambos. Z; Maenhaut. W: Cafmever. J; Schwarz. J; Muller. K; Spindler. G:
        Wieprecht. W: Acker. K; Hillamo. R; Makela. T. (2004). Intercomparison of methods to measure the
        mass concentration of the atmospheric aerosol during INTERCOMP2000 - Influence of
        instrumentation and size cuts. Atmos Environ 38: 6467-6476.
        http://dx.doi.0rg/10.1016/i.atmosenv.2004.08.025

     Hjortenkrans. DST; Bergback. BG: Haggerud. AV. (2007). Metal emissions from brake linings and tires:
        Case studies of Stockholm, Sweden 1995/1998 and 2005. Environ Sci Technol 41: 5224-5230.
        http://dx.doi.org/10.1021/es070198o

     Holsen. TM; Noll. KB; Fang. GC: Lee. WJ: Lin. JM; Keeler. GJ. (1993). Dry deposition and particle size
        distributions measured during the Lake Michigan Urban Air Toxics Study. Environ Sci Technol 27:
        1327-1333. http://dx.doi.org/10.1021/es00044a006

     Hong. S: Candelone. JP; Patterson.  CC: Boutron. CF.  (1994). Greenland ice evidence of hemispheric lead
        pollution two millennia ago by Greek and Roman civilizations.  Science 265: 1841-1843.
        http://dx.doi.org/10.1126/science.265.5180.1841

     Horowitz. AJ; Elrick. KA; Smith. JJ. (2008). Monitoring urban impacts on suspended sediment, trace
        element, and nutrient fluxes within the City of Atlanta, Georgia, USA: Program design,
        methodological considerations, and initial results.  Hydrolog Process 22: 1473-1496.
        http://dx.doi.org/10.1002/hyp.6699

     Horowitz. AJ; Stephens. VC. (2008). The effects of land use on fluvial sediment chemistry for the
        conterminous U.S. Results from the first cycle of the NAWQA Program: Trace and major elements,
        phosphorus, carbon, and sulfur. Sci Total Environ 400: 290-314.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.04.027

     Hossain. MA; Furumai. H; Nakajima. F; Aryal.  RK. (2007). Heavy metals speciation in soakaways
        sediment and evaluation of metal retention properties of surrounding soil. Water Sci Technol 56:
        81-89. http://dx.doi.org/10.2166/wst.2007.746

     Hou. A; Laws. E; Gambrell. R: Tan. M; Delaune. R: Li. Y; Roberts. H. (2006). Pathogen indicator
        microbes and heavy metals in Lake Pontchartrain following Hurricane Katrina. Environ Sci Technol
        40: 5904-5910. http://dx.doi.org/10.1021/es060946u

     Hsu. SC: Chen Liu. S: Jeng. WL; Chou. CC. KG: Hsu. RT; Huang. YT; Chen. YY. (2006). Lead isotope
        ratios in ambient aerosols from Taipei, Taiwan: Identifying long-range transport of airborne Pb from
        the Yangtze Delta. Atmos Environ 40: 5393-5404. http://dx.doi.Org/10.1016/j.atmosenv.2006.05.003

     Hsu. SC: Liu.  SC: Huang. YT: Chou. CCK: Lung. SCC: Liu. TH: Tu. JY: Tsai. FJ. (2009c). Long-range
        southeastward transport of Asian biosmoke pollution: Signature detected by aerosol potassium in
        Northern Taiwan. J Geophys Res 114: D14301. http://dx.doi.org/10.1029/2009jd011725

     Hu. S: Herner. JD; Shafer. M; Robertson. W: Schauer. JJ: Dwver. H; Collins. J: Huai. T; Ayala. A.
        (2009a). Metals emitted from heavy-duty diesel vehicles equipped with advanced PM and NOx
        emission controls. Atmos Environ 43: 2950-2959. http://dx.doi.Org/10.1016/i.atmosenv.2009.02.052

     Huang. S: Conte. MH. (2009). Source/process apportionment of major and trace elements in sinking
        particles in the Sargasso sea. Geochim Cosmo Act 73: 65-90.
        http://dx.doi.0rg/10.1016/i.gca.2008.08.023
November 2012                             3-235                   Draft - Do Not Cite or Quote

-------
     Hur. SD: Cunde. X; Hong. S: Barbante. C: Gabrielli. P; Lee. K; Boutron. CF; Ming. Y. (2007). Seasonal
        patterns of heavy metal deposition to the snow on Lambert Glacier basin, East Antarctica. Atmos
        Environ 41: 8567-8578. http://dx.doi.Org/10.1016/i.atmosenv.2007.07.012

     Huston. R; Chan. YC: Gardner. T; Shaw. G: Chapman. H. (2009). Characterisation of atmospheric
        deposition as a source of contaminants in urban rainwater tanks. Water Res 43: 1630-1640.
        http://dx.doi.0rg/10.1016/i.watres.2008.12.045

     IAEA (International Atomic Energy Agency). (1993). The "GENT" Stacked Filter Unit (SFU) sampler for
        the collection of atmospheric aerosols in two size fractions: Description and instructions for
        installation and use. Vienna, VA.

     Ikegawa. M; Kimura. M; Honda. K; Akabane. I; Makita.  K; Motovama. H; Fujii. Y; Itokawa. Y. (1999).
        Geographical variations of major and trace elements in East Antarctica. Atmos Environ 33:
        1457-1467. http://dx.doi.org/10.1016/S1352-2310(98)00243-X

     Illinois Environmental Protection Agency. (2002). Chicago OHare Airport air toxic monitoring program.
        Springfield, IL. http://www.epa.state.il.us/air/ohare/

     Ishizaka. T; Tohno. S: Ma. CJ: Morikawa. A; Takaoka. M; Nishiyama. F; Yamamoto. K.  (2009).
        Reactivity between PbS04 and CaCO3 particles relevant to the modification of mineral particles and
        chemical forms of Pb in particles sampled at two remote sites during an Asian dust event. Atmos
        Environ 43: 2550-2560. http://dx.doi.Org/10.1016/i.atmosenv.2009.02.041

     Jennings. AA; Ma. J. (2007). Variation in North American regulatory guidance for heavy metal surface
        soil contamination at commercial and industrial sites. Journal of Environmental Engineering and
        Science 6: 587-609. http://dx.doi.org/10.1139/s07-010

     Jensen. MB; Holm. PE; Laursen. J; Hansen. HCB. (2006). Contaminant aspects of blackish surface
        deposits on highway roadsides. Water Air Soil Pollut  175: 305-321.
        http://dx.doi.org/10.1007/sll270-006-9140-z

     Jerez. SB; Zhang. Y; Mcclure. JW: Jacobson. L; Heber. A; Hoff S: Koziel. J: Beaslev. D. (2006).
        Comparison of measured total suspended paniculate matter concentrations using tapered element
        oscillating microbalance and a total suspended paniculate sampler. J Air Waste Manag Assoc 56:
        261-270.

     Jimenez. S: Perez. M; Ballester. J. (2008). Vaporization of trace elements and their emission with
        submicrometer aerosols in biomass combustion. Energy Fuels 22: 2270-2277.
        http://dx.doi.org/10.1021/ef800111u

     Jitaru. P; Infante. HG: Adams. FC. (2004). Simultaneous multi-elemental speciation analysis of
        organometallic compounds by solid-phase microextraction and multicapillary gas chromatography
        hyphenated to inductively coupled plasma-time-of-flight-mass spectrometry. J Anal At Spectrom 19:
        867-875. http://dx.doi.org/10.1039/B404106B

     Johnson. AH; Richter. SL. (2010). Organic-horizon lead,  copper, and zinc contents of Mid-Atlantic forest
        soils, 1978-2004. Soil Sci Soc Am J 74:  1001-1009. http://dx.doi.org/10.2136/sssai2008.0337

     Johnson. KS: de Foy. B; Zuberi. B; Molina. IT; Molina.  MJ; Xie. Y; Laskin. A; Shutthanandan. V
        (2006). Aerosol composition and source apportionment in the Mexico City Metropolitan Area with
        PIXE/PESA/STIM and multivariate analysis. Atmos Chem Phys 6: 4591-4600.
        http://dx.doi.org/10.5194/acp-6-4591-2006

     Johnson. KS: Laskin. A; Jimenez. JL; Shutthanandan. V; Molina. IT; Salcedo. D; Dzepina. K; Molina.
        MJ. (2008). Comparative analysis of urban atmospheric aerosol by particle-induced X-ray emission
        (PIXE), proton elastic scattering analysis (PESA), and aerosol mass spectrometry (AMS). Environ Sci
        Technol 42: 6619-6624. http://dx.doi.org/10.1021/es800393e

     Joshi. UM; Viiavaraghavan. K; Balasubramanian. R. (2009). Elemental composition of urban street dusts
        and their dissolution characteristics in various aqueous media. Chemosphere 77: 526-533.
        http://dx.doi.0rg/10.1016/i.chemosphere.2009.07.043
November 2012                             3-236                  Draft - Do Not Cite or Quote

-------
     Kang. S: Zhang. Q; Kaspari. S: Qin. D; Cong. Z; Ren. J; Mavewski. PA. (2007). Spatial and seasonal
        variations of elemental composition in Mt. Everest (Qomolangma) snow/firn. Atmos Environ 41:
        72087218. http://dx.doi.0rg/10.1016/i.atmosenv.2007.05.024

     Kaste. J: Friedland. A; Sturup. S. (2003). Using stable and radioactive isotopes to trace atmospherically
        deposited Pb in montane forest soils. Environ Sci Technol 37: 3560-3567.
        http://dx.doi.org/10.1021/es026372k

     Kaste. JM; Bostick. BC: Friedland. AJ; Schroth. AW: Siccama. TG. (2006). Fate and speciation of
        gasoline-derived lead in organic horizons of the northeastern USA. Soil Sci Soc Am J 70: 1688-1698.
        http://dx.doi.org/10.2136/sssai2005.0321

     Kastner-Klein. P; Fedorovich. E; Ketzel. M; R. B; Britter. R. (2003). The modelling of turbulence from
        traffic in urban dispersion models part II: Evaluation against laboratory and full-scale  concentration
        measurements in street canyons. Environ Fluid Mech 3: 145-172.
        http://dx.doi.Org/10.1023/A:1022049224166

     Kastner-Klein. P; Fedorovich. E; Rotach. MW. (2001). A wind tunnel study of organised and turbulent air
        motions in urban street canyons. J Wind Eng Ind Aerod 89: 849-861.
        http://dx.doi.org/10.1016/S0167-6105(01)00074-5

     Kastner-Klein. P; Rotach. MW. (2004). Mean flow and turbulence characteristics in an urban roughness
        sublayer. Boundary Layer Meteorol 111: 55-84.
        http://dx.doi.Org/10.1023/B:BOUN.0000010994.32240.bl

     Kayhanian. M; Suverkropp. C: Ruby. A; Tsav. K. (2007). Characterization  and prediction of highway
        runoff constituent event mean concentration. J Environ Manage 85: 279-295.
        http://dx.doi.0rg/10.1016/i.jenvman.2006.09.024

     Kenny. L; Beaumont. G: Gudmundsson. A; Thorpe. A; Koch. W (2005). Aspiration and sampling
        efficiencies of the TSP and louvered paniculate matter inlets. J Environ Monit 7: 481-487.
        http://dx.doi.org/10.1039/b419001g

     Keskinen. J: Pietarinen. K; Lehtimaki.  M. (1992). Electrical low pressure impactor. J Aerosol Sci 23:
        353-360. http://dx.doi.org/10.1016/0021-8502(92)90004-F

     Kidwell. CB; Ondov. JM. (2001). Development and evaluation of a prototype system for  collecting sub-
        hourly ambient aerosol for chemical analysis. Aerosol Sci Technol 35: 596-601.
        http://dx.doi.org/10.1080/02786820118049

     Kidwell. CB; Ondov. JM. (2004). Elemental analysis of sub-hourly ambient aerosol collections. Aerosol
        Sci Technol 38: 205-218. http://dx.doi.org/10.1080/02786820490261726

     Kimbrough. KL; Lauenstein. GG: Christensen. JD; Apeti. DA. (2008). An assessment of two decades of
        contaminant monitoring in the nation's coastal zone. Silver Spring, MD: National Centers for Coastal
        Ocean Science, http://aquaticcommons.org/2232/

     Klaminder. J: Bindler. R; Emteryd.  O: Applebv. P: Grip. H. (2006a). Estimating the mean residence time
        of lead in the organic horizon of boreal forest soils using 210-lead, stable lead and a soil
        chronosequence. Biogeochemistry 78: 31-49. http://dx.doi.org/10.1007/sl0533-005-2230-y

     Klaminder. J: Bindler. R; Emteryd.  O: Renberg. I. (2005).  Uptake and recycling of lead by boreal forest
        plants: Quantitative estimates from  a site in northern Sweden. Geochim Cosmo Act 69: 2485-2496.
        http://dx.doi.0rg/10.1016/i.gca.2004.ll.013

     Klaminder. J: Bindler. R; Laudon. H; Bishop. K; Emteryd. O: Renberg. I. (2006b). Flux rates of
        atmospheric lead pollution within soils of a small catchment in northern Sweden and their
        implications for future  stream water quality.  Environ Sci Technol 40: 4639-4645.
        http://dx.doi.org/10.1021/es0520666

     Klaminder. J: Bindler. R; Renberg. I. (2008a). The biogeochemistry of atmospherically derived Pb in the
        boreal forest of Sweden. Appl Geochem 23:  2922-2931.
        http://dx.doi.0rg/10.1016/i.apgeochem.2008.04.007
November 2012                             3-237                   Draft - Do Not Cite or Quote

-------
     Klaminder. J; Bindler. R; Rydberg. J; Renberg. I. (2008b). Is there a chronological record of atmospheric
        mercury and lead deposition preserved in the mor layer (O-horizon) of boreal forest soils? Geochim
        Cosmo Act 72: 703-712. http://dx.doi.Org/10.1016/i.gca.2007.10.030

     Klaminder. J: Hammarlund. D; Kokfelt. U: Vonk. JE; Bigler. C. (2010). Lead contamination of subarctic
        lakes and its response to reduced atmospheric fallout: Can the recovery process be counteracted by the
        ongoing climate change? Environ Sci Technol 44: 2335-2340. http://dx.doi.org/10.1021/es903025z

     Knowlton. SW: Moran. SB. (2010). Stable Pb isotope ratios in aerosols, precipitation, and size-
        fractionated paniculate matter in the Gulf of Maine, Scotian Shelf, and Labrador Sea. Mar Pollut Bull
        60: 984-989. http://dx.doi.Org/10.1016/i.marpolbul.2010.02.005

     Kumar. P; Hopke. PK; Raja. S: Casuccio. G: Lersch. TL; West. RR. (2012). Characterization and
        heterogeneity  of coarse particles across an urban area. Atmos Environ 46: 449-459.

     Kylander. ME;  Cortizas. AM; Rauch. S: Weiss. DJ. (2008). Lead penetration and leaching in a complex
        temperate soil profile. Environ Sci Technol 42: 3177-3184. http://dx.doi.org/10.1021/es702358e

     Kyotani. T; Iwatsuki. M. (2002). Characterization of soluble and insoluble components in PM25 and
        PM10 fractions of airborne particulate matter in Kofu city, Japan. Atmos Environ 36: 639-649.
        http://dx.doi.org/10.1016/81352-2310(01)00494-0

     Laeudu. URK:  Raia. S: Hopke. PK: Chalupa. DC: Utell. MJ: Casuccio. G: Lersch. TL: West. RR. (2011).
        Heterogeneity of Coarse Particles in an Urban Area. Environ Sci Technol 45: 3288-3296.
        http://dx.doi.org/10.1021/esl03831w

     Lai. X; Norisuve. K; Mikata. M; Minami. T; Bowie. AR; Sohrin. Y. (2008b). Spatial and temporal
        distribution  of Fe, Ni, Cu and Pb along 140 degrees E in the Southern Ocean during austral summer
        2001/02. Mar  Chem 111: 171-183. http://dx.doi.Org/10.1016/i.marchem.2008.05.001

     Laidlaw. MAS: Filippelli. GM. (2008). Resuspension of urban soils as a persistent source of lead
        poisoning in children: A review and new directions. Appl Geochem 23: 2021-2039.
        http://dx.doi.0rg/10.1016/i.apgeochem.2008.05.009

     Laidlaw. MAS: Zahran. S: Mielke. HW: Taylor. MP; Filippelli. GM. (2012). Re-suspension of lead
        contaminated urban soil as a dominant source of atmospheric lead in Birmingham, Chicago, Detroit
        and Pittsburgh, USA. Atmos Environ 49: 302-310. http://dx.doi.Org/10.1016/i.atmosenv.2011.ll.030

     Landers. DH; Simonich. SL; Jaffe. DA; Geiser. LH; Campbell DH; Schwindt. AR; Schreck. CB; Kent.
        ML; Hafner. WD; Taylor. HE; Hageman. KJ; Usenko. S: Ackerman. LK; Schrlau. JE; Rose. NL; Blett.
        TF; Erwav. MM.  (2008). The fate, transport, and ecological impacts of airborne contaminants in
        western national parks (USA). (EPA/600/R-07/138). Corvallis, Oregon: U.S. Environmental
        Protection Agency, NHEERL, Western Ecology Division.
        http://www.nature.nps.gov/air/studies/air toxics/WACAPreport.cfm

     Landers. DH: Simonich. SM; Jaffe. D; Geiser. L; Campbell. DH: Schwindt. A; Schreck. C: Kent. M;
        Hafner. W: Taylor. HE; Hageman. K; Usenko. S: Ackerman. L; Schrlau. J: Rose. N; Blett. T; Erwav.
        MM. (2010). The Western Airborne Contaminant Assessment Project (WACAP): An interdisciplinary
        evaluation of the impacts of airborne contaminants  in western U.S. National Parks. Environ Sci
        Technol 44:  855-859. http://dx.doi.org/10.1021/es901866e

     Landre. AL; Watmough. SA; Dillon. PJ. (2009). The effects of dissolved  organic carbon, acidity and
        seasonality on metal geochemistry within a forested catchment on the Precambrian Shield, central
        Ontario, Canada. Biogeochemistry 93:  271-289. http://dx.doi.org/10.1007/sl0533-009-9305-0

     Landre. AL; Watmough. SA; Dillon. PJ. (2010). Metal pools, fluxes, and budgets in an acidified forested
        catchment on the Precambrian Shield, Central Ontario, Canada. Water Air Soil Pollut 209: 209-228.
        htrp://dx.doi.org/10.1007/sll270-009-0193-7

     Lankev. RL; Davidson. CI; McMichael. FC.  (1998). Mass balance for lead in the California south coast
        air basin: an update. Environ Res 78: 86-93.
November 2012                             3-238                  Draft - Do Not Cite or Quote

-------
     Lasheen. MR; Sharabv. CM; El-Kholy. NG: Elsherif. IY; El-Wakeel. ST. (2008). Factors influencing lead
        and iron release from some Egyptian drinking water pipes. J Hazard Mater 160: 675-680.
        http://dx.doi.0rg/10.1016/i.jhazmat.2008.03.040

     Leal-Granadillo. IA; Alonso. JIG: Sanz-Medel. A. (2000). Determination of the speciation of organolead
        compounds in airborne paniculate matter by gas chromatography-inductively coupled plasma mass
        spectrometry. Anal Chim Acta 423:  21-29. http://dx.doi.org/10.1016/S0003-2670(00)01032-l

     Lee. G: Faure. G: Bigham. JM; Williams. DJ. (2008a). Metal release from bottom sediments of Ocoee
        Lake No. 3, a primary catchment area for the Ducktown Mining District. J Environ Qua! 37: 344-352.
        http://dx.doi.org/10.2134/iea2007.0223

     Lee. K; Hur. SD; Hou. S: Hong. S: Qin. X; Ren. J: Liu. Y; Rosman. KJR; Barbante. C: Boutron. CF.
        (2008b). Atmospheric pollution for  trace elements in the remote high-altitude atmosphere in central
        Asia as recorded in snow from Mt. Qomolangma (Everest) of the Himalayas. Sci Total Environ 404:
        171-181. http://dx.doi.0rg/10.1016/i.scitotenv.2008.06.022

     Lee. RE. Jr. (1972). Size distribution of suspended particulates in air. Res/Dev 23: 18-21.

     Lee. RE. Jr; Patterson. RK; Wagman. J. (1968). Particle-size distribution of metal components in urban
        air. Environ Sci Technol 2: 288-290.

     Leith. D; Sommerlatt. D; Boundv. MG. (2007). Passive sampler for PM10-2.5 aerosol. J Air Waste Manag
        Assoc 57: 332-336. http://dx.doi.org/10.1080/10473289.2007.10465336

     Leland. HV; Kuwabara. JS. (1985). Trace metals. In GM Rand; SR Petrocelli (Eds.), Fundamentals of
        aquatic toxicology: Methods and applications (pp. 374-415). Washington, DC: Hemisphere
        Publishing.

     Levin. R; Brown. MJ; Kashtock. ME; Jacobs. DE; Whelan. EA; Rodman. J: Schock. MR; Padilla. A;
        Sinks. T. (2008). Lead exposures in US children, 2008: Implications for prevention. Environ Health
        Perspect 116: 1285-1293. http://dx.doi.org/10.1289/ehp.11241

     Lewis. J: Sjostrom. J: Skyllberg. U; Hagglund. L. (2010). Distribution, chemical speciation, and mobility
        of lead and antimony originating from small arms ammunition in a coarse-grained unsaturated surface
        sand. J Environ Qual 39: 863-870. http://dx.doi.org/10.2134/ieq2009.0211

     Li. C: Nguyen. Q; Ryan. PH; Lemasters. GK; Spitz. H; Lobaugh. M; Glover. S: Grinshpun. SA. (2009a).
        School bus pollution and changes in the air quality  at schools: A case study. J Environ Monit 11:
        1037-1042. http://dx.doi.org/10.1039/b819458k

     Li. L; Pala. F; Jiang. M; Krahforst. C: Wallace. GT. (2010a). Three-dimensional modeling of Cu and Pb
        distributions in Boston Harbor, Massachusetts and Cape Cod Bays. Estuar Coast Shelf Sci 88:
        450-463. http://dx.doi.0rg/10.1016/i.ecss.2010.05.003

     Lim. JH; Sabin. LD; Schiff. KG: Stolzenbach. KD. (2006). Concentration, size distribution,  and dry
        deposition rate of particle-associated metals in the Los Angeles region. Atmos Environ 40: 7810-7823.
        http://dx.doi.0rg/10.1016/i.atmosenv.2006.07.025

     Lin. CC: Chen. SJ: Huang. KL; Hwang. WI; Chang-Chien. GP; Lin. WY. (2005).  Characteristics of
        metals in nano/ultrafine/fme/coarse particles collected beside a heavily trafficked road. Environ Sci
        Technol 39: 8113-8122.

     Liu. W: Hopke. PK; van Curen. RA. (2003).  Origins of fine aerosol mass in the western United States
        using positive matrix factorization. J Geophys Res  108: 4716.
        http://dx.doi.org/10.1029/2003JD003678

     Lough. GC: Schauer. JJ; Park. JS: Shafer. MM; Deminter. JT; Weinstein. JP (2005). Emissions of metals
        associated with motor vehicle roadways. Environ Sci Technol 39: 826-836.

     Mager. EM. (2012). Lead. In Homeostasis and toxicology of non-essential metals. U.K.: Academic Press.
        http://www.elsevierdirect.com/ISBN/9780123786340/Fish-Phvsiologv-Homeostasis-and-Toxicologv-
        of-NonEssential-Metals
November 2012                             3-239                  Draft - Do Not Cite or Quote

-------
     Magill. N: Sansalone. J. (2010). Distribution of particulate-bound metals for source area snow in the Lake
        Tahoe watershed. J Environ Eng 136: 185-193.
        http://dx.doi.org/10.1061/(ASCE)EE. 1943-7870.0000146

     Maher. BA; Moore. C: Matzka. J. (2008). Spatial variation in vehicle-derived metal pollution identified
        by magnetic and elemental analysis of roadside tree leaves. Atmos Environ 42: 364-373.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.09.013

     Mahler. BJ; Van Metre. PC: Callender. E. (2006). Trends in metals in urban and reference lake sediments
        across the United States, 1970 to 2001. Environ Toxicol Chem 25: 1698-1709.

     Makkonen. U; Hellen. H; Anttila. P; Perm. M. (2010). Size distribution and chemical composition of
        airborne particles in south-eastern Finland during different seasons and wildfire episodes in 2006. Sci
        Total Environ 408: 644-651. http://dx.doi.Org/10.1016/i.scitotenv.2009.10.050

     Marple. VA; Rubow. KL; Behm. SM. (1991). A microorifice uniform deposit impactor (MOUDI):
        Description, calibration, and use. Aerosol Sci Technol 14: 434-446.
        http://dx.doi.org/10.1080/02786829108959504

     Martuzevicius. D; Grinshpun. SA; Reponen. T; Gorny. RL; Shukla. R; Lockev. J: Hu. S: McDonald. R;
        Biswas. P; Kliucininkas. L; LeMasters. G. (2004). Spatial and temporal variations of PM2.5
        concentration  and composition throughout an urban area with high freeway density: The Greater
        Cincinnati study. Atmos Environ 38: 1091-1105. http://dx.doi.Org/10.1016/j.atmosenv.2003.ll.015

     Marx. SK; Kamber. BS: McGowan. HA. (2008). Scavenging of atmospheric trace metal pollutants by
        mineral dusts: Inter-regional transport of Australian trace metal pollution to New Zealand. Atmos
        Environ 42: 2460-2478. http://dx.doi.Org/10.1016/i.atmosenv.2007.12.014

     McConnell. JR; Edwards. R. (2008). Coal burning leaves toxic heavy metal legacy in the Arctic. PNAS
        105: 12140-12144. http://dx.doi.org/10.1073/pnas.08035641Q5

     McGee. CJ: Fernandez. IJ; Norton. SA; Stubbs. CS. (2007). Cd, Ni, Pb, and Zn concentrations in forest
        vegetation and soils in Maine. Water Air Soil Pollut 180: 141-153.
        http://dx.doi.org/10.1007/sll270-006-9257-0

     McKenzie. ER; Money. JE; Green. PG: Young. TM. (2009). Metals associated with stormwater-relevant
        brake and tire samples. Sci Total Environ 407: 5855-5860.
        http://dx.doi.0rg/10.1016/i.scitotenv.2009.07.018

     McKenzie. ER; Wong. CM; Green. PG: Kayhanian. M; Young. TM. (2008). Size dependent elemental
        composition of road-associated particles. Sci Total Environ 398: 145-153.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.02.052

     Mercer. TT; Tillery. MI; Newton. GJ. (1970). A multi-stage, low flow rate cascade impactor. J Aerosol  Sci
        1: 9-15. http://dx.doi.org/10.1016/0021-8502(70)90024-8

     MetOne (Met One Instruments Inc.). (2009). SASS/SUPER SASS speciation samplers: Specifications.
        Available online at http://www.metone.com/documents/SASS0301Particulate.pdf

     Mielke. HW: Gonzales. C. (2008). Mercury (Hg) and lead (Pb) in interior and exterior New Orleans
        house paint films. Chemosphere 72: 882-885. http://dx.doi.Org/10.1016/i.chemosphere.2008.03.061

     Mielke. HW: Gonzales. C: Powell. E; Mielke. PW. Jr.  (2008). Urban  soil-lead (Pb) footprint:
        Retrospective comparison of public and private properties in New Orleans. Environ Geochem Health
        30: 231-242. http://dx.doi.org/10.1007/sl0653-007-9111-3

     Mielke. HW: Gonzales. CR; Powell. E; Jartun. M; Mielke. PW. Jr. (2007a). Nonlinear association
        between soil lead and blood lead of children in metropolitan New Orleans, Louisiana: 2000-2005. Sci
        Total Environ 388: 43-53.  http://dx.doi.Org/10.1016/j.scitotenv.2007.08.012

     Mielke. HW: Laidlaw. MA; Gonzales. C. (2010). Lead (Pb) legacy from vehicle traffic in eight California
        urbanized areas: Continuing influence of lead dust on children's health  [Review]. Sci Total Environ
        408: 3965-3975. http://dx.doi.Org/10.1016/i.scitotenv.2010.05.017
November 2012                             3-240                   Draft - Do Not Cite or Quote

-------
     Mielke. HW: Laidlaw. MA; Gonzales. CR. (2011c). Estimation of leaded (Pb) gasoline's continuing
        material and health impacts on 90 US urbanized areas [Review]. Environ Int 37: 248-257.
        http://dx.doi.0rg/10.1016/i.envint.2010.08.006

     Mielke. HW: Powell. ET; Gonzales. CR: Mielke. PW. Jr. (2007b). Potential lead on play surfaces:
        Evaluation of the "PLOPS" sampler as a new tool for primary lead prevention. Environ Res 103:
        154-159. http://dx.doi.0rg/10.1016/i.envres.2006.08.007

     Moffet. RC: de Foy. B; Molina. LT; Molina. MJ; Prather. KA. (2008a). Measurement of ambient aerosols
        in northern Mexico City by single particle mass spectrometry. Atmos Chem Phys 8: 4499-4516.
        http://dx.doi.org/10.5194/acp-8-4499-2008

     Moffet. RC: Desvaterik. Y; Hopkins. RJ: Tivanski. AV; Gilles. MK; Wang. Y; Shutthanandan. V; Molina.
        LT; Abraham. RG: Johnson. KS: Mugica. V: Molina. MJ: Laskin. A; Prather. KA. (2008b).
        Characterization of aerosols containing Zn, Pb, and Cl from an industrial region of Mexico City.
        Environ Sci Technol 42: 7091-7097. http://dx.doi.org/10.1021/es7030483

     Molnar. P; Bellander. T; Sallsten. G: Boman. J. (2007). Indoor and outdoor concentrations of PM2.5 trace
        elements at homes, preschools and schools in Stockholm, Sweden. J Environ Monit 9: 348-357.
        http://dx.doi.org/10.1039/b616858b

     Moreno. T; Querol. X; Pev. J: Minguillon. MC: Perez. N; Alastuev. A; Bernabe. RM; Blanco. S:
        Cardenas. B; Eichinger. W: Salcido. A; Gibbons. W. (2008). Spatial  and temporal variations in
        inhalable CuZnPb aerosols within the Mexico City pollution plume.  J Environ Monit 10: 370-378.
        http://dx.doi.org/10.1039/b716507b

     Murakami. M; Nakaiima. F; Furumai. H; Tomiyasu. B; Owari. M. (2007). Identification of particles
        containing chromium and lead in road dust and soakaway sediment by electron probe microanalyser.
        Chemosphere 67: 2000-2010. http://dx.doi.Org/10.1016/i.chemosphere.2006.ll.044

     Murphy. DM; Capps. SL; Daniel. JS: Frost. GJ: White. WH. (2008). Weekly patterns of aerosol in the
        United States. Atmos Chem Phys 8: 2729-2739. http://dx.doi.org/10.5194/acp-8-2729-2008

     Murphy. DM; Hudson. PK; Cziczo. DJ; Gallavardin.  S: Froyd. KD; Johnston. MV; Middlebrook. AM;
        Reinard. MS: Thomson. DS: Thornberrv. T; Wexler. AS. (2007). Distribution of lead in single
        atmospheric particles. Atmos Chem Phys 7: 3195-3210. http://dx.doi.org/10.5194/acp-7-3195-2007

     N'guessan. YM; Probst.  JL; Bur. T; Probst. A. (2009). Trace elements in stream bed sediments from
        agricultural catchments (Gascogne region, S-W France): Where do they come from? Sci Total Environ
        407: 2939-2952. http://dx.doi.Org/10.1016/i.scitotenv.2008.12.047

     NAS (National Academy of Sciences). (1980). Lead in the human environment. Washington, DC:
        National Academy Press.

     Nicholson. KW (1988). Areview of particle resuspension [Review]. Atmos Environ 22: 2639-2651.
        http://dx.doi.org/10.1016/0004-6981(88)90433-7

     Nicholson. KW: Branson. JW. (1990). Factors affecting resuspension by road traffic. Sci Total Environ
        93: 349-358. http://dx.doi.org/10.1016/0048-9697(90)90126-F

     Nishiguchi. K; Utani. K; Fujimori. E. (2008).  Real-time multielement monitoring of airborne particulate
        matter using ICP-MS instrument equipped with gas converter apparatus. J Anal At Spectrom 23:
        1125-1129. http://dx.doi.org/10.1039/b802302f

     Noble. SR: Horstwood. MSA; Daw. P; Pashlev. V: Spiro. B; Smith. S. (2008). Evolving Pb isotope
        signatures of London airborne particulate matter (PM10): Constraints from on-filter and solution-
        mode MC-ICP-MS. J Environ Monit 10: 830-836. http://dx.doi.org/10.1039/b80215la

     Noll. KE. (1970). A rotary inertia! impactor for sampling giant particles in the atmosphere. Atmos
        Environ 4: 9-19. http://dx.doi.org/10.1016/0004-6981(70)90050-8

     Norton. SA. (2007). Atmospheric metal pollutants: Archives, methods, and history. Water Air Soil Pollut
        7: 93-98. http://dx.doi.org/10.1007/sll267-006-9089-3
November 2012                             3-241                   Draft - Do Not Cite or Quote

-------
    Novak. M; Erel. Y; Zemanova. L; Bottrell. SH; Adamova. M. (2008). A comparison of lead pollution
        record in sphagnum peat with known historical Pb emission rates in the British Isles and the Czech
        Republic. Atmos Environ 42: 8997-9006. http://dx.doi.Org/10.1016/i.atmosenv.2008.09.031

    NFS (U.S. National Park Service). (2011). WACAP database.
        http://www.nature.nps.gov/air/studies/air toxics/WACAPreport.cfm

    NRCS (Natural Resources Conservation Service). (2011). USDANational Water and Climate Center.
        Available online at http://www.wcc.nrcs.usda.gov/ftpref/downloads/climate/windrose/ (accessed
        January 11,2011).

    Nriagu. JO. (1989). A global assessment of natural sources of atmospheric trace metals. Nature 338:
        47-49. http://dx.doi.org/10.1038/338047aO

    Obernberger. I; Brunner. T; Barnthaler. G. (2006). Chemical properties of solid biofuels—significance and
        impact. Biomass and Bioenergy 30: 973-982. http://dx.doi.Org/10.1016/j.biombioe.2006.06.011

    Oberts. GL. (2000). Influence of snowmelt dynamics on stormwater runoff quality. Watershed Protection
        Techniques 1:  55-61.

    Odigie. KO; Flegal. AR. (2011). Pyrogenic remobilization of historic industrial lead depositions. Environ
        Sci Technol 45: 6290-6295. http://dx.doi.org/10.1021/es200944w

    Ogulei. D; Hopke. PK; Zhou. L; Pancras. JP; Nair. N; Ondov. JM. (2006). Source apportionment of
        Baltimore aerosol from combined size distribution and chemical composition data. Atmos Environ 40:
        396-410. http://dx.doi.0rg/10.1016/i.atmosenv.2005.ll.075

    Ondov. JM; Buckley. TJ; Hopke. PK; Ogulei. D; Parlange. MB; Rogge.  WF; Squibb. KS: Johnston. MV;
        Wexler. AS. (2006). Baltimore Supersite: Highly time- and size-resolved concentrations of urban
        PM2.5 and its  constituents for resolution of sources and immune responses. Atmos Environ 40:
        224-237. http://dx.doi.0rg/10.1016/i.atmosenv.2005.ll.072

    Osterberg. E; Mavewski. P; Kreutz. K; Fisher. D; Handlev. M; Sneed. S: Zdanowicz. C: Zheng. J;
        Demuth. M; Waskiewicz. M; Bourgeois. J. (2008). Ice core record of rising lead pollution in the North
        Pacific atmosphere. Geophys Res Lett 35: L05810. http://dx.doi.org/10.1029/2007gl032680

    Pacyna. EG: Pacyna. JM: Fudala. J: Strzelecka-Jastrzab. E; Hlawiczka. S: Panasiuk. D; Nitter. S: Pregger.
        T; Pfeiffer. H;  Friedrich. R. (2007). Current and future emissions of selected heavy metals to the
        atmosphere from anthropogenic sources in Europe. Atmos Environ 41: 8557-8566.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.07.040

    Pacyna. JM: Pacyna. EG. (2001). An assessment of global and regional emissions of trace metals to the
        atmosphere from anthropogenic sources worldwide. Environ Rev 9: 269-298.
        http://dx.doi.org/10.1139/a01-012

    Pancras. JP: Ondov. JM: Poor. N; Landis. MS: Stevens. RK. (2006). Identification of sources and
        estimation of emission profiles from highly time-resolved pollutant measurements in Tampa, FL.
        Atmos Environ 40: 467-481. http://dx.doi.Org/10.1016/i.atmosenv.2005.12.036

    Pancras. JP: Ondov. JM: Zeisler. R. (2005).  Multi-element electrothermal AAS determination of 11
        marker elements in fine ambient aerosol slurry samples collected with SEAS-II. Anal Chim Acta 538:
        303-312. http://dx.doi.0rg/10.1016/i.aca.2005.01.062

    Paode. RD: Sofuoglu. SC: Sivadechathep. J: Noll. KB; Holsen. TM; Keeler. GJ. (1998). Dry deposition
        fluxes and mass size distributions of Pb, Cu, and Zn measured in southern Lake Michigan during
        AEOLOS. Environ Sci Technol 32: 1629-1635. http://dx.doi.org/10.1021/es970892b

    Pardue. JH; Moe.  WM; Mcinnis. D: Thibodeaux. LJ; Valsaraj. KT; Maciasz. E; van Heerden. I; Korevec.
        N; Yuan. QZ. (2005). Chemical and microbiological parameters in New Orleans floodwater following
        Hurricane Katrina. Environ Sci Technol  39: 8591-8599. http://dx.doi.org/10.1021/es0518631

    Park. JM; Rock. JC: Wang. L; Seo. YC: Bhatnagar. A; Kim. S. (2009a).  Performance evaluation of six
        different aerosol samplers in aparticulate matter generation chamber. Atmos Environ 43: 280-289.
        http://dx.doi.0rg/10.1016/i.atmosenv.2008.09.028
November 2012                             3-242                  Draft - Do Not Cite or Quote

-------
     Patra. A; Colvile. R; Arnold. S: Bowen. E; Shallcross. D; Martin. D; Price. C: Tate. J: ApSimon. H;
        Robins. A. (2008). On street observations of paniculate matter movement and dispersion due to traffic
        on an urban road. Atmos Environ 42: 3911-3926. http://dx.doi.Org/10.1016/j.atmosenv.2006.10.070

     Pedrot. M; Dia. A; Davranche. M; Bouhnik-Le Coz. M; Henin. O; Gruau. G. (2008). Insights into colloid-
        mediated trace element release at the soil/water interface. J Colloid Interface Sci 325: 187-197.
        http://dx.doi.0rg/10.1016/i.icis.2008.05.019

     Pekev. B; Bozkurt. ZB; Pekev. H; Dogan. G: Zararsiz. A; Efe. N; Tuncel. G. (2010). Indoor/outdoor
        concentrations and elemental composition of PM10/PM2.5 in urban/industrial areas of Kocaeli City,
        Turkey. Indoor Air 20: 112-125. http://dx.doi.Org/10.llll/i.1600-0668.2009.00628.x

     Peknev. NJ; Davidson. CI; Bein. KJ; Wexler. AS: Johnston. MV. (2006). Identification of sources of
        atmospheric PM at the Pittsburgh supersite, part I: Single particle analysis and filter-based positive
        matrix factorization. Atmos Environ 40: 411-423. http://dx.doi.Org/10.1016/i.atmosenv.2005.12.072

     Perdrial. N; Liewig. N; Delphin. JE; Elsass. F. (2008). TEM evidence for intracellular accumulation of
        lead by bacteria in subsurface environments. Chem Geol 253: 196-204.
        http://dx.doi.0rg/10.1016/i.chemgeo.2008.05.008

     Perrino. C: Canepari. S: Pappalardo. S: Marconi. E. (2010). Time-resolved measurements of water-
        soluble ions and elements in atmospheric paniculate matter for the characterization of local and long-
        range transport events. Chemosphere 80: 1291-1300.
        http://dx.doi.0rg/10.1016/i.chemosphere.2010.06.050

     Pingitore. NE. Jr; Clague. JW: Amava. MA; Maciejewska. B; Reynoso. JJ. (2009). Urban airborne lead:
        X-ray  absorption spectroscopy establishes soil as dominant source. PLoS ONE 4: e5019.
        http://dx.doi.org/10.1371/iournal.pone.0005019

     Planchon. FAM; van de Velde. K; Rosman. KJR; Wolff. EW: Ferrari. CP; Boutron. CF. (2003). One
        hundred fifty-year record of lead isotopes in Antarctic snow from Coats Land. Geochim Cosmo Act
        67: 693-708. http://dx.doi.org/10.1016/S0016-7037(02)01136-5

     Point. D: Monperrus. M; Tessier. E; Amouroux. D; Chauvaud. L; Thouzeau. G: Jean. F; Amice. E; Grail.
        J; Leynaert. A; Clavier. J: Donard. OFX.  (2007). Biological control of trace metal and organometal
        benthic fluxes in a eutrophic lagoon (Thau Lagoon, Mediterranean Sea, France). Estuar Coast Shelf
        Sci 72: 457-471. http://dx.doi.Org/10.1016/j.ecss.2006.ll.013

     Pokrovsky. OS: Schott. J. (2002). Iron colloids/organic matter associated transport of major and trace
        elements in small boreal rivers and their estuaries (NW Russia). Chem Geol 190: 141-179.
        http://dx.doi.org/10.1016/S0009-2541(02)00115-8

     Pokrovsky. OS: Schott. J: Dupre. B. (2006). Trace element fractionation and transport in boreal rivers and
        soil porewaters of permafrost-dominated basaltic terrain in Central Siberia. Geochim Cosmo Act 70:
        3239-3260. http://dx.doi.Org/10.1016/i.gca.2006.04.008

     Pokrovsky. OS: Viers. J: Shirokova. LS: Shevchenko. VP; Filipov. AS: Dupre. B. (2010). Dissolved,
        suspended, and colloidal fluxes of organic carbon, major and trace elements in the Severnaya Dvina
        River and its tributary. Chem Geol 273: 136-149. http://dx.doi.Org/10.1016/j.chemgeo.2010.02.018

     Polissar. AV; Hopke. PK; Poirot. RL. (2001). Atmospheric aerosol over Vermont: chemical composition
        and sources. Environ Sci Technol 35: 4604-4621.

     Polissar. AV: Paatero. P; Hopke. PK; Malm.  WC: Sisler. JF. (1998). Atmospheric aerosol over Alaska: 2.
        Elemental composition and sources. J Geophys Res  103: 19045-19057.

     Poperechna. N; Heumann. KG. (2005). Species-specific GC/ICP-IDMS for trimethyllead determinations
        in biological and environmental samples. Anal Chem 77: 511-516.
        http://dx.doi.org/10.1021/ac048757m

     Popp. M;  Hann. S: Koellensperger. G.  (2010). Environmental application of elemental speciation analysis
        based  on liquid or gas chromatography hyphenated to inductively coupled plasma mass spectrometry—
        a review. Anal Chim Acta 668: 114-129. http://dx.doi.Org/10.1016/i.aca.2010.04.036
November 2012                             3-243                   Draft - Do Not Cite or Quote

-------
     Potts. PJ: Webb. PC. (1992). X-ray fluorescence spectrometry. J Geochem Explor 44: 251-296.
        http://dx.doi.org/10.1016/0375-6742(92)90052-A

     Poykio. R; Nurmesniemi. H; Dahl. O. (2007). Concentrations of nutrients and heavy metals in cyclone fly
        ash from the grate-fired boiler at a small municipal district heating plant (6 MW). J Resid Sci Tech 4:
        127-136.

     Prather. KA; Nordmever. T; Salt. K. (1994). Real-time characterization of individual aerosol particles
        using time-of-flight mass spectrometry. Anal Chem 66: 1403-1407.
        http://dx.doi.org/10.1021/ac00081a007

     Pratt. C: Lottermoser. BG. (2007). Mobilisation of traffic-derived trace metals from road corridors into
        coastal stream and estuarine sediments, Cairns, northern Australia. Environ Geol 52: 437-448.
        http://dx.doi.org/10.1007/s00254-006-0471-2

     Preciado. HF; Li. LY. (2006). Evaluation of metal loadings and bioavailability in air, water and soil along
        two highways of British Columbia, Canada. Water Air Soil Pollut 172: 81-108.
        http://dx.doi.org/10.1007/sll270-005-9063-0

     Presley. SM; Abel. MT; Austin. GP; Rainwater. TR; Brown. RW: Mcdaniel. LN; Marsland. EJ;
        Fornerette. AM; Dillard. ML; Rigdon. RW: Kendall. RJ: Cobb. GP. (2010). Metal concentrations in
        schoolyard soils from New Orleans, Louisiana before and after hurricanes Katrina and Rita.
        Chemosphere 80: 67-73. http://dx.doi.Org/10.1016/i.chemosphere.2010.03.031

     Presley. SM; Rainwater. TR; Austin. GP: Platt. SG: Zak. JC:  Cobb. GP: Marsland. EJ: Tian. K; Zhang. B;
        Anderson. TA; Cox. SB; Abel. MT; Leftwich.  BD; Huddleston. JR; Jeter. RM; Kendall. RJ. (2006).
        Assessment of pathogens and toxicants in New Orleans, LA following hurricane Katrina. Environ Sci
        Technol 40: 468-474. http://dx.doi.org/10.1021/es052219p

     Prospero. JM. (1999). Assessing the impact of advected African dust on air quality and health in the
        United  States. Hum Ecol Risk Assess 5: 471-479.

     Pruvot C:  Douav. F; Herve. F; Waterlot. C. (2006). Heavy metals in soil, crops and grass as a source of
        human  exposure in the former mining areas. Journal of Soils and Sediments 6: 215-220.
        http://dx.doi.org/10.1065/iss2006.10.186

     Pyrzynska. K. (1996). Organolead speciation in environmental  samples: A review [Review]. Mikrochim
        Acta 122: 279-293. http://dx.doi.org/10.1007/BF01245788

     Qi. J: Feng. L; Li. X; Zhang. M. (2006). An X-ray photoelectron spectroscopy study of elements on the
        surface of aerosol particles. J Aerosol Sci 37: 218-227.
        http://dx.doi.0rg/10.1016/i.jaerosci.2005.04.005

     Quevauviller.  P. (2000). Accuracy and traceability in environmental monitoringpitfalls in methylmercury
        determinations as a case study. J Environ Monit 2: 292-299. http://dx.doi.org/10.1039/b001852j

     Quinton. JN; Catt. JA. (2007).  Enrichment of heavy metals in sediment resulting from soil erosion on
        agricultural fields. Environ Sci Technol 41: 3495-3500. http://dx.doi.org/10.1021/es062147h

     Qureshi. S: Dutkiewicz. VA; Khan. AR; Swami. K; Yang. KX; Husain. L; Schwab. JJ: Demerjian. KL.
        (2006). Elemental composition of PM2.5 aerosols in Queens, New York: Solubility and temporal
        trends. Atmos Environ 40: S238-S251. http://dx.doi.Org/10.1016/i.atmosenv.2005.12.070

     Raabe. OG: Braaten. DA; Axelbaum. RL; Teague. SV: Cahill. TA. (1988). Calibration studies of the drum
        impactor. JAerosol Sci 19:  183-195. http://dx.doi.org/10.1016/0021-8502(88)90222-4

     Rabinowitz. MB. (1986). Stable isotopes of ambient environmental lead in Boston, Massachusetts.
        (Technical Report 1-28). Washington, DC: U.S. Environmental Protection Agency.

     Radakovitch. O; Roussiez. V; Ollivier. P; Ludwig. W: Grenz. C: Probst. JL. (2008). Input of particulate
        heavy metals from rivers and associated sedimentary deposits on the Gulf of Lion continental shelf.
        Estuar Coast Shelf Sci 77: 285-295. http://dx.doi.Org/10.1016/i.ecss.2007.09.028
November 2012                              3-244                   Draft - Do Not Cite or Quote

-------
     Rauch. JN; Pacyna. JM. (2009). Earth's global Ag, Al, Cr, Cu, Fe, Ni, Pb, and Zn cycles. Global
        Biogeochem Cycles 23: GB2001. http://dx.doi.org/10.1029/2008gb003376

     Reinard. MS: Adoua. K; Martini. JM; Johnston. MV. (2007). Source characterization and identification by
        real-time single particle mass spectrometry. Atmos Environ 41: 9397-9409.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.09.001

     Reuer. MK; Weiss. DJ. (2002). Anthropogenic lead dynamics in the terrestrial and marine environment.
        Proc Biol Sci 360: 2889-2904.

     Rice. J. (2007). Memorandum: Summary of method detection limits for ambient lead methods. Research
        Triangle Park, NC: United States Environmental Protection Agency.

     Richter. P; Grino. P; Ahumada. I; Giordano. A. (2007). Total element concentration and chemical
        fractionation in airborne particulate matter from Santiago, Chile. Atmos Environ 41: 6729-6738.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.04.053

     Riediker. M; Williams. R; Devlin. R; Griggs. T; Bromberg. P. (2003). Exposure to particulate matter,
        volatile organic compounds, and other air pollutants inside patrol cars. Environ Sci Technol 37:
        2084-2093. http://dx.doi.org/10.1021/es026264y

     Rodes. CE; Evans. EG. (1985). Preliminary assessment of 10 "mu"m particulate sampling at eight
        locations in the United States. Atmos Environ  19: 293-303.
        http://dx.doi.org/10.1016/0004-6981(85)90097-6

     Roonev. CP; McLaren. RG: Condron. LM. (2007). Control of lead solubility in soil contaminated with
        lead shot: Effect of soil pH. Environ Pollut 149: 149-157.
        http://dx.doi.0rg/10.1016/i.envpol.2007.01.009

     Root. RA. (2000).  Lead loading of urban streets by motor vehicle wheel weights. Environ Health Perspect
        108: 937-940. http://dx.doi.org/10.1289/ehp.00108937

     Ross. JM; Sherrell. PJVI. (1999). The role of colloids in tracemetal transport and adsorption behavior in
        New Jersey. Limnol Oceanogr44: 1019-1034.

     Rothwell. JJ: Evans. MG: Allott. TEH. (2007a). Lead contamination of fluvial sediments in an eroding
        blanket peat catchment. Appl Geochem 22: 446-459.
        http://dx.doi.0rg/10.1016/i.apgeochem.2006.ll.002

     Rothwell. JJ: Evans. MG: Allott. TEH. (2008). In-stream processing of sediment-associated metals in
        peatland fluvial systems. Water Air Soil Pollut 187: 53-64.
        http://dx.doi.org/10.1007/sll270-007-9496-8

     Rothwell. JJ: Evans. MG: Daniels. SM; Allott. TEH. (2007b). Baseflow and stormflow metal
        concentrations  in streams draining contaminated peat moorlands in the Peak District National Park
        (U.K.). JHydrol341: 90-104. http://dx.doi.Org/10.1016/i.ihvdrol.2007.05.004

     Rothwell. JJ: Lindsay. JB; Evans. MG: Allott. TEH.  (2010). Modelling suspended sediment lead
        concentrations  in contaminated peatland catchments using digital terrain analysis. Ecol Eng 36:
        623-630. http://dx.doi.0rg/10.1016/i.ecoleng.2008.10.010

     Roulier. JL; Belaud. S: Coquery. M. (2010). Comparison of dynamic mobilization of Co, Cd and Pb in
        sediments using DGT and metal mobility assessed  by sequential extraction.  Chemosphere 79:
        839-843. http://dx.doi.0rg/10.1016/i.chemosphere.2010.02.056

     Roulier. S: Robinson. B; Kuster. E; Schulin. R. (2008b). Analysing the preferential transport of lead in a
        vegetated roadside soil using lysimeter experiments and a dual-porosity model. Eur J Soil Sci 59:
        61-70. http://dx.doi.0rg/10.llll/i.1365-2389.2007.00954.x

     Sabin. LD; Lim. JH; Stolzenbach. KD; Schiff. KG. (2006a). Atmospheric dry deposition of trace metals in
        the coastal region of Los Angeles, California, USA. Environ Toxicol Chem 25: 2334-2341.
        http://dx.doi.Org/10.1897/05-300R.l
November 2012                              3-245                   Draft - Do Not Cite or Quote

-------
     Sabin. LD; Lim. JH; Venezia. MT; Winer. AM; Schiff. KG: Stolzenbach. KD. (2006b). Dry deposition
        and resuspension of particle-associated metals near a freeway in Los Angeles. Atmos Environ 40:
        7528-7538. http://dx.doi.Org/10.1016/i.atmosenv.2006.07.004

     Sabin. LD; Schiff. KG. (2008). Dry atmospheric deposition rates of metals along a coastal transect in
        southern California. Atmos Environ 42: 6606-6613. http://dx.doi.Org/10.1016/i.atmosenv.2008.04.042

     Sadiq. M. (1992). Lead in marine environments. In Toxic metal chemistry in marine environments. New
        York, NY: Marcel Dekker, Inc.

     Sakata. M; Asakura. K. (2008). Evaluating relative contribution of atmospheric mercury species to
        mercury dry deposition in Japan. Water Air Soil Pollut 193: 51-63.
        http://dx.doi.org/10.1007/sll270-008-9667-2

     Salcedo. D; Onasch. TB; Aiken. AC: Williams. LR; de Foy. B; Cubison. MJ; Worsnop. PR; Molina. LT;
        Jimenez. JL. (2010). Determination of particulate lead using aerosol mass spectrometry:
        MILAGRO/MCMA-2006 observations. Atmos Chem Phys 10: 5371-5389.
        http://dx.doi.org/10.5194/acp-10-5371-2010

     Sammut. ML; Noack. Y; Rose. J: Hazemann. JL; Proux. O; Depoux. M; Ziebel. A; Fiani. E. (2010).
        Speciation of Cd and Pb in dust emitted from sinter plant. Chemosphere 78: 445-450.
        http://dx.doi.0rg/10.1016/i.chemosphere.2009.10.039

     Sanchez-Ccovllo. OR; Ynoue. RY; Martins. LD: Astolfo. R; Miranda. RM; Freitas. ED;  Borges. AS:
        Fornaro. A; Freitas. H; Moreira. A; Andrade. MF. (2009). Vehicular particulate matter emissions in
        road tunnels in Sao Paulo, Brazil. Environ  Monit Assess 149:  241-249.
        http://dx.doi.org/10.1007/sl0661-008-0198-5

     Sansalone. J: Ying. G: Lin. H. (2010). Distribution of metals for particulate matter transported in source
        area rainfall-runoff. J Environ Eng  136: 172-184.
        http://dx.doi.org/10.1061/(ASCE)EE. 1943-7870.0000139

     Sansalone. JJ: Buchberger. SG. (1996). Characterization of metals and solids in urban highway winter
        snow and spring rainfall-runoff. Trans Res  Rec 1523:  147-159. http://dx.doi.org/10.3141/1523-18

     Sato. K; Tamura. T; Furuta. N. (2008). Partitioning between soluble and insoluble fractions of major and
        trace elements in size-classified airborne particulate matter collected in Tokyo. J Environ Monit 10:
        211-218. http://dx.doi.org/10.1039/B709937A

     Schauer. JJ: Lough. GC: Shafer. MM; Christensen. WF; Arndt. MF; DeMinter. JT; Park. JS. (2006).
        Characterization of metals emitted from motor vehicles. (133). Boston, MA: Health Effects Institute.
        http://pubs.healtheffects.org/view.php?id=150

     Scheetz. CD; Rimstidt. JD. (2009). Dissolution, transport, and fate of lead on a shooting range in the
        Jefferson National Forest near Blacksburg, VA, USA.  Environ Geol 58: 655-665.
        http://dx.doi.org/10.1007/s00254-008-1540-5

     Scheid. S: Gunthardt-Goerg. MS: Schulin. R; Nowack. B. (2009). Accumulation and solubility of metals
        during leaf litter decomposition in non-polluted and polluted soil. Eur J Soil Sci 60: 613-621.
        http://dx.doi.0rg/10.llll/i.1365-2389.2009.01153.x

     Schleicher. N; Norra. S: Chai. F; Chen. Y; Wang. S: Stuben. D. (2010). Anthropogenic versus geogenic
        contribution to total suspended atmospheric particulate matter and its variations during a two-year
        sampling period in Beijing, China. J Environ Monit 12: 434-441. http://dx.doi.org/10.1039/b914739j

     Schroth. AW: Bostick. BC: Kaste. JM; Friedland. AJ. (2008). Lead sequestration and species
        redistribution during soil organic matter decomposition. Environ Sci Technol 42: 3627-3633.
        http://dx.doi.org/10.1021/es703002b

     Schulz-Zunkel. C: Krueger. F.  (2009). Trace metal dynamics in floodplain soils of the river Elbe: a review
        [Review]. J Environ Qual 38: 1349-1362. http://dx.doi.org/10.2134/iea2008.0299
November 2012                              3-246                  Draft - Do Not Cite or Quote

-------
     Schuwirth. N; Voegelin. A; Kretzschmar. R; Hofmann. T. (2007). Vertical distribution and speciation of
        trace metals in weathering flotation residues of a zinc/lead sulfide mine. J Environ Qua! 36: 61-69.
        http://dx.doi.org/10.2134/ieq2006.0148

     Schwab. AP; Zhu. DS: Banks. MK. (2008). Influence of organic acids on the transport of heavy metals in
        soil. Chemosphere 72: 986-994. http://dx.doi.Org/10.1016/i.chemosphere.2008.02.047

     Schwab. KJ; Gibson. KE; Williams. PL; Kulbicki. KM; Lo. CP; Mihalic. JN; Brevsse. PN; Curriero. FC:
        Geyh. AS. (2007). Microbial and chemical assessment of regions within New Orleans, LA impacted
        by hurricane Katrina. Environ Sci Technol 41: 2401-2406. http://dx.doi.org/10.1021/es062916x

     Sedefian. L; Rao. ST; Czapski. U. (1981). Effects of traffic-generated turbulence on near-field dispersion.
        Atmos Environ 15: 527-536. http://dx.doi.org/10.1016/0004-6981(81)90182-7

     Sengor. SS: Spvcher. NF; Ginn. TR; Sani. RK; Peyton. B. (2007). Biogeochemical reactive-diffusive
        transport of heavy metals in Lake Coeur d'Alene sediments. Appl Geochem 22: 2569-2594.
        http://dx.doi.0rg/10.1016/i.apgeochem.2007.06.011

     Shacklette. HT; Boerngen. JG.  (1984).  Element concentrations in soils and other surficial materials of the
        conterminous United States. (1270). Washington D.C.: Government Printing Office.
        http://pubs.usgs.gov/pp/1270/

     Shi. H; Witt. E. Ill; Shu. S: Su. T; Wang. J; Adams. C. (2010). Toxic trace element assessment for
        soils/sediments deposited during hurricanes Katrina and Rita from southern Louisiana, USA: A
        sequential extraction analysis. Environ Toxicol Chem 29: 1419-1428.
        http://dx.doi.org/10.1002/etc.218

     Shotbolt. LA; Rothwell. JJ: Lawlor. AJ. (2008). Amass balance approach to quantifying Pb storage and
        fluxes in an upland catchment of the Peak District, north-central England. Earth Surface Processes and
        Landforms33: 1721-1741. http://dx.doi.org/10.1002/esp.1644

     Shotyk. W: Krachler. M. (2007). Lead in bottled waters: Contamination from glass and comparison with
        pristine groundwater. Environ Sci Technol 41: 3508-3513. http://dx.doi.org/10.1021/es062964h

     Shotyk. W: Krachler. M. (2009). Determination of trace element concentrations in natural freshwaters:
        How low is "low," and how low do  we need to go? J Environ Monit 11: 1747-1753.
        http://dx.doi.org/10.1039/b917090c

     Shotvk. W: Krachler. M. (2010). The isotopic evolution of atmospheric Pb in central Ontario since AD
        1800, and its impacts on the soils, waters, and sediments of a forested watershed, Kawagama Lake.
        Geochim Cosmo Act 74: 1963-1981. http://dx.doi.Org/10.1016/j.gca.2010.01.009

     Shotvk. W: Krachler. M; Aeschbach-Hertig. W: Hillier. S: Zheng. JC. (2010). Trace elements in recent
        groundwater of an artesian flow system and comparison with snow: Enrichments, depletions, and
        chemical evolution of the water. J Environ Monit 12: 208-217. http://dx.doi.org/10.1039/B909723F

     Shotvk. W: Le Roux. G. (2005). Biogeochemistry and cycling of lead [Review]. In A Sigel; H Sigel;
        RKO Sigel (Eds.), Biogeochemical  cycles of elements (pp. 239-275). Boca Raton, FL: Taylor &
        Francis. http://dx.doi.org/10.1201/9780824751999.chlO

     Shotvk. W: Weiss. D; Heisterkamp. M; Cheburkin. AK; Applebv. PG: Adams. FC. (2002). New peat bog
        record of atmospheric lead pollution in Switzerland: Pb concentrations, enrichment factors, isotopic
        composition, and organolead species. Environ Sci Technol 36: 3893-3900.
        http://dx.doi.org/10.1021/es010196i

     Sillanpaa. M; Saarikoski. S: Hillamo. R; Pennanen. A; Makkonen. U; Spolnik. Z; Van Grieken. R;
        Koskentalo. TA; Salonen. RO. (2005). Chemical composition, mass size distribution  and source
        analysis of long-range transported wildfire smokes in Helsinki.  Sci Total Environ 350: 119-135.
        http://dx.doi.0rg/10.1016/i.scitotenv.2005.01.024

     Silva. PJ: Prather. KA. (1997).  On-line characterization of individual particles from automobile
        emissions. Environ Sci Technol 31:  3074-3080. http://dx.doi.org/10.1021/es961063d
November 2012                             3-247                  Draft - Do Not Cite or Quote

-------
     Singh. M; Jaques. PA; Sioutas. C. (2002). Size distribution and diurnal characteristics of particle-bound
        metals in source and receptor sites of the Los Angeles Basin. Atmos Environ 36: 1675-1689.
        http://dx.doi.org/10.1016/81352-2310(02)00166-8

     Smichowski. P; Polla. G: Gomez. D. (2005). Metal fractionation of atmospheric aerosols via sequential
        chemical extraction: a review [Review]. Anal Bioanal Chem 381: 302-316.
        http://dx.doi.org/10.1007/s00216-004-2849-x

     Smichowski. P; Polla. G: Gomez. D; Espinosa. AJF; Lopez. AC. (2008). A three-step metal fractionation
        scheme for fly ashes collected in an Argentine thermal power plant. Fuel 87: 1249-1258.
        http://dx.doi.0rg/10.1016/i.fuel.2007.07.011

     Smith. GR.  (2001). Minerals yearbook 1999: lead. Washington, DC: U.S. Department of the Interior, U.S.
        Geological Survey.

     Snyder. DC: Schauer. JJ; Gross. DS: Turner. JR. (2009). Estimating the contribution of point sources to
        atmospheric metals using single-particle mass spectrometry. Atmos Environ 43: 4033-4042.
        http://dx.doi.0rg/10.1016/i.atmosenv.2009.05.011

     Sobanska. S: Ricq. N; Laboudigue. A; Guillermo. R; Bremard. C: Laureyns. J: Merlin. JC: Wignacourt
        JP (1999). Microchemical investigations of dust emitted by a lead smelter. Environ Sci Technol 33:
        1334-1339. http://dx.doi.org/10.1021/es9805270

     Solomon. P; Derrick. M; Movers. J: Hyde. P. (1982). Performance Comparison of Three Samplers of
        Suspended Airborne Paniculate Matter.  J Air Pollut Control Assoc 32: 373-375.
        http://dx.doi.org/10.1080/00022470.1982.10465411

     Song. F; Gao. Y. (2011). Size distributions of trace elements associated with ambient particular matter in
        the affinity of a major highway in the New JerseyNew York metropolitan area. Atmos Environ 45:
        6714-6723. http://dx.doi.Org/10.1016/i.atmosenv.2011.08.031

     Song. XH; Polissar. AV; Hopke. PK. (2001). Sources of fine particle composition in the northeastern US.
        Atmos Environ 35: 5277-5286.

     Soto-Jimenez. MF; Flegal. AR. (2009). Origin of lead in the Gulf of California Ecoregion using stable
        isotope analysis. J Geochem Explor 101: 209-217. http://dx.doi.Org/10.1016/i.gexplo.2008.07.003

     Soto-Jimenez. MF; Paez-Osuna. F.  (2010). A first approach to study the mobility and behavior of lead in
        hypersaline salt marsh sediments: Diffusive and advective fluxes, geochemical partitioning and Pb
        isotopes. J Geochem Explor 104: 87-96. http://dx.doi.Org/10.1016/i.gexplo.2009.12.006

     Soulhac. L; Garbero. V; Salizzoni. P; Mejean. P; Perkins. RJ. (2009). Flow and dispersion in street
        intersections. Atmos Environ 43: 2981-2996. http://dx.doi.Org/10.1016/j.atmosenv.2009.02.061

     Spalinger. SM; von Braun. MC: Petrosvan. V; von Lindern. IH. (2007). Northern Idaho house dust and
        soil lead levels compared to the Bunker Hill superfund site. Environ Monit Assess 130: 57-72.
        http://dx.doi.org/10.1007/sl0661-006-9450-z

     Steinnes. E; Sjobakk. TE; Donisa.  C: Brannvall. ML. (2005). Quantification of pollutant lead in forest
        soils. Soil Sci Soc Am J 69: 1399-1404. http://dx.doi.org/10.2136/sssaj2004.0095

     Stolpe. B; Hassellov. M. (2007). Changes in size distribution of fresh water nanoscale colloidal matter
        and associated elements on mixing with seawater. Geochim Cosmo Act 71: 3292-3301.
        http://dx.doi.0rg/10.1016/i.gca.2007.04.025

     Sturm. A; Crowe. SA; Fowle. DA. (2008). Trace lead impacts biomineralization pathways during
        bacterial iron reduction. Chem Geol 249: 282-293.  http://dx.doi.Org/10.1016/i.chemgeo.2008.01.017

     Swarbrick. JC: Skyllberg. U; Karlsson. T; Glatzel. P. (2009). High energy resolution X-ray absorption
        spectroscopy of environmentally relevant lead(II) compounds. Inorg Chem 48: 10748-10756.
        http://dx.doi.org/10.1021/ic9015299

     Sweet. CW: Gatz. DF. (1998). Summary and analysis of available PM25 measurements in Illinois. Atmos
        Environ 32: 1129-1133.
November 2012                             3-248                   Draft - Do Not Cite or Quote

-------
     Sweet. CW: Weiss. A; Vermette. SJ. (1998). Atmospheric deposition of trace metals at three sites near the
        Great Lakes. Water Air Soil Pollut 103: 423-439. http://dx.doi.Org/10.1023/A:1004905832617

     Taillefert. M; Lienemann. CP; Gaillard. JF; Ferret. D. (2000). Speciation, reactivity, and cycling of Fe and
        Pb in a meromictic lake. Geochim Cosmo Act 64: 169-183.
        http://dx.doi.org/10.1016/S0016-7037(99)00285-9

     Takamatsu. T; Watanabe. M; Koshikawa. MK; Murata. T; Yamamura. S: Havashi. S. (2010). Pollution of
        montane soil with Cu, Zn, As, Sb, Pb, and nitrate in Kanto, Japan. Sci Total Environ 408: 1932-1942.
        http://dx.doi.0rg/10.1016/i.scitotenv.2010.01.016

     Tan. MG: Zhang. GL; Li. XL; Zhang. YX; Yue. WS: Chen. JM; Wang. YS: Li. AG: Li. Y; Zhang. YM;
        Shan. ZC. (2006). Comprehensive study of lead pollution in Shanghai by multiple techniques. Anal
        Chem 78: 8044-8050. http://dx.doi.org/10.1021/ac061365q

     Taylor. MP; Mackav. AK; Hudson-Edwards. KA; Holz. E. (2010). Soil Cd, Cu, Pb and Zn contaminants
        around Mount Isa city, Queensland, Australia: Potential sources and risks to human health. Appl
        Geochem 25: 841-855. http://dx.doi.Org/10.1016/j.apgeochem.2010.03.003

     Tessier. A; Campbell. PGC: Bisson. M. (1979). Sequential extraction procedure for the speciation of
        paniculate trace-metals. Anal Chem 51: 844-851. http://dx.doi.org/10.1021/ac50043a017

     Theodosi. C: Markaki. Z; Tselepides. A; Mihalopoulos. N. (2010). The significance of atmospheric inputs
        of soluble and paniculate major and trace metals to the eastern Mediterranean seawater. Mar Chem
        120:  154-163. http://dx.doi.0rg/10.1016/i.marchem.2010.02.003

     Thurston. GD; Ito. K; Lall. R. (2011). A source apportionment of U.S. fine paniculate matter air pollution.
        Atmos Environ 45: 3924-3936. http://dx.doi.Org/10.1016/i.atmosenv.2011.04.070

     Tipping. E; Lawlor. AJ; Lofts. S: Shotbolt. L. (2006). Simulating the long-term chemistry of an upland
        U.K. catchment: Heavy metals. Environ Pollut 141: 139-150.
        http://dx.doi.0rg/10.1016/i.envpol.2005.08.019

     Tuccillo. ME. (2006). Size fractionation of metals in runoff from residential and highway storm sewers.
        Sci Total Environ 355: 288-300. http://dx.doi.Org/10.1016/j.scitotenv.2005.03.003

     U.S. EPA (U.S. Environmental Protection Agency). (1977). Air quality criteria for lead [EPA Report].
        (EPA-600/8-77-017). Washington, D.C.
        http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=20013GWR.txt

     U.S. EPA (U.S. Environmental Protection Agency). (1986a). Air quality criteria for lead [EPA Report].
        (EPA/600/8-83/028aF-dF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1986b). Air quality criteria for lead: Volume I of IV
        [EPA Report]. (EPA-600/8-83/028aF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1999). Particulate matter (PM2.5) speciation
        guidance: Final draft (Edition 1) [EPA Report]. Research Triangle Park, NC.
        http://www.epa.gov/ttn/amtic/files/ambient/pm25/spec/specpln2.pdf

     U.S. EPA (U.S. Environmental Protection Agency). (2003b). Guidance for developing ecological soil
        screening levels (Eco-SSLs): Review of background concentration for metals - Attachment 1-4 [EPA
        Report]. (OSWER Directive 92857-55). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2004). Air quality criteria for paniculate matter [EPA
        Report]. (EPA/600/P-99/002aF-bF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=87903

     U.S. EPA (U.S. Environmental Protection Agency). (2006a). 2002 National  Emissions Inventory data and
        documentation [EPA Report]. Research Triangle Park, NC.
        http://www.epa.gov/ttnchiel/net/2002inventory.html
November 2012                             3-249                  Draft - Do Not Cite or Quote

-------
     U.S. EPA (U.S. Environmental Protection Agency). (2006b). Air quality criteria for lead: Volume I of II
        [EPA Report]. (EPA/600/R-05/144aF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/CFM/recordisplav.cfm?deid=158823

     U.S. EPA (U.S. Environmental Protection Agency). (2007d). Guidance for developing ecological soil
        screening levels [EPA Report]. (OSWER Directive 92857-55). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2007h). Review of the national ambient air quality
        standards for lead: Policy assessment of scientific and technical Information: OAQPS staff paper [EPA
        Report]. (EPA-452/R-07-013). Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2008a). 2005 National Emissions Inventory data and
        documentation. Available online at http://www.epa.gov/ttn/chief/net/2005inventory.html (accessed
        November 1, 2010).

     U.S. EPA (U.S. Environmental Protection Agency). (2008b). Ambient aquatic life water quality: Lead
        (draft) [EPA Report]. (EPA Contract No. 68-C-04-006). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2008c). California area designations for the 2008
        lead National Ambient Air Quality Standards [EPA Report]. Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2008d). Florida area designations for the 2008 lead
        National Ambient Air Quality Standards [EPA Report]. Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2008f). Missouri area designations for the 2008 lead
        National Ambient Air Quality Standards [EPA Report]. Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2008g). Ohio area designations for the 2008 lead
        National Ambient Air Quality Standards [EPA Report]. Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2008h). Tennessee area designations for the 2008
        lead National Ambient Air Quality Standards [EPA Report]. Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for paniculate
        matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546

     U.S. EPA (U.S. Environmental Protection Agency). (2010).  Integrated science assessment for carbon
        monoxide [EPA Report]. (EPA/600/R-09/019F). Research Triangle Park, NC.
        http ://cfpub. epa. gov/ncea/cfm/recordisplav. cfm?deid=218686

     U.S. EPA (U.S. Environmental Protection Agency). (2011a). 2008 National Emissions Inventory data and
        documentation. Available online at http://www.epa.gov/ttn/chief/net/2008inventory.html (accessed
        February 17,2011).

     U.S. EPA (U.S. Environmental Protection Agency). (2011b). Emission factor supporting documentation
        for the final mercury and air toxics standards [EPA Report]. (EPA-454/R-11-012).  Research Triangle
        Park, NC. http://www.epa.gov/ttn/atw/utilitv/mats  efs casestudies  currentbaseei.pdf

     U.S. EPA (U.S. Environmental Protection Agency). (20111). List of designated reference and equivalent
        methods [EPA Report]. Research Triangle Park, NC.
        http://www.epa.gov/ttn/amtic/Files/ambient/criteria/reference-equivalent-methods-list.pdf

     UC (University of California). (1995). IMPROVE data guide.  Davis, CA: University of California Davis.

     USGS (U.S. Geological Survey). (1996). Mineral commodities summary: Lead. In Mineral commodity
        summaries 1996 (pp. 92-93). U.S. Department of the Interior.
        http://minerals.usgs.gov/minerals/pubs/commoditv/lead/lead mcs96.pdf

     USGS (U.S. Geological Survey). (2005). Historical statistics for mineral and material commodities in the
        United States. Available online at http://minerals.usgs.gov/ds/2005/140/ (accessed May 10, 2010).
November 2012                              3-250                   Draft - Do Not Cite or Quote

-------
     USGS (U.S. Geological Survey). (2006). Mineral commodities summary: Lead. In Mineral commodity
        summaries 2006 (pp. 96-97). U.S. Department of the Interior.
        http://minerals.usgs.gov/minerals/pubs/commoditv/lead/lead  mcs06.pdf

     USGS (U.S. Geological Survey). (2012). 2010 minerals yearbook: Lead [advance release]. Washington,
        DC: U.S. Department of the Interior.

     Uzu. G: Sobanska. S: Aliouane. Y; Pradere. P; Dumat. C. (2009). Study of lead phytoavailability for
        atmospheric industrial micronic and sub-micronic particles in relation with lead speciation. Environ
        Pollut 157: 1178-1185. http://dx.doi.Org/10.1016/i.envpol.2008.09.053

     van Herwijnen. R; Hutchings. TR; Al-Tabbaa. A; Moffat AJ; Johns. ML; Ouki. SK. (2007). Remediation
        of metal contaminated soil with mineral-amended composts. Environ Pollut 150: 347-354.
        http://dx.doi.0rg/10.1016/i.envpol.2007.01.023

     van Lith. SC: Alonso-Ramirez. V; Jensen. PA; Frandsen. FJ; Glarborg. P. (2006). Release to the gas phase
        of inorganic elements during wood combustion: Part 1:  Development and evaluation of quantification
        methods. Energy Fuels 20: 964-978. http://dx.doi.org/10.1021/ef050131r

     van Lith. SC: Jensen. PA; Frandsen. FJ: Glarborg. P. (2008). Release to the gas phase of inorganic
        elements during wood combustion: Part 2. Influence of fuel composition. Energy Fuels 22:
        1598-1609. http://dx.doi.org/10.1021/ef060613i

     Van Metre. PC: Mahler. BJ; Wilson. JT; Callender. E. (2006). Major and Trace Elements in 35 Lake and
        Reservoir Sediment Cores From Across the United States, 19942001. (Data Series 166). Reston, VA:
        U.S. Geological Survey. http://www.ntis.gov/search/product.aspx?ABBR=PB2006113440

     Vermette. SJ: Irvine. KN; Drake. JJ. (1991). Temporal variability of the elemental composition in urban
        street dust. Environ Monit Assess  18:  69-77. http://dx.doi.org/10.1007/BF00394478

     Verstraete. S: Van Meirvenne. M. (2008). A multi-stage sampling strategy for the delineation of soil
        pollution in a contaminated brownfield. Environ Pollut 154: 184-191.
        http://dx.doi.0rg/10.1016/i.envpol.2007.10.014

     Vogl. J: Heumann. KG. (1997). Determination of heavy metal complexes with humic substances by
        HPLC/ICP-MS coupling using on-line isotope dilution technique. Fresenius' Journal of Analytical
        Chemistry 359: 438-441. http://dx.doi.org/10.1007/s002160050606

     Voutsa. D; Samara.  C. (2002). Labile  and bioaccessible fractions of heavy metals in the airborne
        paniculate matter from urban and  industrial areas. Atmos Environ 36: 3583-3590.
        http://dx.doi.org/10.1016/81352-2310(02)00282-0

     Vucetich. JA; Outridge. PM; Peterson. RO; Eide. R; Isrenn. R. (2009). Mercury, lead and lead isotope
        ratios in the teeth of moose (Alces alces) from Isle Royale, US Upper Midwest, from 1952 to 2002. J
        Environ Monit 11: 1352-1359. http://dx.doi.org/10.1039/b903417j

     Wagner. J: Leith. D. (2001). Field tests of a passive aerosol sampler. J Aerosol Sci 32: 33-48.
        http://dx.doi.org/10.1016/S0021-8502(00)00055-0

     Waheed. A; Zhang. YX; Bao. LM; Cao. QC: Zhang. GL; Li. Y; Li. XL. (2010). Study of seasonal
        variation and source characteristic of PM10 of Shanghai urban atmosphere using PIXE. Journal of
        Radioanal Chem 283: 427-432. http://dx.doi.org/10.1007/sl0967-009-0390-2

     Wahlin. P; Berkowicz. R; Palmgren. F. (2006). Characterisation of traffic-generated paniculate matter in
        Copenhagen. Atmos Environ 40: 2151-2159. http://dx.doi.Org/10.1016/i.atmosenv.2005.ll.049

     Walder. AJ: Freedman. PA. (1992). Isotopic ratio measurement using a double focusing magnetic sector
        mass analyser with an inductively coupled plasma as an ion source. J Anal At Spectrom 7: 571-575.
        http://dx.doi.org/10.1039/JA9920700571

     Wang. G: Huang. L; Gao. S: Gao. S: Wang. L. (2002). Characterization of water-soluble species of PM10
        and PM2.5 aerosols in urban area in Nanjing, China. Atmos Environ 36: 1299-1307.
        http://dx.doi.org/10.1016/81352-2310(01)00550-7
November 2012                             3-251                   Draft - Do Not Cite or Quote

-------
     Wang. LJ; Wanjura. JD; Parnell. CB; Lacev. RE; Shaw. BW. (2005b). Performance characteristics of a
        low-volume PM10 sampler. Trans ASAE 48: 739-748.

     Wang. W: Liu. XD; Zhao. LW: Quo. DF; Tian. XD; Adams. F. (2006c). Effectiveness of leaded petrol
        phase-out in Tianjin, China based on the aerosol lead concentration and isotope abundance ratio. Sci
        Total Environ 364: 175-187. http://dx.doi.Org/10.1016/i.scitotenv.2005.07.002

     Wang. XL; Sato. T; Xing. BS. (2006d). Size distribution and anthropogenic sources apportionment of
        airborne trace metals in Kanazawa, Japan. Chemosphere 65: 2440-2448.
        http://dx.doi.0rg/10.1016/i.chemosphere.2006.04.050

     Wanjura. JD: Parnell. CB. Jr; Shaw. BW: Lacev. RE. (2005). Design and evaluation of a low-volume total
        suspended paniculate sampler. Trans ASAE 48: 1547-1552.

     Warnken. KW: Santschi. PH. (2009). Delivery of trace metals (Al, Fe, Mn, V, Co, Ni, Cu, Cd, Ag, Pb)
        from the Trinity River Watershed towards the ocean. Estuaries Coasts 32: 158-172.
        http://dx.doi.org/10.1007/sl2237-008-9088-x

     Watmough. SA; Dillon. PJ. (2007). Lead biogeochemistry in a central Ontario forested watershed.
        Biogeochemistry 84: 143-159. http://dx.doi.org/10.1007/sl0533-007-9110-6

     Watmough. SA; Hutchinson. TC: Dillon. PJ. (2004). Lead dynamics in the forest floor and mineral soil in
        south-central Ontari. Biogeochemistry 71: 43-68. http://dx.doi.org/10.1007/sl0533-004-7661-3

     Watson. JG: Chow. JC: Shah. JJ; Pace. TG. (1983). The effect of sampling inlets on the PM-10 and
        PM-15 to TSP concentration ratios. J Air Waste Manag Assoc 33: 114-119.

     Wedding. JB; McFarland. AR; Cermak. JE. (1977). Large particle collection characteristics of ambient
        aerosol samplers. Environ Sci Technol 11: 387-390. http://dx.doi.org/10.1021/es60127a005

     Wedepohl. KH. (1978). Handbook of geochemistry. In KH Wedepohl (Ed.). New York, NY: Springer.

     Weiss. AL; Caravanos. J: Blaise. MJ; Jaeger. RJ. (2006). Distribution of lead in urban roadway grit and its
        association with elevated steel structures. Chemosphere 65: 1762-1771.
        http://dx.doi.0rg/10.1016/i.chemosphere.2006.04.079

     Weitkamp. EA; Lipsky. EM; Pancras. PJ: Ondov. JM; Polidori. A; Turpin. BJ; Robinson. AL.  (2005). Fine
        particle emission profile for a large coke production facility based on highly time-resolved fence line
        measurements. Atmos Environ 39: 6719-6733. http://dx.doi.Org/10.1016/j.atmosenv.2005.06.028

     Westerlund. C: Viklander. M. (2006). Particles and associated metals in road runoff during snowmelt and
        rainfall. Sci Total Environ 362: 143-156. http://dx.doi.Org/10.1016/i.scitotenv.2005.06.031

     Widory. D. (2006). Lead isotopes decipher multiple origins within  single PM10 samples in the
        atmosphere of Paris. Isotopes Environ Health Stud 42: 97-105.
        http://dx.doi.org/10.1080/10256010500502736

     Wiersma. GB; Elvir. JA; Eckhoff. JD. (2007). Forest vegetation monitoring and foliar chemistry of red
        spruce and red maple at Acadia National Park in Maine. Environ Monit Assess 126: 27-37.
        http://dx.doi.org/10.1007/sl0661-006-9329-z

     Williams. AGB;  Scheckel. KG: Tolaymat. T; Impellitteri. CA. (2006). Mineralogy and characterization of
        arsenic, iron, and lead in a mine waste-derived fertilizer. Environ Sci Technol 40: 4874-4879.
        http://dx.doi.org/10.1021/es060853c

     Witt. M; Baker. AR; Jickells. TD. (2006). Atmospheric trace metals over the Atlantic and South Indian
        Oceans: Investigation of metal concentrations and lead isotope ratios in coastal and remote marine
        aerosols. Atmos Environ 40: 5435-5451. http://dx.doi.Org/10.1016/i.atmosenv.2006.04.041

     Wojas. B; Almquist. C. (2007). Mass concentrations and metals speciation of PM2.5, PM10, and total
        suspended solids in Oxford, Ohio and comparison with those from metropolitan sites in the Greater
        Cincinnati region. Atmos Environ 41: 9064-9078. http://dx.doi.Org/10.1016/i.atmosenv.2007.08.010
November 2012                              3-252                   Draft - Do Not Cite or Quote

-------
     Wong. CSC: Li. XD; Thornton. I. (2006). Urban environmental geochemistry of trace metals [Review].
        Environ Pollut 142: 1-16. http://dx.doi.Org/10.1016/i.envpol.2005.09.004

     Wu. J: Edwards. R; He. XE; Liu. Z; Kleinman. M. (2010). Spatial analysis of bioavailable soil lead
        concentrations in Los Angeles, California. Environ Res 110: 309-317.
        http://dx.doi.0rg/10.1016/i.envres.2010.02.004

     Wu. SC: Luo. YM; Cheung. KG: Wong. MH. (2006). Influence of bacteria on Pb and Zn speciation,
        mobility and bioavailability in soil: A laboratory study. Environ Pollut 144: 765-773.
        http://dx.doi.0rg/10.1016/i.envpol.2006.02.022

     Xie. L; Giammar. DE. (2007).  Equilibrium solubility and dissolution rate of the lead phosphate
        chloropyromorphite. Environ Sci Technol 41: 8050-8055. http://dx.doi.org/10.1021/es071517e

     Xie. RK; Seip. HM; Liu. L; Zhang. DS. (2009). Characterization of individual airborne particles in
        Taiyuan City, China. Air Qual Atmos Health 2: 123-131. http://dx.doi.org/10.1007/sll869-009-0039-x

     Yadav. S: Raiamani. V. (2006). Air quality and trace metal chemistry of different size fractions of aerosols
        in N-NW India - implications for source diversity. Atmos Environ 40: 698-712.
        http://dx.doi.0rg/10.1016/i.atmosenv.2005.10.005

     Yang. HH; Chiang. CF; Lee. WJ: Hwang. KP; Wu. EMY. (1999). Size distribution and dry deposition of
        road dust PAHs. Environ Int 25: 585-597. http://dx.doi.org/10.1016/S0160-4120(99)00036-7

     Yesilonis. ID; Pouvat. RV: Neerchal. NK. (2008). Spatial distribution of metals in soils in Baltimore,
        Maryland: Role of native parent material, proximity to major roads, housing age and screening
        guidelines. Environ Pollut  156: 723-731. http://dx.doi.Org/10.1016/i.envpol.2008.06.010

     Yi. SM; Totten. LA; Thota. S:  Yan. S: Offenberg. JH; Eisenreich. SJ: Granev. J: Holsen. TM. (2006).
        Atmospheric dry deposition of trace elements measured around the urban and industrially impacted
        NY-NJ harbor. Atmos Environ 40: 6626-6637. http://dx.doi.Org/10.1016/i.atmosenv.2006.05.062

     Ying. G: Sansalone. J. (2008).  Granulometric relationships for urban source area runoff as a function of
        hydrologic event classification and sedimentation.  Water Air Soil Pollut 193: 229-246.
        http://dx.doi.org/10.1007/sll270-008-9685-0

     Yu. CH; Fan. ZH; Meng. Q; Zhu. X; Korn. L; Bonanno. LJ. (2011). Spatial/temporal variations of
        elemental carbon, organic carbon, and trace elements in PM10 and the impact of land-use patterns on
        community air pollution in Paterson, NJ. J Air Waste Manag Assoc 61: 673-688.

     Zahran. S: Mielke. HW: Gonzales. CR; Powell. ET; Weiler. S. (2010). New Orleans before and after
        hurricanes Katrina/Rita: A quasi-experiment of the association between soil lead and children's blood
        lead. Environ Sci Technol 44: 4433-4440. http://dx.doi.org/10.1021/esl00572s

     Zak. K; Rohovec. J: Navratil. T. (2009). Fluxes of heavy metals from a highly polluted watershed during
        flood events: A case study of the Litavka River, Czech Republic. Water Air Soil Pollut 203: 343-358.
        htrp://dx.doi.org/10.1007/sll270-009-0017-9

     Zereini. F; Alt. F; Messerschmidt. J: Wiseman. C: Feldmann. I; Von Bohlen. A; Muller. J: Liebl. K;
        Puttmann. W (2005). Concentration and distribution of heavy metals in urban airborne paniculate
        matter in Frankfurt am Main, Germany. Environ Sci Technol 39: 2983-2989.
        http://dx.doi.org/10.1021/es040040t

     Zhang. LF;  Peng. SQ: Wang. S. (2009a). Decreased aortic contractile reaction to 5-hydroxytryptamine in
        rats with long-term hypertension induced by lead (Pb2+) exposure. Toxicol Lett 186: 78-83.
        http://dx.doi.0rg/10.1016/i.toxlet.2009.01.004

     Zheng. RQ; Li. CY. (2009). Effect of lead on survival, locomotion and sperm morphology of Asian
        earthworm, Pheretima guillelmi. J Environ Sci 21: 691-695.
        http://dx.doi.org/10.1016/sl001-0742(08)62325-6

     Zota. AR; Willis. R; Jim. R; Norris. GA; Shine. JP; Duvall. RM; Schaider. LA; Spengler. JD. (2009).
        Impact of mine waste on airborne respirable particulates in northeastern Oklahoma, United States. J
        Air Waste Manag Assoc 59: 1347-1357.  http://dx.doi.org/10.3155/1047-3289.59.ll.1347
November 2012                              3-253                  Draft - Do Not Cite or Quote

-------
     Zuna. M; Mihaljevic. M; Sebek. O; Ettler. V; Handlev. M; Navratil. T; Golias. V. (2011). Recent lead
        deposition trends in the Czech Republic as recorded by peat bogs and tree rings. Atmos Environ 45:
        4950-4958. http://dx.doi.Org/10.1016/i.atmosenv.2011.06.007
November 2012                              3-254                   Draft - Do Not Cite or Quote

-------
       4  EXPOSURE,  TOXICOKINETICS,  AND
          BIOMARKERS
      4.1       Exposure Assessment

 1                  The purpose of this section is to present recent studies that provide insight about human
 2                  exposure to Pb through various pathways. Pb is considered to be a multimedia
 3                  contaminant with multiple pathways of exposure. The relative importance of various
 4                  media in affecting Pb exposure changes with source strength and location, location and
 5                  time activity of the exposed individuals, behavior of the exposed individuals, and risk
 6                  factors such as age and socioeconomic factors (risk factors are discussed in detail in
 7                  Chapter_5). Blood Pb and bone Pb biomarkers (discussed in Sections 4.3. 4.4. and 4.5).
 8                  are often used to indicate composite Pb exposure resulting from multiple media and
 9                  pathways of exposure.

10                  The recent information provided here builds upon the conclusions of the 2006 Pb AQCD
11                  (U.S. EPA. 2006b). which found that air Pb concentrations and blood Pb levels have
12                  decreased substantially following the restrictions on Pb in on-road vehicle gasoline, Pb in
13                  household paints, the use of Pb solder, and reductions in industrial Pb emissions that have
14                  occurred since the late 1970s. Nevertheless, detectable quantities of Pb have still been
15                  observed to be bioaccessible in various media types. It was reported in the
16                  2006 Pb AQCD (U.S. EPA. 2006b) that airborne maximum quarterly Pb concentrations
17                  in the U.S. were in the range of 0.03-0.05 ug/m3 for non-source-oriented monitors for the
18                  years 2000-2004 and were 0.10-0.22 ug/m3 for source-oriented monitors during that time
19                  period, while blood Pb levels reached a median of 1.70 ug/dL among children (1-5 years
20                  of age) in 2001-2002. It was also observed that Pb exposures were associated with nearby
21                  industrial Pb sources, presence of Pb-based paint, and Pb deposited onto food in several
22                  of the studies described in the  2006 Pb AQCD.
      4.1.1        Pathways for Pb Exposure

23                  Pathways of Pb exposure are difficult to disentangle because Pb has multiple sources in
24                  the environment and passes through various environmental media. These issues are
25                  described in detail in Sections 3.2 and 3.3. Air-related pathways of Pb exposure are the
26                  focus of this ISA. Pb can be emitted to air, soil, or water and then cycle through any or all
27                  of these media. In addition to primary emission of particle-bound or gaseous Pb to the
28                  atmosphere, Pb can be resuspended to the air from soil or dust. Additionally, Pb-bearing
29                  PM can be deposited from the air to soil or water through wet and dry deposition. Air-

      November 2012                             4-1                   Draft - Do Not Cite or Quote

-------
 1                  related Pb exposures also include inhalation and ingestion of Pb-contaminated food,
 2                  water or other materials following atmospheric deposition of Pb; these exposures include
 3                  dust and soil via hand-to-mouth contact. In general, air-related pathways include those
 4                  pathways where Pb passes through ambient air on its path from a source to human
 5                  exposure. Some non-air-related exposures of Pb include ingestion of indoor Pb paint, Pb
 6                  in diet as a result of inadvertent additions during food processing, and Pb in drinking
 7                  water attributable to Pb in distribution systems, as well as other generally less prevalent
 8                  pathways.

 9                  The complicated nature of Pb exposure is illustrated in Figure 4-1. in which the Venn
10                  diagram depicts how Pb can cycle through multiple environmental media prior to human
11                  exposure. The "air/soil/water" arrows illustrate Pb exposures to plants, animals, and/or
12                  humans via contact with Pb-containing media. The exposures  are air-related if Pb passed
13                  through the air compartment. When animals consume plant material or water exposed to
14                  Pb that has at some point passed through the air compartment, and when human diet
15                  includes animals, plants or drinking water exposed to Pb that has passed through the air
16                  compartment, these are also considered air-related Pb exposures. As a result of the
17                  multitude of possible air-related exposure scenarios and the related difficulty of
18                  constructing Pb exposure histories, most studies of Pb exposure through air, water,  and
19                  soil can be informative to this review. Figure 4-1 also illustrates other exposures, such as
20                  occupational exposures, contact with consumer goods in which Pb has been used, or
21                  ingestion of Pb in drinking water conveyed through Pb pipes.  Most Pb biomarker studies
22                  do not indicate speciation or isotopic signature, and so exposures that are not related to
23                  Pb in ambient air are also reviewed in this section because they can contribute to Pb body
24                  burden. Many of the studies presented in the subsequent material focus on observations
25                  of Pb exposure via one medium: air, water, soil and dust, diet, or occupation.
      November 2012                              4-2                   Draft - Do Not Cite or Quote

-------
                                            Newly Emitted Pb
                                          Historically Emitted Pb
                            OOTDOORSOIL
                                   NDDUST
                  NATURAL WATERS
                  AND SEDIMENTS
Non-air Pb
  eleases

•— --.
AIR
SOIL
WATER
1 	 — - -

**• —
PLANT
EXPOSURE
-+
^^.
AIR
SOIL
WATER
--. — — -

**• — — -— ,
r ANIMAL
t EXPOSURE]]"
AIR
SOIL
WATER
•——_—--

.-•— •
HUMAN
^EXPOSURE
                                                             ( COSMETICS  )
                                                                 OYS et
      Note: The Venn diagram is used to illustrate the passage of Pb through multiple environmental media compartments through which
      exposure can occur.

      Figure 4-1      Conceptual model of multimedia Pb exposure.

 1                  The relative importance of different sources or pathways of potential exposure to Pb in
 2                  the environment is often difficult to discern. Individual factors such as home
 3                  environment, location, and risk factors (described in more detail in Chapter_5) may
 4                  influence exposures. The National Human Exposure Assessment Survey (NHEXAS)
 5                  study sampled Pb, as well  as other pollutants and VOCs, in multiple exposure media from
 6                  subjects across six states in EPA Region 5 (Illinois, Indiana, Michigan, Minnesota, Ohio,
 7                  and Wisconsin) (Clayton et al.. 1999) as well  as in Arizona (O'Rourke et al.. 1999) and
 8                  Maryland (Egeghy et al., 2005). Results from NHEXAS indicate that personal exposure
 9                  concentrations of Pb are higher than indoor or outdoor concentrations of Pb, perhaps
10                  suggesting a personal cloud effect; see Table 4-1. Pb levels in windowsill dust were
11                  higher than Pb levels in surface dust collected from other surfaces. Clayton et al. (1999)
12                  suggested that higher windowsill levels could be attributed to the presence of Pb-based
13                  paint and/or to accumulation of infiltrated outdoor Pb-bearing PM. Pb levels in food were
14                  higher than in beverages, and Pb levels in standing tap water (also referred to as "first
15                  flush" or "first draw") were higher than Pb levels obtained after allowing water to run for
16                  three minutes to flush out pipes. Layton and Beamer (2009) estimated that 34-66% of Pb
17                  in floor dust was tracked in from outdoors and originated as ambient air Pb, based on
     November 2012
4-3
Draft - Do Not Cite or Quote

-------
                  1992 levels in Sacramento; in 1992, phase-out of Pb usage in gasoline was near complete,

                  but industrial emissions were still higher than current levels; see Section 3.2.
Table 4-1 Estimates of Pb measurements for EPA Region 5 from the
NHEXAS study.
Medium3
Personal air
(ng/m3)d
Indoor air
(ng/m3)d
Outdoor air
(ng/m3)d
Surface dust
(ng/cm2)
Surface dust
(mg/kg)
Window sill dust
(ng/cm2)
Window sill dust
(mg/kg)
Standing tap water
(ug/L)
Flushed tap water
(ug/L)
Solid food
(ug/kg)
Beverages
(ug/kg)
Food + Beverages
(ug/kg)
Food intake
(ug/day)
Beverage intake
(ug/day)
Food + Beverage intake
(ug/day)
Blood
(ug/dL)
N
167
213
87
245
244
239
239
444
443
159
160
156
159
160
156
165
Percentage above
LODb (CLs)c
81.6
(71 .3; 92.0)
49.8
(37.2; 62.3)
73.8
(56.3; 91 .3)
92.1
(87.4; 96.8)
92.1
(87.4; 96.8)
95.8
(92.5; 99.0)
95.8
(92.5; 99.0)
98.8
(97.6; 100.0)
78.7
(70.7; 86.7)
100.0
(100.0; 100.0)
91.5
(85.2; 97.8)
100.0
(100.0; 100.0)
100.0
(100.0; 100.0)
91.5
(85.2; 97.8)
100.0
(100.0; 100.0)
94.2
(88.2; 100.0)
Mean (CLs)c
26.83
(17.60; 36.06)
14.37
(8.76; 19.98)
11.32
(8.16; 14.47)
514.43
(-336.6; 1365.5)
463.09
(188.1 5; 738.04)
1,822.6
(481.49; 3,163.6)
954.07
(506.70; 1 ,401 .4)
3.92
(3.06; 4.79)
0.84
(0.60; 1.07)
10.47
(6.87; 14.07)
1.42
(1.13; 1.72)
4.48
(2.94; 6.02)
7.96
(4.25; 1 1 .68)
2.15
(1.66; 2.64)
10.20
(6.52; 13.89)
2.18
(1.78; 2.58)
50th (CLs)c
13.01
(11.13; 18.13)
6.61
(4.99; 8. 15)
8.50
(7. 14; 10.35)
5.96
(3.37; 10.94)
120.12
(83.85; 160.59)
16.76
(10.44; 39.41)
191.43
(140.48; 256.65)
1.92
(1 .49; 2.74)
0.33
(0.23; 0.49)
6.88
(6.44; 8.04)
0.99
(0.84; 1.21)
3.10
(2.66; 3.52)
4.56
(3.68; 5.36)
1.41
(1.18; 1.60)
6.40
(5.21 ; 7.78)
1.61
(1.41; 2. 17)
90th (CLs)c
57.20
(31. 18; 85.10)
18.50
(12.69; 30.31)
20.36
(12.60; 34.91)
84.23
(26.52; 442.63)
698.92
(411.84; 1,062.8)
439.73
(106.34; 4,436.2)
1 ,842.8
(1,1 51 .3; 2,782.5)
9.34
(7.87; 12.35)
1.85
(1.21; 3.04)
14.88
(10.78; 19.08)
2.47
(2.06; 3.59)
6.37
(4.89; 8.00)
12.61
(9.27; 16.38)
4.45
(3.1 5; 5.65)
16.05
(13.31; 18.85)
4.05
(3.24; 5.18)
Note: EPA Region 5 includes six states: Illinois, Indiana, Ohio, Michigan, Minnesota, and Wisconsin. Participants were enrolled
using a stratified, four-stage probability sampling design, and submitted questionnaire and physical measurements data. Summary
statistics (percentage above limit of detection (LOD), mean, median, 90th percentile) were computed using weighted sample data
analysis. The estimates apply to the larger Region 5 target population (all non-institutionalized residents residing in households).
"Estimates for indoor air, outdoor air, dust media, and water media apply to the target population of Region 5 households; estimates
for other media apply to the target population of Region 5 residents.
bPercentage of the target population of residents (or households) estimated to have Pb levels above limit of detection (LOD).
°The lower and upper bounds of the 95% confidence limits (CL) are provided.
dPM50.
Source: Reprinted with permission of Nature Publishing Group, Clayton et al. (1999)
November 2012
4-4
Draft - Do Not Cite or Quote

-------
                    4.1.1.1      Particle Size Distributions for Airborne-Pb, Dust-Pb, and
                                 Soil-Pb

 1                  The size distribution of ingestible dust particles differs from the size distribution of
 2                  inhalable ambient air Pb particles and therefore cannot be directly compared. The
 3                  inhalability of airborne PM is a gradually decreasing function of particle size. Inhalability
 4                  criteria established from experimental data, obtained at wind speeds of 1-8
 5                  meters/second, describe PM inhalability of 77% for particles <10 jam (dae, aerodynamic
 6                  diameter). Inhalability of particles ranging in size from 40 to 100 (im dae is 50%; above
 7                  100 (im, inhalability data are lacking (Soderholm. 1989: ACGIH. 1985). The particles
 8                  that are not inhaled may settle to surfaces, making them available for subsequent
 9                  ingestion. The size distribution of soil and house dust particles tends to be much larger
10                  than airborne PM. Que Hee et al. (1985) and U.S. EPA (1990b) observed that 50% or
11                  more of the mass of house dust tends to be comprised of particles smaller than 150 pirn.
12                  Gulson et al. (1995b) observed that the mode of the Pb house dust size distribution was in
13                  the 38-53  nm range; they did not report the overall house  dust size distribution. Given the
14                  house dust Pb size distributions documented, dust Pb brought into homes with foot traffic
15                  may be aerosolized but is likely to stay airborne for only a few seconds, since particles
16                  larger than PM2 5 tend to settle from the air quickly; see Section 3.3.1.3 and Section
17                  4.1.3.1. Siciliano et al. (2009) observed different size distributions for different types of
18                  soils: agricultural sites had median soil Pb of 34 nm, and brownfields had median soil Pb
19                  of 105 nm. These observations of larger particle sizes for  soil and dust Pb support the
20                  notion that exposure to Pb in dusts and soils would occur by ingestion rather than
21                  inhalation following resuspension.

22                  The main pathway for Pb ingestion by children is by hand to mouth contact (Lanphear et
23                  al.. 1998). In a playground environment in London, U.K., Duggan et al. (1985) reported
24                  that hand to mouth transfer was effectively limited  to particles smaller than 10 urn, even
25                  when the soil itself exhibited a much larger particle size distribution. More recently,
26                  Yamamoto et al. (2006) reported for a cohort of children in Kanagawa Prefecture, Japan
27                  (greater Tokyo area) that the mode of size distributions of particles adhering to children's
28                  hands was 39 ± 26 (im, with the upper tail ranging  from 200-300 (im. Kissel et al. (1996)
29                  measured three size fractions of soil adhered to a hand via a hand press: < 150 pirn,
30                  150-250 nm, and > 250 pirn and observed that, when soil was dry (<2% moisture
31                  content), 43%-69% of the soil was in the smallest fraction. When the moisture content
32                  was higher than 2%, 28-81% of the adhered soil was in larger than 250 pirn. Percentage
33                  and mass adhered per area (mg/cm2) depended on soil type, with wet sand and loamy
34                  sand adhering more to hands than sandy loam or silt loam. For dry soil, silt loam mass
35                  produced the largest adherence in terms of mass per area.  Differences among the size
      November 2012                             4-5                   Draft - Do Not Cite or Quote

-------
 1                   distribution results may be related to differences in the soil type, soil moisture levels
 2                   between the locations, and/or to differences between the analytical methods used to
 3                   measure size distribution; Duggan and Inskip (1985) used optical microscopy of the dust
 4                   wipes, while Yamamoto et al. (2006) used a laser scattering device measuring sampled
 5                   particles suspended in an aqueous solution. Siciliano et al. (2009) regressed adhered
 6                   average soil size on hands bulk soil particle size and found a log-log relationship with
 7                   (3 = 0.66 using both brownfield and agricultural soils; the proportion of soil adhered
 8                   depended on organic content.

 9                   Several studies have found that Pb is enriched in the smaller fractions of the soil or
10                   house-dust size distribution. Davies and White (1981) observed that enrichment
11                   decreased linearly with increasing dust size bin, with dust particles smaller than 64 nm
12                   having a Pb concentration of 76.1 mg/kg and particles in the 1,000-2,000 urn size range
13                   having a Pb concentration of 16.4 mg/kg. Sheets and Bergquist (1999) also found that Pb
14                   content decreased with increasing particle size.  More recently, Ljung et al.  (2006)
15                   investigated childhood exposures to trace metals on playgrounds in Uppsala,  Sweden and
16                   observed that the Pb content in soil in the <50 pirn size fraction was 1.5 times higher than
17                   that in the >4 mm or 50-100 pirn size fractions. Sheppard et al. (1995) measured
18                   enrichment in different types of soils (sand and clay) and found that enrichment was
19                   substantially higher in the sand.

20                   Studies focusing on particle size distributions of house dust adhered to the hands are
21                   lacking. Ingestion of house dust has been reported to be the major source of Pb  intake
22                   during  early childhood (Lanphear et al.. 2002). If a similar particle size distribution holds
23                   for household dust, then ingestion of indoor Pb  of atmospheric origin could also be
24                   strongly dependent on dust particle size. Therefore, larger particles of atmospheric origin,
25                   which may not be considered relevant for exposure by inhalation exposure, are  still
26                   relevant for Pb exposure by ingestion. However, no studies in the literature have
27                   presented information on the relative contributions of Pb from different PM size fractions
28                   to blood Pb concentrations.

29                   It should be noted that different measurement techniques are used for different
30                   environmental media. For example, ambient air Pb-PM size distribution is measured by
31                   one of the non-FRM instruments such as a MOUDI, described in Section 3.4. and its
32                   measurement is subject to errors specific to the technique. Dust and soil size distribution
33                   are typically measured with graduated sieves, and errors associated with these methods
34                   occur more often in the smaller size fractions that are subject to agglomeration and
35                   clogging if the particle shape is nonspherical.
      November 2012                              4-6                   Draft - Do Not Cite or Quote

-------
                    4.1.1.2      Estimating Pb Exposure in the Integrated Exposure
                                 Uptake Biokinetic (IEUBK) Model

 1                  Several studies have used a combination of measured values and default model values to
 2                  represent exposures and determine their relative contributions to blood Pb. For example,
 3                  Cornells et al. (2006) used the Integrated Exposure Uptake Biokinetic model (IEUBK),
 4                  described in detail in the 2006 Pb AQCD (U.S. EPA. 2006b) to model children's
 5                  exposures to Pb emissions from a non-ferrous smelter in Hoboken, Belgium. In deriving
 6                  the model input (annual averages) for ambient air Pb concentration, as well as soil and
 7                  dust, they employed weighting based on children's time spent in different locations in the
 8                  study area and air, soil and indoor dust measurements in those areas. In their results for
 9                  the area of the smelter, the ingestion of dust and soil pathways accounted for more than
10                  70% of the exposure, while the inhalation pathway accounted for less than 2%. Similarly,
11                  Carrizales et al. (2006) analyzed exposures to children living near a copper (Cu) smelter
12                  in San Luis Potosi, Mexico. They employed the IEUBK default options for assignment of
13                  Pb dust concentration as 70% of the soil Pb concentration, while air Pb concentration was
14                  assigned based on measurements by the Mexican government. Based on these
15                  assumptions, they attributed 87% of blood Pb to soil and dust exposure. These studies did
16                  not estimate the air Pb contribution to the soil/dust Pb concentrations and consequently
17                  did not estimate the portion of the ingestion pathway that derives from ambient air Pb.

18                  Appendix I of the 2007 Pb Risk Assessment (U.S. EPA. 2007f) provides estimates of the
19                  contribution of various pathways to the blood Pb of children simulated in several case
20                  studies. Simulations provided estimates of contributions from outdoor ambient air Pb by
21                  inhalation and by ingestion of indoor dust, including the fraction of indoor dust Pb
22                  associated with recent penetration of ambient air Pb into the residence. Although ambient
23                  air Pb may also contribute to Pb ingestion through other pathways (i.e., diet, soil), data
24                  and tools to support a simulation of the linkage between air Pb concentrations and
25                  concentrations in other media were limited. Accordingly, Pb concentrations pertaining to
26                  other pathways (e.g., diet, outdoor soil, the component of indoor dust Pb other than that
27                  derived from Pb recently in ambient air) were held constant across the different air
28                  quality scenarios simulated. Table 4-2 provides estimates for the General Urban Case
29                  Study in the 2007 Pb Risk Assessment (U.S. EPA. 2007f). The General Urban Case
30                  Study, unlike the various location-specific case studies, was not based on any specific
31                  urban location and reflected several simplifying assumptions including uniform ambient
32                  air Pb levels across the simulated, hypothetical study area and a uniform study
33                  population.
      November 2012                              4-7                   Draft - Do Not Cite or Quote

-------
      Table 4-2    Predicted concurrent blood Pb levels and source contributions for
                     children in  their seventh year of life.

                                                        Pathway Contribution (%)
Air Pba
(ug/m3)
0.05
0.149
0.2
0.87 h
Median
Blood Pbb
(ug/dL)
1.7(5.7)'
1.9(6.5)
2.0 (6.9)
2.1 (7.2)

Diet0
32
28
26
25
Ingestion
Outdoor
Soil/Dust
44
38
36
33
Indoor
Otherd
11
6
5
4

Dust
Pb recently in
aire
12.6
28.3
32.7
37.2
Inhalation
Ambient
Air
0.1
0.5
0.7
0.9
      "Concentrations are maximum calendar quarter averages of Pb in TSP with exception of 0.05 ug/m which is a maximum
      monthly average
      ""Average of blood Pb concentrations at 75 and 81 months, assuming exposure concentrations were constant through
      7 years of life
      Includes food and drinking water
      Includes indoor dust with Pb contributions from sources other than Pb recently in the air (e.g., indoor paint, outdoor
      soil/dust, and additional sources including historical air Pb)
      Includes contributions associated with outdoor ambient air Pb from ingestion of indoor dust predicted to be associated
      with outdoor ambient Pb levels
      Values in parentheses are the 95th percentile blood Pb for a geometric standard deviation of 2.1
      9Mean of the maximum quarterly average concentrations of Pb in TSP (for period 2003 to 2005) among monitor locations
      in urban areas having more than one million residents
      h95th percentile of the maximum quarterly average concentration of Pb in TSP (for period 2003 to 2005) among monitor
      locations in urban areas having more than one million residents

      Source: Based on General Urban Case Study (Hybrid Dust Model) in Appendix I, 2007 Pb Risk Assessment (U.S. EPA,
      2007 f}.
      4.1.2        Environmental Exposure Assessment Methodologies

 1                   A number of monitoring and modeling techniques have been employed for exposure
 2                   assessment. These are detailed in either the 2006 Pb AQCD (U.S. EPA. 2006b) or in the
 3                   subsequent Risk and Exposure Assessment performed as part of the same NAAQS
 4                   review (U.S. EPA. 2007g). Some of these methods are briefly described here to provide a
 5                   context for the exposure studies described in Section 4.1.3. Blood Pb sampling is
 6                   described in detail in Section 4.3.2.

 7                   Data collection to assess Pb exposure pathways may involve air, soil, and dust samples.
 8                   Methods used for digesting air Pb samples are described in Section 3.4. as are ambient air
 9                   Pb monitoring techniques. Factors affecting collection of ambient air Pb samples are
10                   described in detail in Section 3.4. For the monitors in the FRM network, the primary role
11                   is compliance assessment. Accordingly, this network includes monitors in locations near
12                   sources of air Pb emissions which are expected to or have been shown to contribute to
13                   ambient air Pb concentrations in excess of the Pb NAAQS. In such locations, Pb may be
      November 2012
4-8
Draft - Do Not Cite or Quote

-------
 1                  associated with relatively larger size particles, contributing to air Pb concentration
 2                  gradients with distance from the source and greater deposition in the near-source
 3                  locations. The FRM network also includes non-source-oriented monitors for which the
 4                  main objective is to gather information on neighborhood-scale Pb concentrations that are
 5                  typical in urban areas so to better understand ambient air-related Pb exposures for
 6                  populations in these areas. This part of the Pb NAAQS network, was required to be
 7                  operational as of December  27, 2011. These monitor locations are distributed across a
 8                  broad geographic area, representing approximately 63 large urban areas which contain
 9                  approximately half of the total U.S. population (based on recently published 2010 Census
10                  Bureau data). In lieu of more detailed analysis of population proximity for these newly
11                  established monitors, population counts were calculated near previously existing
12                  monitors for which data are  presented in Section 3.5. For the monitors in that limited
13                  dataset, among the total population of 311,127,619 people in the 2010 Census (ESRI.
14                  2011). 181,100 (0.06%) lived within 1 km of a source-oriented monitor, while 918,351
15                  (0.30%) lived within 1 km of a non-source-oriented monitor.

16                  Dust sampling has not changed drastically since it was first proposed by Sayre et al.
17                  (1974). in which a disposable paper towel was soaked in 20% denatured alcohol and
18                  1:750 benzalkonium chloride and then used to wipe a 1 ft2 sampling area in a systematic
19                  fashion. Que Hee et al. (1985) and Sterling et al. (1999) compared wipe testing with
20                  vacuum methods. Sampling efficiency for the first attempt varied between 53-76% with
21                  vacuum pump flow rate and tube type and was 52% for the wipe method for the Que Hee
22                  et al. (1985) study, with 100% efficiency after five consecutive samples were obtained.
23                  Sterling et al. (1999) observed that two of three vacuuming methods had significantly
24                  higher geometric mean collection (vacuum 1: 94.3 (ig/ft2; vacuum 2: 23.5 (ig/ft2)
25                  compared with dust wipes (5.6 (ig/ft2).

26                  Models may also be used in exposure assessment.  For example, two dispersion models,
27                  the American Meteorological Society/Environmental Protection Agency Regulatory
28                  Model (AERMOD), and Industrial Source Complex-Plume Rise Model Enhancements
29                  (ISC-PRIME) were employed to model dispersion of Pb emissions from specific
30                  industrial facilities (Cimorelli et al.. 2005; Perry etal.  2005; EPRI. 1997). and to
31                  estimate ambient air Pb concentrations at some of the case studies included in the 2007
32                  Risk and Exposure Assessment (U.S. EPA. 2007g). These models assume plume
33                  dispersion follows a Gaussian  distribution from a point source. For the two point source
34                  case studies included in the 2007 risk assessment, the plume models were used to track
35                  emissions to ambient air near homes located within a few miles of emitting facilities.
36                  However, dispersion models can also be used to track long distance transport of Pb
37                  emissions, as performed by Krell and Roeckner (1988) to model the dispersion and
38                  deposition of Pb and Cd from European nations into the North Sea.
      November 2012                             4-9                  Draft - Do Not Cite or Quote

-------
 1                  Several models estimate blood Pb levels resulting from estimated exposure to Pb in
 2                  environmental media. These models, which are described in detail in the 2006 Pb AQCD
 3                  (U.S. EPA. 2006b) include the IEUBK model, and the EPA All Ages Lead Model
 4                  (AALM), which combines and expands the thorough exposure and absorption modules of
 5                  the IEUBK model with the comprehensive biokinetic model of Leggett (1993). As of the
 6                  writing of this assessment, the AALM is still in development.

 7                  The Stochastic Human Exposure and Dose (SHEDS) and NORMTOX models also are
 8                  capable of modeling metals exposures through various routes including inhalation,
 9                  ingestion, and dermal exposure (Loos etal. 2010; Burke et al.. 2002). Pb exposure
10                  modeling can also be accomplished using the  Modeling Environment for Total Risk
11                  (MENTOR) framework, in which airborne Pb levels could be modeled using AQS,
12                  dispersion modeling, or chemical transport modeling, while human exposure is modeled
13                  with SHEDS or a similar exposure model (Georgopoulos and Lioy. 2006). Additionally,
14                  housing data and time-activity data from the Consolidated Human Activity Database
15                  (CHAD) are incorporated into MENTOR to develop refined estimates of Pb exposure  and
16                  tissue burden. However, a literature search did not produce any Pb exposure studies using
17                  the SHEDS, NORMTOX, or MENTOR modeling systems. In general, these models take
18                  input for several environmental Pb exposure media including soil, dust, food and water,
19                  outdoor air, and indoor air. The models are designed to evaluate different exposure
20                  scenarios based on specification of particular  conditions.
      4.1.3       Exposure Studies
                    4.1.3.1      Airborne Pb Exposure

21                  Limited personal exposure monitoring data for airborne Pb were available for the
22                  2006 Pb AQCD (U.S. EPA. 2006b). As described above, the NHEXAS study showed
23                  personal air Pb concentrations to be significantly higher than indoor or outdoor air Pb
24                  concentrations (Clayton et al.. 1999). Indoor air Pb concentration was moderately
25                  correlated with floor dust and residential yard soil Pb concentration (Rabinowitz et al..
26                  1985). Egeghy et al. (2005) performed multivariate fixed effects analysis of the
27                  NHEXAS-Maryland data and found that Pb levels measured in indoor air were
28                  significantly associated with log-transformed outdoor air Pb levels, ambient temperature,
29                  number of hours in which windows were open, whether homes were built before 1950,
30                  and frequency of fireplace usage (Table 4-3).
     November 2012                            4-10                 Draft - Do Not Cite or Quote

-------
Table 4-3     Estimates of fixed effects multivariate modeling of Pb levels
                measured during the NHEXAS-MD study.
Pb in Indoor Air
Fixed Effect pa p-value
Intercept -0.50 0.0051
Pb in Dust Dermal Pb Blood Pb
pa p-value pa p-value pa p-value
6.22 <0.0001 6.23 <0.0001 0.02 0.91
Outdoor Pb concentration11 0.51 <0.0001
Average weekly temperature (°F) 0.01 0.046
Open window periods (hr) 0.01 0.035
-0.03 0.0082
House pets (yes) -0.15 0.078
Air filter use (yes) -0.28 0.087
Home age (<1 950) 0.25 0.025
Fireplace (frequency of use) 0.11 0.045
Pb concentration in soilb
Interior Pb paint chipping/peeling (yes)
Cement at primary entryway (yes)
Indoor pesticide usage last 6 mo (yes)
Electrostatic air filter usage (yes)
Sex of participants (male)
Ethnic minority participants (yes)
Washing hands after lawn mowing (no)
Gasoline power- equipment usage (yes)
Bathing or showering activities (yes)
Dust level indoors (scale: 1-3)
Residing near commercial areas (yes)
Age of participants (yr)
Number cigarettes smoked (count)
Burning wood or trash (days)
Showering frequency (avg # days)
Work outside home (yes)
Health status (good)
Adherence to high fiber diet (yes)
Gas or charcoal grill usage (yes)
-0.12 0.088
0.96 0.029
0.46 0.0054
0.27 0.037
0.43 0.091
1 .97 0.0064
-0.78 0.0003
-0.91 0.062
0.41 0.0012 0.43 <0.0001
0.41 0.0063
1.04 0.0010
0.61 0.0072
-0.43 0.019
0.22 0.019
0.32 0.0087
0.02 <0.0001
0.03 <0.0001
0.58 0.0099
-0.29 0.0064
-0.26 <0.0001
0.23 0.0009
-0.15 0.040
-0.17 0.0002
 Estimates of fixed effects in final multiple regression analysis models for Pb in the Maryland investigation data in the National Human Exposure Assessment Survey
(NHEXAS-MD).
bLog transform

Source: Reprinted with permission of Nature Publishing Group, Egeghy et al. (20051.
November 2012
4-11
Draft - Do Not Cite or Quote

-------
 1                  Some recent studies have shown that the ratio of indoor to outdoor Pb-PM varies from
 2                  site to site depending on factors including infiltration, indoor and outdoor Pb sources, and
 3                  meteorology. Adgate et al. (2007) measured the concentrations of several trace elements
 4                  in personal, indoor, and outdoor air samples of PM2 5 and found that average personal
 5                  Pb-PM2 5 concentration was roughly three times higher than outdoor air Pb-PM2 5
 6                  concentration and two times higher than indoor Pb-PM2 5 concentration (Table 4-4).
 7                  Another study of indoor and outdoor air concentrations of Pb was carried out by Molnar
 8                  et al. (2007). PM2 5 trace element concentrations were determined in homes, preschools
 9                  and schools in Stockholm, Sweden. In all sampled locations, Pb-PM2 5 concentrations
10                  were higher in the outdoor environment than in the proximal indoor environment. The
11                  indoor/outdoor ratios for Pb-PM2 5 suggest an outdoor Pb-PM2 5 net infiltration of -0.6 for
12                  these buildings. Outdoor air Pb concentrations did not differ between the central and
13                  more rural locations. Indoor air Pb concentrations were higher in spring than in winter,
14                  which the authors attributed to greater resuspension of elements that had accumulated in
15                  road dust over the winter period and increased roadwear on days with dry surfaces. Pekey
16                  et al. (2010) measured indoor and outdoor trace element composition of PM2 5 and PMi0
17                  in Kocaeli, an industrial region of Turkey, and found that average airborne Pb
18                  concentrations were higher outdoors than indoors for both PM2 5 and PMi0 during
19                  summer and for PMi0 during winter, but that indoor Pb concentration was higher than
20                  outdoor Pb concentration for PM2 5 during winter. The indoor-to-outdoor ratio of airborne
21                  Pb varied by environment; it tended to be less than one, but the ratio varied from one
22                  microenvironment to another. In a pilot study in Windsor, Ontario, Rasmussen et al.
23                  (2007) observed that the concentration of Pb in PM2 5 from a personal exposure sample
24                  was roughly 40% higher than the concentration of Pb in outdoor PM2 5 and  150% higher
25                  than Pb in indoor PM2 5. The three studies that included personal samples recorded
26                  measurements that were consistently higher than indoor or outdoor levels, and outdoor
27                  concentrations were higher than indoor concentrations.

28                  Domestic wood burning is a potential source of Pb compounds (Section 3.2.2.5). Alves et
29                  al. (2011) measured trace metals in woodstove and fireplace emissions and found that
30                  PM25 contained Pb, with concentrations from wood burning ranging from 3.3-12.2 (ig/g
31                  and 2.89-30.3 for woodstoves and fireplaces, respectively. When burning briquettes, the
32                  PM2 5 measurements showed Pb enrichment above all other metal elements other than
33                  potassium (woodstove: 1361 (ig/g; fireplace: 616 (ig/g). Molnar et al. (2005) measured
34                  trace element concentration in indoor and personal exposure PM2 5 samples for homes in
35                  which wood is burned and in a reference group where no wood burning occurs in the
36                  home. For both indoor and personal samples, Molnar et al. (2005) observed that Pb
37                  concentrations were higher for the wood burning group and nearly statistically significant
38                  for the personal exposure samples (indoor concentration:  6.0 (ig/m3 versus 4.3 (ig/m3,
39                  p = 0.26; personal exposure: 4.6 (ig/m3 versus 3.0 (ig/m3, p = 0.06).

      November 2012                             4-12                  Draft - Do Not Cite or Quote

-------
 1                   Indoor activity has been associated with resuspension of settled dust, which could cause
 2                   airborne contact with particle-bound Pb. Qian et al. (2008) estimated a PMi0 resuspension
 3                   rate of 1.4xlO"4/hr for one person walking across a carpeted floor. Measurements of
 4                   submicron particles illustrated a roughly two-fold increase of airborne particle
 5                   concentration for particles smaller than 1.8 (im for activity versus low activity periods,
 6                   with maximum concentrations reaching 4-11 times the maximum value during low
 7                   activity periods. For PM10, average concentration was 2.5  times higher than background
 8                   levels during activity periods, while peak concentration was 4.5 times higher. Qian and
 9                   Ferro (2008) observed that resuspension rates depend on particle size, floor material, and
10                   ventilation position.  Increases in walking speed and weight of the walker did not
11                   consistently produce increases in resuspension. 5-10 (im particles produced a higher
12                   resuspension rate compared with smaller particles. Newly carpeted areas produced
13                   significantly higher resuspension rates than vinyl floors. Zhang et al. (2008) modeled and
14                   conducted experiments of particle dispersion from walking and observed that human
15                   activity did affect resuspension. They found that larger particles were more readily
16                   detached from the carpet by walking motion, but that smaller particles are more  easily
17                   resuspended once detached. Hunt and Johnson (2012) studied the duration and spatial
18                   extent of resuspension of 0.3-5.0 (im particles following walking by a soiled shoe.
19                   0.3-0.5 (im particle concentration remained increased over atime period of 23 min, while
20                   1-5  (im particles declined in concentration over the same time period. Experiments and
21                   computational fluid dynamics simulations by Eisner et al.  (2010) for a mechanical foot
22                   moving on carpeting suggested that the rotating motion of the moving foot on the carpet
23                   induced rotating air movement beneath the foot that re-entrained the particles.

24                   Several of the studies can be used to develop an understanding of how personal exposure
25                   to PM-bound Pb varies with other exposures. Molnar et al. (2007) reported Spearman
26                   correlations of Pb with PM25 and NO2 in three outdoor microenvironments (residence,
27                   school, and preschool) and found that Pb and other trace metals were generally well
28                   correlated with PM2 5 (r = 0.72-0.85), but Pb was only statistically significantly correlated
29                   with NO2 in one of the three outdoor microenvironments (r = 0.24-0.75). Pb was
30                   attributed by Molnar et al. (2007) to long range transport.  Table 3-29 illustrates that Pb
31                   concentrations in the four studies (summarized in the Chapter 3 Appendix [Section 3.81)
32                   are typically well below the level of the NAAQS. The higher personal air concentratoins
33                   occurred in a heavily industrialized area of Kocaeli, Turkey with an incinerator and
34                   several industrial facilities including metal processing, cement, petroleum refining, and
35                   agriculture processing. Otherwise, concentrations were all between 0.002 and
36                   0.006  ug/m3. The proportion of Pb compared with other trace metals varied with location
37                   and component. It was typically several times lower than S as well as crustal elements
38                   such as Ca2+ and Fe. In the industrial area of Kocaeli, Pb comprised a greater proportion
39                   of the PM compared with other areas.

      November 2012                             4-13                  Draft - Do Not Cite or Quote

-------
1
2
o
J
4
5
6
7
Table 4-4
Study
Clayton et al.
(1999)
Adgate et al.
(2007)
Molnar et al.
(2007)
Tovalin-
Ahumada et al.
(2007)
Pekey et al.
(2010)
Rasmussen et
al. (2007)

Comparison of personal, indoor, and outdoor Pb-PM measurements
from several studies.
Location
IL, IN, Ml, MN,
OH, Wl
Minneapolis-
St. Paul, MN
Stockholm,
Sweden
Mexico City,
Mexico
Puebla, Mexico
Kocaeli, Turkey
Windsor,
Ontario,
Canada

4.1.3.2
The 2006 Pb
Pb Metric
Med. Pb-PMso
(ng/m3)
Avg. Pb-PM25
(ng/m3)
Avg. Pb-PM25
(ng/m3)
Med. Pb-PM25
(ng/m3)
Med. Pb-PM2.5
(ng/m3)
Avg. Pb-PM25
(ng/m3)
Avg. Pb-PM10
(ng/m3)
Med. Pb-PM2.5
(mg/kg)

Sampling
Period Personal Pb
July, 1995- 13
May, 1997
Spring, 6.2
Summer,
Fall, 1999
December,
2003-
July, 2004
April-May,
2002
April-May,
2002
May-June,
2006,
December,
2006-
January 2007
May-June,
2006,
December,
2006-
January 2007
April, 2004 31 1

Exposure to Pb in Soil and Dust
AOCD (U.S. EPA. 2006b) lists indoor Pb dust
Indoor Pb Outdoor Pb
6.6 8.5
3.4 2.0
Homes: 3.4 Homes: 4.5
Schools: 2.5 Schools: 4.6
Preschools: 1 .8 Preschools: 2.6
26 56
4 4
Summer: 34 Summer: 47
Winter: 85 Winter: 72
Summer: 57 Summer: 78
Winter: 125 Winter: 159
124 221

infiltrated from outdoors as a
potential source of exposure to Pb soil and dust. Thus, outdoor soil Pb may present an
inhalation exposure if resuspended indoors or an ingestion exposure during hand-to-
mouth contact. A detailed description of studies of outdoor soil Pb concentration is
provided in Section 3.6.1. Indoor measurements can reflect infiltrated Pb as well as Pb
dust derived from debrided paint, consumer products, or soil that has been transported
into the home via foot traffic. Table 4-5 presents indoor dust Pb concentrations for
2006-2011 observational studies in which indoor dust Pb was measured.
     November 2012
                            4-14
Draft - Do Not Cite or Quote

-------
Table 4-5
Reference
Caravanos et al.
(2006b)
Khoder et al.
(2010)
Brattin and
Griffin (2011)
Yu et al. (2006)
Turner and
Simmonds
(2006)
Gaitenset al.
(2009)
Wilson et al.
(2007)
Zota et al. (201 1 )
Spalingeret al.
(2007)
Measurements of indoor dust Pb concentration from 2006-2011
studies.
Study Location
New York City, New York
Giza, Egypt (extensive leaded
gasoline use; industrial area)
Eureka, Utah near Eureka Mills
Superfund Site
Denver, CO, near VBI70 Superfund
Site
East Helena, MT, near East Helena
Superfund Site
Syracuse, New York
Birmingham, Plymouth,
and 2 rural sites, U.K.
U.S. (nationwide)
Milwaukee, Wisconsin
Ottawa County, Oklahoma
(area surrounding the Tar Creek
Superfund Site)
Rural towns, Idaho
Bunker Hill, Idaho Superfund site
Metric (units)
Weekly dust
loading (ug/m2)
Weekly dust
loading (ug/m2)
Dust concentration
(mg/kg)
Dust concentration
range (mg/kg)
Dust concentration
(mg/kg)
Dust loading
(ug/m2)
Dust concentration
(ug/m2)
Dust concentration
(mg/kg)
Dust concentration
(mg/kg)
Dust concentration
(mg/kg)
Sample Site
Glass plate next to open
window of academic
building
Glass plate in second-
floor living room of
apartments
Indoor home site (not
specified)
Indoor home site (not
specified)
Indoor home site (not
specified)
Floor
Floor
Smooth floor
Rough floor
Smooth windowsill
Rough windowsill
Central perimeter
Entry
Window
Indoor (site not specified)
Vacuum
Floor
Vacuum
Floor
Indoor Pb
Concentration
Median: 52
Median: 408
1 60-2000
11-660
68-1 000
Range: 209-1770
Median: 178
Median: 1.7
Avg.: 4.4
Median: 5.6
Avg.: 16
Median: 2.5
Avg.: 190
Median: 55
Avg.: 480
Avg.: 107
Avg.: 140
Avg.: 151
Avg.: 109
Median: 63
Max.: 881
Median: 120
Max: 830
Median: 95
Max: 1 ,300
Median: 470
Max: 2,000
Median: 290
Max: 4,600
November 2012
4-15
Draft - Do Not Cite or Quote

-------
 1                  Several studies suggested the infiltration of Pb dust into buildings. For example,
 2                  Caravanos et al. (2006b) collected dust on glass plates at an interior location near an open
 3                  window, a sheltered exterior location, and an open exterior location for a two-year period
 4                  in Manhattan, NY. Median weekly dust loading was reported to be 52 (ig/m2 for the
 5                  indoor site, 153 (ig/m2 for the unsheltered outdoor site, and 347 (ig/m2 for the sheltered
 6                  outdoor site. This paper demonstrated the likely role of outdoor Pb in influencing indoor
 7                  dust Pb loading and indicated that under quiescent conditions (e.g., no cleaning) near an
 8                  open second-story window, the indoor dust Pb level might exceed EPA's hazard level for
 9                  interior floor dust of 430 ug/m2 (40 ug/ft2). Khoder et al. (2010) used the same
10                  methodology to study Pb dust deposition in residential households in the town of Giza,
11                  Egypt, located between two industrial areas and where leaded gasoline is still in use; the
12                  investigators reported a median weekly deposition rate of 408 (ig/m2 and an exterior
13                  median deposition rate of 2,600 (ig/m2. In the latter study, Pb deposition rate correlated
14                  with total dust deposition rate  (R=0.92), Cd deposition rate (R=0.95), and Ni deposition
15                  rate (R=0.90). Statistically significant differences in Pb deposition rates were observed
16                  between old and new homes (p <0.01) in the Khoder et al. (2010) study, although the
17                  only quantitative information provided regarding home age stated that the oldest home
18                  was 22 years old when the study was performed in 2007. Khoder et al. (2010) found no
19                  statistically significant difference between Pb loadings when segregating the data by
20                  proximity to roadways. Recently, Brattin and Griffin (2011) performed linear regressions
21                  of dust Pb on soil Pb based on data collected previously for outdoor soil Pb and indoor
22                  dust Pb near mining and/or smelting Superfund sites in Utah, Colorado, and Montana
23                  (U.S. EPA. 2005f: SRC. 2002: U.S. EPA. 2001). They observed that the dust Pb
24                  concentration was 4-35% of outdoor soil Pb. Excluding outliers on the regression, dust
25                  Pb concentration ranged from 160-2,000 mg/kg,  11-660 mg/kg, and 68-1,000 mg/kg at
26                  three sites.

27                  Correlations between indoor and outdoor Pb content in dust can be partially explained
28                  with speciation. Beauchemin et al. (2011) used XANES to speciate in-home paint
29                  samples to assess the contributions of indoor paint and outdoor material to indoor dust Pb
30                  concentrations. In indoor dust samples of particles <150 (im in size, Pb oxide, Pb sulfate,
31                  and Pb carbonate  were measured. These materials commonly were used in white paint. In
32                  the size fraction of particles <36 (im, half of the measured Pb was associated with
33                  Fe-oxyhydroxides such as ferrihydrite and goethite and presumably adsorbed onto these
34                  species. This finding suggested that a mix of indoor and outdoor sources may affect the
35                  composition of dust in the smaller size fraction in houses with leaded paint.

36                  Residual Pb dust contamination following cleaning activities has been documented. For
37                  instance, Hunt et al. (2008) estimated Pb deposition and concentration from experiments
38                  in which Herculaneum, MO yard soil  samples that had been dried, ground, and sieved
      November 2012                             4-16                   Draft - Do Not Cite or Quote

-------
 1                  were tracked onto a tile test surface and then repeatedly cleaned until visual inspection of
 2                  the tiles uncovered no surface discoloration. Cleaning resulted in a 5-6 fold decrease in
 3                  residual Pb, with 7,100 (ig/m2 measured after multiple walks across the sample floor prior
 4                  to cleaning. Yu et al. (2006) analyzed dust samples from 50 homes in northern
 5                  New Jersey (typically of older housing stock, although the study does not specify housing
 6                  age). The investigators found that total Pb concentration in carpet dust ranged from 209
 7                  to 1,770 mg/kg dust. Wilson et al. (2007) studied Pb dust samples from homes in
 8                  Milwaukee, WI, and in resident children with and without elevated blood Pb > 10 (ig/dL.
 9                  They found that Pb dust samples obtained from the floor were always significantly higher
10                  in residences of children with elevated blood Pb, with the exception of samples from the
11                  bathroom floor. Windowsill dust was not significantly higher in residences of children
12                  with elevated blood Pb. Residual Pb dust in homes is a potential exposure source for
13                  small children who use touch to explore their environments.

14                  Pb dust on floors, windowsills, and other accessible surfaces is related to several
15                  demographic, socioeconomic, and housing conditions. Gaitens et al.  (2009) used National
16                  Health and Nutrition Examination Survey (NHANES) data from 1999 through 2004 to
17                  examine Pb in dust in homes of children ages 12-60 months. Floor Pb dust loading value
18                  was modeled against several survey covariates and was significantly associated with
19                  several covariates but with mixed sign (p <0.05). Floor Pb dust was positively associated
20                  with windowsill Pb dust loading, being of non-Hispanic black race/ethnicity,  and
21                  presence of smokers in the home. Floor Pb dust was negatively associated with presence
22                  of carpeting, poverty-to-income ratio, and living in a home built after 1950. It was nearly
23                  significantly and positively associated (p = 0.056) with renovations made to pre-1950
24                  homes. Windowsill Pb dust level was also significantly associated (p <0.05) with several
25                  covariates. It was positively associated with being of non-Hispanic black race/ethnicity,
26                  negatively associated with living in a home built after 1950, positively associated with
27                  not smooth and cleanable window surface condition, positively associated with presence
28                  of smokers in the home, and positively associated with deterioration of indoor paint. It
29                  was nearly statistically significantly and positively associated (p = 0.076) with
30                  deterioration of outdoor paint when homes were built prior to 1950. Dust Pb loading was
31                  found by Egeghy et al. (2005) to be significantly and positively associated with the log-
32                  transform of soil Pb concentration, cement content in the home entryway, frequency of
33                  fireplace usage, and homes built before 1950. Dust Pb loading was significantly and
34                  negatively associated with indoor pesticide use and number of hours in which windows
35                  were open (Table 4-3).

36                  Building demolition and renovation activities can create dust from interior and exterior
37                  paints with Pb content. Mielke and Gonzales (2008) measured Pb content in paint chips
38                  from paint applied prior to 1992 and found that median Pb levels were 420 mg/kg for
      November 2012                             4-17                  Draft - Do Not Cite or Quote

-------
 1                  interior paint and 77,000 mg/kg for exterior paint. Maximum levels were 63,000 mg/kg
 2                  and 120,000 mg/kg for interior and exterior paint, respectively. Mielke et al. (2001)
 3                  compared dust samples from two New Orleans houses that were prepared for painting.
 4                  One home was power sanded without any confinement or control of removed material,
 5                  while the other was hand-scraped with containment and collection of paint chips.
 6                  Immediately after sanding, Pb dust samples ranged from <3 to 28,000  mg/kg at the
 7                  sanded house. Pb dust samples from the scraped house ranged from 7 to 1,200 mg/kg.

 8                  Dust Pb concentrations have also been reported for homes in the vicinity of historic and
 9                  active metals mining and smelting sources. As described in Section 3.6.1. soil Pb has
10                  been found to be elevated near source of ambient air Pb. Near an active smelter in Port
11                  Pirie, Australia, median hand dust Pb loadings increased with age among a cohort of
12                  fourteen children followed over age 0-36 months (2-5 months: 54 (ig/m2, >15 months:
13                  336 (ig/m2) (Simon et al.. 2007). Zota et al. (2011) studied Pb dust and indoor Pb-PM2 5
14                  concentration in Ottawa County, OK near the Tar Creek Superfund Site, in which a
15                  metals mine had closed. Statistically significant correlations among outdoor soil Pb
16                  concentration, indoor dust Pb concentration, indoor dust Pb loading, and indoor air
17                  Pb-PM2 5 concentrations were observed (r = 0.25-0.65), with an average dust Pb
18                  concentration of 109 mg/kg, dust Pb loading of 54 (ig/m2, soil Pb concentration of
19                  201 mg/kg, and indoor Pb-PM2 5 concentration of 1 ng/m3. House dust Pb concentrations
20                  were found to increase significantly with residential proximity to two chat (i.e., dry
21                  mining waste) sources and to decrease with distance to the street and presence of central
22                  air conditioning. Spalinger et al. (2007) measured Pb in dust in homes in a  34 km2 area
23                  surrounding a designated Superfund site where a Pb and Zn smelter formerly operated at
24                  Bunker Hill, ID. During spring of 1999, vacuum and floor mat samples were taken from
25                  homes in three towns within the 34 km2 area and five "background" towns  further from
26                  the  Superfund site. For the background towns, Pb concentration in vacuum dust had a
27                  median of 120 mg/kg, and Pb concentration in floor dust had a median of 95 mg/kg. The
28                  median Pb dust loading rate was measured to be 40 (ig/m2 per day.  In contrast, Pb in
29                  vacuum dust and floor mats for the towns contained within the Bunker Hill Superfund
30                  site had a median Pb concentration of 470 mg/kg and 290 mg/kg, respectively. The
31                  median Pb loading rate for indoor dust in houses in these towns was 300 (ig/m2 per day.
32                  These results suggest that those living in close proximity to large Pb and Zn smelters or
33                  mines that are now Superfund sites are at much greater risk of exposure to Pb dust
34                  compared to the general population.

35                  Pb exposure has been reported on children's playgrounds. Mielke et al. (2011 a) reported
36                  median soil Pb concentration of 558 mg/kg on playground soils at eleven New Orleans
37                  day care or community centers. Following remediation efforts to cover playground soil
38                  with clean soil, median concentration dropped to 4.1 mg/kg. Duggan et al. (1985)
      November 2012                            4-18                  Draft - Do Not Cite or Quote

-------
 1                  reported on the concentration and size distribution of wipe samples on the hands of 368
 2                  pre-school children from eleven schools in London, U.K.. Hand Pb residue (PbH) values
 3                  were modeled as linear (p <0.05) and power functions (p <0.001) of Pb dust; linear slope
 4                  was 0.0064 (ig hand Pb residue per mg/kg Pb dust. Given that the Duggan et al. (1985)
 5                  study was performed when Pb additives were used in gasoline, dust Pb concentration
 6                  values are not reported here. As described in Section 4.1.1.1. exposure to Pb in soil and
 7                  dust may be related to size distribution of the soil or dust particles, with higher Pb
 8                  enrichment in the smaller particles.
                    4.1.3.3     Dietary Pb Exposure

 9                  This subsection covers several dietary Pb exposures from a diverse set of sources.
10                  Included among those are drinking water, fish and meat, agriculture, urban gardening,
11                  dietary supplements, tobacco, cultural food sources, and breastfeeding. The breadth of
12                  dietary Pb exposures is illustrated in Figure 4-2. which illustrates the data obtained in the
13                  2008 FDA Total Diet Study market basket survey (FDA. 2008). Among the highest Pb
14                  concentrations were those for noodles, baby food carrots, baby food oatmeal, Swiss
15                  cheese, beef tacos from a Mexican restaurant, and fruit-flavored cereal. Possible sources
16                  of Pb in food samples include introduction during processing or preparation with drinking
17                  water contaminated with Pb, deposition of Pb onto  raw materials for each food, and Pb
18                  exposure in  livestock that produce dairy or meat ingredients. Manton et al. (2005) used
19                  Pb isotope ratios to estimate sources of dietary Pb among a cohort of mothers and
20                  children from Omaha, NE using a combination of food samples, hand wipes, house dust
21                  wipes, and aerosol samples collected between 1990 and  1997. Drinking water Pb was not
22                  included in this study. The authors cited results from Egan et al. (2002) that imported
23                  vegetables contributed 55% of Pb dietary intake for infants, 30% for 2-6 year old
24                  children, and 20% for 25-30 year old women. Imported candy contributed  10% of Pb
25                  dietary intake for 2-6 year old children and  9% for  25-30 year old women.  Isotopic data
26                  from Manton et al. (2005) suggested that, with the exception of children age 0-12 mos,
27                  house dust is a large contributor to dietary Pb. The  pattern of certain Pb-isotope ratios
28                  observed in  the diet of children 0-12 mos are suggested to derive from Ca salts in
29                  limestone that may have been used in dietary supplements in baby formula. The
30                  contribution of ambient Pb aerosols to dietary Pb samples was not statistically significant
31                  for this urban exposure study.
      November 2012                             4-19                   Draft - Do Not Cite or Quote

-------


IL 1
INQOQISS ~~
BF(carrot) -
BF(oatmeal) -
Cheese -
Taco —
Cereal -
Egg'n'chs —
Peas -
Milk(2%) -
BF(grnbean) —
Cabbage -
BF(lamb) -
Beans —
Rice -
Coffee -
Strawberries —
Squash —
Potato —
Chicken -
Biscuits —
BF(cobbler) -
BF(beef) -



L J 1
M 1
i— i H
ED
r-|~|H
\L
h[J}--H
i nil
i ii
rflO
EH
\L
r| | |- - H
(0
h|~~f1-H
1 1 1 1
u
EH







i i i i i i i
0.00 0.02 0.04 0.06 0.08 0.10 0.12 014
                                      Concentration (mg/kg)
Note: from the 2008 FDA Total Diet Study. "BF" denotes baby food.
Source Data: (FDA. 2008)
Figure 4-2
Market basket survey results for Pb concentration in foods.
November 2012
                      4-20
Draft - Do Not Cite or Quote

-------
                     Drinking Water

 1                   Pb concentrations in drinking water vary substantially. For example, Shotyk and Krachler
 2                   (2009) measured the Pb concentration in tap water, commercially bottled tap water and
 3                   bottled natural water. They found that, in many cases, tap water contained less Pb than
 4                   bottled water. Excluding bottled water in glass containers because Pb can be leached
 5                   from the glass, the median Pb concentration in the bottled water samples was 8.5 ng/L
 6                   (range < 1 to 761 ng/L). Pb in drinking water supplies can derive from atmospheric
 7                   deposition onto surface waters, runoff of atmospheric deposition as described in Section
 8                   3.3. or via corrosion of Pb in the distribution network exacerbated by contact with acidic
 9                   disinfection byproducts, as described in the following paragraphs.

10                   It is now recognized that environmental nanoparticles (NPs) (-1-100 nm) can play a key
11                   role in determining the chemical characteristics of treated drinking water as well as
12                   natural waters (Wigginton et al., 2007). An important question is whether or not NPs
13                   from source waters affect the quality of drinking water. For example, if Fe-oxide NPs are
14                   not removed during the flocculation/coagulation stage of the treatment process, they may
15                   become effective transporters of contaminants such as Pb, particularly if these
16                   contaminants are leached from piping in the distribution system.

17                   Corrosion byproducts can influence Pb concentrations in drinking  water. Schock et al.
18                   (2008) characterized Pb pipe scales from 91 pipes made available from 26 different
19                   municipal water systems from across the northern U.S. They found a wide range of
20                   elements including Cu, Zn and V as well as Al, Fe and Mn. Interestingly, V was present
21                   at nearly one percent levels in pipes from many geographically diverse systems. In a
22                   separate study, Gerke et al. (2009) identified the corrosion product, vanadinite
23                   (Pb5(VO4)3Cl) in Pb pipe corrosion byproducts collected from 15 Pb or Pb-lined pipes
24                   representing 8 different municipal drinking water distribution systems in the Northeastern
25                   and Midwest regions of the U.S. Vanadinite was most frequently found in the surface
26                   layers of the corrosion products. The vanadate ion, VO43", essentially replaces the
27                   phosphate ion in pyromorphite and hydroxyapatite structures. It is not known whether the
28                   application of orthophosphate as a corrosion inhibitor would destabilize vanadinite, but
29                   this substitution would have implications for V release into  drinking water. The stability
30                   of vanadinite in the presence of monochloramine is also not known, and its stability
31                   might have implications for both Pb and V release into drinking water.

32                   In recent years, drinking water treatment plants in many municipalities have switched
33                   from using chlorine to other disinfecting agents because their disinfection byproducts
34                   may be less carcinogenic. However, chloramines are more acidic than chlorine and can
35                   increase Pb  solubility (Raab et al.. 1991) and increase Pb concentrations in tap water. For
36                   example, after observing elevated Pb  concentrations in drinking water samples, Kim and
      November 2012                             4-21                  Draft - Do Not Cite or Quote

-------
 1                  Herrera (2010) observed Pb oxide corrosion scales occurring after using acidic alum as a
 2                  disinfection agent. Edwards and Dudi (2004) observed a red-brown particle-bound Pb in
 3                  Washington, B.C. water that could be confused with innocuous Fe. The source of the
 4                  particle-bound Pb was not known but was thought to originate from the source water. The
 5                  high Pb concentrations were attributed to leaching of Pb from Pb-bearing pipes promoted
 6                  by breakdown products of disinfection agents (Edwards and Dudi. 2004). Maas et al.
 7                  (2007) tested the effect of fluoridation and chlorine-based (chlorine and chloramines)
 8                  disinfection agents on Pb leaching from plumbing soldered with Pb. When using chlorine
 9                  disinfection agents alone, the Pb concentration in water samples doubled during the first
10                  week of application (from 100 to 200 ppb) but then decreased over time. When adding
11                  fluorosilicic acid and ammonia, the Pb concentration spiked to 900 ppb and increased
12                  further over time. However, Macek et al.  (2006) regressed blood Pb among children ages
13                  1-16 years on fluoride treatment, adjusted for several demographic and socioeconomic
14                  factors, and found no association when all data were combined into one model; when
15                  stratifying by housing age, Macek et al. (2006) found statistically significant odds ratios
16                  for those living in housing built before 1946 or for housing age unknown. Similarly,
17                  Lasheen et al. (2008) observed Pb leaching from pipes in Egypt when exposed to an acid
18                  of pH = 6. Exposure to basic solutions actually resulted in reduction of Pb concentration
19                  in the drinking water. Leaching of Pb from pipes following disinfection with acidic
20                  agents can lead to increased Pb exposure; Miranda et al. (2007b) observed a statistically
21                  significant association between blood Pb  levels among children living in Wayne County,
22                  NC and use of chloramines (p <0.001) in a log-linear model, although the study did not
23                  control for the presence of Pb paint in the dwellings, so it is difficult to distinguish the
24                  influence  of Pb pipes from Pb in paint on blood Pb levels.

25                  Several chemical mechanisms may  contribute to release of Pb during use of chloramine
26                  disinfection agents.  Edwards and Dudi (2004) hypothesized that Pb leaching occurs when
27                  chloramines cause the breakdown of brass alloys and solder containing Pb. After
28                  observing that nitrification also leads to increased Pb concentrations in water, they also
29                  proposed that chloramines may trigger nitrification and hence cause decreasing pH,
30                  alkalinity, and dissolved oxygen that leads to corrosion after observing that nitrification
31                  also leads to increased Pb concentrations  in water. However, Zhang et al. (2009b) found
32                  no evidence that nitrification brought about significant leaching of Pb from Pb pipes.
33                  Lytle et al. (2009) suggested that a lack of increased Pb(II) concentrations in drinking
34                  water following a change from free chlorine to chloramine disinfection is attributed to the
35                  formation of the Pb(II) mineral hydroxypyromorphite (Pb5(PO4)3OH) instead of Pb(IV)
36                  oxide. Xie et al. (2010) further investigated the mechanisms by which Pb(II) release is
37                  affected by chloramines. Two opposing mechanisms were proposed:  Pb(IV)O2 reduction
38                  by an intermediate species from decomposition of monochloramine; and increasing redox
39                  potential which decreases the thermodynamic driving force for reduction. They suggest

      November 2012                             4-22                   Draft - Do Not Cite or Quote

-------
 1                  that the contact time of monochloramine with PbO2 and the C12:N ratio in
 2                  monochloramine formation will determine which mechanism is more important. Free
 3                  chlorine can control Pb concentrations from dissolution under flowing conditions but for
 4                  long stagnation periods, Pb concentrations can exceed the action level:  4-10 days were
 5                  required for Pb concentrations to exceed 15  (ig/L (for relatively high loadings of PbO2 of
 6                  1 g/L). Thus, under less extreme conditions, it was concluded that chloramination was
 7                  unlikely to have a major effect on the release of Pb into drinking water.


                    Agriculture

 8                  The 2006 Pb AQCD (2006b) states that surface deposition "represents a significant
 9                  contribution to the total  Pb in and on the plant", while uptake through a plant's roots can
10                  also contribute to a plant's Pb concentration. Consequently, Pb content in plants may
11                  contribute to human dietary exposure. Uptake of Pb by plants growing  in contaminated
12                  soil has been repeatedly demonstrated in some species during controlled potted plant
13                  experiments (Del Rio-Celestino et al.. 2006). In this study, most species retained Pb in
14                  the roots with little mobilization to the shoots of the plants. However, certain species of
15                  grasses were able to mobilize Pb from the roots to the shoots of the plant; these specific
16                  species could lead  to human exposures through consumption of grazing animals. Lima et
17                  al. (2009) conducted similar greenhouse experiments with several vegetable crops grown
18                  in soil contaminated by Pb-containing residue from battery recycling waste. In this study,
19                  carrots had high bioaccumulation, measured as the percent of Pb concentration measured
20                  in the plant compared with the Pb concentration in the soil, with little translocation of the
21                  Pb to the shoots. Conversely, beets, cabbages, sweet peppers, and collard greens had low
22                  bioaccumulation but moderate to high translocation. Okra, tomatoes, and eggplants had
23                  moderate  bioaccumulation and moderate to high translocation.  Sesli et  al. (2008) also
24                  noted uptake  of Pb within wild mushrooms.  Vandenhove et al.  (2009) reviewed
25                  bioaccumulation data  for plant groupings and found that grasses had the highest uptake,
26                  followed by leafy vegetables and root crops  grown in sandy soils; see Table 4-6. These
27                  references also suggested high transfer from roots to shoots among root crops, with
28                  shoots having roughly four times higher Pb bioaccumulation than roots.

29                  Sources of atmospheric Pb can lead to vegetable contamination. For example, Uzu et al.
30                  (2010) found that Pb deposition from smelter emissions caused a linear increase in Pb
31                  concentrations of 7.0 mg/kg per day (R2=0.96) in lettuce plants cultivated in the
32                  courtyard of a smelter. They reported that lettuce grown 250-400 meters from the smelter
33                  had concentrations that were 10-20 times lower, which is consistent with findings
34                  described in Section 3.3 that deposition of Pb containing material drops off with distance
35                  from a source. Pb contamination of crops may also occur through piston-engine aircraft
36                  Pb emissions during aerial application of fertilizers and pesticides. In 2010, the U.S.

      November 2012                             4-23                  Draft - Do Not Cite or Quote

-------
 1                  Federal Aviation Administration (FAA) recorded 396,000 hours of flight time for aerial
 2                  application. This term encompasses crop and timber production including seeding
 3                  cropland and fertilizer and pesticide application. It is estimated that 86% of these flight-
 4                  hours involved piston engine aircraft utilizing leaded fuel (FAA. 2010).

 5                  Some land use and soil characteristics have been shown to increase bioaccessibility of Pb
 6                  in soil, which could then lead to plant contamination. Fernandez et al. (2010; 2008; 2007)
 7                  measured Pb from atmospheric deposition in two adjacent plots of land having the same
 8                  soil composition but different uses: one was pasture land and one was agricultural. In the
 9                  arable land, size distributions of soil particle-bound Pb, were uniformly distributed. In
10                  pasture land, size distributions of soil particle-bound Pb were bimodal with peaks around
11                  2-20 urn and 50-100 urn (Fernandez etal., 2010). For the agricultural plot, Pb
12                  concentration was constant around 70 mg/kg in samples taken over the first 30 cm of soil,
13                  at which time it dropped below  10 mg/kg at soil depths between 35 and 100  cm. In
14                  contrast, Pb concentration in pasture land peaked at a depth of 10 cm at a concentration
15                  of roughly 70 mg/kg and then dropped off gradually to approach zero concentration at a
16                  depth of approximately 50 cm. The sharp change in concentration for the arable land was
17                  attributed to a combination of plowing the soil and use of fertilizers to increase the
18                  acidity of the soil and solubility of Pb into the soil (Fernandez et al..  2007). They found
19                  that the surface layer was acidic (pH: 3.37-4.09), as was the subsurface layer (pH:
20                  3.65-4.38). Jin et al. (2005) examined how soil characteristics affect Pb contamination of
21                  crops by testing  soil Pb, bioaccessibility of soil Pb  (determined by CaCl2 extraction),  and
22                  Pb in tea samples from tea gardens. They observed that the Pb concentration in tea leaves
23                  was proportional to the bioaccessible Pb in soil.

24                  There is some evidence that Pb contamination of crops can originate with treatment of
25                  crops. For example, compost produced from wastewater sludge has the potential to add
26                  Pb to crops. Cai et al. (2007) demonstrated that production of compost from  sludge
27                  enriched the Pb content by 15-43% compared with the Pb content in sludge prior to
28                  composting. Chen et al. (2008b) observed that the median concentration of Pb in
29                  California crop soil samples was 16.2 mg/kg (range: 6.0-62.2 mg/kg). Chen  et al. (2008a)
30                  further observed that in three of the seven California agricultural regions sampled,
31                  concentrations of Pb increased following addition of fertilizer, but the increase was less
32                  than that for phosphorous (P) and Zn indicators of fertilizer. In four regions, there was no
33                  increase of Pb at all. Furthermore, Tu et al. (2000) observed a decrease in Pb fraction
34                  with increasing P application. Nziguheba and Smolders (2008) also surveyed phosphate-
35                  based fertilizers  sold in European markets to determine the contribution of these
36                  fertilizers to heavy metal concentrations in agricultural products. They reported a median
37                  fertilizer Pb concentration of 2.1 mg/kg based on total weight of the fertilizer, with a 95th
38                  percentile concentration of 7.5 mg/kg. Across Europe, Nziguheba and Smolders (2008)
      November 2012                             4-24                  Draft - Do Not Cite or Quote

-------
 1                  estimated that the amount of Pb applied via fertilizers to be only 2.6% of that resulting
 2                  from atmospheric deposition.

 3                  Although Pb in on-road vehicle gasoline has been phased out in the U.S., if imported
 4                  crops are produced in countries that still use Pb antiknock agents in on-road gasoline,
 5                  they have the potential to introduce dietary Pb to U.S. consumers. For example, high
 6                  concentrations of Pb have been found in chocolate from beans grown in Nigeria, during
 7                  the time when leaded gasoline was still legally sold. Rankin et al. (2005) observed that
 8                  the ratios of 207Pb to 206Pb and 208Pb to 207Pb were similar to those of Pb in gasoline.
 9                  Although this study showed that Pb concentration in the shelled cocoa beans was low (~1
10                  ng/g), manufactured cocoa powder and baking chocolate had Pb concentrations similar to
11                  those of the cocoa bean shells, on the order of 200 ng/g, and Pb concentration in
12                  chocolate products was roughly 50 ng/g (Rankin et al.. 2005). It is possible that the
13                  increases were attributed to contamination of the cocoa by the shells during storage or
14                  manufacture, but the authors note that more research is needed to verify the source of
15                  contamination.

16                  Findings from Pb uptake studies have implications for urban gardening if urban soils may
17                  be contaminated with Pb. For instance, Clark et al. (2006) tested the soil in 103 urban
18                  gardens in two Boston neighborhoods. Using isotopic analysis, they found that Pb-based
19                  paint contributed 40-80% of Pb in the urban garden  soil samples, with the rest  coming
20                  from historical gasoline emissions. Furthermore, Clark et al. (2006) estimated that Pb
21                  consumption from urban gardens can be equivalent to 10-25% of the exposure to Pb from
22                  drinking water for children  living in the Boston neighborhoods studied. Because soil Pb
23                  levels in urban areas will depend on surrounding sources (Pruvot et al.. 2006).  Pb
24                  exposures in urban garden vegetables will vary.
      November 2012                             4-25                  Draft - Do Not Cite or Quote

-------
Table 4-6 Pb bioaccumulation data for various plants. Bioaccumulation is
expressed as percent of Pb concentration in the plant to the Pb
concentration in the soil.
Plant
Plant Group Compartment
All
Cereals Grain
Straw
Maize Grain
Straw
Rice Grain
Leafy Vegetables
Non-Leafy Fruits
vegeiaoies ^^
Legumes Pods
Shoots
Root Crops Roots
Shoots
Tubers Tubers
Fruits Fruits
Leaves
Grasses
Natural Pastures
Leguminous Fodder
All Cereals
Pastures/Grasses
Fodder
Soil

All
All
All
All
All
All
Sand
Loam
Clay
All
All
All
Sand
Loam
Clay
All
All
Sand
Loam
All
All
Sand
Loam
All
All
All
All
All
All
Sand
Loam
Clay
All
All
Sand
Clay
n
210
9
4
9
3
2
31
4
3
7
5
2
17
3
5
4
1
27
5
5
12
30
5
17
5
1
17
34
1
20
5
8
6
51
24
4
4
GM
2.0%
1.0%
2.3%
0.12%
0.28%

8.0%
7.3%
82%
2.8%
1.5%

0.53%
0.27%
0.14%
0.080%

1.5%
6.4%
2.3%
6.3%
0.15%
0.64%
0.052%
0.77%

31%
92%

0.43%
0.61%
0.17%
0.90%
14%
2.5%
4.5%
0.82%
GSD
14
3.6
3.5
2.3
6.6

13
1.5
1.0
4.1
26

12
3.2
4.4
1.0

16
1.6
4.7
15
7.4
3.5
2.4
2.6

1.8
4.8

4.7
5.3
3.9
4.0
4.2
12
2.3
5.7
AM
63%
1.8%
3.8%
0.17%
0.85%
2.2%
210%
7.8%
82%
5.1%
78%
0.88%
34%
0.42%
0.42%
0.33%
0.080%
41%
7.0%
0.50%
250%
9.1%
1.2%
0.073%
1.0%
25%
36%
23%
1.6%
1.1%
1.3%
0.53%
1.8%
27%
130%
5.6%
2.7%
SD
290%
1.6%
4.0%
0.14%
1.3%
1.4%
610%
3.3%
3.5%
4.8%
170%
0.42%
120%
0.34%
0.34%
0.47%

98%
3.4%
0.68%
570%
48%
1.6%
0.062%
0.60%

22%
29%

1.4%
1.3%
1.1%
1.8%
27%
420%
4.0%
4.6%
Min
0.015%
0.19%
0.51%
0.052%
0.060%
1.2%
0.32%
4.9%
79%
0.41%
0.15%
0.58%
0.046%
0.065%
0.065%
0.046%

0.024%
4.2%
0.024%
0.30%
0.015%
0.16%
0.015%
0.15%

11%
0.22%

0.052%
0.052%
0.059%
0.22%
0.22%
0.060%
1.6%
0.16%
Max
2,500%
4.8%
9.6%
0.38%
2.3%
3.2%
2,500%
11%
86%
12%
390%
1.17%
490%
0.89%
0.89%
1.0%

330%
12%
1.7%
16%
260%
3.9%
0.23%
1.7%

100%
100%

4.8%
3.2%
3.2%
4.8%
100%
1 ,600%
11%
9.6%
Source: Reprinted with permission of Elsevier Publishers, Vandenhove et al.
November 2012
4-26
Draft - Do Not Cite or Quote

-------
                    Game

 1                  Atmospheric sources of Pb have also been shown to contaminate game meat, thus
 2                  potentially posing a risk of Pb exposure. In Pb mining or smelting areas, several studies
 3                  have documented Pb concentrations in game [e.g., (Nwude et al., 2010; Reglero et al.,
 4                  2009b)1.

 5                  Potential Pb exposure through consumption of animals exposed to or killed with Pb shot
 6                  has also been well documented (Hunt et al., 2009; Tsuji et al., 2009; Tsuji et al., 2008;
 7                  Hunt et al.. 2006). For example, Martinez-Haro et al. (2010) observed Pb in the feces of
 8                  mallards that ingested gunshot of 34-13,930 mg/kg with a median of 1,104 mg/kg, while
 9                  mallards that did not ingest gunshot had feces Pb levels <12.5 mg/kg. Mateo et al. (2011)
10                  studied Pb bioaccessibility as a function of cooking method for breast meat from
11                  partridges killed with gunshot. They observed that preparation in cold or hot vinegar
12                  increased bioaccessibility compared with total Pb in the samples.


                    Fish

13                  Pb content in fish could also lead to human exposure to Pb  (U.S. EPA. 2006b. 1986a).
14                  Ghosh et al. (2007) demonstrated in laboratory experiments that exposure to Pb in water
15                  can lead to linearly increasing Pb levels in the kidneys, liver, gills, skeleton, and muscle
16                  offish. Several studies have documented the potential for human Pb exposure through
17                  fish and seafood. Welt et al. (2003) conducted a survey of individuals who fished in
18                  Bayou St. John, Louisiana in conjunction with sampling Pb content in sediment. They
19                  found that median sediment Pb concentrations ranged from 43 to 330 mg/kg in different
20                  locations, while maximum sediment Pb concentrations ranged from 580 to 6,500 mg/kg.
21                  In total, 65% of the surveyed individuals fished for food from the Bayou, with 86%
22                  consuming fish from the Bayou each week. In a study of the effect of coal mining on
23                  levels of metals in fish (measured as blood Pb) in northeastern Oklahoma, Schmitt et al.
24                  (2005) found that fish blood Pb levels varied with respect to species of fish, but blood Pb
25                  levels were higher in fish in areas close to  mining activities. Similarly, Besser et al.
26                  (2008) observed higher levels  offish blood Pb close to mining activities in southeastern
27                  Missouri.  In a related study offish species in the same region of Missouri, fish blood Pb
28                  levels were found to be statistically significantly higher in sites within  10 km downstream
29                  of active Pb-Zn mines (p <0.01) compared with  fish located further from the mines
30                  (Schmitt et al.. 2007a). and elevated fish blood Pb levels were again noted near a Pb-Zn
31                  mine (Schmitt et al.. 2009). It was noted that the Ozark streams where these studies were
32                  performed were often used for recreational fishing.
      November 2012                             4-27                  Draft - Do Not Cite or Quote

-------
                    Breast Milk

 1                  Studies of breastfeeding women suggest that infants may be exposed to Pb in breast milk.
 2                  Ettinger et al. (2004a) observed in a 1994-1995 study of Mexico City women that at
 3                  1 month postpartum, 88 women breastfeeding exclusively (with mean blood Pb level of
 4                  9.4 (ig/dL) had breast milk Pb concentrations of 1.4 ± 1.1 (ig/L, and 165 women
 5                  breastfeeding partially (with mean blood Pb level of 9.5 (ig/dL) had breast milk Pb
 6                  concentrations of 1.5 ± 1.2 (ig/L. During the same time period, Ettinger et al. (2006)
 7                  studied breastfeeding women in Mexico City over a child's first year of life and sampled
 8                  Pb concentration in breast milk at 1, 4, and 7 mo post-partum. They observed that mean
 9                  breast milk concentrations dropped from 1.4 (ig/L at 1 mo (mean maternal blood
10                  Pb = 9.3 (ig/dL) to a mean of 1.2 (ig/L at 4 mo (mean maternal blood Pb = 9.0 (ig/dL) to
11                  0.9 (ig/L at 7 mo (mean maternal blood Pb = 8.1  (ig/dL); this reduction was statistically
12                  significant (p <0.00001). Among the 310 women included in the study, 181 had previous
13                  pregnancies. In one study of nursing mothers living in Port Pirie, Australia near a Pb
14                  smelter, 10 of the 11 mothers had breast milk concentrations <5 (ig/L (Simon et al..
15                  2007). The authors hypothesized that breast milk concentration was too low to be a major
16                  contributor to blood Pb level in these infants relative to other factors such as hand loading
17                  of Pb. However, one mother with a blood Pb level of 25 (ig/dL had a breast milk Pb level
18                  of 28 (ig/L (Simon et al.. 2007).

19                  In summary, several sources of dietary Pb can originate from atmospheric Pb emissions,
20                  including  drinking water, vegetables, game, fish, and breast  milk. Drinking  water Pb
21                  levels are  affected by source strength and proximity, runoff, and water treatment
22                  processes  and chemicals. Among plants grown for agriculture, Pb content is highest in
23                  grasses, followed by leafy vegetables, then root vegetables. Pb in soil or dust can also
24                  collect on the surfaces of vegetables. Pb contamination of vegetables depends on a
25                  number of factors, including presence of nearby sources of atmospheric Pb, soil type and
26                  chemistry, land use, and land treatment. Other  sources of Pb, such as international
27                  consumer products or historic emissions, also have the potential to introduce Pb into the
28                  U.S. diet.  Pb contamination through the food chain potentially leads to elevated Pb levels
29                  in meat. Likewise, Pb contamination of surface waters can lead to elevated levels of Pb in
30                  fish used for consumption. Breastfeeding also presents a potential Pb exposure to
31                  newborn babies, and that exposure drops off as the mothers nurse and as the babies age
32                  and add more food to their diet.
      November 2012                             4-28                   Draft - Do Not Cite or Quote

-------
                    4.1.3.4      Occupational

 1                  Occupational environments have the potential to expose individuals to Pb. Some modern
 2                  day occupational exposures are briefly discussed below in the context of understanding
 3                  potential exposures that are not attributed to ambient air. For example, Miller et al.
 4                  (2010) obtained personal and area samples of particle-borne Pb in a precious metals
 5                  refinery; year of the study was not reported. It was not stated explicitly, but it is likely
 6                  that Miller et al. (2010) measured the PM as TSP because the Occupational Safety and
 7                  Health Administration (OSHA) permissible exposure limit (PEL) for Pb is based on TSP
 8                  rather than a smaller size cut, and the OSHA PEL was used for comparison.
 9                  Concentrations measured by personal samples ranged from 2 to 6 (ig/m3, and
10                  concentrations from area samples ranged from 4 to 14 (ig/m3. The OSHA PEL is 5 (ig/m3.
11                  In steel production, sintering was found to be the largest source of airborne Pb exposure
12                  in a survey of operations (Sammut et al.. 2010). with Pb enrichment in PM reported to be
13                  20,000 mg/kg. Although total PM concentration was not reported by the authors, the PM
14                  was reported to have 75% of its particulate mass at below the 2.5 (im diameter size.

15                  Operations involving Pb-containing materials in various industries are a source of
16                  occupational Pb exposure, in addition to a residential exposure. Rodrigues et al. (2010)
17                  reported exposures to airborne Pb among New England painters, who regularly use
18                  electric grinders to prepare surfaces for painting. Two-week averaged airborne Pb
19                  concentrations, sampled with an Institute of Medicine inhalable PM sampler designed to
20                  capture PM smaller than 100 (im, were reported to be 59 (ig/m3, with a maximum daily
21                  value of 210 (ig/m3. The Pb concentrations reported here were corrected by the National
22                  Institute for Occupational Safety and Health (NIOSH) respirator protection factors,
23                  although the respirator protection factors were not reported by Rodrigues et al. (2010).
24                  Information on the air Pb-blood Pb relationship can  be found in Section 4.5.1.
                    4.1.3.5      Exposure to Pb from Consumer Products

25                  Pb is present in varying amounts in several consumer products including alternative
26                  medicines, candies, cosmetics, pottery, tobacco, toys, and vitamins (Table 4-7). Several
27                  of these categories suggest children may incur regular exposures. Pb concentrations were
28                  reported to range from non-detectable levels up to 77% by mass, for the case of one
29                  medicinal product. Exposure to these products, which originate in a range of different
30                  countries, can account for substantial influence on Pb body burden (Miodovnik and
31                  Landrigan. 2009; Levin et al.. 2008).
      November 2012                            4-29                  Draft - Do Not Cite or Quote

-------
Table 4-7    Pb content in various consumer products.
Product
Category
Alternative
and
Traditional
Medicines







Product
Cissus quadrangularis, Caulophyllum thalictroides,
Turners diffuse, Centella asiatica, Hoodia gordonii,
Sutherlandia frutescens, Curcuma longa,
fucoxanthin, Euterpe oleracea
(dietary supplements claimed to be from Hoodia
gordonii)
Malva sylvestris
Yugmijihwang-tang, Bojungigki-tang, Sibjeondaebo-
tang, Kuibi-tang, Ojeogsan
Lemongrass, licorice, holy basil, cloves, ginger





B-Success 28, Operation Sweep, Aloe Vera Plus
Location of
Purchase
U.S.
(Mississippi)3

Turkey
Korea
India





Nigeria

Pb Content (units)
Not detected (N.D.)b
to 4.21 mg/kg

1.1-2.0 mg/kg
7.9x1 0"6 to 2.5x1 0"5
mg/kg body weight/day
Average:
Lemongrass &
Holy Basil Leaves:
6.1 mg/kg;
Licorice Stolons:
6.1 mg/kg,
Clove Dried Flower
Buds: 7.8 mg/kg,
Ginger Rhizome:
5.8 mg/kg
925-27,000 ug

Reference
Avula et al.
(201 0)

Hicsbmnez et
al. (2009)
Kim et al.
(2009a)
Naithani and
Kakkar (2006)





Obi et al. (2006)
             Bitter Aloes, Zarausmacine, Virgy-Virgy Computer
             Worm-Expeller, Dorasine Powder, Sexual Energy,
             U&DEE Infection Cleansing Powder, U&DEE Sweet
             Bitter, Natural Power Stone, Chama Black Stone,
             Portugal Antiseptic Soap, Edysol Antiseptic Soap,
             H-Nal, M-Reg, Veins Flocher, Diabor, C-Candi, C-
             Cysta, Firas, D-Diab, P-Pile, Infecta, Ribacin Forte,
             Aloe Vera Cure Formula
Shell of Hen's Egg
Berberis (6. aristata, B. chitria, B. asiatica,
B. lyceum), Daruharidra
Greta powder
Candy Tamarind Candy
India
India
U.S.
(California)
U.S.
(Oklahoma)
1 4 mg/kg
Berberis:
Roots: 3.1-24.7 mg/kg
Stems: 8.0-23.8 mg/kg
Daruharidra:
16.9-49.8 mg/kg
770,000 ppm
Product: 0.15-3.61
mg/kg
Sharma et al.
(2009)
Srivastava et al.
(2006)
CDC (2002)
Lynch et al.
(2000)
                                                                       Stems: 0.36-2.5 mg/kg
                                                                       Wrappers: 459-27,125
                                                                       mg/kg
             Tamarind Candy
           U.S.
           (California)
 Product: 0.2-0.3 mg/kg
 Stems: 400 mg/kg
 Wrappers:
 16,000-21,000 mg/kg
Cosmetics
Lipsticks
Eye Shadows
U.S.
Nigeria
Average: 1.07 mg/kg
N.D. to 55 mg/kg
Heppet al.
(2009)
Omolaoye et al.
(201 Oa)
November 2012
4-30
Draft - Do Not Cite or Quote

-------
Product
Category
Pottery
Tobacco
Toys
Vitamins
Product
Foods prepared in Pb-glazed pottery
Smokeless Tobacco
Cigarette Tobacco (210Pb concentrations)
Red and yellow painted toy vehicles and tracks
535 PVC and non-PVC toys from day care centers
Soft plastic toys
Toy necklace
Soft plastic toys
Vitamins for young children, older children, and
pregnant or lactating women
Location of
Purchase
Mexico
U.K.
Pakistan
Brazil
U.S. (Nevada)
India
U.S.
Nigeria
U.S.
Pb Content (units)
N.D. to 3, 100 mg/kg
0.15-1.56mg/kg
Activity cone.: 7-20
Bq/kg
500-6,000 mg/kg
PVC: avg. 325 mg/kg
Non-PVC: avg. 89
mg/kg Yellow: 216
mg/kg
Non-yellow: 94 mg/kg
Average (by city):
21 -280 mg/kg
388,000 mg/kg
2.5-1 ,445 mg/kg
Average:
Young children:
2.9 ug/day
Older children:
1 .8 ug/day
Pregnant and lactating
women: 4.9 ug/day
Reference
Villalobos et al.
(2009)
McNeill et al.
(2006)
Tahir and
Alaamer (2008)
Godoi et al.
(2009)
Greenway and
Gerstenberger
(201 0)
Kumar and
Pastore (2007)
Meyer et al.
(2008)
Omolaoye et al.
(201 Ob)
Mindak et al.
(2008)
      "Hoodia gordonii, from Eastern Cape, South Africa Euterpe oleracea from Ninole Orchard, Ninole, Hawaii
      *Note that the country of origin is not provided because it was not published in the references cited.
      4.2        Kinetics

 1                   This section summarizes the empirical basis for understanding Pb toxicokinetics in
 2                   humans. The large amount of empirical information on Pb biokinetics in humans and
 3                   animal models has been integrated into mechanistic biokinetics models (U.S. EPA.
 4                   2006b). These models support predictions about the kinetics of Pb in blood and other
 5                   selected tissues based on the empirically-based information about Pb biokinetics. In
 6                   Section 4.3 (and Section 4.2.2.1). Pb biokinetics is described from the context of model
 7                   predictions.

 8                   The discussion of Pb toxicokinetics emphasizes inorganic Pb since this comprises the
 9                   dominant forms of Pb to which humans in the U.S. are currently exposed as a result of
10                   releases of Pb to the atmosphere and historic surface deposition of atmospheric Pb (see
11                   Section 3.2.2). The toxicokinetics of organic Pb is only briefly described and a more
12                   extensive discussion can be found in the 2006 Pb AQCD. Human exposures to organic
      November 2012
4-31
Draft - Do Not Cite or Quote

-------
 1                   Pb could occur in occupational settings (e.g., during manufacturing of tetraethyl Pb or
 2                   aviation fuels); however, environmental exposures to organic Pb compounds rarely occur
 3                   in the U.S. other than in the limited circumstances of those involved in fueling piston-
 4                   driven aircraft that use leaded aviation gasoline.
      4.2.1        Absorption

 5                   The major exposure routes of Pb in humans are inhalation and ingestion. Therefore, these
 6                   exposure routes are important in the discussion of Pb absorption (see Sections 4.2.1.1 and
 7                   4.2.1.2). The term "absorption" refers to the fraction of the amount of Pb ingested or
 8                   inhaled that is absorbed from the respiratory or gastrointestinal tract. The term
 9                   bioavailability, as it is used in this section, refers to the fraction of the amount of Pb
10                   ingested or inhaled that enters the systemic circulation. If properly measured (e.g., time-
11                   integrated blood Pb),  under most conditions Pb bioavailability is equivalent (or nearly
12                   equivalent) to Pb absorption. The time-integrated blood Pb (i.e., the integral of blood Pb
13                   over time) provides a useful measure of bioavailability because it reflects both recent Pb
14                   absorption as well as  contributions from Pb sequestered in soft tissue and bone.
15                   Bioaccessibility is a measure of the physiological solubility of Pb in the respiratory or
16                   gastrointestinal tract.  Pb must become bioaccessible in order for absorption to occur.
17                   Processes that contribute to bioaccessibility include physical transformation of Pb
18                   particles and dissolution of Pb compounds into forms that can be absorbed (e.g., Pb2+).
19                   Bioaccessibility is typically assessed by measuring the fraction of Pb in a sample that can
20                   be extracted into a physiological or physiological-like solution (e.g., gastric juice or
21                   solution similar to gastric juice).

22                   The 2006 Pb AQCD (U.S.  EPA. 2006c) also presented dermal absorption of inorganic
23                   and organic Pb compounds, which is generally considered to be much less than by
24                   inhalation or ingestion. A study published subsequent to the 2006 Pb AQCD measured
25                   rates of absorption of Pb in skin patches harvested from nude mice (PanetaL 2010).
26                   Following application of 12 mg Pb as Pb acetate  or Pb nitrate, the absorption rate
27                   (measured over a 10-hour observation period) was approximately 0.02 (ig Pb/cm2 per
28                   hour. Absorbed Pb was detected in liver and kidney of nude mice following a 120-hr
29                   occluded dermal application of approximately  14 mg Pb as either Pb acetate or Pb nitrate.
30                   Uptake of Pb into the skin at the site of application was greater when Pb acetate was
31                   applied to the skin compared to Pb nitrate; however, liver and kidney Pb concentrations
32                   observed at the conclusion of the study (120 hours following the application of Pb) were
33                   not different for the two Pb compound. No additional information provides evidence of
34                   dermal absorption being a major exposure route of environmental Pb.
      November 2012                              4-32                   Draft - Do Not Cite or Quote

-------
                     4.2.1.1      Inhalation

 1                   Systemic absorption of Pb deposited in the respiratory tract is influenced by particle size
 2                   and solubility, as well as by the pattern of regional deposition within the respiratory tract.
 3                   Fine particles (<1 (im) deposited in the bronchiolar and alveolar region can be absorbed
 4                   after extracellular dissolution or can be ingested by phagocytic cells and transported from
 5                   the respiratory tract (Bailey and Roy. 1994). Larger particles (>2.5 (im) that are primarily
 6                   deposited in the ciliated airways (nasopharyngeal and tracheobronchial regions) can be
 7                   transferred by mucociliary transport into the esophagus and swallowed, thus being
 8                   absorbed via the gut.

 9                   Inhaled Pb lodging deep in the respiratory tract seems to be absorbed equally  and totally,
10                   regardless of chemical form (Morrow et al..  1980; Chamberlain et al.. 1978; Rabinowitz
11                   et al.. 1977). Absorption half-times (tl/2) have been estimated for radon decay  progeny in
12                   adults who inhaled aerosols of Pb and bismuth isotopes generated from decay of 220Rn or
13                   222Rn. The absorption half-time for Pb from the respiratory tract to blood was  estimated
14                   to be approximately  10 hours in subjects who inhaled aerosols having an activity median
15                   particle diameter of approximately 160 nm (range 50-500 nm) (Marsh and Birchall.
16                   1999). and approximately 68 min for aerosols having diameters of approximately 0.3-
17                   3 nm (Butterweck et al.. 2002). Given the submicron particle size of the exposure, these
18                   rates are thought to represent, primarily, absorption from the  bronchiolar and  alveolar
19                   regions of the respiratory tract.

20                   Several studies have quantified the bioaccessibility of Pb in atmospheric PM,  based on
21                   various in vitro extraction methods. In a study of PMi0 and PM2 5 samples from
22                   downtown Vienna, Austria, Falta et al.  (2008) used synthetic gastric juice to investigate
23                   the bioaccessibility of metals including Pb. The rationale was that inhaled particles in the
24                   2.5-10 (im size range are mostly deposited in the tracheal and bronchial regions of the
25                   lung from where they are transported within hours by mucociliary clearance, i.e., they are
26                   mainly swallowed. In contrast, the <2.5 (im particles are deposited in the pulmonary
27                   alveoli  where they can stay for months  to years. The study aimed to determine the
28                   bioaccessibility of the 2.5-10 (im PM. It is important to note that they do not isolate the
29                   2.5-10 (im size range; instead, they infer the characteristics from the difference between
30                   the PM2 5 and PMi0 fractions. The Pb concentrations associated with the two fractions
31                   were almost identical, as was the percentage extracted by synthetic gastric juice (86% and
32                   83% Pb for PM2 5 and PMi0 fractions, respectively). The mean daily bioavailable mass
33                   was calculated to be  16 ng for the PM25.10 size range. Since the quantitative clearance of
34                   these particles to the stomach was assumed, this value  represents an upper estimate for
35                   the amount of bioavailable Pb. Niu et al. (2010) determined the bioaccessibility of Pb in
36                   fine (100-1,000 nm)  and ultrafine-sized (<100 nm) urban airborne PM from two sites
      November 2012                              4-33                   Draft - Do Not Cite or Quote

-------
 1                   within the city of Ottawa, Canada. For all size fractions, the median Pb concentrations for
 2                   particles smaller than 10 um were 8,800 and 7,800 mg/kg for the two different locations.
 3                   The bioaccessibility was based on ammonium acetate extractability and it was found that,
 4                   within the fine and ultrafme-size ranges, 13-28% Pb was extracted. The Falta et al.
 5                   (2008) and Niu et al. (2010) results illustrate that different extraction techniques result in
 6                   different bioaccessible fractions. The main finding from Niu et al. (2010) was that the
 7                   highest values (-28% and -19% for the two different locations) were found for the
 8                   <57 nm particles, with percent bioaccessibility decreasing with increasing particle size.
 9                   This result indicated that Pb was potentially most bioaccessible in the ultrafme-size
10                   range.

11                   A recent study by Barrett et al. (2010) investigated the solid phase speciation of Pb in
12                   urban road dust in Manchester, U.K., and considered the health implications of inhalation
13                   and ingestion of such material. Human exposure via inhalation is likely to involve only
14                   the finest grained fractions (up to 10 (im) and unfortunately this study characterized only
15                   the <38 (im fraction. Pb-goethite and PbCrO4 comprised the largest fractions, 45% and
16                   21% respectively, of Pb in the <38 (im fraction. These forms tend to be less bioaccessible
17                   if ingested compared with PbO or Pb acetate because they are less soluble.

18                   The above considerations indicate that the relationship between air Pb exposure and
19                   blood Pb will depend on numerous exposure variables (e.g., particle size, solubility,
20                   exposure frequency and duration)  and physiological variables (age, activity level,
21                   transport and absorption in the respiratory tract, blood Pb kinetics). For a detailed
22                   discussion of factors affecting particle deposition and retention in the human respiratory
23                   tract the reader is referred to Chapter 4 of 2009 PM ISA (U.S.  EPA. 2009a). Section 4.2.4
24                   of that document specifically addresses biological factors affecting particle deposition
25                   such as activity level and age with an emphasis on children. Mechanistic models provide
26                   one means for integrating these variables into predictions of blood Pb - air Pb
27                   relationships; although, predictions are subject to simplifications and generalizations
28                   made in constructing the models. As an example, the ICRP (Pounds and Leggett 1998;
29                   ICRP. 1994; Leggett. 1993) model (see Section 4.3 for a brief description) can be used to
30                   predict blood Pb - air Pb slopes for specific direct Pb inhalation exposure scenarios. For a
31                   long-term continuous (24 hours/day) exposure of a typical adult male engaged in light
32                   exercise (ventilation rate 20-22 mVday) to Pb-bearing particles having a 1 (im uniform
33                   particle size, the predicted blood Pb  - air Pb slopes range from 0.7 (ig/dL per (ig/m3 (for
34                   low solubility particles; e.g., Pb oxide) to 3 (ig/dL per (ig/m3 (for highly soluble Pb;
35                   e.g., Pb salts). These slopes were calculated by running ICRP model simulations with
36                   varying air concentrations (0.1-6 (ig/m3) to achieve a range of blood Pb concentrations
37                   up to 10 (ig/dL, starting with a baseline of 1.6 (ig/dL, and estimating the linear slope of
38                   the relationship between blood Pb concentration and air Pb. Empirical estimates of blood
      November 2012                              4-34                  Draft - Do Not Cite or Quote

-------
 1                   Pb - air Pb slopes for various populations, derived from epidemiological studies, are
 2                   summarized in Section 4.5.1.
                     Organic Pb

 3                   Alkyl Pb compounds can exist in ambient air as vapors. Inhaled tetraalkyl Pb vapor is
 4                   nearly completely absorbed following deposition in the respiratory tract. As reported in
 5                   the 2006 Pb AQCD (U.S. EPA. 2006c). a single exposure to vapors of radioactive (203Pb)
 6                   tetraethyl Pb resulted in 37% initially deposited in the respiratory tract, of which -20%
 7                   was exhaled in the subsequent 48 hours (Heard etal.. 1979). In a similar experiment
 8                   conducted with 203Pb tetramethyl Pb, 51% of the inhaled 203Pb dose was initially
 9                   deposited in the respiratory tract, of which -40% was exhaled in 48 hours (Heard et al..
10                   1979).

11                   Estimation of bioavailability of organic Pb is relevant to some aviation fuel exposures
12                   (e.g., persons exposed to leaded gasoline used in piston-engine aircraft). Mahaffey (1977)
13                   estimated that 40% of inhaled Pb in urban air (largely attributed to combustion of
14                   gasoline containing tetraethyllead) is bioavailable to adults. Chamberlain et al. (1975)
15                   suggested that 35% of inhaled combustion products of tetraethyl 203Pb fuel [likely to have
16                   been a mixture dominated by inorganic Pb halides, but may also have include alkly Pb
17                   species (U.S. EPA. 2006b)1 are deposited and then retained in adult lungs with a half-life
18                   of 6 hours. Fifty percent of that 203Pb was detectable in the blood within 50 hours of
19                   inhalation, and the rest was found to deposit in bone or tissue. Chamberlain et al.  (1975)
20                   estimated that continuous inhalation of Pb in engine exhaust from fuel containing
21                   tetraethyllead at a concentration of 0.001 ug/m3 for a period of months could produce a
22                   1  ug/dL increment in blood Pb.
                     4.2.1.2      Ingestion

23                   The extent and rate of GI absorption of ingested inorganic Pb are influenced by
24                   physiological states of the exposed individual (e.g., age, fasting, nutritional calcium
25                   (Ca2+) and iron (Fe) status, pregnancy) and physicochemical characteristics of the
26                   Pb-bearing material ingested (e.g., particle size, mineralogy, solubility). Pb absorption in
27                   humans may be a capacity-limited process, in which case the percentage of ingested Pb
28                   that is absorbed may decrease with increasing rate of Pb intake. Numerous observations
29                   of nonlinear relationships between blood Pb concentration and Pb intake in humans
30                   provide support for the likely existence of a saturable absorption mechanism or some
31                   other capacity-limited process in the distribution of Pb in humans  (Sherlock and Quinn,
32                   1986: Sherlock et al.. 1984: Pococketal.. 1983: Sherlock et al.. 1982). While evidence


      November 2012                             4-35                   Draft - Do Not Cite or Quote

-------
 1                  for capacity-limited processes at the level of the intestinal epithelium is compelling, the
 2                  dose at which absorption becomes appreciably limited in humans is not known.

 3                  In adults, estimates of absorption of ingested water-soluble Pb compounds
 4                  (e-g-, Pb chloride, Pb nitrate, Pb acetate) range from 3 to 10% in fed subjects (Maddaloni
 5                  etal.. 1998: Watson etal.. 1986: James etal.. 1985: Heard and Chamberlain. 1982:
 6                  Rabinowitz et al.. 1980). The absence of food in the GI tract increases absorption of
 7                  water-soluble Pb in adults. Reported estimates of soluble Pb absorption range from 26 to
 8                  70% in fasted adults (Maddaloni et al.. 1998: James etal.. 1985: Blake etal.. 1983: Heard
 9                  and Chamberlain. 1982: Rabinowitz et al.. 1980). Reported  fed:fasted ratios for soluble
10                  Pb absorption in adults range from 0.04 to 0.2 (James etal.. 1985: Blake etal.. 1983:
11                  Heard and Chamberlain. 1982: Rabinowitz et al.. 1980).

12                  Limited evidence demonstrates that GI absorption of water-soluble Pb is higher in
13                  children than in adults. Estimates derived from dietary balance studies conducted in
14                  infants and children (ages 2 weeks to 8 years) indicate that -40-50% of ingested Pb is
15                  absorbed (Ziegler et al.. 1978: Alexander etal.. 1974). Experimental studies provide
16                  further evidence for greater absorption of Pb from the gut in young animals compared to
17                  adult animals (Aungst et al.. 1981: Kostial et al.. 1978: Pounds etal.. 1978: Forbes  and
18                  Reina. 1972). The mechanisms for an apparent age difference in GI absorption of Pb have
19                  not been completely elucidated and may include both physiological and dietary factors
20                  (Mushak. 1991). To further investigate the effects of the presence of food in the GI tract
21                  on Pb absorption, children (3-5 years old) who ate breakfast had lower blood Pb levels
22                  compared to children who did not eat breakfast (Liu etal.. 201 la). This difference
23                  persisted after controlling for nutritional variables (blood iron [Fe], calcium [Ca2+],
24                  copper [Cu], magnesium [Mg], zinc [Zn]). This observation may be explained by lower
25                  GI absorption of Pb ingested with or in close temporal proximity to meals. Direct
26                  evidence for meals lowering GI absorption of Pb has also been reported for adults
27                  (Maddaloni et al.. 1998: James etal.. 1985).

28                  Nutritional interactions of Pb with dietary elements (e.g., Fe, Ca2+, Zn) are complex. Pb
29                  competes with other elements for transport and binding sites that can result in
30                  adjustments of homeostatic regulators to absorb and retain needed elements.
31                  Additionally, low levels of macronutrients may alter Pb bioaccessibility in the GI tract.
32                  Genetic variation in absorption and metabolism may modify all of the above.

33                  Children who are iron-deficient have higher blood Pb concentrations than similarly
34                  exposed iron-replete children, suggesting that iron deficiency may result in higher Pb
35                  absorption or, possibly, other changes in Pb biokinetics that contribute to altered blood
36                  Pb concentrations (Schell et al.. 2004: Marcus and Schwartz. 1987:  Mahaffey and
37                  Annest. 1986). Studies conducted in animal models have provided direct evidence for


      November 2012                             4-36                   Draft - Do Not Cite or Quote

-------
 1                  interactions between iron deficiency and increased Pb absorption, perhaps by enhancing
 2                  binding of Pb to iron-binding proteins in the intestine (Bannon et al.. 2003; Morrison and
 3                  Ouarterman.  1987; Barton etal.. 1978b). An analysis of data from a sample 448 woman
 4                  (age 20-55 years) did not find a significant association between iron body stores
 5                  (indicated from serum ferritin concentration) and blood Pb concentrations, although
 6                  depleted irons stores (serum ferritin of <12 (ig/L) was associated with higher blood
 7                  concentrations of Cd, cobalt (Co) and manganese (Mn) higher (Meltzer et al.. 2010).The
 8                  effects of iron nutritional status on blood Pb include changes in blood Pb concentrations
 9                  in association with genetic variation in genes involved in iron metabolism. For example,
10                  genetic variants in the hemochromatosis (HFE) and transferrin genes are associated with
11                  higher blood  Pb  concentrations in children (Hopkins et al.. 2008). In contrast, HFE gene
12                  variants are associated with lower bone and blood Pb levels in elderly men (Wright et al..
13                  2004).

14                  Several studies have suggested that dietary Ca2+ may have a protective role against Pb by
15                  decreasing absorption of Pb in the GI tract and by decreasing the mobilization of Pb from
16                  bone stores to blood. In experimental studies of adults, absorption of a single dose of Pb
17                  (100-300 (ig Pb chloride) was lower when the Pb was ingested together with
18                  Ca2+ carbonate (0.2 g Ca2+ carbonate) than when the Pb was ingested without additional
19                  Ca2+ (Blake and  Mann. 1983; Heard and Chamberlain. 1982). A similar effect of Ca2+
20                  occurs in rats (Barton et al.. 1978a).  Similarly, an inverse relationship was observed
21                  between dietary  Ca2+ intake and blood Pb concentration in children, suggesting that
22                  children who are Ca2+-deficient may absorb more Pb than Ca2+-replete children (Elias et
23                  al.. 2007; Schell et al.. 2004; Mahaffev et al.. 1986; Ziegleretal.. 1978). These
24                  observations  suggest that Ca2+  and Pb share and may compete for common binding and
25                  transport mechanisms in the small intestine which are regulated in response to dietary
26                  Ca2+ and Ca2+-body stores (Fullmer and Rosen. 1990; Bronner et al.. 1986). However,
27                  animal studies have also shown that multiple aspects  of Pb toxicokinetics are affected by
28                  Ca2+ nutritional status. For example, feeding rats a Ca2+ deficient diet is associated with
29                  increased Pb  absorption, decreased whole body Pb clearance, and increased volume of
30                  distribution of Pb (Aungst and Fung. 1985).  These studies suggest that associations
31                  between Ca2+ nutrition and blood Pb that have been observed in human populations may
32                  not be solely  attributable to effects of Ca2+ nutrition on Pb absorption. Other potential
33                  mechanisms by which Ca2+ nutrition may affect blood Pb and Pb biokinetics include
34                  effects on bone mineral metabolism  and renal function.

35                  Blood Pb concentrations in young children have also been shown to increase in
36                  association with lower dietary Zn levels (Schell et al., 2004). Mechanisms for how Zn
37                  affects blood Pb concentration, i.e., whether it involves changes  in absorption or changes
38                  in distribution and/or elimination of Pb, have not been determined.
      November 2012                             4-37                  Draft - Do Not Cite or Quote

-------
 1                   Dissolution of Pb from the soil/mineralogical matrix in the stomach appears to be the
 2                   major process that renders soil Pb bioaccessible for absorption in the GI tract. Absorption
 3                   of Pb has been shown to vary depending upon the Pb mineralogy and physical
 4                   characteristics of the Pb in the soil (e.g., encapsulated or exposed) and size of the
 5                   Pb-bearing grains. GI absorption of larger Pb-containing particles (MOO urn) tends to be
 6                   lower than smaller particles (Healy et al.. 1992; Barltrop and Meek. 1979). Absorption of
 7                   Pb in soils and dust has been most extensively studied in the in vivo swine model. Gastric
 8                   function of swine is thought to be sufficiently similar to that of humans to justify use of
 9                   swine as a model for assessing factors that may affect GI absorption of Pb from soils in
10                   humans (Juhasz et al.. 2009; U.S. EPA. 2007b: Casteel et al.. 2006; Casteel et al.. 1997;
11                   Weis and Lavelle.  1991). Other practical advantages of the swine model over rodent
12                   models have been described, and include: absence of coprophagia; ease with which Pb
13                   dosing can be administered and controlled; and higher absorption fraction of soluble Pb
14                   (e-g-, Pb acetate) in swine, which is more similar to humans than rats (Smith et al.,
15                   2009a). The swine studies measure blood and/or tissue Pb (e.g., kidney, liver, bone)
16                   concentrations following oral dosing of swine with  either soil or with a highly water
17                   soluble and fully bioaccessible form of Pb (e.g., Pb  acetate). A comparison of the internal
18                   concentrations of Pb under these two conditions provides a measure of the bioavailability
19                   (i.e., absorption) of Pb in soil relative to that of Pb acetate, which is typically referred to
20                   as relative bioavailability (RBA).  Relative bioavailability measured in the swine assay is
21                   equivalent to the ratio of the absorbed fraction (AF) of ingested dose of soil Pb to that of
22                   water-soluble Pb acetate (e.g., RBA = AFSoiipb/AFPbacetate)-

23                   Collectively, published studies conducted in swine have provided 39 estimates of Pb
24                   RBA for 38 different soil or "soil-like" test materials (Bannon et al.. 2009; Smith et al..
25                   2009a; Casteel et al.. 2006; Marschner et al.. 2006). The mean of RBA estimates from 25
26                   soils is 0.49 (± 0.29[SD]), median is 0.51, and 5th to 95th percentile range is 0.12 to
27                   -0.89. RBA estimates for soils collected from 8 firing ranges were approximately 1.0
28                   (Bannon et al.. 2009). The relatively high RBA for the firing range soils may reflect the
29                   high abundance  of relatively un-encapsulated Pb carbonate (30-90% abundance) and Pb
30                   oxide (1-60%) in these soils. Similarly, a soil sample (low Pb concentration) mixed with
31                   a NIST paint standard (55% Pb carbonate, 44% Pb oxide) also had a relatively high
32                   bioavailability (0.72) (Casteel et al.. 2006). Samples of smelter slag, or soils in which the
33                   dominant source of Pb was smelter slag, had relatively low RBA (0.14 - 0.40, n = 3), as
34                   did a sample from a mine tailings pile (RBA =  0.06), and a sample of finely ground
35                   galena mixed with soil (Casteel et al.. 2006).

36                   Based on data for 18 soil materials assayed in swine, RBA of Pb mineral phases were
37                   categorized into "low" (<0.25 [25%]), "medium" (0.25-0.75 [25 to 75%]), and "high"
38                   (>0.75 [75%]) categories (Casteel et al.. 2006). Figure 4-3 shows some of the materials
      November 2012                             4-38                  Draft - Do Not Cite or Quote

-------
1

2

3

4

5
that fall into these three categories. Mineral phases observed in mineralogical wastes can

be expected to change overtime (i.e., weathering), which could change the RBA over

time. The above observations in swine are supported by various studies conducted in rats

that have found RBA of Pb in soils to vary considerably and to be less than 1.0 (Smith et

al.. 2009a. 2008; Freeman et al.. 1996; Freeman et al.. 1994; Freeman et al.. 1992).
                CO
                T3
                 re
                 0)
                 o
                 0)
                 Q.


                 Q.
                 3
                 O
                O

                -o
                 Q)

                "re

                 E
                m
                                                 Group
     Note: based on results from juvenile swine assays.

     Source: Casteel et al. (2006).
     Figure 4-3      Estimated relative bioavailability (RBA, compared to Pb acetate)
                      of ingested Pb in mineral groups.
    November 2012
                          4-39
Draft - Do Not Cite or Quote

-------
 1                  Drexler and Brattin (2007) developed an in vitro bioaccessibility (IVBA) assay for soil
 2                  Pb that utilizes extraction fluid comprised of glycine, deionized water, and hydrochloric
 3                  acid at a pH of 1.50 that is combined with sieved test material (<250 urn) for 1 hour. The
 4                  assay was tested for predicting in vivo RBA of 18 soil-like test materials that were
 5                  assayed in a juvenile  swine assay (Casteel et al., 2006). A regression model relating
 6                  IVBA and RBA was  derived based on these data (Equation 4-1):


                                            RBA = (0.878 X IVBA) - 0.028

                                                                                          Equation 4-1

 7                  where RBA and IVBA are expressed as  fractions (i.e., not as percent). The weighted r2
 8                  for the relationship (weighted for error in the IVBA and RBA estimates) was 0.924
 9                  (p <0.001). The IVBA assay reported in Drexler and Brattin (2007) has been identified by
10                  the U.S. EPA as a validated method for predicting RBA of Pb in soils for use in risk
11                  assessment (U.S. EPA. 2007e). A review of soil Pb RBA estimates made using the IVBA
12                  assay described above and Equation 4-1 identified 270 estimates of Pb RBA in soils
13                  obtained from 11 hazardous waste sites. The mean for the site-wide RBA estimates
14                  (n = 11 sites) was 0.57 (SD 0.15), median was 0.63, and 5th to 95th percentile range was
15                  0.34 to 0.71.

16                  Equation 4-1 cannot be reliably extrapolated to other in vitro assays that have been
17                  developed for estimating Pb bioaccessibility without validation against in vivo RBA
18                  measurements made on the same test materials. Comparisons of outcomes among
19                  different in vitro assays applied to the same soil test materials have found considerable
20                  variability in IVBA estimates (Juhasz etal., 2011; Smith etal., 2011; Saikat et al., 2007;
21                  Van de Wiele et al.. 2007). This variability has been attributed to differences in assay
22                  conditions, including pH, liquid:soil ratios, inclusion  or absence of food material, and
23                  differences in methods used to separate dissolved and particle-bound Pb
24                  (e.g., centrifugation versus filtration). Smith et al. (2011) found that algorithms for
25                  predicting RBA based on two different IVBA assays  did not yield similar predictions of
26                  RBA when applied to the same material. Given the dependence of IVBA outcomes on
27                  assay conditions, in vitro assays used to  predict in vivo RBA should be evaluated against
28                  in vivo RBA estimates to quantitatively  assess uncertainty in RBA predictions (U.S.
29                  EPA. 2007e).

30                  Absorption of Pb in house dusts has not been rigorously evaluated quantitatively in
31                  humans or in experimental animal models. The RBA for paint Pb mixed with soil was
32                  reported to be approximately 0.72 (95% CI: 0.44, 0.98) in juvenile swine, suggesting that
33                  paint Pb dust reaching the gastrointestinal tract maybe highly bioavailable (Casteel et al.,


      November 2012                             4-40                   Draft - Do Not Cite or Quote

-------
 1                  2006). The same material yielded a bioaccessibility value (based on IVBA assay) of 0.75
 2                  (Drexler and Brattin. 2007). which corresponds to a predicted RBA of 0.63, based on
 3                  Equation 4-1. A review of indoor Pb RBA estimates made using the IVBA assay and
 4                  Equation 4-1 identified 100 estimates of Pb RBA in dusts obtained from two hazardous
 5                  waste sites. Mean Pb RBAs for the Herculaneum site were 0.47 (SD 0.07, 10 samples)
 6                  for indoor dust and 0.69 (SD 0.03, 12 samples) for soil. At the Omaha site, mean Pb
 7                  RBAs were 0.73 (SD 0.10, 90 samples) for indoor dust and 0.70 (SD 0.10, 45 samples)
 8                  for soil. Yu et al. (2006) applied an IVBA method to estimate bioaccessibility of Pb in
 9                  house dust samples collected from 15 urban homes. Homes were selected for inclusion in
10                  this study based on reporting to the state department of health of at least on child with a
11                  blood Pb concentration >15 (ig/dL and Pb paint dust may have contributed to indoor dust
12                  Pb. The mean  IVBA was 0.65 (SD 0.08, age: 52.5 to 77.2 months).

13                  The above results, and the IVBA assays used in studies of interior dust, have not been
14                  evaluated against in vivo RBA estimates for dust samples. Although, expectations are
15                  that a validated IVBA methodology for soil would perform well for predicting RBA of
16                  interior dust, this validation has not actually been experimentally confirmed. Factors that
17                  may affect in vivo predictions of RBA of interior dust Pb could include particle size
18                  distribution of interior dust Pb and the composition of the dust matrix, which may be
19                  quite different from that of soil.

20                  Other estimates of bioaccessibility of Pb in house dusts have been reported, based on
21                  results from in vitro extraction assays that have not been validated for predicting in vivo
22                  bioavailability. Bioaccessibility assays that sequentially extract soil at gastric pH
23                  followed by intestinal pH tend to show higher bioaccessibility of soil and dust Pb when
24                  incubated at gastric conditions (Juhasz et al., 2011; Lu et al.. 2011; Smith et al.. 2011;
25                  Roussel et al.. 2010; Yu et al.. 2006). Yu et al. (2006)  dissolved Pb dust,  obtained from
26                  vacuuming carpet samples into simulated gastric and intestinal acids (also
27                  Section 4.1.3.2). The carpet samples were obtained from homes located in northern
28                  New Jersey. Pb concentration in carpet ranged from 209 to 1,770 mg/kg dust, with
29                  52-77% of Pb  dissolving in simulated gastric acid and 5-32% dissolving in simulated
30                  intestinal acids. In a similar test in the U.K., Turner and Simmonds (2006) observed
31                  median Pb dust concentrations of 178 mg/kg with approximately 80% bioaccessibility in
32                  simulated gastric acid. Jin et al. (2005) observed that bioaccessibility of Pb in soil was
33                  proportional to the soil acidity and organic matter content of the soil.
      November 2012                             4-41                  Draft - Do Not Cite or Quote

-------
      4.2.2        Distribution

 1                  A simple conceptual representation of Pb distribution is that it contains a fast turnover
 2                  pool, comprising mainly soft tissue, and a slow pool, comprising mainly skeletal tissues
 3                  (Rabinowitz et al.. 1976). The highest soft tissue concentrations in adults occur in liver
 4                  and kidney cortex (Gerhardsson et al.. 1995; Oldereid et al.. 1993; Gerhardsson et al..
 5                  1986; Barry. 1975; Gross et al..  1975). Pb in blood (i.e., plasma) exchanges with both of
 6                  these compartments.
                    4.2.2.1      Blood

 7                  Blood comprises ~1% of total Pb body burden. Pb in blood is found primarily (>99%) in
 8                  the RBCs (Smith et al.. 2002; Manton et al.. 2001; Bergdahl et al.. 1999; Bergdahl et al..
 9                  1998; Hernandez-Avila et al.. 1998; Bergdahl et al.. 1997a; Schutzetal. 1996).
10                  5-aminolevulinic acid dehydratase (ALAD) is the primary binding ligand for Pb in
11                  erythrocytes (Bergdahl et al.. 1998; Xieetal.. 1998; Bergdahl et al.. 1997a; Sakai et al..
12                  1982). Two other Pb-binding proteins have been  identified in the RBC, a 45 kDa protein
13                  (Kmax 700 (ig/dL; Kd 5.5 (ig/L) and a smaller protein band having a molecular weight of
14                  <10 kDa (Bergdahl et al.. 1998; Bergdahl et al.. 1997a; Bergdahl et al.. 1996). Of the
15                  three principal Pb-binding proteins identified in RBCs, ALAD has the strongest affinity
16                  for Pb (Bergdahl etal.. 1998) and appears to dominate the ligand distribution of Pb (35 to
17                  84% of total erythrocyte Pb) at blood Pb levels below 40 (ig/dL (Bergdahl  et al.. 1998;
18                  Bergdahl et al.. 1996; Sakai etal.. 1982). Pb binding to ALAD is saturable; the binding
19                  capacity was estimated to be -850 (ig/dL RBCs (or ~40 (ig/dL whole blood) and the
20                  apparent dissociation constant has been estimated to be ~1.5 (ig/L (Bergdahl et al.. 1998).
21                  Hematocrit is somewhat higher in the neonate at birth (51%) than in later infancy (35% at
22                  6 months), which may lead to a decrease in the total binding capacity of blood over the
23                  first 6 months of life that results in a redistribution of Pb among other tissues (Simon et
24                  al.. 2007).

25                  Saturable binding to RBC proteins contributes to an increase in the plasma/blood Pb ratio
26                  with increasing blood Pb concentration and curvature to the  blood Pb-plasma Pb
27                  relationship (Rentschler etal.. 2012;  Kang et al..  2009; Jin et al.. 2008; Barbosaet al..
28                  2006b; Smith et al.. 2002; Manton etal.. 2001; Bergdahl et al.. 1999; Bergdahl et al..
29                  1998; Bergdahl et al.. 1997b; DeSilva. 1981). An example of this is shown in Figure 4-4.
30                  Saturable binding of Pb to RBC proteins has several important consequences. As blood
31                  Pb increases and the higher affinity binding sites  for Pb in RBCs become saturated, a
32                  larger fraction of the blood Pb is available in plasma to distribute to brain and other
33                  Pb-responsive tissues. This change in distribution of Pb contributes to a curvature in the


      November 2012                             4-42                 Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5

 6
 7
 8
 9
10
11
12
13
relationship between Pb intake (at constant absorption fraction) and blood Pb
concentration. Plasma Pb also exhibits faster kinetics. Following exposures of 5 adults
that resulted in relatively high blood Pb concentrations (56-110 (ig/dL), the initial (fast-
phase) elimination half-time for plasma Pb (38 ± 20 [SD] days) was approximately half
that of blood (81 ± 25 days) (Rentschler etal. 2012).

Typically, at blood Pb concentrations <100 (ig/dL, only a small fraction (<1%) of blood
Pb is found in plasma (Marcus. 1985; Manton and Cook. 1984; DeSilva. 1981). However,
as previously noted, plasma Pb may be the more biologically labile and lexicologically
effective fraction of the circulating Pb. Approximately 40-75% of Pb in the plasma is
bound to proteins, of which albumin appears to be the dominant ligand (Al-Modhefer et
al., 1991; Ong and Lee. 1980a). Pb in serum that is not bound to protein exists largely as
complexes with low molecular weight sulfhydryl compounds (e.g., cysteine,
homocysteine) and other ligands (Al-Modhefer et al.,  1991).
                           0
                                          oAdults  • Children
                  20         40         60
                            Blood Pb (Mg/dL)
    80
100
     Source: Adapted with permission of Elsevier Publishing and the Finland Institute of Occupational Health, Bergdahl et al. (1999:
     1997b).
     Figure 4-4       Plot of blood and plasma Pb concentrations measured in adults
                        and children.
     November 2012
                           4-43
Draft - Do Not Cite or Quote

-------
1
2
3
4
5
6
7
As shown in Figure 4-4. the limited binding capacity of Pb binding proteins in RBCs
produces a curvilinear relationship between blood and plasma Pb concentration. The
limited binding capacity of RBC binding proteins also confers, or at least contributes, to a
curvilinear relationship  between Pb intake and blood Pb concentration. A curvilinear
relationship between Pb intake and blood Pb concentration has been observed in children
(Sherlock and Quinn. 1986; Lacevetal.. 1985; Ryuetal.. 1983). As shown in Figure 4-5.
the relationship becomes pseudo-linear at relatively low daily Pb intakes
(i.e., <10 ug/day/kg) and at blood Pb concentrations <25 (ig/dL.
                          0
                    100            200           300
                           Pb Intake (|jg/day)
              400
     Data represent mean and standard errors for intake; the line is the regression model (blood Pb = 3.9 + 2.43 (Pb intake [|jg/week]  ).
     Source: Adapted with permission of Taylor & Francis Publishing, Sherlock and Quinn (1986).
     Figure 4-5       Relationship between Pb  intake and blood Pb concentration in
                       infants  (n = 105, age 13 weeks, formula-fed).
     November 2012
                           4-44
Draft - Do Not Cite or Quote

-------
 1                  Figure 4-6 shows the predicted relationship between quasi-steady state blood and plasma
 2                  Pb concentrations in a 4-year old child using the ICRP model [(Pounds and Leggett.
 3                  1998; ICRP. 1994; Leggett.  1993). see Section 43 for a brief description of the ICRP
 4                  model]. The abrupt inflection point that occurs at approximately 25 (ig/dL blood Pb is an
 5                  artifact of the numerical approach to simulate the saturation of binding using
 6                  discontinuous first-order rate constants for uptake and exit of Pb from the RBC. A
 7                  continuous function of binding sites and affinity, using empirical estimates of both
 8                  parameters, yield a similar but continuous curvature in the relationship (Bergdahl et al.
 9                  1998; O'Flaherty. 1995). Nevertheless, either approach predicts a pseudo-linear
10                  relationship at blood Pb concentrations below approximately 25 (ig/dL which, in this
11                  model, corresponds to an intake of approximately 100 (ig/day (absorption rate
12                  ~ 30 (ig/day) (upper panel). An important consequence of the limited Pb-binding capacity
13                  of RBC proteins is that the plasma Pb concentration will continue to grow at a linear rate
14                  above the saturation point for RBC protein binding. One implication of this is that a
15                  larger fraction of the Pb in blood will become  available to distribute to brain and other
16                  Pb-responsive tissues  as blood Pb increases. This could potentially contribute to
17                  non-linearity in dose-response relationships in studies in which blood Pb is the used as
18                  the internal dose metric.
      November 2012                             4-45                  Draft - Do Not Cite or Quote

-------
                  0.60
                  0.50 -
               3) 0.40 -
               °-  0.30 -
               ra
               E

               ra  0.20 -
               o.
                  0.10 -


                  0.00
                10      20       30      40

                        Blood Pb (ng/dL)
                                                                 50
                  60
   50 -

IT

1) 40 -
^
si

£ 30 H
o

m 20 -


   10 -
                            	Blood
                              •Plasma
                                                     1.0
                                                                    0.8
                                                                        0)
                                                                        tn
                                                                    0.6  3
                                                                        0)
                                                                        TJ
                                                                    0-4
                                                                    0.2
                                                                    0.0
                               100       200       300

                                        Intake (tig/day)
                                                 400
Note: Model simulations are for a 4-year old having from birth a constant Pb intake of between 1 and 400 ug/day. Simulation based
on ICRP Pb biokinetics model (Leggett. 1993).


Figure 4-6      Simulation of quasi-steady state blood and plasma Pb
                 concentrations in a child (age 4 years) associated with varying Pb

                 ingestion rates.
November 2012
                        4-46
Draft - Do Not Cite or Quote

-------
 1                  Studies conducted in swine provide additional evidence in support of RBC binding
 2                  kinetics influencing distribution of Pb to tissues. In these studies, the relationship
 3                  between the ingested dose of Pb and tissue Pb concentrations (e.g., liver, kidney, bone)
 4                  was linear, whereas, the relationship between dose and blood Pb was curvilinear with the
 5                  slope decreasing as the dose increased (Casteel et al.. 2006). Saturable binding of Pb to
 6                  RBC proteins also contributes to a curvilinear relationship between urinary Pb excretion
 7                  and plasma Pb concentration (Section 4.2.3) (Besser et al., 2008; Bergdahl et al., 1997b).
                    4.2.2.2     Bone

 8                  The dominant compartment for Pb in the body is in bone. In human adults, more than
 9                  90% of the total body burden of Pb is found in the bones, whereas bone Pb accounts for
10                  just under 60% of the body burden in infants less than a year old and just over 70% of the
11                  body burden in older children (Barry. 1975). Bone is comprised of two main types,
12                  cortical (or compact) and trabecular (or spongy or cancellous). The proportion of cortical
13                  to trabecular bone in the human body varies by age, but on average is about 80 to 20
14                  percent (O'Flahertv. 1998; Leggett 1993; ICRP.  1973).

15                  The exchange of Pb from plasma to the bone surface is a rapid process (i.e., adult ti/2
16                  =0.19 and 0.23  hours for trabecular and cortical bone, respectively)  (Leggett, 1993).
17                  Some Pb diffuses from the bone surface to deeper bone regions (adult ti/2=150 days)
18                  where it is relatively inert (in adults) and part of a "nonexchangeable" (removed only
19                  through bone resorption/remodeling) pool of Pb in bone (Leggett. 1993).

20                  Pb distribution  in bone includes uptake into cells that populate bone (e.g., osteoblasts,
21                  osteoclasts, osteocytes) and exchanges with proteins and minerals in the extracellular
22                  matrix (Pounds etal. 1991). Pb forms highly stable complexes with phosphate and can
23                  replace calcium in the calcium-phosphate salt,  hydroxyapatite, which comprises the
24                  primary crystalline matrix of bone (Meirer et al.. 2011; Bres etal.. 1986; Miyake. 1986;
25                  Verbeeck et al., 1981). Several intracellular kinetic pools of Pb have been described in
26                  isolated cultures of osteoblasts and osteoclasts  which appear to reflect physiological
27                  compartmentalization within the cell, including membranes, mitochondria, soluble
28                  intracellular binding proteins, mineralized Pb (i.e., hydroxyapatite) and inclusion bodies
29                  (Long etal.. 1990; Pounds and Rosen. 1986; Rosen. 1983). Approximately 70-80% of Pb
30                  taken up into isolated primary cultures of osteoblasts or osteocytes is associated with
31                  mitochondria and mineralized Pb (Pounds etal.. 1991).

32                  Pb accumulates in bone regions having the most active calcification at the time of
33                  exposure. Pb accumulation is thought to occur predominantly in trabecular bone during
34                  childhood and in both cortical and trabecular bone in adulthood (Aufderheide and

      November 2012                             4-47                   Draft - Do Not Cite or Quote

-------
 1                  Wittmers. 1992). Early Pb uptake in children is greater in trabecular bone due to its larger
 2                  surface area and higher metabolic rate. With continued exposure, Pb concentrations in
 3                  bone may increase with age throughout the lifetime beginning in childhood, indicative of
 4                  a relatively slow turnover of Pb in adult bone (Park et al.. 2009c: Barry and Connolly.
 5                  1981; Barry. 1975; Gross etal.. 1975; Schroeder and Tipton. 1968). The cortical and
 6                  trabecular bones have distinct rates of turnover and Pb release. For example, tibia has a
 7                  turnover rate of about 2% per year whereas trabecular bone has a turnover rate of more
 8                  than 8% per year in adults (Rabinowitz. 1991).

 9                  A high bone formation rate in early childhood results in the rapid uptake of circulating Pb
10                  into mineralizing bone; however, bone Pb is also recycled to other tissue compartments
11                  or excreted in accordance with a high bone resorption rate (O'Flaherty. 1995). Thus, most
12                  of the Pb acquired early in life  is  not permanently fixed in the bone (60-65%)
13                  (O'Flahertv. 1995; Leggett. 1993; ICRP. 1973). However, some  Pb accumulated in bone
14                  does persist into later life. McNeill et al. (2000) compared tibia Pb levels and cumulative
15                  blood Pb indices in a population of 19- to 29-year-olds who had been highly exposed to
16                  Pb in childhood from the Bunker Hill, Idaho smelter; they concluded that Pb from
17                  exposure in early childhood had persisted in the bone matrix until adulthood.

18                  Additional discussion of the Pb in bone and its mobilization are provided in other
19                  sections of this chapter. Maternal mobilization of Pb from the bone to the fetus is
20                  discussed in Section 4.2.2.4. The  relationship between Pb in blood and bone is discussed
21                  in Section 4.3.5.
                    4.2.2.3     Soft Tissues

22                  Most of the Pb in soft tissue is in liver and kidney (Gerhardsson et al.. 1995; Oldereid et
23                  al.. 1993; Gerhardsson et al.. 1986; Barry. 1981. 1975; Gross etal.. 1975). Presumably,
24                  the Pb in these soft tissues (i.e., kidney, liver, and brain) exists predominantly bound to
25                  protein. High affinity cytosolic Pb-binding proteins have been identified in rat kidney and
26                  brain (DuVal and Fowler. 1989; Fowler. 1989). The Pb-binding proteins in rat are
27                  cleavage products of a2(i globulin, a member of the protein superfamily known as
28                  retinol-binding proteins that are generally observed only in male rats (Fowler and DuVal.
29                  1991). Other high-affinity Pb-binding proteins (Kd ~14 nM) have been isolated in human
30                  kidney, two of which have been identified as a 5 kDa peptide, thymosin 4 and a 9 kDa
31                  peptide, acyl-CoA binding protein (Smith etal.. 1998). Pb also binds to metallothionein,
32                  but does not appear to be a significant inducer of the protein in comparison with the
33                  inducers Cd and Zn (Waalkes and Klaassen. 1985; Eaton et al.. 1980).
      November 2012                             4-48                  Draft - Do Not Cite or Quote

-------
 1                  The liver and kidneys rapidly accumulate systemic Pb (ti/2=0.21 and 0.41 hours,
 2                  respectively), which amounts to 10-15% and 15-20% of intravenously injected Pb,
 3                  respectively (Leggett 1993). A linear relationship in dose-tissue Pb concentrations for
 4                  kidney and liver has been demonstrated in swine, dogs, and rats (Smith et al.. 2008;
 5                  Casteel et al.. 2006; Casteel et al., 1997; Azaretal.. 1973). In contrast to Pb in bone,
 6                  which accumulates Pb with continued exposure in adulthood, concentrations in soft
 7                  tissues (e.g., liver and kidney) are relatively constant in adults (Treble and Thompson.
 8                  1997; Barry. 1975). reflecting a faster turnover of Pb in soft tissue relative to bone.
                    4.2.2.4      Fetus

 9                  Evidence for maternal-to-fetal transfer of Pb in humans is derived from cord blood Pb to
10                  maternal blood Pb ratios (i.e., cord blood Pb concentration divided by mother's blood Pb
11                  concentration). Group mean ratios range from about 0.7 to 1.0 at the time of delivery for
12                  mean maternal blood Pb levels ranging from 1.7 to 8.6 (ig/dL (Amaral et al.. 2010;
13                  Kordas et al.. 2009; Patel and Prabhu. 2009; Carbone et al.. 1998; Gover. 1990; Graziano
14                  et al.. 1990). In a study of 159 mothers having blood Pb levels of less than 14 (ig/dL,
15                  based on a linear regression of maternal blood Pb and cord blood Pb, the ratio of cord
16                  blood Pb to maternal blood Pb appeared to decrease with decreasing maternal blood Pb
17                  from 1.0 at 10 (ig/dL to 0.34 at 3 (ig/dL (Carbone et al.. 1998). A ratio of 0.34 is lower
18                  than reported based on mean data in other studies. However, consistent with other
19                  studies, the ratio of mean cord blood Pb (4.87 ug/dL) to mean maternal blood Pb (5.81
20                  ug/dL) was 0.84. In addition, the similarity of isotopic ratios in maternal blood and in
21                  blood and urine of newly-born infants provide further evidence for placental transfer of
22                  Pb to the fetus (Gulson et al.. 1999).

23                  Transplacental transfer of Pb may be facilitated by an increase in the plasma/blood Pb
24                  concentration ratio during pregnancy (Montenegro et al.. 2008; Lamadrid-Figueroa et al..
25                  2006). Maternal-to-fetal transfer of Pb appears to be related partly to the mobilization of
26                  Pb from the maternal skeleton. Evidence for transfer of maternal bone Pb to the fetus has
27                  been provided by stable Pb isotope studies in cynomolgus monkeys exposed during
28                  pregnancy. Approximately  7-39% of the maternal Pb burden transferred to the fetus was
29                  derived from the maternal skeleton, with the remainder derived from contemporaneous
30                  exposure (O'Flaherty. 1998; Franklin et al.. 1997). The upper value  in the range (39%)
31                  represented the one monkey with historical Pb exposure, but received only small amounts
32                  of environmental Pb exposure during pregnancy; for the monkeys that received high
33                  doses of Pb during pregnancy, the range was lower (7-25%) (O'Flahertv. 1998; Franklin
34                  etal.. 1997).
      November 2012                             4-49                  Draft - Do Not Cite or Quote

-------
                    4.2.2.5     Organic Pb

 1                  Information on the distribution of Pb in humans following exposures to organic Pb is
 2                  extremely limited. However, as reported in the 2006 Pb AQCD (U.S. EPA. 2006c). the
 3                  available evidence demonstrates near complete absorption following inhalation of
 4                  tetraalkyl Pb vapor and subsequent transformation to trialkyl Pb metabolites. One hour
 5                  following brief inhalation exposures to 203Pb tetraethyl or tetramethyl Pb (1 mg/m3),
 6                  -50% of the 203Pb body burden was associated with liver and 5% with kidney; the
 7                  remaining 203Pb was widely distributed throughout the body (Heard etal.. 1979). The
 8                  kinetics of 203Pb in blood showed an initial declining phase during the first 4 hours
 9                  (tetramethyl Pb) or 10 hours (tetraethyl Pb) after the exposure, followed by a
10                  reappearance of radioactivity back into the blood after ~20 hours. The high level of
11                  radioactivity initially in the plasma indicates the presence of tetraalkyl/trialkyl Pb. The
12                  subsequent rise in blood radioactivity, however, probably represents water-soluble
13                  inorganic Pb and trialkyl and dialkyl Pb compounds that were formed from the metabolic
14                  conversion of the volatile parent compounds (Heard etal.. 1979).

15                  Alkyl Pb compounds undergo oxidative dealkylation catalyzed by cytochrome P450 in
16                  liver and, possibly, in other tissues. Trialkyl Pb metabolites have been found in the liver,
17                  kidney, and brain following exposure to the tetraalkyl compounds in workers
18                  (Bolanowska et al.. 1967); these metabolites have also been detected in brain tissue of
19                  nonoccupational subjects (Nielsen et al.. 1978).
      4.2.3        Elimination

20                  The rapid-phase (30-40 days) of Pb excretion amounts to 50-60% of the absorbed
21                  fraction (Chamberlain et al.. 1978; Rabinowitz et al., 1976; Kehoe. 1961a. b. c).
22                  Absorbed Pb is excreted primarily in urine and feces, with sweat, saliva, hair, nails, and
23                  breast milk being minor routes of excretion (Kehoe. 1987; Chamberlain et al..  1978;
24                  Rabinowitz et al.. 1976; Griffin etal.. 1975; Hurshetal.. 1969; Hursh and Suomela.
25                  1968).

26                  Approximately 30% of intravenously injected Pb in humans (40-50% in beagles and
27                  baboons) is excreted via urine and feces during the first 20 days following administration
28                  (Leggett 1993). The kinetics of urinary excretion following a single dose of Pb is similar
29                  to that of blood (Chamberlain et al.. 1978). likely due to the fact that Pb in urine derives
30                  largely from Pb in plasma. Evidence for this is the observation that urinary Pb  excretion
31                  is strongly correlated with the rate of glomerular filtration of Pb (Araki etal.. 1986) and
32                  plasma Pb concentration (Rentschler et al., 2012; Bergdahl et al., 1997b) (i.e.,  glomerular
      November 2012                             4-50                  Draft - Do Not Cite or Quote

-------
 1                   filtration rate x plasma Pb concentration), and both relationships are linear. While the
 2                   relationship between urinary Pb excretion and plasma Pb concentration is linear, the
 3                   plasma Pb relationship to blood Pb concentration is curvilinear (as described in Section
 4                   4.2.2.1 and demonstrated in Figure 4-6). This relationship contributes to an increase in
 5                   the renal clearance of Pb from blood with increasing blood Pb concentrations
 6                   (Chamberlain. 1983). Similarly, a linear relationship between plasma Pb concentration
 7                   and urinary excretion rate predicts a linear relationship between Pb intake (at constant
 8                   absorption fraction) and urinary Pb excretion rate, whereas the relationship with blood Pb
 9                   concentration would be expected to be curvilinear (Section 4.3.6).

10                   Estimates of urinary filtration of Pb from plasma range from 13-22 L/day, with a mean of
11                   18 L/day (Arakietal.  1986; Manton and Cook. 1984; Manton and Mallov.  1983;
12                   Chamberlain et al.. 1978). which corresponds to half-time for transfer of Pb from plasma
13                   to urine of 0.1-0.16 days for a 70-kg adult who has a plasma volume of ~3 L. The rate of
14                   urinary excretion of Pb was less than the rate of glomerular filtration of ultrafilterable Pb,
15                   suggesting that urinary Pb is the result of incomplete renal tubular re-absorption of Pb in
16                   the glomerular filtrate (Araki etal. 1986): although, net tubular secretion of Pb has been
17                   demonstrated in animals (Victery etal.. 1979; Vander et al..  1977). On the other hand,
18                   estimates of blood-to-urine clearance range from 0.03-0.3 L/day with a mean of 0.18
19                   L/day (Diamond. 1992; Arakietal.. 1990; Bergeretal. 1990; Kosteretal.. 1989;
20                   Manton and Malloy. 1983; Ryu etal.. 1983; Chamberlain et al.. 1978; Rabinowitz et al..
21                   1973). consistent with a plasma Pb to blood Pb concentration ratio of-0.005-0.01 L/day
22                   (Klotzback et al.. 2003). Based on the above differences, urinary excretion of Pb can be
23                   expected to  reflect the concentration of Pb in plasma and variables that affect delivery of
24                   Pb from plasma to urine (e.g., glomerular filtration and other transfer processes in the
25                   kidney).

26                   The value for fecal:urinary excretion ratio (-0.5) was observed during days 2-14
27                   following intravenous injection of Pb in humans (Chamberlain et al.. 1978; Booker et al..
28                   1969; Hurshetal.. 1969). This ratio is slightly higher (0.7-0.8) with inhalation of
29                   submicron Pb-bearing PM due to ciliary clearance and subsequent ingestion. The transfer
30                   of Pb from blood plasma to the small intestine by biliary secretion in the liver is rapid
31                   (adult ti/2 =10 days), and accounts for 70% of the total plasma clearance (O'Flahertv.
32                   1995).
      November 2012                             4-51                  Draft - Do Not Cite or Quote

-------
                    Organic Pb
 1                  Pb absorbed after inhalation of tetraethyl and tetramethyl Pb is excreted in exhaled air,
 2                  urine, and feces (Heard et al.. 1979). Fecal:urinary excretion ratios were 1.8 following
 3                  exposure to tetraethyl Pb and 1.0 following exposure to tetramethyl Pb (Heard et al.,
 4                  1979). Occupational monitoring studies of workers exposed to tetraethyl Pb showed that
 5                  tetraethyl Pb is excreted in the urine as diethyl Pb, ethyl Pb, and inorganic Pb (Vural and
 6                  Duvdu. 1995: Zhang etal.. 1994: Turlakiewicz and Chmielnicka. 1985).
      4.3        Pb Biomarkers

 7                  This section describes the biological measurements of Pb and their interpretation as
 8                  indicators of exposure or body burden.

 9                  For any health endpoint of interest, the most useful biomarker of exposure is one that
10                  provides information about the Pb dose at the critical target organ and, moreover, reflects
11                  the exposure averaging time that is appropriate to the underlying pathogenetic processes
12                  (e.g., instantaneous, cumulative over lifetime, or cumulative over a circumscribed age
13                  range). In recent studies of Pb and health, the exposure biomarkers most frequently used
14                  are Pb in blood and bone. For outcomes other than those relating to hematopoiesis and
15                  bone health, these biomarkers provide information about Pb dose that is some distance
16                  from the target organ. For example, given that the central nervous  system is considered
17                  the critical target organ for childhood Pb toxicity, it would be most helpful to be able to
18                  measure, in vivo, the Pb concentrations at the cellular site(s) of action in the brain.
19                  However, because such measurements are not currently feasible, investigators must rely
20                  on measurements of Pb in the more readily accessible but peripheral tissues. The
21                  relationship  between brain Pb and Pb in each of these surrogate tissues is still poorly
22                  understood,  although the pharmacokinetics clearly differs among these compartments.

23                  As an exposure biomarker, blood Pb concentration has other limitations. Only about 1%
24                  of an individual's total body Pb burden resides in blood. Furthermore, blood consists of
25                  several subcompartments. More than 90% of Pb in whole blood is bound to red cell
26                  proteins such as ALAD, with the balance in plasma. From a toxicological perspective, the
27                  unbound fraction is likely to be the most important subcompartment of blood Pb because
28                  it distributes into soft tissues. The concentration of Pb in plasma is much lower than in
29                  whole blood (<1%). The greater relative abundance of Pb in whole blood makes its
30                  measurement much easier (and more affordable) than measurement of Pb in plasma. The
31                  use of whole blood Pb as a surrogate for plasma Pb could be justified if the ratio of whole
32                  blood Pb to plasma Pb were well characterized, but this is not so. At least some studies
33                  suggest that  it varies several-fold among individuals with the same blood Pb level.

      November 2012                            4-52                  Draft - Do Not Cite or Quote

-------
 1                   Moreover, binding Pb in red blood cells is limited, so the ratio of blood Pb to plasma Pb
 2                   would be expected to be nonlinear. Thus, interpreting whole blood Pb level as a proxy for
 3                   plasma Pb level, which, itself, is a proxy for brain Pb level, will result in some exposure
 4                   misclassification.

 5                   Another limitation of blood Pb as an exposure biomarker is that the kinetics of Pb in
 6                   blood is relatively fast compared to the kinetics of Pb in bone, and therefore, of the whole
 7                   body burden. Thus, a high blood Pb concentration measured at any given time does not
 8                   necessarily indicate a high body Pb burden. Similarly, individuals who have the same
 9                   blood Pb level will not necessarily  have similar body burdens or exposure histories. The
10                   rate at which blood Pb changes with time/age depends on exposure history due to re-
11                   equilibration of Pb stored in the various body pools.

12                   The development of X-ray-fluorescence (XRF) methods for measuring Pb in mineralized
13                   tissues offers another approach for characterization and reconstruction of exposure
14                   history. Such tissues are long-term Pb storage sites, with a half-life measured in decades
15                   and contain -90% of the total body Pb burden in adults and 70% in children. Thus, bone
16                   Pb reflects a long exposure averaging time.

17                   Mechanistic models are used throughout the section as a means to describe basic
18                   concepts that derive from the wealth of information on Pb toxicokinetics. Although
19                   predictions from models are inherently uncertain, models can serve to illustrate expected
20                   interrelationships between Pb intake and tissue distribution that are important in
21                   interpreting human  clinical and epidemiologic studies. Thus, models serve as the only
22                   means available for synthesizing the extensive, but incomplete, knowledge of Pb
23                   biokinetics into a holistic representation of Pb biokinetics. Furthermore, models can also
24                   be used to make predictions about biokinetics relationships that have not been thoroughly
25                   evaluated in experiments or epidemiologic studies. In this way, models can serve as
26                   heuristic tools for shaping data collection to improve understanding of Pb biokinetics.

27                   Mechanistic toxicokinetics models can make predictions about hypothetical populations
28                   and exposure scenarios. When a model is run as a single simulation, the output represents
29                   average outcomes from what is in reality a distribution of possible outcomes that would
30                   be expected in the population (or in any single individual) where intra-individual and
31                   inter-individual variability in exposure and toxicokinetics exist. More realistic predictions
32                   for the population can be developed by running a series of model simulations in which
33                   ranges (i.e., distributions) of parameter values are considered that may better represent
34                   the population of interest. In this section, only single simulations are used to demonstrate
35                   relationships between various biomarkers (e.g., blood Pb and bone Pb) that would apply
36                   to a population having "typical" or "average" exposure and toxicokinetics. These single
      November 2012                             4-53                  Draft - Do Not Cite or Quote

-------
 1                  simulations are used for illustrative purposes to describe general concepts and patterns.
 2                  Variability would be expected in real populations.

 3                  Numerous mechanistic models of Pb biokinetics in humans have been proposed, and
 4                  these are described in the 2006 Pb AQCD (U.S. EPA. 2006b) and in the supporting
 5                  literature cited in that report. In this section, for simplicity and for internal consistency,
 6                  discussion is limited to predictions from a single model, the ICRP Pb biokinetics model
 7                  (Pounds and Leggett. 1998: ICRP. 1994: Leggett. 1993). The ICRP model consists of a
 8                  systemic biokinetics model (Leggett, 1993) and a human respiratory tract model (ICRP.
 9                  1994). The Leggett model simulates age-dependent kinetics of tissue distribution and
10                  excretion of Pb ingestion and inhalation intakes. This model was originally developed for
11                  the purpose of supporting radiation dosimetry predictions and it has been used to develop
12                  cancer risk coefficients for internal radiation exposures to Pb and other alkaline earth
13                  elements that have biokinetics similar to those of calcium (ICRP. 1993). Although the
14                  ICRP model has not been validated by U.S. EPA as a regulatory model for Pb risk
15                  assessment, it has been applied in Pb risk assessment (Abrahams et al.. 2006: Lorenzana
16                  et al.. 2005: Khoury and Diamond. 2003). Portions of the model have been incorporated
17                  into an AALM that is being developed by EPA (2005a). In addition to the above
18                  considerations regarding previous  applications of the ICRP model, the model was
19                  selected for use in the ISA because it has several useful features for predicting exposure-
20                  body burden relationships. The model simulates blood Pb and tissue Pb concentration
21                  dynamics associated with the uptake and elimination phases of exposures of > 1 day in
22                  duration; and it simulates age-dependent and particle size-dependent deposition and
23                  clearance of inhaled Pb in the respiratory tract. These types of simulations can only be
24                  approximated with the U.S. EPA IEUBK Model for Pb in children because it simulates
25                  exposures in time steps of 1 year (i.e., age-year average exposures); lumps the simulation
26                  of deposition, mechanical clearance, and absorption of inhaled Pb into a single absorption
27                  term representing the combined processes of gastrointestinal and respiratory tract
28                  absorption of inhaled Pb; simulates steady state blood Pb concentrations and was does
29                  not allow access to the underlying simulations of tissue Pb concentrations which serve as
30                  intermediate variables in the model for predicting steady state blood Pb concentrations.
31                  Other models have been developed that allow simulations of tissue Pb concentrations
32                  (e.g., O'Flahertv. 1995: Leggett. 1993) and comparisons of these models have been
33                  previously described (Maddaloni et al.. 2005).
      November 2012                             4-54                   Draft - Do Not Cite or Quote

-------
 1                  Pb biokinetics in adolescents is poorly characterized by all existing Pb biokinetics
 2                  models. Individuals undergo rapid changes in sexual development, growth, food and
 3                  water intake, bone growth and turnover, behavior, etc. during adolescence. There is a
 4                  paucity of experimental measurements of Pb biomarkers during this time developmental
 5                  window. The individual biological and kinetic parameters for adolescents are largely
 6                  interpolated rather than based on solid experimental and toxicological measurements.
 7                  These deficiencies limit the validity of model predictions in this age group.
      4.3.1       Bone-Pb Measurements

 8                  For Pb measurements in bone, the most commonly examined bones are the tibia,
 9                  calcaneus, patella, and finger bone. For cortical bone, the midpoint of the tibia is
10                  measured. For trabecular bone, both the patella and calcaneus are measured. The tibia
11                  consists of more than 95% cortical bone, the calcaneus and patella comprise more than
12                  95% trabecular bone, and finger bone is a mixed cortical and trabecular bone although
13                  the second phalanx is dominantly cortical. Recent studies favor measurement of the
14                  patella for estimating trabecular bone Pb, because it has more bone mass and may afford
15                  better measurement precision than the calcaneus.

16                  Bone Pb measurements are typically expressed in units of ug Pb  per g bone mineral. This
17                  convention may potentially introduce variability into the bone Pb measurements related
18                  to variation in bone density.  Typically, potential associations between bone density and
19                  bone Pb concentration are not evaluated in epidemiologic studies (Theppeang et al..
20                  2008a; Hu et al., 2007a). An important consequence of expressing bone Pb measures
21                  relative to bone mineral content is that lower bone mineral density is associated with
22                  greater measurement uncertainty in bone Pb. This can have important implications for
23                  studies in older women for whom low bone mineral density is more common than in
24                  other populations including men and younger adults.

25                  Methods of direct analysis of bone tissue samples include flame atomic absorption
26                  spectrometry (AAS), anode stripping voltammetry (ASV), inductively coupled plasma
27                  atomic emission spectroscopy (ICP-AES), inductively coupled plasma mass spectrometry
28                  (ICP-MS), laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS),
29                  thermal ionization mass spectrometry (TIMS), synchrotron radiation induced X-ray
30                  emission (SRIXE), particle induced X-ray emission (PIXE), and  X-ray fluorescence
31                  (XRF). Non-invasive, in vivo measurements of bone Pb is achieved with XRF. The
32                  upsurge in popularity of the XRF method has paralleled a decline in the use of the other
33                  methods. More information on the precision, accuracy, and variability in bone Pb
34                  measurements can be found in the 2006 Pb AQCD (U.S. EPA. 2006b).
      November 2012                             4-55                  Draft - Do Not Cite or Quote

-------
 1                  Two main approaches for XRF measurements have been used to measure Pb
 2                  concentrations in bone, the K-shell and L-shell methods. The K-shell method is the most
 3                  widely used, as there have been relatively few developments in L-shell devices since the
 4                  early 1990s. However, Nie et al. (2011 a) recently reported on the use of a new portable
 5                  L-shell device for human in vivo Pb measurements. Advances in L-shell device
 6                  technology resulted in much higher sensitivity than previous L-shell devices. The new
 7                  L-shell device showed sensitivity similar to that of K-shell methods (detection limit was
 8                  approximately 8 (ig/g bone mineral with 2 mm of soft tissue overlay targeted bone) and a
 9                  high correlation with results obtained from K-shell methods (intraclass
10                  correlation = 0.65). Behinaen et al. (2011) described application of a 4-detector system
11                  ("clover leaf array ") for the K-shell method that provided higher precision and lower
12                  minimum detection limits (MDL) for tibia and calcaneus Pb measurements (3.25 and
13                  4.78 (ig/g bone mineral, respectively) compared to measurements made with single
14                  detectors (8-12 (ig/g and 14-15 (ig/g, respectively).

15                  Since 1986, several investigators have reported refinements to hardware and software to
16                  improve the precision and accuracy of XRF measurements and there have been a number
17                  of investigations into the precision, accuracy and variability in XRF measurements [e.g.,
18                  (Todd et al.. 2002: Toddetal.. 2001: Aro et al.. 2000: Todd et al.. 2000)1. Todd et al.
19                  (2000) provided a detailed discussion of factors that influence the variability and
20                  measurement uncertainty, including repositioning, sample measurement duration,
21                  overlying tissue, operator expertise, detector resolution, and changes to measurement
22                  process over time. Some of these aspects were also discussed by Hu et al. (1995). From
23                  their cadaver and in vivo measurements, Todd et al. (2000) concluded that the uncertainty
24                  in an individual measurement was an underestimate of the standard deviation of replicate
25                  measurements, suggesting a methodological deficiency probably shared by most current
26                  109Cd-based K-shell XRF Pb measurement systems. In examining the reproducibility of
27                  the bone Pb measurements over a 4!/2 month period, Todd et al. (2000) also found the
28                  average difference between the XRF results from short term and longer term
29                  measurements was 1.2 (ig/g, indicating only a small amount of variability in the XRF
30                  results over a sustained period of time.

31                  In the epidemiologic  literature, XRF bone Pb data have typically been reported in two
32                  ways: one that involves a methodological approach to assessing the minimum detection
33                  limit and the other termed an epidemiologic approach by Rosen and Pounds (1998). In
34                  the former approach, a minimum detection limit is defined using various methods,
35                  including two or three times the square root of the background counts; one, two, or three
36                  times the standard deviation (SD) of the background; or two times the observed median
37                  error. This approach relies upon the minimum detection limit to define a quantitative
38                  estimate that is of sufficient precision to be included in the statistical analysis, as
      November 2012                             4-56                  Draft - Do Not Cite or Quote

-------
 1                  demonstrated by Bellinger et al. Q994a), Gerhardsson et al. (1993). and Christoffersson
 2                  et al. (1986).

 3                  With the epidemiologic approach, all values are used (including negative values) to
 4                  determine the minimum detection limit of an instrument that results in extremely low
 5                  detection limits. Rosen and Pounds (1998) noted that this approach yields population
 6                  bone Pb averages that were artificially low. However, not including values that are
 7                  negative or below the detection limit, or assigning these values a fixed number is also of
 8                  concern. Using the epidemiologic approach of retaining all point estimates of measured
 9                  bone Pb concentrations provided the least amount of bias and the greatest efficiency in
10                  comparing the mean or median levels of bone Pb of different populations (Kim et al..
11                  1995).
      4.3.2       Blood-Pb Measurements

12                  Analytical methods for measuring Pb in blood include AAS, graphite furnace atomic
13                  absorption spectrometry (GFAAS), ASV, ICP-AES, and ICP-MS. GFAAS and ASV are
14                  generally considered to be the methods of choice (Flegal and Smith. 1995). Limits of
15                  detection for Pb using AAS are on the order of 5-10 (ig/dL for flame AAS measurements
16                  and approximately 0.1 (ig/dL for flameless AAS measurements (Flegal and Smith. 1995;
17                  NIOSH. 1994). A detection limit of 0.005 (ig/dL has been achieved for Pb in blood
18                  samples analyzed by GFAAS.

19                  For measurement of Pb in plasma, ICP-MS provides sufficient sensitivity (Schutz et al..
20                  1996). While the technique has been applied to assessing Pb exposures in adults, ICP-MS
21                  has not received widespread use in epidemiologic studies.

22                  The primary binding ligand for Pb in RBC, ALAD, is encoded by a single gene in
23                  humans that is polymorphic in two alleles (ALAD1 and ALAD2) (Scinicariello et al..
24                  2007). Since the ALAD1 and ALAD 2 alleles can be co-dominantly expressed, 3
25                  different genotypes (ALAD 1-1, ALAD 1-2, and ALAD 2-2) are possible. The ALAD
26                  1-1 genotype is the most common. Scinicariello et al. (2010) tested genotypes in civilian,
27                  non-institutionalized U.S. individuals that participated as part of NHANES III from
28                  1988-1994 and found that 15.6% of non-Hispanic whites, 2.6% non-Hispanic blacks, and
29                  8.8% Mexican Americans carried the ALAD2 allele.

30                  The 2006 Pb AQCD (U.S. EPA. 2006c) reported that many studies have shown that, with
31                  similar exposures to Pb, individuals with the ALAD-2 allele have higher blood Pb levels
32                  than those without (Kim et al.. 2004; Perez-Bravo et al.. 2004; Bergdahl et al.. 1997b:
33                  Smith et al.. 1995a: Wetmur. 1994; Wetmuretal. 1991b: Astrinetal.. 1987). More
      November 2012                            4-57                  Draft - Do Not Cite or Quote

-------
 1                  recent meta analyses provide further support for ALAD2 carriers having higher blood Pb
 2                  levels than ALAD1 -1 homozygotes (Scinicariello et al.. 2007; Zhao et al.. 2007). The
 3                  mechanism for this association may be higher Pb binding affinity of ALAD2. Although,
 4                  this interpretation would be consistent with the structural differences that result in greater
 5                  electronegativity of ALAD 1 compared to ALAD2 (Wetmur. 1994; Wetmur et al.. 199 la).
 6                  measurements of Pb binding affinity to ALAD1 and ALAD2 (i.e., Pb2+ displacement of
 7                  Zn2+ binding to recombinant ALAD 1 and ALAD2) have not revealed differences in Pb
 8                  binding affinity (Jaffe et al.. 2000). In a meta-analysis of 24 studies, Scinicariello et al.
 9                  (2007). observed the greatest differences for ALAD2 compared to ALAD1 in highly
10                  exposed adults with little difference among environmentally-exposed adults; large
11                  differences were also observed for children at low exposures. However, there are few
12                  studies that evaluated children and the largest study contributing to the  meta analysis may
13                  have been influenced by selection bias (Scinicariello et al.. 2007). Individual studies find
14                  similar results in occupationally-exposed adults, with blood Pb levels being higher in
15                  individuals with ALAD2 alleles (Miyaki et al.. 2009; Shaik and Jamil. 2009). A
16                  subsequent meta analysis of adult data from NHANES III did not find any differences in
17                  blood Pb level between all carriers of either the ALAD 1-1 or ALAD 1-2/2-2 allele
18                  (Scinicariello et al.. 2010). Other studies provide further support for no blood Pb
19                  differences among ALAD1 and ALAD2 carriers (Sobin et al..  2009; Rabstein et al..
20                  2008; Montenegro et al.. 2006; Wananukul et al.. 2006) or lower blood Pb levels for
21                  individuals with ALAD 1 -2/2-2 (Krieg et al.. 2009; Chia et al.. 2006).

22                  Genetic polymorphism in the gene that encodes for peptide transporter  2 (PEPT2) has
23                  been associated with variability in blood Pb concentrations in children (Sobin et al..
24                  2009). PEPT2 expression in the brain and renal proximal tubule has been associated with
25                  transport of di- and tri-peptides and may function in the transport of 5-ALA into brain
26                  and renal tubular re-absorption of peptides. The PRPT2*2 polymorphism was associated
27                  with increased blood Pb concentrations in a sample of 116 children of Mexican-
28                  American/Hispanic heritage (age 4-12 years, mean blood Pb concentration 3-6  (ig/dL).

29                  Analyses of serial blood Pb concentrations measured in longitudinal epidemiologic
30                  studies found relatively strong correlations  (e.g., r = 0.5-0.8) among each child's
31                  individual blood Pb concentrations measured after 6-12 months of age (Schnaas et al..
32                  2000; Dietrich et al.. 1993a; McMichael et al.. 1988; Ottoetal..  1985; Rabinowitz et al..
33                  1984). These observations  suggest that, in general, exposure characteristics of an
34                  individual child (e.g., exposure levels and/or exposure behaviors) tend to be relatively
35                  constant across age. However, a single blood Pb measurement may not distinguish
36                  between a history of long-term lower-level  Pb exposure from a history that includes
37                  higher acute exposures (Mushak. 1998). This concept is illustrated in Figure 4-7. Two
38                  hypothetical children are simulated. Child A has a relatively constant Pb intake from
      November 2012                             4-58                   Draft - Do Not Cite or Quote

-------
 1                  birth, whereas Child B has the same Pb intake as Child A for the first two years of life,
 2                  then a 1-year elevated intake beginning at age 24 months (Figure 4-7. upper panel) that
 3                  returns to the same intake as Child A at 36 months. The absorption fraction is assumed to
 4                  be the same for both children. Blood Pb samples 1 and 5 for Child A and B, or 2 and 4
 5                  for Child B, will yield similar blood Pb concentrations (~3 or 10 (ig/dL, respectively), yet
 6                  the exposure contexts for these samples are very different. Two samples (e.g., 1 and 2, or
 7                  4 and 5), at a minimum, are needed to ascertain if the blood Pb concentration is changing
 8                  over time. The rate of change can provide information about the magnitude of change in
 9                  exposure, but not necessarily about the time history of the change (Figure 4-7. lower
10                  panel). Time-integrated measurements of Pb concentration may provide a means for
11                  accounting for some of these factors and, thereby, provide a better measure of long-term
12                  Pb exposure.
     November 2012                             4-59                   Draft - Do Not Cite or Quote

-------
                          0       12      24     36      48     60   72    84
                                                Age (months)
                   O)
                   o>
                       40
                       30
                       20 ^
                  m
                        0
                          0      12       24     36     48      60

                                                 Age (months)
                              72    84
Note: Child A and Child B had a constant basal Pb intake (10 |jg/day) from birth; Child B experienced an elevated intake of
5.5 |jg/day/kg for 1 year beginning at 24 months of age (upper panel). Blood Pb measurements 1 and 5 for Child A and B, or 2 and
4 for Child B, will yield similar blood Pb concentrations (~3 or 10 ug/dL, respectively), yet the exposure scenarios for these samples
are very different. As shown in the example of Child C and Child D, two blood Pb measurements can provide information about the
magnitude of change in exposure, but not necessarily the temporal history of the change (lower panel). Child C and D had a
constant basal Pb intake (10 ug/day) from birth. Child C experienced an elevated intake of 13 ug/day starting at 12 months of age
for 1 year, whereas, Child D experienced an elevated intake of 5.5 ug/day starting at 24 months of age for 1 year. Simulation based
on ICRP Pb biokinetics model (Leggett. 1993).


Figure 4-7        Simulation of temporal relationships between Pb exposure and

                    blood Pb concentration  in  children.
November 2012
4-60
Draft - Do Not Cite or Quote

-------
      4.3.3        Urine-Pb Measurements

 1                  Standard methods that have been reported for urine Pb analysis are, in general, the same
 2                  as those analyses noted for determination of Pb in blood. Reported detection limits are
 3                  -50 (ig/L for AAS, 5-10 (ig/L for ICP AES, and 4 (ig/L for ASV for urine Pb analyses.

 4                  The concentration of Pb in urine is a function of the urinary Pb excretion (Section 4.2.3)
 5                  and the urine flow rate. Urine flow rate requires collection of a timed urine sample, which
 6                  is often problematic in epidemiologic studies. Collection of untimed ("spot") urine
 7                  samples, a common alternative to timed samples, requires adjustment of the Pb
 8                  measurement in urine to account for variation in urine flow (Diamond. 1988). Several
 9                  approaches to this adjustment have been explored, including adjusting the measured urine
10                  Pb concentration by the urine creatinine concentration, urine osmolality, or specific
11                  gravity (Fukui et al., 1999; Araki et al., 1990). Urine flow rate can vary by a factor or
12                  more than 10, depending on the state of hydration and other factors that affect glomerular
13                  filtration rate and renal tubular reabsorption of the glomerular filtrate. All of these factors
14                  can be affected by Pb exposure at levels that produce nephrotoxicity (i.e., decreased
15                  glomerular filtration rate, impaired renal tubular transport function). Therefore, urine Pb
16                  concentration measurements provide little reliable information about exposure (or Pb
17                  body burden), unless they can be adjusted to account for unmeasured variability in urine
18                  flow rate (Araki etal. 1990).

19                  Urinary Pb concentration reflects, mainly, the concentration of Pb in the blood. As such,
20                  urinary concentrations reflect both recent and past exposures to Pb (see Section 4.3.5). A
21                  single urinary Pb measurement cannot distinguish between a long-term low level of
22                  exposure or a higher acute exposure. Urinary Pb measurements would be expected to
23                  correlate with concurrent blood Pb (see Section 4.3.6 for additional discussion of the
24                  relationship between blood and urine Pb). Chiang et al. (2008) reported a significant, but
25                  relatively weak correlation between urinary Pb levels ((ig/dg creatinine) and individual
26                  Pb intakes ((ig/day) estimated  in a group of 10- to 12-year-old children (|3: 0.053,
27                  R = 0.320, p = 0.02, n = 57). A contributing factor to the relatively weak correlation may
28                  have been the temporal displacement between the urine  sampling and measurements used
29                  to estimate  intake, which may have been as long as 6 months for some children.

30                  Thus, a single urine Pb measurement, or a series of measurements taken over short-time
31                  span, is likely a relatively poor index of Pb body burden for the same reasons that blood
32                  Pb is not a good indicator of body burden. On the other hand, long-term average
33                  measurements  of urinary Pb can be expected to be a better index of body burden (Figure
34                  4;8.
      November 2012                             4-61                  Draft - Do Not Cite or Quote

-------
                       10
                    re
                    5

                    O)
                        8 -
                        6 -
                    £1
                    0.
                    0)

                    •I   4
                        2 -
                                      ""           '
                            — j
                               20
                               15 CD
                                  o
                                  Q.
                                  <

                                  CD


                               10 I.
                                  CD
                                  3



                                  (Q
                                                                        -  5
                          25   30   35   40   45   50   55  60   65   70

                                             Age (year)
                   o
                   I    4


                   Its

                   SI
                   o
                        2  -
                        1  -I
                                                                           20
                               15 oo
                                   o
                                   0.
                                  <

                                   03

                               10 a
                                   o
                                   3


                                  (Q
                                                                         -  5
                          25  30   35   40   45   50   55   60   65  70

                                             Age (year)
Note: A change in Pb uptake results in a relatively rapid change in urinary excretion of Pb, to a new quasi-steady state, and a
relatively small change in body burden (upper panel). Baseline ingestion was 20 ug/day from age 0 to 30 yrs, then intake increased
to 120 ug/day from age 30 to 50 with a subsequent decrease in intake to the baseline of 20 ug/day at age 50. The long-term
average urinary Pb excretion more closely tracks the pattern of change in body burden (lower panel). Simulation based on ICRP Pb
biokinetics model (Leggett. 1993).



Figure 4-8       Simulation of relationship between  urinary Pb excretion and body

                   burden in  adults.
November 2012
4-62
Draft - Do Not Cite or Quote

-------
      4.3.4        Pb in Other Biomarkers

 1                  There was extensive discussion in the 2006 Pb AQCD (U.S. EPA. 2006c) regarding the
 2                  utility of other Pb biomarkers as indicators of exposure or body burden. Due to the fact
 3                  that most epidemiologic studies continue to use blood Pb or bone Pb, and other potential
 4                  biomarkers (i.e., teeth, hair, and saliva) have not been established to the same extent as
 5                  blood or bone Pb, only summaries are provided below.
                    4.3.4.1      Teeth

 6                  Tooth Pb is a minor contributor to the total body burden of Pb. As teeth accumulate Pb,
 7                  tooth Pb levels are generally considered an estimate of cumulative Pb exposure. The
 8                  tooth Pb-blood Pb relationship is more complex than the bone Pb-blood Pb relationship
 9                  because of differences in tooth type, location, and analytical method. Although
10                  mobilization of Pb from bone  appears well established, this is not the case for Pb in teeth.
11                  Conventional wisdom has Pb fixed once it enters the tooth. Although that may be the case
12                  for the bulk of enamel, it is not true for the surface of the enamel and dentine (Gulson et
13                  al.. 1997; Rabinowitz et al., 1993). Limited studies have demonstrated moderate-to-high
14                  correlations between tooth Pb  levels and blood Pb levels (Rabinowitz.  1995; Rabinowitz
15                  etal.. 1989).

16                  Teeth are composed of several tissues formed pre- and postnatal. Therefore, if a child's
17                  Pb exposure during the years of tooth formation varied widely, different amounts of Pb
18                  would be deposited at different rates (Rabinowitz et al., 1993). This difference may allow
19                  investigators to elucidate the history of Pb exposure in a child. Robbins et al. (2010)
20                  found a significant association between environmental Pb measures that correlated with
21                  leaded gasoline use and tooth enamel Pb in permanent teeth. Costa de Almeida et al.
22                  (2007) discerned differences between tooth enamel Pb concentration in biopsy samples
23                  from children who lived in areas having higher or lower levels of Pb contamination.
24                  Gulson and Wilson (1994) advocated the use of sections of enamel and dentine to obtain
25                  additional information compared with analysis of the whole tooth (e.g., (Tvinnereim et
26                  al.. 1997; Fosse etal.. 1995). For example, deciduous tooth Pb in the enamel provides
27                  information about in utero exposure whereas that in dentine from the same tooth provides
28                  information about postnatal exposure until the tooth exfoliates at about 6-7 years of age.
      November 2012                             4-63                  Draft - Do Not Cite or Quote

-------
                    4.3.4.2     Hair

 1                  The 2006 Pb AQCD (U.S. EPA. 2006c) discussed applications of hair Pb measurements
 2                  for assessing Pb body burden or exposure and noted methodological limitations
 3                  (e.g., external contamination) and lack of a strong empirical basis for relating hair Pb
 4                  levels to body burden or exposure. No new methodological or conceptual advances
 5                  regarding hair Pb measurements have occurred since 2006, and widespread application of
 6                  hair Pb measurements in epidemiologic studies has not occurred.

 7                  Pb is incorporated into human hair and hair roots (Bos et al.. 1985; Rabinowitz et al.,
 8                  1976) and has been explored as a noninvasive approach for estimating Pb body burden
 9                  rWilhelm et al.. 2002; Gerhardsson et al.. 1995; Wilhelm et al.. 1989). Hair Pb
10                  measurements are subject to error from contamination of the surface with environmental
11                  Pb and contaminants in artificial hair treatments (i.e., dyeing, bleaching, permanents) and
12                  are a relatively poor predictor of blood Pb concentrations, particularly at blood Pb levels
13                  less than 10-12  (ig/dL (Rodrigues et al.. 2008; Campbell and Toribara. 2001; Esteban et
14                  al.. 1999; Drasch et al.. 1997). Temporal relationships between Pb exposure and hair Pb
15                  levels, and kinetics of deposition and retention of Pb in hair have not been evaluated.
16                  Although hair Pb measurements have been  used in some epidemiologic studies (Shah et
17                  al.. 2011; U.S. EPA. 2006b). an empirical basis for interpreting hair Pb measurements in
18                  terms of body burden or exposure has not been firmly established.
                    4.3.4.3     Saliva

19                  A growing body of literature on the utility of measurements of salivary Pb has developed
20                  since the completion of the 2006 Pb AQCD (U.S. EPA. 2006b). Earlier reports suggested
21                  a relatively strong correlation between salivary Pb concentration and blood Pb
22                  concentration (Omokhodion and Crockford. 1991; Brodeur et al.. 1983; P'an. 1981);
23                  however, more recent assessments have shown relatively weak or inconsistent
24                  associations (Costa de Almeida et al.. 2011; Costa de Almeida et al.. 2010; Costa de
25                  Almeida et al., 2009; Barbosa et al., 2006a; Nriagu et al., 2006). The differences in these
26                  outcomes may reflect differences in blood Pb concentrations, exposure history and/or
27                  dental health (i.e., transfer of Pb between dentin and saliva) and possibly methods for
28                  determining Pb in saliva. Barbosa et al. (2006a) found a significant but relatively weak
29                  correlation (logfblood PB] versus logfsaliva Pb], r = 0.277, p = 0.008) in a sample of
30                  adults, ages 18-60 years (n = 88). The correlation was similar for salivary and plasma Pb.
31                  Nriagu et al. (2006) found also found a relatively weak association (R2 ~ 0.026) between
32                  blood Pb ((ig/dL) and salivary Pb ((ig/L) in a sample of adults who resided in Detroit, MI
33                  (n = 904). Costa de Almeida et al. (2009) found a significant correlation between salivary
      November 2012                             4-64                   Draft - Do Not Cite or Quote

-------
 1                  and blood Pb concentrations in children in a Pb-contaminated region in Sao Paulo State,
 2                  Brazil (r = 0.76. p = 0.04, n = 7) prior to site remediation; however, the correlation
 3                  degenerated (r = 0.03, p = 0.94, n = 9) following remediation. Nevertheless, salivary Pb
 4                  concentrations in the group of children who lived in the contaminated area were
 5                  significantly elevated compared to a reference population. It is possible, that salivary Pb
 6                  measurements may be a useful non-invasive biomarker for detecting elevated Pb
 7                  exposure; however, it is not clear based on currently available data, if salivary Pb
 8                  measurements would be a more reliable measure of exposure than blood Pb
 9                  measurements.
                    4.3.4.4      Serum 5-ALA and ALAD

10                  The association between blood Pb and blood ALAD activity and serum 5-aminolevulinic
11                  acid (5-ALA) levels was recognized decades ago as having potential use as a biomarker
12                  of Pb exposure (Mitchell et al., 1977; Hernberg et al., 1970). More recently reference
13                  values for blood ALAD activity ratio (the ratio of ALAD activity in the blood sample to
14                  that measured after fully activating the enzyme in the sample) have been reported
15                  (Gultepe et al.. 2009). Inhibition of erythrocyte ALAD by Pb results in a rise in the
16                  plasma concentration of the ALAD substrate 5-ALA. The 5-ALA biomarker can be
17                  measured in serum and has been used as a surrogate  for Pb measurements in studies in
18                  which whole blood samples or adequately prepared plasma or serum samples were not
19                  available for Pb measurements (Opleretal., 2008; Opler et al.. 2004).
      4.3.5       Relationship between Pb in Blood and Pb in Bone

20                  The kinetics of elimination of Pb from the body reflects the existence of multiple pools of
21                  Pb in the body that have different elimination kinetics. The dominant washout phase of
22                  Pb from the blood, exhibited shortly after a change in exposure occurs, has a half-life of
23                  -20-30 days (Leggett. 1993; Rabinowitz et al.. 1976). Studies of a limited number of
24                  adults (four individuals with hip or knee replacement, a married couple, and 10 female
25                  Australian immigrants) in which the Pb exposure was from historical environmental
26                  sources have found that bone Pb stores can contribute 40-70% to blood Pb (Smith et al..
27                  1996; Gulsonetal. 1995a: Manton. 1985). Bone Pb burdens in adults are slowly lost by
28                  diffusion (heteroionic exchange) as well as by resorption (O'Flaherty. 1995). Half-times
29                  for the release of Pb in bone are dependent on age and intensity of exposure. Bone
30                  compartments are  much more labile in infants and children than in adults as reflected by
31                  half-times for movement of Pb from bone into the plasma (e.g., cortical ti/2 = 0.23 years
      November 2012                             4-65                 Draft - Do Not Cite or Quote

-------
 1                  at birth, 1.2 years at 5 years of age, 3.7 years at 15 years of age, and 23 years in adults;
 2                  trabecular ti/2 = 0.23 years at birth, 1.0 years at 5 years of age, 2.0 years at 15 years of
 3                  age, and 3.9 years in adults) (Leggett 1993). Slow transfer rates for the movement of Pb
 4                  from nonexchangeable bone pools to the plasma are the dominant transfer process
 5                  determining long-term accumulation and elimination of bone Pb burden.

 6                  When blood Pb concentrations are monitored in individuals over periods of years
 7                  following a cessation or decrease in exposure, the decrease in blood Pb concentration
 8                  exhibits complex kinetics that can be disaggregated into components having faster and
 9                  slower rates. The slower rates of clearance of Pb from the blood over months and years
10                  following the cessation or reduction in exposures is thought to primarily reflect
11                  elimination of Pb stores in bone. Nilsson et al. (1991) reported a tri-exponential decay in
12                  the blood Pb concentrations of 14 individuals having a median occupational exposure
13                  period of 26 years. Thirteen of these 14 individuals had been temporarily removed from
14                  work because of excessive exposures (blood levels > 70 ug/dL or high urinary
15                  5-aminolevulinic acid levels). Representing 22% of blood Pb, the fast compartment had a
16                  clearance half time of 34 days. The intermediate compartment, 27% of blood Pb, had a
17                  clearance half time of 1.12 year. And, the slow compartment, 50% of blood Pb, had a
18                  clearance half time of 13 years. The authors attributed the fast, intermediate, and slow
19                  compartment clearance to elimination of Pb from blood and some soft tissues, from
20                  trabecular bone, and cortical bone, respectively. Rentschler et al. (2012) also observed a
21                  slow terminal phase of Pb elimination from blood in five adults who had Pb poisoning
22                  due to either occupational or non-occupational exposures that ranged from approximately
23                  1 month to 12 years and resulted in blood Pb concentrations of 70-110 (ig/dL. In this
24                  study, the blood Pb monitoring period extended from 1 to 74 days following cessation of
25                  exposure to approximately 800 days following the diagnosis of poisoning; however, it
26                  was not of sufficient duration to estimate the terminal half-time. When the terminal half-
27                  time estimated by Nilsson et al. (1991) was used (13 years) to fit data  for these Pb
28                  poisoning cases to a two-component exponential decay model, the  initial faster phase
29                  represented approximately 80%  of the blood Pb and the half-time was estimated to range
30                  from 60 to 120 days. The relatively longer fast phase half-time reported by Rentschler et
31                  al. (2012) compared to Nilsson et al. (1991) may reflect the relatively  high blood Pb
32                  concentrations in these poisoning cases that resulted in temporary anemia and subsequent
33                  reestablishment of a normal erythrocyte levels. Also, the use of a two-compartment
34                  model, with an assumed slow half-time of 13 years, as well as uncertainty about the
35                  actual time of cessation of exposure may have prevented discerning a third, faster
36                  elimination compartment in these data.
      November 2012                             4-66                   Draft - Do Not Cite or Quote

-------
 1                  Children who have been removed from a relatively brief exposure to elevated
 2                  environmental Pb also exhibit faster slow-phase kinetics than children removed from
 3                  exposures that lasted several years, with half-times of 10 and 20-38 months, respectively
 4                  (Manton et al.. 2000). Rothenberg et al. (1998) also showed that exposures in the first 6
 5                  months of life could contribute to elevated blood Pb levels through at least 3 years
 6                  relative to children with lower early life exposures, despite similar environmental
 7                  exposures at later time points. In both adults and children, the longer half-times measured
 8                  under the latter conditions reflect the contribution of bone Pb stores to blood Pb
 9                  following a change in exposure.

10                  The longer half-life of Pb in bone compared to blood Pb, allows a more cumulative
11                  measure of long-term Pb exposure. Pb in adult bone can serve to maintain blood Pb levels
12                  long after external exposure has ceased (Fleming et al.. 1997; Inskip etal. 1996; Smith et
13                  al.. 1996; Kehoe. 1987; O'Flaherty et al.,  1982). even for exposures that occurred during
14                  childhood (McNeill et al.. 2000). The more widespread use of in vivo XRF Pb
15                  measurements in bone and indirect measurements of bone processes with stable Pb
16                  isotopes have enhanced the use of bone Pb as a biomarker of Pb body burden.

17                  Several studies have found a stronger relationship between patella Pb and blood Pb than
18                  tibia Pb and blood Pb (Park et al.. 2009c; Huetal.. 1998; Hernandez-Avilaetal.. 1996;
19                  Hu et al.. 1996a). Hu et al. (1998) suggest that trabecular bone is the predominant bone
20                  type providing Pb back into circulation under steady-state and pathologic conditions. The
21                  stronger relationship between blood Pb and trabecular Pb compared with  cortical bone is
22                  probably associated with the larger surface area of trabecular bone allowing for more Pb
23                  to bind via ion exchange mechanisms and more rapid turnover making it more  sensitive
24                  to changing patterns of exposure.

25                  Relationships between Pb in blood and bone in children and adults are discussed in
26                  greater detail below (Sections 4.3.5.1. and 4.3.5.2). In these discussions, simulations
27                  based on a biokinetics model are shown to illustrate general patterns in the relationships
28                  between bone Pb and blood Pb that can be predicted based on the current understanding
29                  of Pb biokinetics in children and adults. However, these simulations reflect assumptions
30                  in the model and may not accurately represent the observed blood Pb kinetics in
31                  individuals or variability in blood Pb kinetics observed in specific populations. The
32                  simulations  include two metrics of blood Pb, the blood  Pb concentration at each time
33                  point in the  simulation and the time-integrated blood Pb for the period preceding each
34                  time point in the simulation (also referred to as the cumulative blood Pb index  [CBLI]).
35                  The time-integrated blood Pb metric has been used to estimate long-term  average and
36                  cumulative absorbed Pb doses in epidemiologic studies (e.g., Nie et al.. 20lib; Healev et
      November 2012                             4-67                  Draft - Do Not Cite or Quote

-------
 1                  al.. 2008; Hu et al. 2007a: Chuang et al.. 2000; McNeill et al.. 2000; Fleming et al..
 2                  1997: Roelsetal. 1995: Gerhardsson et al.. 1993: Armstrong et al.. 1992).
                    4.3.5.1      Children

 3                  As mentioned in Section 4.2.2.2. bone growth in children will contribute to accumulation
 4                  of Pb in bone, which will comprise most of the Pb body burden. As a result, Pb in bone
 5                  will more closely reflect Pb body burden than blood Pb. However, changes in blood Pb
 6                  concentration in children (i.e., associated with changing exposures) are thought to more
 7                  closely parallel changes in total body burden. Figure 4-9 shows a biokinetics model
 8                  simulation of the temporal profile of Pb in blood and bone in a child who experiences a
 9                  period of constant Pb intake (from age 2-5) via ingestion (|ig Pb/day) followed by an
10                  abrupt decline in intake. The figure illustrates several important general concepts about
11                  the relationship between Pb in blood and bone. While blood Pb approaches a quasi-steady
12                  state after a period of a few months with a constant rate of Pb intake (as demonstrated by
13                  the vertical dashed line), Pb continues to accumulate in bone with continued Pb intake
14                  after the quasi-steady state is achieved in blood. The model also predicts that the rate of
15                  release of Pb from bone after a reduction in exposure is faster than in adults. This
16                  difference has been attributed to accelerated growth-related bone mineral turnover in
17                  children, which is the primary mechanism for release of Pb that has been  incorporated
18                  into the bone mineral matrix.

19                  Empirical evidence in support of this conclusion comes from longitudinal studies in
20                  which relatively high correlations were found between concurrent (r = 0.75) or average
21                  lifetime (obtained at 6-month intervals from birth to age  10 or 12) blood Pb
22                  concentrations (r = 0.85) and tibia bone  Pb concentrations (measured by XRF) in a
23                  sample of children in which the group mean concurrent blood Pb concentration exceeded
24                  20 (ig/dL; the correlations was much weaker (r <0.15) among the group of children with
25                  a mean concurrent blood Pb concentration <10 (ig/dL (Wasserman et al..  2003).

26                  Time-integrated blood Pb metrics display rates of change in response to the exposure
27                  event that more closely approximate the slower kinetics of bone Pb and body burden,
28                  than the kinetics of blood Pb concentration, with notable differences (Figure 4-9). The
29                  time-integrated blood PB concentration  is  a cumulative function and increases throughout
30                  childhood; however,  the slope of the increase is higher during the exposure event than
31                  prior to or following the event. Following  cessation of the enhanced exposure period, the
32                  time-integrated blood Pb and body burden diverge. This result is expected, as the time-
33                  integrated blood Pb curve is a cumulative function which cannot decrease over time and
34                  bone Pb levels will decrease with reduction in exposure.
      November 2012                             4-68                   Draft - Do Not Cite or Quote

-------
1                   The time-integrated blood Pb concentration will be a better reflection of the total amount
2                   of Pb that has been absorbed, than the body burden at any given time. The time-
3                   integrated blood Pb concentration will also reflect cumulative Pb absorption, and
4                   cumulative exposure if the absorption fraction is constant. This is illustrated in the
5                   hypothetical simulations of an exposure event experienced by a child (Figure 4-10). This
6                   pattern is similar for adults.
     November 2012                              4-69                   Draft - Do Not Cite or Quote

-------
                       10
                                                                     10
                        50
                                    246

                                             Age (year)
                         10
Note: Blood Pb concentration is thought to parallel body burden more closely in children than in adults, due to more rapid turnover of
bone and bone-Pb stores in children (upper panel). Baseline Pb intake is 3.2 ug/day from birth until age 2, followed by a period of
increased intake (38.2 ug/day) from age 2 until age 5, with a return to baseline intake of 3.2 ug/day at age 5.  The time-integrated
blood Pb concentration increases overtime (lower panel). Simulation based on ICRP Pb biokinetics model (Leggett. 1993).


Figure 4-9       Simulation of relationship between blood Pb  concentration and
                   body burden in children, with an elevated constant Pb intake from

                   age 2 to 5 years.
November 2012
4-70
Draft - Do Not Cite or Quote

-------
                  (0
                  o>
                  _i
                  T3
                  O
                  CD
                  •o
                         50
                         40 -
                         20 -
                         10 -
                                         •Blood
                                         •Absorption
                                       246
                                                Age (year)
                                                8
                                                            20
                                                                 O

                                                            15  I
                                                                 Q)
                                                                 S-.
                                                                 (D
                                                                 >
                                                                 CT
                                                                 (/>
                                                                 O
                                                                 •3
                                                                 a-.
                                                                 o
                                                                 3
        --  10
                                                                              --  5   --
       10
                                                                                     (Q
      Note: The simulations include a 3-year period of elevated constant Pb intake during ages 2-5 years. Baseline Pb intake is
      3.2 ug/day from birth until age 2, followed by a period of increased intake (38.2 ug/day) from age 2 until age 5, with a return to
      baseline intake of 3.2 ug/day at age 5. The time-integrated blood Pb concentration closely parallels cumulative Pb absorption.
      Simulation based on ICRP Pb biokinetics model (Leggett. 1993).

      Figure 4-10     Simulation  of relationship between time-integrated blood Pb
                        concentration and cumulative Pb absorption in children.
                    4.3.5.2
             Adults
 i
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11
12
In adults, where a relatively large fraction of the body burden residing in bone has a
slower turnover compared to blood, a constant Pb uptake (or constant intake and
fractional absorption) gives rise to a quasi-steady state blood Pb concentration, while the
body burden continues to increase over a much longer period, largely as a consequence of
continued accumulation of Pb in bone. This pattern is illustrated in Figure 4-11 that
depicts a hypothetical simulation of an exposure consisting of a 20-year period of daily
ingestion of Pb in an adult. The exposure shown in the simulations gives rise to a
relatively rapid increase in blood Pb concentration from a baseline of approximately
2 (ig/dL, to a new quasi-steady state of approximately 9 (ig/dL, achieved in -75-100 days
(i.e., approximately 3-4 times the blood elimination half-life). In contrast, the body
burden exhibits a steady increase across the full exposure period of 70 yr. Following
cessation of the enhanced exposure period, blood Pb concentration declines relatively
      November 2012
                            4-71
Draft - Do Not Cite or Quote

-------
 1                  rapidly compared to the slower decline in body burden. Careful examination of the
 2                  simulation shown in Figure 4-11 reveals that the accumulation and elimination phases of
 3                  blood Pb kinetics are not symmetrical; elimination is slower than accumulation as a result
 4                  of the gradual release of bone Pb stores to blood. This response, known as the prolonged
 5                  terminal elimination phase of Pb from blood, has been observed in retired Pb workers and
 6                  in workers who continued to work after improved industrial hygiene standards reduced
 7                  their exposures. In the adult simulation shown in Figure 4-11, following cessation of the
 8                  enhanced exposure period at age 50, the blood Pb concentration is reduced by half in
 9                  approximately 75 days. Following this relatively short elimination period, the half-time
10                  of the subsequent 4-year period is approximately 14 years; however, the half-time
11                  increases to approximately 50 years during the period 5-20 years after the reduction in
12                  exposure.

13                  These model predictions are consistent with the slow elimination of Pb from blood and
14                  elimination half-times of several decades for bone Pb (e.g., 16-98 years) that have been
15                  estimated from observations made on Pb workers (Wilker et al., 2011; Fleming et al.,
16                  1997; Gerhardsson et al.. 1995). Based on this hypothetical simulation, a blood Pb
17                  concentration measured 1 year following cessation of a period of increased Pb uptake
18                  would be elevated by only a relatively small amount over the baseline measured prior to
19                  the exposure (3 (ig/dL versus the 2 (ig/dL), whereas, the body burden would remain
20                  elevated. These simulations in Figure 4-11 illustrate how a single blood Pb concentration
21                  measurement, or a series of measurements taken over a short-time span, could be a
22                  relatively poor index of Pb body burden. The simulation shown in Figure 4-11 represents
23                  an exposure that resulted in a quasi-steady state blood Pb concentration of approximately
24                  10 (ig/dL. Exposures that achieve higher blood Pb concentrations, more indicative of
25                  poisoning or historic occupational exposures will result in a more prolonged elevation of
26                  blood Pb concentration following cessation of the enhanced exposure period. Figure 4-12
27                  shows a model simulation of an adult exposed to Pb that results in a quasi-steady state
28                  blood Pb concentration of approximately 90 (ig/dL.  In this case, the blood Pb
29                  concentration remains substantially elevated 1  year following the exposure event
30                  (42 (ig/dL versus 2 (ig/dL) and 20 years following the exposure event (11 (ig/dL).

31                  One important potential implication of the profoundly different kinetics of Pb in blood
32                  and bone is that, for a constant Pb exposure, Pb in bone will increase with increasing
33                  duration of exposure and, therefore, with age. In contrast, blood Pb  concentration will
34                  achieve a quasi-steady state. As a result, the relationship between blood Pb and bone Pb
35                  will diverge with increasing exposure duration and age. This divergence can impart
36                  different degrees of age-confounding when either blood Pb or bone  Pb is used as an
37                  internal dose metric in dose-response models. In a review of epidemiologic studies that
38                  evaluated the associations between blood Pb, bone Pb and cognitive function, the
      November 2012                             4-72                  Draft - Do Not Cite or Quote

-------
 1                   association was stronger for bone Pb than blood Pb (particularly for longitudinal studies)
 2                   for older individuals with environmental Pb exposures and low blood Pb levels (Shih et
 3                   al., 2007). In contrast, occupational workers with high current Pb exposures had the
 4                   strongest associations for blood Pb levels with cognitive function, thus providing
 5                   evidence for this divergence (Shih et al., 2007).

 6                   The aforementioned expectation for an increase  in bone Pb and body burden with age
 7                   applies to scenarios of constant exposure but not necessarily to real world populations in
 8                   which individual and population exposures have changed overtime. Longitudinal studies
 9                   of blood and bone Pb trends have not always found strong dependence on age (Nie et al.
10                   2009; Kimetal.  1997). Kim et al. (1997) found that bone Pb levels increased with
11                   increasing age in elderly adults (age 52-83) years), only when the data were analyzed
12                   cross-sectionally. When analyzed longitudinally, the trend for individual patella Pb was a
13                   23% decrease over a 3-year period (approximate ti/2 of 8 years), whereas tibia Pb levels
14                   did not change  with over the same period. Therefore, although older individuals tended to
15                   have higher bone Pb levels, the 3-year temporal  trend for individuals was a loss of Pb
16                   from the more labile Pb stores in trabecular bone. Nie et al. (2011 a) observed that
17                   longitudinal observations of blood and bone Pb in elderly adults did not show a
18                   significant age  effect on the association between blood Pb and bone Pb (patella and tibia),
19                   when the sample population (n=776) was stratified into age tertiles (mean age 62, 69 or
20                   77 years). Nie et al. (2009) did find that regressed function bone Pb and appeared to level
21                   off at bone Pb levels >20 (ig/g bone mineral.
      November 2012                             4-73                  Draft - Do Not Cite or Quote

-------
                     10
 O)

~  "H
CL
i   "^
                      2 -
                                                            20
                                                                                  CD

                                                                             «   1
                                                                                  O
                                                                                  •^
                                                                                  CD
                                                                                  O
                                                                             10  <
                                                                                  CD
                                                                                  c.
                                                                                  a
                                                                                  (D
                                                                                  3
                                                                          -5   •=•
                                                                                 CD
                        25   30   35   40   45   50   55   60   65   70

                                             Age (year)
                       350
                 .0
                 Q.
      300  -


•o    250  -


^ > 200  -


||15°-

•T    100  -
E
'£     50  H
                                                                           -   15
                                                                      C>-i   10
                                                                           -  5
                                                            20
               03
               o
               D
               (D
               O

               CD
               O

              <
               03
               c.
               a
               (D
                                                                                  3
                                                                                  CD
                            25   30   35  40  45   50   55   60   65   70

                                             Age (year)
Note: A constant baseline intake of 20 |jg/day from age 0-30 results in a quasi-steady state blood Pb concentration and body
burden. An increase in Pb intake to 120 ug/day from age 30 to 50 gives rise to a relatively rapid increase in blood Pb, to a new
quasi-steady state, and a slower increase in body burden (upper panel). At age 50, intake returns to the baseline of 20 ug/day.
Following the long period of elevated Pb intake, there is a rapid decline in blood Pb from 9 to 3 ug/dL over the first year and a more
gradual decline in blood Pb to less than 2 ug/dL by age 60.  The time-integrated blood Pb concentration increases over the lifetime,
with a greater rate of increase during periods of higher Pb uptake (lower panel). Simulation based on ICRP Pb biokinetics model
(Leggett. 1993).


Figure 4-11      Simulation of relationship between blood Pb concentration,  bone

                   Pb and body burden  in adults with  relatively low Pb  intake.
November 2012
                           4-74
Draft - Do Not Cite or Quote

-------
                                                                           600
                             25   30   35   40  45   50   55   60   65  70

                                               Age (year)
                             2500
                              600
                                  25  30  35  40  45  50  55  60  65  70

                                               Age (year)
Note: A constant baseline intake of 20 |jg/day from age 0-30 results in a quasi-steady state blood Pb concentration and body
burden. A increase in Pb intake to 6020 ug/day from age 30 to 50 gives rise to a relatively rapid increase in blood Pb, to a new
quasi-steady state, and a slower increase in body burden (upper panel). At age 50, intake returns to the baseline of 20 ug/day.
Following the long period of high Pb intake, there is a rapid decline in blood Pb from 90 to 40 ug/dL over the first a year followed by
a more gradual decline in blood Pb to 20 ug/dL by age 60 and 10 ug/dL at age 70. The time-integrated blood Pb concentration
increases over the lifetime, with a greater rate of increase during periods of higher Pb uptake (lower panel). Simulation based  on
ICRP Pb biokinetics model (Leggett. 1993).


Figure 4-12      Simulation of relationship between blood Pb concentration, bone
                    Pb and body burden  in adults with relatively high Pb intake.
November 2012
4-75
Draft - Do Not Cite or Quote

-------
 1                  Tibia bone Pb is correlated with time-integrated blood Pb concentration (i.e., CBLI).
 2                  McNeill et al. (2000) compared tibia Pb levels and cumulative blood Pb indices in a
 3                  population of 19- to 29-year-olds who had been highly exposed to Pb in childhood from
 4                  the Bunker Hill, Idaho smelter. They concluded that Pb from exposure in early childhood
 5                  had persisted in the bone matrix until adulthood. The bone Pb/CBLI slopes from various
 6                  studies range from 0.022 to 0.067 ug/g bone mineral per ug-year/dL (Healev et al.. 2008;
 7                  Hu et al.. 2007a). Because the CBLI is a cumulative function which cannot decrease over
 8                  time, CBLI and bone Pb would be expected to diverge following cessation of exposure,
 9                  as bone Pb levels decrease. This divergence was observed as a lower bone Pb/CBLI slope
10                  in retired Pb workers compared to active workers and in worker populations whose
11                  exposures declined over time as a result of improved industrial hygiene (Fleming et al..
12                  1997; Gerhardsson et al.. 1993).

13                  Although differences in kinetics of blood and bone Pb degrade the predictive value of
14                  blood Pb as a metric of Pb body burden, within a population that has similar exposure
15                  histories and  age demographics, blood and bone Pb may show relatively strong
16                  associations.  A recent analysis of a subset of data from the Normative Aging Study (an
17                  all male cohort) showed that cross-sectional measurements of blood Pb concentration
18                  accounted for approximately 9% (tibia) to 13% (patella) of the variability in bone Pb
19                  levels. Inclusion of age in the regression model accounted for an additional 7-10% of the
20                  variability in  bone Pb (Park et al.. 2009c).


                    Mobilization of Pb from Bone in Adulthood

21                  In addition to changes in exposure (e.g., declines in exposure discussed in prior sections),
22                  there are physiological processes during different life circumstances that can increase the
23                  contribution of bone Pb to blood Pb. These life circumstances include times of
24                  physiological stress associated with enhanced bone remodeling such as during pregnancy
25                  and lactation  (Hertz-Picciotto et al.. 2000; Silbergeld. 1991; Manton. 1985). menopause
26                  or in the elderly (Silbergeld et al.. 1988). extended bed rest (Markowitz and Weinberger.
27                  1990). hyperparathyroidism (Kessler et al.. 1999) and severe weight loss (Riedt et al..
28                  2009).

29                  During pregnancy, bone Pb can serve as a Pb source as maternal bone is resorbed for the
30                  production of the fetal skeleton (Gulson et al.. 2003; Gulson et al.. 1999; Franklin et al..
31                  1997; Gulson etal.. 1997). Increased blood Pb during pregnancy has been demonstrated
32                  in numerous studies and these changes have been characterized as a "U-shaped" pattern
33                  of lower blood Pb concentrations during the second trimester compared to the first and
34                  third trimesters (Lamadrid-Figueroa et al.. 2006; Gulson et al.. 2004a: Hertz-Picciotto et
35                  al.. 2000; Gulson etal.. 1997; Lagerkvist et al.. 1996; Schuhmacher et al..  1996;
      November 2012                             4-76                  Draft - Do Not Cite or Quote

-------
 1                  Rothenberg et al., 1994a). The U-shaped relationship reflects the relatively higher impact
 2                  of hemodilution in the second trimester versus the rate of bone Pb resorption
 3                  accompanying Ca2+ releases for establishing the fetal skeleton. In the third trimester, fetal
 4                  skeletal growth on calcium demand is greater, and Pb released from maternal skeleton
 5                  offsets hemodilution. Gulson et al. Q998b) reported that, during pregnancy, blood Pb
 6                  concentrations in the first immigrant Australian cohort (n = 15) increased by an average
 7                  of about 20% compared to non-pregnant migrant controls (n = 7). Skeletal contribution to
 8                  blood Pb, based on the isotopic composition for the immigrant subjects, increased in an
 9                  approximately linear manner during pregnancy. The mean increases for each individual
10                  during pregnancy varied from 26% to 99%. Interestingly, the percent change in blood Pb
11                  concentration was significantly greater during the post-pregnancy period than during the
12                  second and third trimesters. This is consistent with Hansen et al. (20 lib) that
13                  demonstrated the greatest blood Pb levels at 6 weeks postpartum compared to the second
14                  trimester in 211 Norwegian women. Increased calcium demands of lactation (relative to
15                  pregnancy) may contribute to the greater change in blood Pb observed post pregnancy
16                  compared to the second and third trimesters. The contribution of skeletal Pb to blood Pb
17                  during the post-pregnancy period remained essentially constant at the increased level of
18                  Pb mobilization.

19                  Gulson et al. (2004a) observed that calcium supplementation was found to delay
20                  increased mobilization of Pb from bone during pregnancy and halved the  flux of Pb
21                  release from bone during late pregnancy and postpartum. In another study, women whose
22                  daily Ca2+ intake was 850 mg per day showed lower amounts of bone resorption during
23                  late pregnancy and postpartum than those whose intake was 560 mg calcium per day
24                  (Manton et al.. 2003). Similarly, calcium supplementation (1,200 mg/day) in pregnant
25                  Mexican women resulted in an 11% reduction in blood Pb level compared to placebo and
26                  a 24% average reduction for the most compliant women  (Ettinger et al.. 2009). When
27                  considering baseline blood Pb levels in women who were more compliant in taking
28                  calcium supplementation, the reductions were similar for those <5 (ig/dL  and those
29                  > 5  (ig/dL (14% and 17%, respectively).  This result is in contrast to a study of women
30                  who had blood Pb concentrations <5 (ig/dL, where calcium supplementation had no
31                  effect on blood Pb concentrations (Gulson et al., 2006b). These investigators attributed
32                  their results to changes in bone resorption with decoupling of trabecular and cortical bone
33                  sites.

34                  Miranda et al. (2010) studied blood Pb level among pregnant women aged 18-44 years
35                  old. The older age segments in the study presumably had greater historic Pb exposures
36                  and associated stored Pb than the younger age segments. Compared with the blood Pb
37                  levels of a reference group in the 25-29 years old age category, pregnant women
38                  > 30 years old had significant odds of having higher blood  Pb levels (aged 30-34:
      November 2012                            4-77                  Draft - Do Not Cite or Quote

-------
 1                  OR = 2.39, p <0.001; aged 35-39: OR = 2.98, p <0.001; aged 40-44: OR = 7.69,
 2                  p <0.001). Similarly, younger women had less chance of having higher blood Pb levels
 3                  compared with the reference group (aged 18-19: OR = 0.60, p = 0.179; aged 20-24:
 4                  OR = 0.54, p = 0.015). These findings indicate that maternal blood Pb levels are more
 5                  likely the result of Pb mobilization of bone stores from historic exposures as opposed to
 6                  contemporaneous exposures.

 7                  Blood Pb levels increase during lactation due to alterations in the endogenous bone Pb
 8                  release rate. After adjusting for patella Pb concentration, an increase in blood Pb levels of
 9                  12.7% (95% CI: 6.2, 19.6) was observed for women who practiced partial lactation and
10                  an increase of 18.6% (95% CI: 7.1, 31.4) for women who practiced exclusive lactation
11                  compared to those who stopped lactation (Tellez-Rojo et al., 2002). In another Mexico
12                  City study, Ettinger et al. (2006; 2004b) concluded that an interquartile increase in patella
13                  Pb was associated with a 14% increase in breast milk Pb, whereas for tibia Pb the
14                  increase was ~5%. Breast milk:maternal blood Pb concentration ratios are generally <0.1,
15                  although values of 0.9 have been reported (Kovashiki et al., 2010; Ettinger et al., 2006;
16                  Gulsonet al.. 1998a). Dietary intake of polyunsaturated fatty acids (PUFA) has been
17                  shown to weaken the association between Pb levels in patella and breast milk, perhaps
18                  indicating decreased transfer of Pb from bone to breast milk with PUFA consumption
19                  (Arora et al., 2008). Breast milk as a source of infant Pb exposure was also discussed in
20                  Section 4.1.3.3 on dietary Pb exposure.

21                  The Pb content in some bones (i.e., mid femur and pelvic bone) plateau at middle age and
22                  then decreases at older ages (Drasch etal..  1987). This decrease is most pronounced in
23                  females and may be due to osteoporosis and release of Pb from resorbed bone to blood
24                  (Gulson et al., 2002). Two studies indicate  that the endogenous release rate in
25                  postmenopausal women ranges from 0.13-0.14 (ig/dL in blood per (ig/g bone and is
26                  nearly double the rate found in premenopausal women (0.07-0.08 (ig/dL per (ig/g bone)
27                  (Popovic et al.. 2005; Garrido Latorre et al.. 2003). An analysis of data on blood Pb
28                  concentrations and markers of bone formation (serum alkaline phosphatase) and
29                  resorption (urinary cross-linked N-telopeptides, NTx) in a sample of U.S. found that
30                  blood Pb concentrations were higher in women (pre- or post-menopausal) who exhibited
31                  the highest bone formation or resorption activities (Jackson et al.. 2010). Calcium or
32                  vitamin D supplementation decreased the blood Pb concentrations in the highest bone
33                  formation and resorption tertiles of the population of post-menopausal women.
34                  Significant associations between increasing NTx and increasing blood Pb levels
35                  (i.e., increased intercept of regression model relating the change in blood Pb per change
36                  in bone Pb) has also been observed in elderly males (Nie et al., 2009).
      November 2012                             4-78                  Draft - Do Not Cite or Quote

-------
 1                  Studies of the effect of hormone replacement therapy on bone Pb mobilization have
 2                  yielded conflicting results (Popovic et al.. 2005; Berkowitz et al.. 2004; Garrido Latorre
 3                  et al.. 2003; Korrick et al.. 2002; Webber etal., 1995). In women with severe weight loss
 4                  (28% of BMI in 6 months) sufficient to increase bone turnover, increased blood Pb levels
 5                  of approximately 2.1 (ig/dL (250%) were reported, and these blood Pb increases were
 6                  associated with biomarkers of increased bone turnover (e.g., urinary pyridinoline cross-
 7                  links) (Riedt et al.. 2009).
      4.3.6        Relationship between Pb in Blood and Pb in Soft Tissues

 8                  Figure 4-13 shows simulations of blood and soft tissues Pb (including brain) for the same
 9                  exposure scenarios previously displayed.  Pb uptake and elimination in soft tissues is
10                  much faster than bone. As a result, following cessation of a period of elevated exposure,
11                  Pb in soft tissues is more quickly returned to blood. The terminal elimination phase from
12                  soft tissue mimics that of blood, and it is similarly influenced by the contribution of bone
13                  Pb returned to blood and being redistributed to soft tissue.

14                  Information on Pb levels in human brain is limited to autopsy data. These data indicate
15                  brain/blood Pb ratios of approximately 0.5 in infancy which remain relatively constant
16                  over the lifetime (range 0.3 to 1.1) (Barry. 1981. 1975). The simulation of brain Pb
17                  shown in Figure 4-14 reflects general concepts derived from observations made in
18                  non-human primates, dogs and rodents. These observations suggest that peak Pb levels in
19                  the brain are reached 6 months following a bolus exposure and within two months
20                  approximately 80% of steady state brain Pb levels are reached (Leggett 1993). There is a
21                  relatively slow elimination of Pb from brain (ti/2 ~ 2 years) compared to other soft tissues
22                  (Leggett 1993). This slow elimination rate is reflected in the slower elimination phase
23                  kinetics is shown in Figure 4-14. Although in this model, brain Pb to blood Pb transfer
24                  half-times are assumed to be the same in children and adults, uptake kinetics are assumed
25                  to be faster during infancy and childhood, which achieves a higher fraction of the soft
26                  tissue burden in brain, consistent with higher brain/body mass relationships. The uptake
27                  half times predicted by Leggett (1993) vary from 0.9 to 3.7 days, depending on age.
28                  Brain Pb kinetics represented in the simulations are simple outcomes  of modeling
29                  assumptions and cannot currently be verified with available observations in humans.
      November 2012                             4-79                  Draft - Do Not Cite or Quote

-------
                       10
                        8 -
                    O)
                    ^  6 H
                    .a
                    Q.
                    •o
                    m
                        2 -
                                               ,	Blood
                                            "" \
                                                       •Soft Tissue
                             1.0
                          -- 0.8
                                                                        0.0
                                            4       6
                                             Age (year)
                          10
                        10
                                                                         0.0
                           25  30  35   40   45   50  55   60   65   70
                                             Age (year)
Note: For the child simulation (upper panel), baseline Pb intake is 3.2 ug/day from birth until age 2, followed by a period of increased
intake to 38.2 ug/day from age 2 until 5, with a return to baseline intake at age 5. For the adult simulation (lower panel), baseline
intake is 20 ug/day from age 0-30, followed by a 20-year period of increased intake to 120 ug/day from age 30 to 50, with a return to
baseline intake at age 50. Simulation based on ICRP Pb biokinetics model (Leggett. 1993).


Figure 4-13     Simulation of blood and soft tissue (including brain) Pb in

                   children and adults who experience a period of increased Pb
                   intake.
November 2012
4-80
Draft - Do Not Cite or Quote

-------
                       10
                    „  8
                    en
                        6
                    CL
                    •D

                    8   4H
                    CO
                        2 -
                     m
                                                       	Blood
                                                        — Brain
                                            4        6
                                             Age (year)
                             30


                             25

                                  o
                             20   sr
                                  03

                             15   I
                                  T3
                                  tr
                             10   g


                             5
                          10
                           25   30   35   40  45  50   55   60   65   70

                                              Age (year)
Note: For the child simulation (upper panel), baseline Pb intake is 3.2 ug/day from birth until age 2, followed by a period of increased
intake to 38.2 ug/day from age 2 until 5, with a return to baseline intake at age 5. For the adult simulation (lower panel), baseline
intake is 20 ug/day from age 0-30, followed by a 20-year period of increased intake to 120 ug/day from age 30 to 50, with a return to
baseline intake at age 50. Simulation based on ICRP Pb biokinetics model (Leggett. 1993).


Figure 4-14     Simulation of blood and brain Pb in children and adults who
                   experience a period of increased Pb intake.
November 2012
4-81
Draft - Do Not Cite or Quote

-------
 1                  Urinary filtering and excretion of Pb is associated with plasma Pb concentrations. Given
 2                  the curvilinear relationship between blood Pb and plasma Pb, a secondary expectation is
 3                  for a curvilinear relationship between blood Pb and urinary Pb excretion that may
 4                  become evident only at relatively high blood Pb concentrations (e.g., >25  (ig/dL). Figure
 5                  4-15 shows these relationships predicted from the model. In this case, the  exposure
 6                  scenario shown is for an adult (age 40 years) at a quasi-steady state blood Pb
 7                  concentration; the same relationships hold for children. At lower blood Pb concentrations
 8                  (<25 (ig/dL), urinary Pb excretion is predicted to closely parallel plasma Pb concentration
 9                  for any given blood Pb level (Figure 4-15. top panel). It follows from this that, similar to
10                  blood Pb, urinary Pb will respond much more rapidly to an abrupt change in Pb exposure
11                  than will bone Pb. One important implication of this relationship is that, as described
12                  previously for blood Pb, the relationships between urinary Pb and bone Pb will diverge
13                  with increasing exposure duration and age, even if exposure remains constant.
14                  Furthermore, following an abrupt cessation of exposure, urine Pb will quickly decrease
15                  while bone Pb will remain elevated (Figure 4-15. lower panel).
      November 2012                             4-82                  Draft - Do Not Cite or Quote

-------
                        80
                                            20      30      40

                                              Blood Pb (ng/dL)
                              0.00
                           50
IZ. •

10 •
"> j"
WTJ
5 o> 8
0 3.
3.—
""""^
£ °- 6 •
a>^
•El
Dm 4 .


2
n

	 Blood

^x"**1!'.1- • • • • 	 J •'











	
f**'*'
* •
t
1 •
*
/
r
*
r
*

^— Urine

. 	 Bone

•
•
•
• .




V
^-*— 	 	 „

• ^u


- 15



CD
0
- 10 m

(Q
*™*

- 5

n
                            25  30   35   40   45  50   55   60   65  70

                                              Age (year)
Note: Upper panel, model simulations are for a 40-year old having a constant intake from birth of between 1 and 1,000 ug/day. For
the lower panel, baseline intake is 20 ug/day from age 0-30, followed by a 20-year period of increased intake to 120 ug/day from
age 30 to 50, with a return to baseline intake at age 50. Simulation based on ICRP Pb biokinetics model (Leggett, 1993).


Figure  4-15      Relationship between  Pb in urine, plasma,  blood and bone.
November 2012
4-83
Draft - Do Not Cite or Quote

-------
      4.4       Studies of Pb Biomarker Levels
      4.4.1       Pb in Blood

 1                  Overall, trends in blood Pb levels have been decreasing among U.S. residents over the
 2                  past 35 years. Blood Pb concentrations in the U.S. general population have been
 3                  monitored in the NHANES. Analyses of these data show a progressive downward trend
 4                  in blood Pb concentrations during the period 1976-2010, with the most dramatic declines
 5                  coincident with the phase out of leaded gasoline and reductions in point source Pb
 6                  emissions described in Section 3.2 (Pirkle et al.. 1998; Brody et al.. 1994; Pirkle et al..
 7                  1994; Schwartz and Pitcher.  1989). The temporal trend for the period 1988-2010 is
 8                  shown in  Figure 4-16. Summary statistics from the most recent publicly available data
 9                  (1999-2010) are presented in Table 4-8 (CDC. 2011 a). The geometric mean Pb
10                  concentration among children 1-5 years of age, based on the  sample collected during the
11                  period 2009-2010, was 1.17 (ig/dL (95% CI: 1.08, 1.26), which was decreased from
12                  2007-2008(1.51 jig/dL, 95% CI:  1.37, 1.66). Figure 4-17 uses NHANES data to illustrate
13                  temporal trends in the distribution of blood Pb levels among  U.S. children aged  1-5 years.
14                  For 2005-2010, the 95th percentile of blood Pb levels for children aged 1-5 years was less
15                  than 5 ng/dL. The geometric mean blood Pb concentration among adults > 20 years of
16                  age was 1.23 (ig/dL (95% CI: 1.19, 1.28) for the sample collected during the period
17                  2009-2010 (CDC. 2011 a). Based on these same data, the geometric mean for all males
18                  (aged > 1  year) was 1.31 (ig/dL (95% CI: 1.25, 1.36), and for females (aged > 1 year) was
19                  0.97 (ig/dL (95% CI: 0.93, 1.01).

20                  There has been a steep decline in mean blood Pb levels from 1975 through 2010 among
21                  all birth cohorts from  1975 to 2010 (Figure 4-18). For all cohorts, blood Pb generally
22                  decreases with age during childhood until adolescence; following adolescence (in the
23                  early 20s), blood Pb generally levels off or even increases with age. It is possible that
24                  bone growth in young people and occupational exposure for  adults influences the shape
25                  of these curves. For the 1960 to 1970 birth cohort, the mean blood Pb is the highest of the
26                  cohorts in the 1970s, but beginning in 1993 the mean blood Pb is one of the lowest of the
27                  cohorts. This interaction between time and cohort may be due to the faster release of Pb
28                  from bone in younger people (Rabinowitz. 1991). This interaction is also apparent for
29                  some of the other more recently born cohorts. In comparison, the slopes of blood Pb over
30                  time are nearly parallel among the cohorts born before 1960. This suggests that the time-
31                  cohort-interaction diminishes among older people. Also, the  leveling of the blood Pb in
32                  the 2000s could be due to aging of the birth cohort and consequent slowing of their Pb
33                  release from bone.
      November 2012                             4-84                  Draft - Do Not Cite or Quote

-------
1
2
3
4
5
6
7
When race/ethnicity groups were compared for years 1999-2004, geometric means (GM)
of blood Pb levels in children were highest in the ethnicity category non-Hispanic black
(GM 2.8, 95% CI: 2.5, 3.0) compared to the categories Mexican-American (GM 1.9, 95%
CI:  1.7, 2.0) and non-Hispanic white (GM 1.7, 95% CI: 1.6, 1.8) (Jones et al.. 2009a).
Figure 4-19 demonstrates the change in percent of children (aged 1-5 years) with various
blood Pb levels by race/ethnicity between the survey during 1988-1991 and that during
1999-2004. When these data for children aged 1-5 years were aggregated for all survey
years from 1988 to 2004, residence in older housing, poverty, age, and being
non-Hispanic black were significant predictors of higher Pb levels (Jones et al., 2009a).
                   O)
                   T3
                   TO
                   0)
                       4 -
                       3 -
                       1  -
                                                      Children 1-5 yrs
                                                      Children 6-11 yrs
                                                      Teens 12-19 yrs
                                                      Adults>20 yrs
                           88-91    91-94   99-00   01-02   03-04   05-06   07-08   09-10
                                                    Survey Period
     Note: Shown are geometric means and 95% CIs based on data from NHANES III Phase 1 (Brodvet al.. 1994: Pirkleetal.. 1994):
     NHANES III Phase 2 (Pirkleetal.. 1998): and NHANES IV (CDC. 2011 a). Data for adults during the period 1988-1994 are for ages
     20-49 years, and > 20 years for the period 1999-2008.

     Figure 4-16      Temporal trend in blood Pb concentration.
     November 2012
                            4-85
Draft - Do Not Cite or Quote

-------
Table 4-8 Blood Pb concentrations in the U.S. population.
Survey Stratum Period Geometric Mean (ug/dL) 95% Confidence Interval
All 1999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
1-5yr 1999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
6-1 1 yr 1 999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
12-1 9 yr 1999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
> 20 yr 1 999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
Males 1 999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
1.66
1.45
1.43
1.29
1.27
1.12
2.23
1.70
1.77
1.46
1.51
1.17
1.51
1.25
1.25
1.02
0.99
0.84
1.10
0.94
0.95
0.80
0.80
0.68
1.75
1.56
1.52
1.41
1.38
1.23
2.01
1.78
1.69
1.52
1.47
1.31
1.60, 1.72
1.39, 1.51
1.36, 1.50
1.23, 1.36
1.21, 1.34
1.08, 1.16
1 .96, 2.53
1.55, 1.87
1.60, 1.95
1.36, 1.57
1.37, 1.66
1.08, 1.26
1.36, 1.66
1.14, 1.36
1.12, 1.39
0.95, 1.01
0.91, 1.07
0.79, 0.89
1.04, 1.17
0.90, 0.99
0.88, 1.02
0.75, 0.85
0.74, 0.86
0.64, 0.73
1.68, 1.81
1.49, 1.62
1.45, 1.60
1 .34, 1 .48
1.31, 1.46
1.19, 1.28
1 .93, 2.09
1.71, 1.86
1.62, 1.75
1.42, 1.62
1.39, 1.56
1.25, 1.36
Number of Subjects
7,970
8,945
8,373
8,407
8,266
8,793
723
898
911
968
817
836
905
1,044
856
934
1,011
1,009
2,135
2,231
2,081
1,996
1,074
1,183
4,207
4,772
4,525
4,509
5,364
5.765
3,913
4,339
4,132
4,092
4,147
4,366
November 2012
4-86
Draft - Do Not Cite or Quote

-------
Survey Stratum     Period     Geometric Mean (ug/dL)   95% Confidence Interval   Number of Subjects
Females 1999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
Mexican - Americans 1999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
Non-Hispanic blacks 1999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
Non-Hispanic whites 1 999-2000
2001-2002
2003-2004
2005-2006
2007-2008
2009-201 0
1.37
1.19
1.22
1.11
1.11
0.97
1.83
1.46
1.55
1.29
1.25
1.14
1.87
1.65
1.69
1.39
1.39
1.24
1.62
1.43
1.37
1.28
1.24
1.10
1.32,
1.14,
1.14,
1.05,
1.06,
0.93,
1.75,
1.34,
1.43,
1.21,
1.15,
1.03,
1.75,
1.52,
1.52,
1.26,
1.30,
1.18,
1.55,
1.37,
1.32,
1.19,
1.16,
1.04,
1.43
1.25
1.31
1.17
1.16
1.01
1.91
1.60
1.69
1.38
1.36
1.28
2.00
1.80
1.89
1.53
1.48
1.30
1.69
1.48
1.43
1.37
1.33
1.16
4,057
4,606
4,241
4,315
4,119
4.427
2,742
2,268
2,085
2,236
1,712
1,966
1,842
2,219
2,293
2,193
1,746
1,593
2,716
3,806
3,478
3,310
3,461
3,760
Age strata correspond to the NHANES study design.
Source: Adapted from data from the NHANES (CDC. 2011 a).
November 2012
4-87
Draft - Do Not Cite or Quote

-------
                      10 -
Note: Top: all data. Bottom: data for subjects having blood Pb levels less than 15 ug/dL.
Source: Adapted from data from the NHANES (NCHS. 2010)

Figure 4-17     Box plots of blood Pb levels among U.S. children (1-5 years old at
                 baseline) from the NHANES survey, 1988-2010.
November 2012
4-8
Draft - Do Not Cite or Quote

-------
    o
   .2 LO
    C
    ro
    0) CO
    o
    o
    0)
    O)
cohort birth yrs
 1900 to 1930<
 1930to1940<
 1940 to 1950 '
 1950 to 1960i
 1960 to 1970<
 1970 to 1975<
 1975 to 1980J
 1980 to 1985i
 1985 to 1990
 1990 to 1995
 1995 to 2000
 2000 to 2005
 2005 to 2008 «
        1970
                1980
   1990
exam year
2000
2010
Note: The means of logged blood Pb were weighted to represent national averages. Data were from the publicly available
NHANES II, NHANES II for 1988-1991 and 1992-1994, and the continuous NHANES in 1999-2000, 2003-2004, 2005-2006,
2007-2008. Continuous NHANES data from 2001-2002 and 2009-2010 are not included because there were only 551 blood Pb
samples in each of those data sets. The year plotted for exam year was the reported exam year for NHANES II, the middle year of
each of the phases of NHANES III, and the second year of each of the continuous NHANES.
Source: Adapted from data from the NHANES (NCHS. 2010)


Figure 4-18      Blood  Pb cohort means versus year of exam.
November 2012
                             4-89
                 Draft - Do Not Cite or Quote

-------
                  10
                                                      1999 - 2004
                     0
                           <1     l-<2.5   2.5-<5   5-<7.5   7.5 - <10    > 10
                                      Blood Pb Level (ug/dL)
                 ...+.. Non-Hispanic black^^^—Mexican American —^— Non-Hispanic white

Source: Data used with permission of the American Academy of Pediatrics, Jones et al. (2009a)

Figure 4-19     Percent distribution of blood Pb levels by race/ethnicity among
                 U.S. children (1-5 years) from the NHANES survey, 1988-1991 (top)
                 and 1999-2004 (bottom).
November 2012
4-90
Draft - Do Not Cite or Quote

-------
 1                  In agreement with the 1986 AQCD (U.S. EPA. 1986a). several studies have shown
 2                  seasonal variation in blood Pb concentrations in children (e.g., Havlena et al.. 2009;
 3                  Gulson et al.. 2008; Kemp et al.. 2007; Laidlaw et al.. 2005; Haley and Talbot. 2004;
 4                  Johnson and Bretsch. 2002; Yiin et al.. 2000; Johnson et al.. 1996). with elevated
 5                  concentrations during the warm season and lower levels in the cold season. Seasonal
 6                  dynamics of blood Pb concentrations in children appear to be  caused at least in part by
 7                  seasonal patterns in access of children to soils and soil properties (e.g., moisture content)
 8                  that may contribute to seasonal variation in entrainment of soil and dust Pb into breathing
 9                  zone air (Laidlaw et al.. 2012; Laidlaw et al.. 2005; Johnson and Bretsch. 2002). Seasonal
10                  variation in blood Pb concentrations occur with strong associations with soil Pb
11                  concentrations (Johnson and Bretsch. 2002). Laidlaw et al. (2012) observed that air Pb in
12                  the PM2 5 fraction and PM2 5 attributed to soil were elevated in the warm season compared
13                  with the cold season. Yiin et al.  (2000) found that geometric mean for blood Pb, floor Pb
14                  loading and concentration, and carpet dust loading were statistically significantly higher
15                  in the cold season compared with the hot season. However, regression of blood Pb on
16                  floor and windowsill dust with and without adjustment  for the hot, warm, and cool
17                  seasons showed no  statistically significant effect of the  seasons directly on blood Pb.
18                  Meteorological factors  appear to contribute to blood Pb seasonality. Laidlaw et al. (2005)
19                  analyzed the temporal relationships between child blood Pb concentrations and various
20                  atmospheric variables in three cities (Indianapolis, IN:  1999-2002; Syracuse, NY:
21                  1994-1998; New Orleans, LA:  1998-2003). Blood Pb data was obtained from public
22                  health screening programs conducted in the three cities. Blood Pb samples were
23                  dominated by children  <5 years of age and age distribution varied across the three cities.
24                  The temporal variation in blood Pb concentrations in each city was predicted by
25                  multivariate regression models that included the following significant variables: PMi0,
26                  wind speed, air temperature, and soil moisture; as well  as dummy variables accounting
27                  for temporal displacement of the effects of each independent variable on blood Pb.
28                  Laidlaw et al. (2005) reported R2 values for the regression models, but did not report the
29                  actual  regression coefficients. The R2 values were as follows:  Indianapolis 0.87
30                  (p = 0.004); Syracuse 0.61 (p = 0.0012); New Orleans 0.59 (p O.OOOOl).

31                  Studies have examined the change in blood Pb with changes in potential Pb sources.
32                  Gulson et al. (2004b) observed that children living near a Zn-Pb smelter in Australia had
33                  blood Pb levels ranging from 10 to 42 (ig/dL, with 55-100% of Pb attributed to the
34                  smelter based on isotope ratio analysis. Rubio-Andrade et al. (2011) followed a cohort of
35                  6-8 year old children living within 3.5 km of a Mexican smelter at 0, 6, 12, and
36                  60 months after environmental intervention including removal of 100,000 kg of
37                  Pb-containing dust from roads and homes using high efficiency vacuums. Soil Pb was
38                  concurrently obtained but not reported at 6, 12, or 60 months.  Median blood Pb level at
39                  initiation of the study was  10.1 (ig/dL for the 598 initial participants (average age: 7.2 y),

      November 2012                             4-91                  Draft - Do Not Cite or Quote

-------
 1                  and median soil Pb was 3,300 mg/kg at the start of the study. After 60 months, median
 2                  blood Pb level was 4.4 ug/dL for the remaining 232 participants (average age: 12.2 y),
 3                  and median soil Pb concentration was 370 mg/kg at that time. Bonnard and McKone
 4                  (2009) modeled blood Pb of French children ages 21-74 months living within a village
 5                  containing a Pb smelter and estimated blood Pb levels of 3.2-10.9 ug/dL. Lanphear et al.
 6                  (1998) noted that the probability of children having blood Pb > 10 ug/dL increases both
 7                  with exterior soil Pb content and interior Pb dust loading. Mielke et al. (20lib) noted
 8                  significant increases in percentages of children younger than 7-years old with blood Pb
 9                  level > 10 ug/dL for those living in inner city New Orleans housing developments
10                  (22.9%) compared with children living in communities located on the city outskirts
11                  (9.1%). At the same time, median soil Pb was significantly higher in the inner city
12                  (438 mg/kg) compared with the city outskirts (117 mg/kg).

13                  For infants <1 year old, very little data are available on blood Pb levels. Simon et al.
14                  (2007) followed a cohort of 13 children living near an Australian smelter from birth
                    V	/                                  O
15                  through 36 months. In general, except for children born with low blood Pb levels of ~1 to
16                  2 ug/dL, immediately after birth blood Pb levels fell for 1-2 months to approximately
17                  47% of birth blood Pb level. After this initial fall, all infants' blood Pb levels rose with
18                  age until approximately  12 months old for children living in a high risk area and until
19                  approximately 18 months for children living in a low  risk area (Simon et al.. 2007).
20                  Median blood Pb level among the children was 1.9 ug/dL at 2 months and increased to
21                  13.6 ug/dL at 16 months. Geometric mean hand-Pb loading of the child and the mother
22                  were significant contributors to the area under the curve for infant blood Pb, with 46%
23                  (infant hand loading) and 60% (mother hand loading) of the variance being explained by
24                  these variables, respectively; geometric mean of the mothers' blood Pb explained 46% of
25                  the variance (Simon et al.. 2007). Across all the data,  there was a good correlation
26                  between child blood Pb level and child hand Pb loading (R2 = 0.70). In another study
27                  (Carbone et al.. 1998). blood Pb levels of 15 infants aged 6-12 months were statistically
28                  significantly lower than their neonatal cord blood Pb levels (2.24 ug/dL versus
29                  4.87 ug/dL). Additionally, 3 infants born with blood Pb levels of greater than 7 ug/dL
30                  were followed for a week, there was  a dramatic drop in the blood Pb of from an average
31                  of 7.6 ug/dL on Day  1 to 2.4 ug/dL on Day 7 (Carbone et al.. 1998).

32                  Pb body burden has been reported among individuals known to consume wild game
33                  hunted with Pb shot.  For example, fifty men from Nuuk, Greenland participated in a
34                  study in which they recorded their diet and produced blood samples (Johansen et al..
35                  2006). Men who regularly ate hunted sea birds had an average blood Pb concentration of
36                  12.8 ug/dL, in contrast with those who did not and had an average blood Pb
37                  concentration of 1.5 ug/dL. Umbilical cord blood was collected from a cohort of Inuit
38                  newborns from northern Quebec, where the Inuit population consumes game killed with
      November 2012                             4-92                   Draft - Do Not Cite or Quote

-------
 1                  Pb shot (Levesque et al., 2003). The geometric mean cord blood Pb level was
 2                  0.19 (imol/L [3.9 (ig/dL], with a range of 0.01-1.31 (imol/L [0.2-27 (ig/dL]; the Canadian
 3                  level of concern for cord blood Pb is 0.48 (imol/L [10 jig/dL]. The authors contrasted the
 4                  finding that 7% of Inuit newborns had cord blood Pb concentration > 0.48 (imol/L
 5                  [10 (ig/dL] in contrast with 0.16% of the Caucasian population in southern Quebec.

 6                  Recent studies have sought to characterize human exposure to Pb from piston-engine
 7                  aircraft emissions. Section 3.2.2.1 describes a study by Carr et al. (2011) in which Pb
 8                  concentrations, both modeled and monitored, extended beyond airport property. Miranda
 9                  et al. (2011) used GIS to study the association between blood Pb level and distance from
10                  airports in six North Carolina Counties. They observed that the trend in blood Pb level
11                  decreases monotonically with distance class from the airports, with subjects within 500
12                  meters of the airports having significantly increased blood Pb levels  ((3 = 0.043, 95% CI:
13                  (0.006,0.080), p <0.05) compared with the general population for a given county after
14                  controlling for proportion of black, Hispanic, percent receiving public assistance, and
15                  household median income at the census block group level and including dummy variables
16                  for season during which the children were screened for blood Pb. In this study, children
17                  living within 500 meters of an airport had blood Pb levels that were, on average, 4.4%
18                  higher than those at distance. Note that the authors did not include Pb emissions in their
19                  model.

20                  Trends in blood Pb levels have been accompanied by changes in Pb isotope ratios for
21                  blood Pb. Isotopic ratios, described in  Sections 3.2 and 3.3 as a tool for source
22                  apportionment, have been used to associate blood Pb measurements with anthropogenic
23                  sources of Pb in the environment. Changes in Pb isotopic ratios in blood samples reflect
24                  the changing influence of sources of Pb following the phase-out of tetraethyl Pb
25                  antiknock agents in automotive gasoline and changes in Pb usage in paints and other
26                  industrial and consumer products (Gulson et al., 2008; Ranft et al.. 2008; Gulson et al..
27                  2006a; Ranft et al.. 2006). Gulson et al. (2006a) illustrated how a linear increase in the
28                  isotopic ratio 206Pb/204Pb occurred in concert with a decrease in blood Pb levels among
29                  selected study populations in Australia during the period  1990-2000  (Figure 4-20).
30                  Gulson et al. (2006a) point out that the isotopic signature of 206Pb/204Pb derived from
31                  Australian mines (median -16.8) differs from that of European and Asian mines, where
32                  206Pb/204Pb varies between -17.4 and -18.1. Liang et al. (2010) also examined the trends
33                  in blood Pb  level over the period 1990 to 2006 in Shanghai and saw a reduction
34                  corresponding to the phase out of Pb in gasoline. A plot of 208Pb/206Pb to 207Pb/206Pb  for
35                  blood and environmental samples showed overlap between the isotopic signature for coal
36                  combustion ash and that measured in blood. This result suggests a growing influence of
37                  Pb from coal ash in Shanghai  in the absence of Pb in automobile emissions. Oulhote et al.
38                  (2011) examined Pb isotope ratios in blood Pb samples of 125 French children aged 6
      November 2012                             4-93                  Draft - Do Not Cite or Quote

-------
1 mo-6 yr. The study found that Pb isotope ratios could be used to attribute Pb exposure to
2 one source for 32% of children and to eliminate an unlikely source of Pb exposure in
3 30% of children.






source
O Adelaide
-s, BK Adult
" females
D BK children
A Broken Hill
V Hobart
<| PbinCa
[> Port Pirie
r Sydney
' children



17.60-

.a 17.20-
Q.
1
*d
B.
1
16.80-



16.40-



D
Q
+ 1
\yf*°
I I 4
R Sq Linear = 0.53
A A

12.5-

10.0-

7.5-
ffl
£
5.0-
2.5-


0.0-
1 1 1 1 1 1
1990 1992 1994 1996 1998 2000

>

>

O
> 1
Q" * 0 ^
^'"""""'"•••^ ^
7 § "~~~~~
-------
      4.4.2        Pb in Bone

 1                  An extensive national database (i.e., NHANES) is available for blood Pb concentrations
 2                  in children and adults, as described in Section 4.4.1. Bone Pb concentrations are less well
 3                  characterized. Table 4-9 and Table 4-10 are compilations of data from epidemiologic
 4                  studies that provided bone Pb concentrations by K-XRF and/or variability in
 5                  concentrations among individuals without reported occupational exposure and those with
 6                  occupational exposures, respectively. In non-occupationally exposed individuals, typical
 7                  group mean tibia bone Pb concentrations ranged from 10 to 30 ug/g. Patella bone Pb
 8                  levels are typically higher than tibia bone Pb levels in the studies considered (Table 4-9).
 9                  For example, in the Normative Aging Study, patella bone Pb concentrations were
10                  approximately 32 ug/g, whereas tibia bone Pb concentrations were about 22 ug/g.
11                  Occupationally exposed individuals generally had greater bone Pb concentrations than
12                  seen in control groups (i.e., unexposed). Bone Pb data in Table 4-10 for occupationally
13                  exposed individuals were also generally higher compared to non-occupationally exposed
14                  individuals (Table 4-9).
     November 2012                             4-95                   Draft - Do Not Cite or Quote

-------
Table 4-9



Reference
Bandeen-
Roche et al.
(2009)



Bellinger et
al. (|994a)





Cheng et al.
(2001 )


















Coon et al.
(2006)













Epidemiologic studies that provide bone Pb
occupationally exposed populations.

Prior Pb Bone Pb
Study Methods Exposure biomarker
Cohort: Cumulative Tibia
Baltimore Memory Study cohort
Age (yrs): 50-70
N:1,140
Location: Baltimore, MD
Study Period: 2001 -2005
Cohort: Not reported Cumulative Tibia
Age (yrs): 5-8 (recruited); Patella
19-20 (follow-up)
N:79
Location: Boston, MA

Study Period: 1989-1990
Cohort: Cumulative Tibia
Normative Aging Study cohort Patella
Age (yrs): Mean ± SD:
Normotensive:
65.49 ±7.1 7
Borderline hypertension:
68.3 ± 7.79
Definite hypertension:
67.93 ± 6.79
N: 833 males
Location: Boston, MA
Study Period:
8/1/1991-12/31/1997







Cohort: Cumulative Tibia
Participants from Henry Ford Calcaneus
Health System (HFHS)
Age (yrs): > 50; Mean: 69.9
N: 121 cases; 41 4 controls
Location: Southeastern Michigan
Study Period:
1995-1999 (participants received
primary health care services)






measurements


Bone Pb Cone.
(Hg/g)
Meant SD
Tibia: 18.8 ± 11.6




Mean (Range):
Tibia: 5.4 (3-1 6)

Patella: 9.2 (4-1 8)



Meant SD
Tibia:
Normotensive:
20.27 ±11. 55
Borderline
hypertension:
23.46 ±15.02
Definite
hypertension:
22.69 ± 14.71

Patella:
Normotensive:
28.95 ±18.01
Borderline
hypertension:
33.73 ± 21 .76
Definite
hypertension:
32.72 ± 19.55
Mean± SD:
Tibia: 12.5 ±7.8
Calcaneus: 20.5 ±
10.2











for non-

Distribution
of Bone Pb
(ug/g)
Not reported





High
exposure:
>24

Low exposure:
<8.7

Lowest
quintile: Tibia:
8.5
Patella: 12.0

Highest
quintile:
Tibia: 36.0
Patella: 53.0











Tibia
Q1 : 0-5.91
Q2:
5.92-10.40
Q3:
10.41-15.50
Q4:> 15.51

Calcaneus
Q1: 0-1 1.70
Q2:
11.71-19.07
Q3:
19.08-25.28
Q4: > 25.29
November 2012
4-96
Draft - Do Not Cite or Quote

-------


Reference
Elmarsafawy
et al. (2006)





Glass et al.
(2009)








Hsieh et al.
(2009b)

Huetal.
(1996a)
[As reported
in Navas-
Acien et al.,
(2008)1


Study Methods
Cohort:
Normative Aging Study
Age (yrs): Not reported
N: 471 elderly males
Location: Greater Boston area,
MA
Study Period: 6/1 991-1 2/1 994
Cohort:
Baltimore Memory Study
Age (yrs): Mean: 59.4;
Range: 50-70
N: 1,001
Location: Baltimore, MD
Study Period: 2001 -2005





Cohort:Not reported
Age (yrs): Mean: Control: 46.06
N: 18 controls
Location: Not reported
Study Period: Not reported
Cohort:
Normative Aging Study
Age (yrs): 48-92; Mean ± SD:
66.6 ± 7.2
N: 590 males
Location: Boston, MA
Study Period: 8/1 991-1 2/1 994

Prior Pb
Exposure
Not reported





Cumulative
(lifetime)








Control group for
occupational
exposure group

Cumulative

Bone Pb Bone Pb Cone.
biomarker (ug/g)
Tibia Meanni SD:
Patella Tibia: 21.6±1 2.0
Patella: 31 .7 ±18.3




Tibia Mean ± SD:
Tibia: 18.8 ±11.1








Tibia Mean ± SD
Patella Tibia Control: 18.51
± 22.40
Patella Control: 7.14
±9.81
Tibia Mean ± SD:
Patella Tibia: 21.8 ± 12.1
Patella: 32.1 ±18.7
Range:
Tibia: <1-96
Patella: 1-142
Distribution
of Bone Pb
(ug/g)
Not reported





NPH Scale:
Lowest tertile:
Mean Tibia
level: 16.3±
11.0

Middle tertile:
Mean Tibia
level: 19.3±
10.7
Highest tertile:
Mean Tibia
level: 20.3 ±
11.4
Not reported

Figures 1 and
2 show both
types of bone
Pb levels
increasing
with age
November 2012
4-97
Draft - Do Not Cite or Quote

-------


Reference
Jain et al.
(2007)

































Kamel et al.
(2002):
Kamel et al.
(2005):
Kamel et al.
(2008)






Prior Pb
Study Methods Exposure
Cohort: Not reported
VA-Normative Aging Study
Age (yrs): Not reported
N: 837 males
Location: Greater Boston, MA
Study Period:
9/1/1991-12/31/2001





























Cohort:Not reported Cumulative
Age (yrs): 30-80 Control group for
N: 256 controls (Bone samples occupational
collected from 41 controls) exposure group
Location: New England (Boston,
MA)
Study Period: 1993-1996





Bone Pb Bone Pb Cone.
biomarker (ug/g)
Tibia Mean ± SD
Patella Tibia:
Non-Cases: 21 .4 ±
13.6
Cases: 24.2 ± 15.9
Patella:
Non-cases:
30.6±19.7
Cases: 36.8 ± 20.8
Range:
Tibia:
Noncases:
-3-1 26
Cases: -5-75

Patella:
Noncases: -10-165
Cases: 5-101



















Tibia Mean ± SE
Patella Tibia Controls: 11.1 ±
1.6
Patella Controls:
16.7 ±2.0





Distribution
of Bone Pb
(ug/g)
Mean ± SD
(Range):

Tibia:
Non-cases:
Tertile 1: 10.2
±3.8 (-3-1 5)
Tertile 2: 19.1
±2.3(16-23)
Tertile 3: 35.5
± 14.4
(24-1 26)
Cases:
Tertile 1: 10.1
±5.3 (-5-1 5)
Tertile 2: 19.8
±2.2(16-23)
Tertile 3: 39.5
±14.9(25-75)
Patella:
Non-cases:
Tertile 1 :
13.9±4.9
(-10-20)
Tertile 2:
27.1±4.1
(21 -34)
Tertile 3:
52.5± 20.7
(35-165)
Cases:
Tertile 1 :
15.3±4.3
(5-19)
Tertile 2: 25.7
±3.8(21-33)
Tertile 3: 53.3
±17.3
(35-101)
Controls
Tibia: N (%)
-7-7: 14(34)
8-1 4: 1 2 (29)
15-61: 15(37)
Patella: N (%)
-4-9: 14(34)
10-20:14(34)
21-107: 13
(32)
November 2012
4-98
Draft - Do Not Cite or Quote

-------


Reference
Khalil et al.
(2009a)



Korrick et al.
(1999) [As
reported in
Navas-Acien
etal.,
(2008)]



Study Methods
Cohort:
1982 Lead Occupational Study
Age (yrs): Control mean: 55
N: 51 controls
Location: Eastern Pennsylvania
Study Period: 1982-2004
Cohort:
Nurses' Health Study
Age (yrs): Combined: 47-74;
Mean ± SD:
Combined: 58.7 ± 7.2;
Cases: 61.1 ±7.1;
High controls: 61.1 ± 7.2;
Low controls: 58. 7 ± 7.1

Prior Pb
Exposure
Control group for
occupational
exposure group



Nonoccupationally
exposed


Bone Pb Bone Pb Cone.
biomarker (ug/g)
Tibia Median (IQR)
Tibia Control: 12
(-8-32)



Tibia Mean ± SD
Patella Tibia:
Combined: 13.3 ±
9.0
Cases: 13.0 ±9.4
High controls: 14.7 ±
10
Distribution
of Bone Pb
(ug/g)
Not reported



Patella:
10th
percentile: 6
90th
percentile: 31

              N: 284 females; (89 cases;
              195 controls)
              Location: Boston, MA
              Study Period: 7/1993-7/1995
                              Low controls: 12.7 ±
                              8.1
                              Patella:
                              Combined: 17.3 ±
                              11.1
                              Cases: 19.5± 12.9
                              High controls: 17.2 ±
                              9
                              Low controls: 15.8 ±
                              10.6




Lee et al.
(2001 a) [As
reported in
Navas-Acien
etal.,
(2008)1




Martin et al.
(2006)




Cohort: Not reported
Age (yrs):
22.0-60.2
Mean ± SD: Controls:
34.5 ±9.1
N: 135 controls

Location: South Korea
Study Period:
10/24/1997-8/19/1999
Cohort:
Baltimore Memory Study




Control group for Tibia
occupational
exposure group







Cumulative Tibia
(lifetime)
Range
Tibia Combined:
-5-69
Patella Combined:
-5-87
Mean ± SD
Tibia Controls: 5.8 ±
7.0


Range
Tibia Controls:
-11-27


Mean ± SD
Tibia: 18.8 ±12.4




Not reported








Tibia IQR:
1 1 .9-24.8
              Age (yrs): 50-70; Mean: 59.4
              N:964
              Location: Baltimore, MD
              Study Period: 5/2001-9/2002
              (1st study visit)
              8/2002-3/2004 (2nd study visit -
              tibia Pb measured)
November 2012
4-99
Draft - Do Not Cite or Quote

-------


Reference
Needleman
et al. (2002)










Osterberg et
al. (1997)
[As reported
in Shih et
al., (2007)1

Parket al.
(2006)





















Prior Pb
Study Methods Exposure
Cohort: Not reported Not reported
Age (yrs):
12-18; Mean age ± SD:
African American cases:
15.8 ±1.4
African American controls:
15.5 ± 1 .1;
White cases: 15.7± 1.3;
White controls: 1 5.8 ± 1.1
N: 194 male youth cases; 146
male youth controls
Location: Allegheny County, PA
(cases); Pittsburgh, PA (controls)
Study Period: 4/1 996-8/1 998
Cohort: Not reported Control group for
Age (yrs): Median: 41 .5 occupational
exposure group
N: 19 male controls
Location: Not reported
Study Period: Not reported
Cohort: Not reported
Normative Aging Study
Age (yrs): Mean: 72.9 ± 6.5
N: 41 3 males
Location: Greater Boston, MA
Study Period:
11/14/2000-12/22/2004; (HRV
measurements taken); 1991-2002
(bone Pb measurements taken)














Bone Pb Bone Pb Cone.
biomarker (ug/g)
Tibia Mean ± SD
Tibia Cases (ppm):
All subjects: 1 1 .0 ±
32.7
African American:
9.0 ±33.6
White: 20 ± 27.5
Tibia Controls (ppm):
All subjects: 1 .5 ±
32.1
African American:
-1 .4± 31 .9
White: 3.5 ± 32.6
Finger bone Median (range)
Finger Bone
Controls:
4 (-19-1 8)


Tibia Median (IQR)
Patella
Tibia: 19.0(11-28)
Patella: 23.0 (15-34)
Estimated Patella3:
16.3(10.4-25.8)
















Distribution
of Bone Pb
(ug/g)
Table 4 of
paper
distributes
bone Pb by
> 25 or <25 for
race, two
parental
figures, and
parent
occupation





Not reported



Median (IQR)
for No. of
metabolic
abnormalities:

Tibia:

0: 18.5
(10.5-23)
1:19(11 -28)
2: 19(12-26)
Patella:
0:22(13.5-32)
1:25(16-36)
2:20(15-32)
Estimated
Patella:
0:16.3
(10.8-24.8)
1: 17.1
(11-29.3)
2:15.1
(9.4-22.1)
November 2012
4-100
Draft - Do Not Cite or Quote

-------


Reference
Parket al.
(2009b)



Parket al.
(2010)








Payton et al.
(1998)




Peters et al.
(2007)




Rajan et al.
(2007)





Prior Pb
Study Methods Exposure
Cohort: Not reported
Normative Aging Study
Age (yrs): Mean: 67.3 ± 7.2
N: 613 males
Location: Greater Boston, MA
Study Period: 8/1 991-1 2/1 995



Cohort: Cumulative (chronic
VA Normative Aging Study cohort exposure)
Age (yrs): Mean: 64.9 (at bone
Pb measurement)
N: 448 males
Location: Eastern Massachusetts
Study Period: 1991-1996





Cohort: Not reported
VA Normative Aging Study cohort
Age (yrs): Mean: 66.8
N: 141 males
Location: Boston, MA
Study Period: 4/1 993-3/1 994
Cohort: Cumulative
Normative Aging Study cohort
Age (yrs): Mean: 66.9
N: 513 male cases
Location: Boston, MA
Study Period: 1991 -1996
Cohort: Not reported
VA Normative Aging Study Cohort
Age (yrs): Mean: 67.5 (at bone
scan)
N: 1075 males
Location: Boston, MA
Study Period: 1991 -2002

Bone Pb Bone Pb Cone.
biomarker (ug/g)
Tibia Median (IQR)
Patella
Tibia: 19(14-27)
Patella: 26 (18-37)



Tibia Mean ± SD
Patella Tibia: 22.5 ± 14.2
Patella: 32.5 ±20.4







Tibia Mean ± SD
Patella Tibia: 22.5 ± 12.2
Patella: 31 .7 ± 19.2



Tibia Mean ± SD
Patella Tibia: 21 .5 ± 13.4
Patella: 31 .5 ± 19.3



Tibia Mean ± SD
Patella Tibia: 22.1 ± 13.8
Patella: 31 .4 ± 19.6



Distribution
of Bone Pb
(ug/g)
Table 1 of
paper
distributes
tibia and
patella Pb by
genotype;
Table 2 of
paper
distributes
tibia and
patella Pb by
number of
gene variants
Tibia IQR: 15
Patella IQR:
21
Table 2 of
paper
provides age-
adjusted mean
bone Pb levels
(age, race,
education,
smoking
[pack-yr],
occupational
noise, noise
notch, BMI,
hypertension,
diabetes)
Not reported




Not reported




Not reported




November 2012
4-101
Draft - Do Not Cite or Quote

-------


Reference
Rajan et al.
(2008)





Rhodes et
al. (2003)











Roelset al.
(1994)




Rothenberg
etal.
(2002a) [as
reported in
Navas-Acien
et al. (2008)1









Study Methods
Cohort:
VA Normative Aging Study Cohort
Age (yrs): a 45
N: 720 males
Location: Boston, MA
Study Period: 1993-2001

Cohort:
VA Normative Aging Study Cohort
Age (yrs): Mean: 67.1
N: 526 males
Location: Boston, MA
Study Period:
1/1/1991-12/31/1995









Cohort: Not reported
Age (yrs): 30-60
N: 68 males
Location: Belgium
Study Period: Not reported



Cohort: Not reported
Age (yrs): 15-44; Mean ± SD:
31.0 ±7.7
N: 720 females
Location: Los Angeles, CA
Study Period: 6/1995-5/2001







Prior Pb
Exposure
Current and
cumulative





Not reported











Control group for
occupational
exposure group




Not reported









Bone Pb Bone Pb Cone.
biomarker (ug/g)
Tibia Mean ± SD
Patella ALAD1-1
Tibia: 21 .9 ±13.8
Patella: 29.3 ± 19.1
ALAD 1 -2/2-2
Tibia: 21 .2 ±11. 6
Patella: 27.9 ± 17.3
Tibia Mean ± SD
Patella Tibia: 21 .9 ± 13.5
Patella: 32.1 ±19.8










Tibia Geometric Mean
(Range)
Tibia Controls:
Normotensive: 21.7
(<1 5.2-69.3)
Hypertensive: 20.2
(<1 5.2-52.9)
Total: 21 .4
(<1 5.2-69.3)
Tibia Mean ± SD
Calcaneus Tibia: 8.0 ± 11 .4
Calcaneus: 10.7 ±
11.9








Distribution
of Bone Pb
(ug/g)
Not reported





No. of
participants
Tibia:
<1-15: 173
(33)
16-24: 186
(35)
25-126: 167
(32)
Patella:
<1-22: 189
(36)
23-35: 165
(31)
36-165:172
(33)
Not reported




Tibia quartiles:
Q1 : -33.7-0.9
Q2: 1 .0-8.0
Q3: 8. 1-16.1
Q4: 16.2-42.5

Calcaneus
quartiles:
Q1 : -30.6-3.0
Q2: 3. 1-10.0
Q3: 10.1 -18.7
Q4: 18.8-49.0
November 2012
4-102
Draft - Do Not Cite or Quote

-------
Reference   Study Methods
                               Prior Pb
                               Exposure
                   Bone Pb
                   biomarker
                                                                            Bone Pb Cone.
                    Distribution
                    of Bone Pb
                    (ug/g)
Shihetal.,
(2006)
Cohort:
Baltimore Memory Study cohort
Age (yrs): Mean: 59.39
N:985
Location: Baltimore, MD
Study Period: Not reported
Not reported
                                                               Tibia
Meant SD:
Tibia: 18.7 ± 11.2
Not reported
Stokes et al.
(1998) [as
reported in
Shihetal.
/onnvM
(ZUU/ )l






Cohort: Not reported
Age (yrs): 19-29 (in 1994); Mean
± SD: Cases: 24.3 ± 3.18
Control: 24.2 ± 3.02
Cases: 9 months-9 yr
(during 1 /1 /1 974-1 2/31 /1 975)
N: 257 cases; 276 controls

Location: Silver Valley, ID;
Spokane, WA
Study Period:
7/10/1994-8/7/1994
Cumulative Tibia
(lifelong)
Environmental
(resided near Pb
smelter during
childhood)






Mean (Range):
Tibia Cases: 4.6
(-28.9-37)
Tibia Controls: 0.6
(-46.4-1 7.4)






Tibia
No. of Cases:
<1 ug/g:
31 .5%
1-5 ug/g:
24.4%

5-10 ug/g:
22.3%
>10 ug/g:
21 .8%

                                                                                                No. of
                                                                                                Controls:
                                                                                                <1 ug/g:
                                                                                                50.4%
                                                                                                1-5 ug/g:
                                                                                                25.6%
                                                                                                5-10 ug/g:
                                                                                                19.4%
                                                                                                >10 ug/g:
                                                                                                4.7%
                                                                                                Mean ± SD
                                                                                                Tibia
                                                                                                concentration
                                                                                                by age group:
                                                                                                Cases:
                                                                                                19-21: 1.47±
                                                                                                8.35
                                                                                                22-24: 4.48 ±
                                                                                                7.45
                                                                                                25-27: 4.82 ±
                                                                                                8.92
                                                                                                28-30: 6.64 ±
                                                                                                9.53
                                                                                                Controls:
                                                                                                19-21: 1.27±
                                                                                                6.60
                                                                                                22-24: -0.61 ±
                                                                                                6.19
                                                                                                25-27: 0.60 ±
                                                                                                8.60
                                                                                                28-30: 1.74 ±
                                                                                                6.42
November 2012
                                  4-103
                            Draft - Do Not Cite or Quote

-------


Reference
Van
Wijngaarden
et al. (2009)


Wasserman
et al. (2003)





Weisskopf et
al. (2004).
[as reported
in Shih et al.
(2007)1




Weisskopf et
al. (2007a)




Prior Pb
Study Methods Exposure
Cohort: Not reported Cumulative
Age (yrs): Mean: 61 .5
N:47
Location: Rochester, NY
Study Period: Not reported
Cohort: Cumulative
Yugoslavia Prospective Study of (lifetime)
Environmental Pb Exposure Environmental
Age (yrs): 1 0-1 2 (Pb smelter,
N: 167 children refinery, battery
plant)
Location: Kosovska, Mitrovica,
Kosovo, Yugoslavia;
Pristina, Kosovo, Yugoslavia
Study Period: 5/1 985-1 2/1 986
(mother's enrollment); 1986-1999
(follow-up through age 1 2 yr);
Tibia Pb measured 1 1 -1 3 yr old



Cohort: Environmental
Normative Aging Study
Age (yrs): Mean ± SD: 67.4 ± 6.6
N: 466 males
Location: Boston, MA
Study Period: 1991 -2002




Cohort: Not reported
VA Normative Aging Study cohort
Age (yrs): Mean:
Lowest Patella quintile: 73.2;
Highest Patella quintile: 80.7
N: 31 males
Location: Boston, MA
Study Period: Bone Pb
measured: 1994-1999 Scans
performed: 2002-2004


Bone Pb Bone Pb Cone.
biomarker (ug/g)
Tibia Mean ± SD
Calcaneus Tibia: 2.0 ± 5.2
Calcaneus: 6.1 ± 8.5


Tibia Mean ± SD:
Tibia
Pristina: 1.36 ±6.5
Mitrovica: 39.09±
24.55





Tibia Median (IQR)
Patella Tibia: 19 (12,26)
Patella: 23 (15, 35)




Tibia Median (IQR)
Patella Tibia
Lowest quintile: 13
(9-17)
Highest quintile: 41
(38-59)
Patella
Lowest quintile: 9
(5-15)
Highest quintile: 63
(43-86)
Distribution
of Bone Pb
(ug/g)
Not reported


Tibia quartiles:
Q1: -14.4-1 .85
Q2: 1.85-1 0.5
Q3: 10.5-35
Q4: 35-1 93.5
Table 3 of
paper
distributes
tibia Pb by
sex, ethnicity,
address at
birth relative to
factory, and
maternal
education
Tibia IQR: 14
Patella IQR:
20
Table 3 of
paper shows
mean Pb
levels across
categorical
variables (yr of
education,
smoking
status,
computer
experience,
first language
English)
Not reported



November 2012
4-104
Draft - Do Not Cite or Quote

-------


Reference
Weisskopf et
al. (2007b)





Weisskopf et
al. (2009)






Weisskopf et
al. (201 0)







Prior Pb
Study Methods Exposure
Cohort: Concurrent and
VA Normative Aging Study cohort cumulative
Age (yrs): Mean: 68.7
N: 1,089 males
Location: Boston, MA
Study Period: 1993-2001




Cohort: Cumulative
Normative Aging Study; (95%
white)
Age (yrs): Mean ± SD (at Patella
baseline); Tertile 1 : 65.2 ± 7.1 ;
Tertile 2: 66.5 ± 6.5
Tertile 3: 70.2 ± 7.2
N: 868 males
Location: Greater Boston area,
MA
Study Period: 1991-1999
Cohort: Cumulative
BUMC, BWH, BIDMC, HVMA,
Normative Aging Study (NAS),
Harvard Cooperative Program on
Aging (HCPOA)
Age (yrs): Mean:
Cases: 66.5; Controls: 69.4
N: 330 cases; 308 controls
Location: Boston, MA
Study Period: 2003-2007
1991-1999 (NAS patients bone Pb
measured)

Distribution
Bone Pb Bone Pb Cone. of Bone Pb
biomarker (ug/g) (ug/g)
Tibia Median (IQR) Table 1 of
Patella Tibia: 20 (13-28) paper shows
distribution of
Patella: 25 (17-37) Pb biomarkers
by categories
of covariates
(age,
education,
smoking
status, alcohol
intake,
physical
activity,
computer
experience,
first language
English)
Tibia Mean ± SD Patella tertiles:
Patella Tibia: 21 .8 ± 13.6 1:<22
Patella: 31 .2 ±19.4 2:22-35
3: >35






Tibia Mean ± SD: Tibia quartiles:
Patella Tibia: 10.7 ± 12.1 Q1:<3.1
Patella: 13.6 ±15.9 Q2: 3.5-9.6
Q3: 10.0-17.0
Q4:>17.3

Patella
quartiles:
Q1:<2.7
Q2: 3.5-1 1.0
Q3:1 1.3-20.9
Q4: >20.9
November 2012
4-105
Draft - Do Not Cite or Quote

-------


Reference
Weuve et al.
(2006)






Weuve et al.
(2009)




Wright et al.
(2003) [as
reported in
Shihetal.
(2007)1





Prior Pb
Study Methods Exposure
Cohort: Cumulative
VA Normative Aging Study cohort
Age (yrs): a 45
N: 720 males
Location: Boston, MA
Study Period: 1991 (measuring
bone Pb levels)
End date not reported





Cohort: Recent and
Nurses' Health Study cohort cumulative
Age (yrs): 47-74
N: 587 females
Location: Boston, MA
Study Period: 1995-2005
Cohort: Environmental
Normative Aging Study
Age (yrs): Mean ± SD: 68.2 ± 6.9
N: 736 males
Location: Boston, MA
Study Period: 1991 -1997





Bone Pb Bone Pb Cone.
biomarker (ug/g)
Tibia Median (1st-3rd
Patella quartile):
Tibia: 19(13-28)
Patella: 27 (18-39)






Tibia Mean ± SD:
Patella Tibia: 10.5 ± 9.7
Patella: 1 2.6 ± 11.6



Tibia Mean ± SD:
Patella
Tibia: 22.4 ±15.3
Patella: 29.5 ±21 .2




Distribution
of Bone Pb
(ug/g)
Table 1 of
paper shows
distribution of
mean Pb bio-
marker levels
by
characteristics
of participants
(age,
education,
computer
experience,
smoking
status, alcohol
consumption,
fertile of Ca2+
intake, fertile
of physical
activity,
diabetes)
Not reported




Tibia:
Difference in
mean from
Lowest-
highest
quartile: 34.2
Patella:
Difference in
mean from
lowest-highest
quartile: 47
November 2012
4-106
Draft - Do Not Cite or Quote

-------
Table 4-10 Epidemiologic studies that provide bone Pb measurements for
occupationally exposed populations.


Reference
Bleecker et
al. (1997)
[as reported
in Shin et
al. (2007)1











Bleecker et
al.(2QQ7b)



Caffo et al.
(2008)







Dorsey et
al. (2006)





Glenn et al.
(2003) [as
reported in
Navas-
Acien et al.
(2008)1








Study Methods
Cohort: Canada Lead Study
Age (yrs): Cumulative: 24-64
Younger: 24-43
Older: 44-64
Mean ± SD:

Cumulative: 44.1 ± 8.36
Younger: 37.2 ± 4.57
Older: 50.9 ± 4.86

N: 80 males

Location: Canada

Study Period: Not Reported
Cohort: Not reported
Age (yrs): Mean: 39.7
N:61
Location: Northern Canada
Study Period: Not Reported
Cohort: Not reported
Age (yrs): Mean: 60.39
N: 51 3 males
Location: Delaware and New
Jersey, U.S.
Study Period: 1994-1997
(Phase 1 recruitment);
2001 -2003 (Phase 2
recruitment)
Cohort: Not reported
Age (yrs): Mean: 43.4
N:652
Location: Korea
Study Period:
10/24/1997-8/19/1999
(enrolled)
Cohort: Not reported
Age (yrs): 40-70; Mean: 55.8
(baseline)
N: 496 males
Location: Eastern U.S.
Study Period: 6/1 994-6/1 996
(enrolled); 6/1998 (follow-up
period ended)




Prior Pb
Exposure
Occupational
(Pb smelter
workers)












Occupational
(primary Pb smelter
workers)


Cumulative
Occupational
(Former organolead
manufacturing
workers)




Occupational (Pb
workers)





Occupational
(Chemical
manufacturing
facility; inorganic
and organic Pb)






Bone Pb
Bone Pb Concentration
biomarker (ug/g)
Tibia Mean ± SD (Tibia):
Cumulative: 41 .0 ±
24.44
Younger: 35 ±
24.11

Older: 46.9 ± 23.59


Range (Tibia):
Cumulative: -12-90

Younger: -12-80

Older: 3-90

Tibia Mean:
Tibia: 38.6



Tibia Mean ± SD:
Peak Tibia: 23.99 ±
18.46






Tibia Mean ± SD:
Patella Tibia: 33.5 ± 43.4
Patella: 75.1 ±
101.1



Tibia Mean ± SD:
Tibia: 14.7 ±9.4 (at
yr3)
Peak Tibia: 24.3 ±
18.1

Range:
Tibia: -1.6-52 (at
year 3)
Peak Tibia:
-2.2-118.8
Distribution of
Bone Pb
(ug/g)
Not reported














Not reported




Not reported








Not reported






Not reported









November 2012
4-107
Draft - Do Not Cite or Quote

-------


Reference
Glenn et al.
(2006)









Hanninen et
al. (1998)
[as reported
in Shin et
al. (2007)1








Study Methods
Cohort: Not reported
Age (yrs): 0-36.2 (baseline);
Mean ± SD: 41 .4 ±9.5
(baseline)
N: 575; (76% male; 24%
female)
Location: South Korea
Study Period: 10/1997-6/2001



Cohort: Not reported
Age (yrs): Mean ± SD:
Male: 43; Female: 48
Blood Pb (max) < 2.4 umol/L:
41 .7 ±9.3
Blood Pb (max) >2.4 umol/L:
46.6 ±6.2
N: 54; (43 males, 11 females)
Location: Helsinki, Finland
Study Period: Not reported

Prior Pb
Exposure
Cumulative and
recent
Occupational (Pb-
using facilities)







Occupational (Pb
acid battery factory
workers)







Bone Pb Distribution of
Bone Pb Concentration Bone Pb
biomarker (ug/g) (ug/g)
Tibia Mean ± SD: Not reported
Tibia: 38.4 ±42.9

Tibia-Women:

Visit 1: 28.2±19.7
Visit 2: 22.8±20.9

Tibia-Men:
Visit 1 : 41 .7±47.6
Visit 2: 37.1 ±48.1
Tibia Mean ± SD: Not reported
Calcaneus
Tibia:
Blood Pb (max)
< 2.4 umol/L: 19.8 ±
13.7
Blood Pb (max)
>2.4 umol/L: 35.3 ±
16.6

f'alr^a nai ic'
                                                                        Blood Pb (max)
                                                                        < 2.4 umol/L: 78.6 ±
                                                                        62.4
Blood Pb (max)
>2.4 umol/L: 100.4
±43.1
Hsiehetal.
2009
(2009b)



Cohort: Not reported Occupational Tibia
Age (yrs): Mean: (Pb paint factory Patella
Cases: 45.71 workers)
Controls: 46.06
N: 22 cases; 18 controls
Location: Not Reported
Study Period: Not reported
Mean ± SD
Tibia
Case: 61 .55 ± 30.21
Control: 18.51 ±
22.40
Patella
Not reported



                                                                        Case: 66.29 ± 19.48
                                                                        Control: 7.14 ±9.81
November 2012
4-108
Draft - Do Not Cite or Quote

-------
Reference
Kamel et al.
(2002):
Kamel et al.
(2005):
Kamel et al.
(2008)
Khalil et al.
(2009a)
Prior Pb
Study Methods Exposure
Cohort: Not reported Cumulative
Age (yrs): 30-80 Occupational (Pb
N: 109 cases; 256 controls; fumes' dust' or
(Bone samples collected from particles)
104 cases and 41 controls)
Location: New England
(Boston, MA)
Study Period: 1993-1996
Cohort: 1982 Pb Occupational Occupational (Pb
Study cohort battery plant
Age (yrs): Mean: workers)
Cases: 54
Bone Pb
Bone Pb Concentration
biomarker (ug/g)
Tibia Mean ± SE
Patella Tibia
Cases: 14.9± 1.6
Controls: 11.1 ±1.6
Patella
Cases: 20. 5 ± 2.1
Controls: 16.7± 2.0
Tibia Median (IQR)
Tibia
Cases: 57 (20-86)
Controls: 12 (-8-32)
Distribution of
Bone Pb
(ug/g)
Cases
Tibia Pb: N (%)
-7-7: 21 (20)
8-14:35(34)
15-61: 48(46)
Patella Pb: N (%)
-4-9: 27 (26)
1 0-20: 40 (38)
21-107:37(36)
Controls
Tibia Pb: N (%)
-7-7: 14(34)
8-14: 12(29)
15-61: 15(37)
Patella Pb: N (%)
-4-9: 14(34)
10-20: 14(34)
21-107:13(32)
Not reported
Osterberg
et al. (1997)
[as reported
in Shih et
al. (2007)1
  Controls: 55
N: 83 cases; 51 controls
Location: Eastern
Pennsylvania
Study Period: 1982-2004
Cohort: Not reported
Age (yrs): Median: 41.5
N: 38 male cases; 19 male
controls
Location: Not reported
Study Period: Not Reported
Occupational
(secondary Pb
smelter- inorganic
Pb)
Finger bone
Median
Finger Bone:
High Cases: 32
Low cases: 16
Control: 4

Range
Finger Bone:
High Cases: 17-101
Low cases: -7-49
Control:-19-18
Not reported
November 2012
                                    4-109
                                Draft - Do Not Cite or Quote

-------


Reference
Roelset al.
(1994)








Schwartz
etal.
(2000b) [as
reported in
Shihetal.,
(2007)1



Schwartz et
al. (2000c)
[as reported
in Navas-
Acien et al.
(2008)1

Schwartz et
al. (2001):
Lee et al.
(2001 a)



Study Methods
Cohort: Not reported
Age (yrs): 30-60
N: 76 male cases; 68 male
controls
Location: Belgium
Study Period: Not Reported








Cohort: U.S. Organolead Study
Age (yrs): Mean ± SD:
Cases: 55.6 ± 7.4
Controls: 58.6 ± 7.0
N: 535 male cases
118 male controls
Location: Eastern U.S.
Study Period: 6/1 994-1 0/1 997
(enrolled); Completed 2-4
annual follow-up visits; Tibia
Pb taken in 3rd year
Cohort: Not reported
Age (yrs): 41 .7-73.7
(Combined)
Mean ± SD:
Combined: 57.6 ± 7.6
Hypertensive: 60.2 ± 6.9
Nonhypertensive: 56.6 ± 7.5
N: 543 males
Location: Eastern U.S.
Study Period: 1995 (recruited);
1996-1 997 (Tibia Pb
taken during the 3rd yr)
Cohort: Not reported
Age (yrs): Mean:
Exposed: 40.4
Control: 34.5
N: 803 cases; 135 controls
Location: South Korea
Study Period:
10/24/1997-8/19/1999

Prior Pb
Exposure
Occupational (Pb
smelter workers)
Mean case
exposure: 18 yr
(range: 6 to 36 yr)








Occupational
(tetraethyl and
tetramethyl Pb
manufacturing
facility)



Occupational
(former organolead
manufacturing
workers)


Occupational
(battery
manufacturing,
secondary smelting,
Pb oxide
manufacturing, car
radiator
manufacturing)
Bone Pb
Bone Pb Concentration
biomarker (ug/g)
Tibia Geometric Mean
(Range)
Tibia Cases:
Normotensive: 64.0
(19.6-167.1)
Hypertensive: 69.0
(21.7-162.3)
Total: 65.8
(19.6-167.1)
Tibia Controls:
Normotensive: 21.7
(<1 5.2-69.3)
Hypertensive: 20.2
(<1 5.2-52.9)
Total: 21 .4
(<1 5.2-69.3)
Tibia Mean ± SD
Current Tibia:
Cases: 14.4 ±9.3
Peak Tibia:
Cases: 22.6 ± 16.5


Tibia Mean ± SD
Tibia:
Combined: 14.4±
9.3
Hypertensive: 15.4
±9.1
Nonhypertensive:
14.0 ±9.3
Range Tibia:
Combined: -1.6-52
Tibia Mean ± SD
Tibia
Cases: 37.1 ±40.3
Control: 5.8 ± 7.0
Range:
Tibia
Cases: -7-338
Controls: -11-27
Distribution of
Bone Pb
(ug/g)
Not reported








Not reported



Not reported


Not reported

November 2012
4-110
Draft - Do Not Cite or Quote

-------


Reference
Schwartz et
al. (2005)


Stewart et
al. (1999)
[as reported
in Shih et
al., (2007)1














Stewart et
al. (2006)
Weaver et
al. (2008)



Study Methods
Cohort: Not reported
Age (yrs): Mean at 1st visit:
41.4
N:576
Location: South Korea
Study Period: 10/1997-6/2001
Cohort: U.S. Organolead Study
Age (yrs): 40-70 (in 1995)
38% > 60 yrs
Mean: 58
N: 534 males
Location: Eastern U.S.
Study Period: Not Reported














Cohort: Not reported
Age (yrs): Mean: 56.1
N: 532 males
Location: Eastern U.S.
Study Period: 1994-1997;
2001-2003
Cohort: Not reported
Age (yrs): Mean ± SD: 43.3 ±
9.8
N:652
Location: South Korea
Study Period: 12/1999-6/2001

Prior Pb
Exposure
Occupational
(current and former
Pb workers)


Occupational
(tetraethyl and
tetramethyl Pb
manufacturing
facility)














Cumulative
Occupational
(Organolead
workers - not
occupationally
exposed to Pb at
time of enrollment)
Occupational
(Current and former
Pb workers; plants
produced Pb
batteries, Pb oxide,
Pb crystal, or
radiators, or were
secondary Pb
smelters)
Bone Pb
Bone Pb Concentration
biomarker (ug/g)
Tibia Mean ± SD
Tibia: 38.4 ±43


Tibia Mean ± SD
Tibia:
Current: 14.4 ±9.3
Peak: 23.7 ± 17.4
Range: Tibia
Current: -1 .6-52
Peak: -2.2-105.9













Tibia Mean ± SD
Current Tibia: 14.5
±9.6
Peak Tibia: 23.9 ±
18.3
Patella Mean ± SD
Patella: 37.5 ± 41 .8

Distribution of
Bone Pb
(ug/g)
Tibia:
25th percentile at
V1: 14.4
75th percentile at
V1:47.1

Current Tibia Pb:
N (%)
<5: 77 (14.2)
5-9.99: 113
(20.8)
10-14.99: 119
(21.9)
15-19.99: 117
(21.5)
>20: 118(21.7)
Peak Tibia Pb: N
<5:49(9.1)
5-9.99: 64 (1 1 .8)
10-14.99: 70
(12.9)
15-19.99:87
(16.1)
20-24.99: 79
(14.6)
25-29.99: 55
(10.2)
>30: 137(26.1)
Not reported
Not reported

November 2012
4-111
Draft - Do Not Cite or Quote

-------
    4.4.3
Pb in Urine
i
2
o
J
4
5
6
 Urine Pb concentrations in the U.S. general population have been monitored in the
 NHANES. Data from the most recent survey (CDC. 2011 a) are shown in Table 4-11. The
geometric mean for the entire sample
creatinine (95% CI: 0.48, 0.55). The §
(n = 1,300) were 0.50 (ig/g creatinine
CI: 0.49, 0.57), respectively.
for the period 2007-2008 (n =
geometric means for males (n =
(95% CI: 0.47, 0.53) and 0.53
2,627) was 0.52 ug/g
= 1,327) and females
(ig/g creatinine (95%
Table 4-11 Urine Pb concentrations in the U.S. population.
Survey Stratum Period Geometric Mean (ug/g CR)a 95% Confidence Interval
All 1999-2000 0.721
2001-2002 0.639
2003-2004 0.632
2005-2006 0.546
2007-2008 0.515
6-1 1 yr 1999-2000 1.170
2001-2002 0.918
2003-2004 0.926
2005-2006 0.628
2007-2008 0.644
12-1 9 yr 1999-2000 0.496
2001-2002 0.404
2003-2004 0.432
2005-2006 0.363
2007-2008 0.301
> 20 yr 1 999-2000 0.720
2001-2002 0.658
2003-2004 0.641
2005-2006 0.573
2007-2008 0.546
Males 1 999-2000 0.720
2001-2002 0.639
2003-2004 0.615
2005-2006 0.551
2007-2008 0.502
Females 1999-2000 0.722
2001-2002 0.639
0.700, 0.742
0.603, 0.677
0.603, 0.662
0.502, 0.573
0.483, 0.549
0.975, 1.41
0.841 , 1 .00
0.812,1.06
0.563, 0.701
0.543, 0.763
0.460, 0.535
0.380, 0.428
0.404, 0.461
0.333, 0.395
0.270, 0.336
0.683, 0.758
0.617,0.703
0.606, 0.679
0.548, 0.600
0.513, 0.580
0.679, 0.763
0.607, 0.673
0.588, 0.644
0.522, 0.582
0.471,0.534
0.681, 0.765
0.594, 0.688
Number of Subjects
2,465
2,689
2,558
2,576
2,627
340
368
290
355
394
719
762
725
701
376
1,406
1,559
1,543
1,520
1,857
1,227
1,334
1,281
1,271
1,327
1,238
1,355
    November 2012
                           4-112
Draft - Do Not Cite or Quote

-------
Survey Stratum

Mexican -Americans
Non-Hispanic blacks
Non-Hispanic whites
Period Geometric Mean (ug/g CR)a 95% Confidence Interval
2003-2004
2005-2006
2007-2008
1 999-2000
2001-2002
2003-2004
2005-2006
2007-2008
1 999-2000
2001-2002
2003-2004
2005-2006
2007-2008
1 999-2000
2001-2002
2003-2004
2005-2006
2007-2008
aValues are in ug Pb/g creatinine (CR)
Source: Based on data from the NHANES
0.648
0.541
0.527
0.940
0.810
0.755
0.686
0.614
0.722
0.644
0.609
0.483
0.452
0.696
0.615
0.623
0.541
0.506
(CDC. 2011 a)
0.601,0.698
0.507, 0.577
0.489, 0.568
0.876,1.01
0.731, 0.898
0.681, 0.838
0.638, 0.737
0.521,0.722
0.659, 0.790
0.559, 0.742
0.529, 0.701
0.459, 0.508
0.414, 0.492
0.668, 0.725
0.579, 0.654
0.592, 0.655
0.500, 0.585
0.466, 0.550

Number of Subjects
1,277
1,305
1,300
884
682
618
652
515
568
667
723
692
589
822
1,132
1,074
1,041
1,095

     4.4.4       Pb in Teeth

 1                  The influence of historical Pb exposures was recently studied by Robbins et al. (2010).
 2                  Tooth enamel samples from 127 subjects born between 1936 and 1993 were analyzed for
 3                  Pb concentration and Pb isotope ratios of the tooth enamel and compared with those
 4                  parameters for sediment cores and estimates of Pb emissions from gasoline during the
 5                  years when 50% enamel formation was estimated to occur. They found that the log-
 6                  transform of tooth enamel concentration was significantly predicted by the  log-transform
 7                  of Lake Erie sediment core data obtained by Graney et al. (1995) (p <0.00001) and by the
 8                  log-transform of U.S.  consumption of Pb in gasoline (p <0.00001); see Figure 4-21.
 9                  Additionally, Robbins et al. (2010) found that 207Pb/206Pb was significantly  predicted by
10                  the 207Pb/206Pb observed in the Lake Erie sediment cores obtained by Graney et al. (1995)
11                  (p O.OOOl) and forthis study (p <0.0002).
     November 2012
4-113
Draft - Do Not Cite or Quote

-------
                  100
                g
                I
                8.   50
                    25
                             o-
                     1930    1940    1950
                             1960    1970     1980     1990
                                Year
      Note: The lines and symbols on the plot represent Pb in study participant teeth (solid line), newly obtained Pb sediment Lake Erie
      cores (open triangles), Pb in previously obtained Lake Erie sediment [open circles, Graney et al. (1995)1, and U.S. gasoline usage
      (closed circles). All values are normalized by the peak observation for that parameter.
      Source: Reprinted with permission of Elsevier Publishing, Robbins et al. (2010).
      Figure 4-21     Comparison of relative temporal changes in tooth enamel Pb
                        concentration.
 i
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
Several Brazillian studies have found increased levels of Pb in teeth in areas where Pb
sources are present. For example, Costa de Almeida et al. (2007) reported Pb
concentration in tooth enamel among 4-6 year old kindergarteners in Sao Paulo, Brazil to
be significantly higher for children living near a Pb-acid battery processing plant in the
Baruru neighborhood compared with 4-6 year old children in other parts of the city
(non-exposed median: 206 mg/kg, n = 247; exposed median: 786 mg/kg, n = 26;
p <0.0001). Subsequent analysis revealed that 55% of 4-6 year old children from Baruru
had tooth enamel Pb concentrations greater than 600 mg/kg, forming a significant
comparison with other neighborhoods having 0-33% of 4-6 year old children with  tooth
enamel Pb greater than 600 mg/kg (p <0.0001) (de Almeida et al., 2008). The authors did
not describe controlling for additional factors, such as socioeconomics or housing
conditions. Arruda-Neto et al. (2009) studied Pb in tooth samples among Sao Paulo
children to compare exposures of children age 4-12 years, living near a dam with heavy
metal sediments with those of children ages 4-13 years, living in a control area thought to
have few exposures. They observed a significant comparison (near dam: avg 1.28 ± 0.11
mg/kg, n = 50; control region: avg 0.91 mg/kg, n = 24). In a related study of Pb measures
in teeth among the general population ages  7-60 years, Arruda-Neto et al.(2010) observed
that 10-year old children had the highest teeth Pb concentrations, which were  115% of
the teeth Pb concentrations in 7-year olds. Twenty-year old subjects had teeth
Pb concentrations at roughly 50% of the 7-year olds' teeth Pb concentrations. Tooth Pb
      November 2012
                           4-114
Draft - Do Not Cite or Quote

-------
 1                  concentrations stayed fairly constant throughout adulthood but then dropped to just above
 2                  30% among 65-year old subjects. Note that the authors did not clarify if average or
 3                  median values were presented, nor did they adjust for potentially confounding factors.
      4.5        Empirical Models of Pb  Exposure-Blood Pb Relationships

 4                  Multivariate regression models, commonly used in epidemiology, provide estimates of
 5                  the contribution of variance in the internal dose metric to various determinants or control
 6                  variables (e.g., air Pb concentration, surface dust Pb concentration). Structural equation
 7                  modeling links several regression models together to estimate the influence of
 8                  determinants on the internal dose metric. Regression models can provide estimates of the
 9                  rate of change of blood or bone Pb concentration in response to an incremental change in
10                  exposure level (i.e., slope factor). One strength of regression models for this purpose is
11                  that they are empirically verified within the domain of observation and have quantitative
12                  estimates of uncertainty imbedded  in the model structure. However, regression models
13                  are based on (and require) paired predictor-outcome data, and, therefore, the resulting
14                  predictions are confined to the domain of observations and are typically not generalizable
15                  to other populations. Regression models also frequently exclude numerous parameters
16                  that are  known to influence human Pb exposures (e.g., soil and dust ingestion rates) and
17                  the relationship between human exposure and tissue Pb levels, parameters which are
18                  expected to vary spatially and temporally. Thus, extrapolation of regression models to
19                  other spatial or temporal contexts, which is often necessary for regulatory applications of
20                  the models, can be problematic.

21                  A variety of factors may potentially affect estimates of blood Pb-air Pb slope factors.
22                  Simultaneous changes in other (nonair) sources of Pb exposure can affect the relationship
23                  indicated for air Pb. For example, remedial programs (e.g., community and home-based
24                  dust control  and education) may be responsible for partial blood Pb reduction seen in
25                  some studies. The effect of remedial programs may lead to an overestimation of declines
26                  in blood Pb due to changes in air Pb and a corresponding positive bias in blood Pb-air Pb
27                  slopes. However, model adjustment for remedial programs and other factors (e.g., soil Pb
28                  concentrations) may also cause a negative bias in blood Pb-air Pb slopes. A tendency
29                  over time for children with lower blood Pb levels to not return for follow-up testing has
30                  been reported. The follow-up of children with higher blood Pb levels would likely lead to
31                  an underestimation of reductions in blood Pb following reductions in air Pb and cause a
32                  negative bias in blood Pb-air  Pb slopes. Another factor is the extent to which all the air
33                  Pb exposure pathways are captured by the data set and its analysis. For example, some
34                  pathways (such as exposure through the diet or surface soils) may respond more slowly to
35                  changes in air Pb than others  (such as inhalation). Additionally, some studies may include

      November 2012                            4-115                 Draft - Do Not Cite or Quote

-------
 1                  adjustments for variables that also reflect an influence from air Pb (e.g., SES or soil Pb).
 2                  Studies may also vary in the ages of subjects, which given age-related changes in blood
 3                  Pb can also influence estimates. Many studies have utilized TSP measurements of air Pb
 4                  concentrations. The sampling efficiency of TSP samplers is affected by particle size
 5                  distribution, wind speed, and wind direction as described in Section 3.4.1. For example,
 6                  especially for larger particles (aerodynamic diameter of 20 urn or more), TSP sampling
 7                  efficiency decreases with increasing wind speed. Such effects on TSP sampling
 8                  efficiency can, in areas where such large particles are a substantial portion of airborne Pb,
 9                  lead to uncertainties in the comparability of air Pb concentrations between samples within
10                  a study and across studies. A uniformly low bias in air Pb concentrations in a study could
11                  positively bias estimated blood Pb-air Pb slopes for that study. Moreover, variability in
12                  TSP samples is likely to result from temporal variation in wind speed, wind direction, and
13                  source strength; see Sections 3.3 and 3.5. Such temporal variability would tend to
14                  increase uncertainty and reduce the statistical strength of the relationship between air Pb
15                  and blood Pb but may not necessarily affect the slope of this relationship. A number of
16                  factors including those described above cause uncertainty in the magnitude of estimated
17                  blood Pb-air Pb  slope factors and may lead to both positive and negative biases in the
18                  estimates from individual studies.
      4.5.1        Air Pb-Blood Pb Relationships in Children

19                  The 1986 Pb AQCD (U.S. EPA. 1986a) described epidemiological studies of
20                  relationships between air Pb and blood Pb. Of the studies examined, the aggregate blood
21                  Pb-air Pb slope factor (when considering both air Pb and Pb in other media derived from
22                  air Pb) was estimated to be approximately double the slope estimated from the
23                  contribution due to inhaled air alone (U.S. EPA. 1986a).

24                  Much of the pertinent earlier literature (e.g., prior to 1984) on children's blood Pb levels
25                  was summarized by Brunekreef (1984). Based on meta-analysis of data from studies of
26                  urban or industrial-urban populations in 18 different locations, Brunekreef (1984)
27                  estimated the blood Pb-air Pb slope for children to be 0.3485 ln[(ig/dL blood Pb] per
28                  ln[(ig/m3 air Pb] (R2 = 0.69; see Figure 4-22). This slope corresponds to an increase of
29                  4.6 (ig/dL blood Pb per (ig/m3 air Pb at an air Pb concentration of 1.5 (ig/m3 for all
30                  groups included in the analysis. The 1.5 (ig/m3 value is the median of the air Pb
31                  concentrations that match the blood Pb concentrations in 96 different child populations in
32                  Figure 3 of Brunekreef et al. (1984). taken from the  Appendix to the same paper. When
33                  the analysis was limited to child populations whose  mean blood Pb concentrations were
34                  <20 (ig/dL (n=43), the slope was 0.2159 (R2=0.33), which corresponds to an increase of
35                  4.8 (ig/dL blood Pb per (ig/m3 air Pb at the median air concentration (0.54 (ig/m3).
      November 2012                             4-116                  Draft - Do Not Cite or Quote

-------
 1                   Newer studies that provide estimates for the blood Pb-air Pb slope factor are described in
 2                   the sections that follow. Those studies that have at least three data points are included, as
 3                   fewer than that contributes little to the understanding of the shape of the blood Pb-air Pb
 4                   relationship. A tabular summary of the major outcomes is provided in Table 4-12. In
 5                   some studies, the blood Pb-air Pb  relationship was described with a nonlinear regression
 6                   function, in which the blood Pb-air Pb slope factor varied with air Pb concentration.
 7                   Studies also varied with regard to  the use of simple or multivariate regression and, for the
 8                   latter, with regard to variables included. In Table 4-12. with the exception of Ranft et al.
 9                   (2008). slopes corresponding to a  central estimate of the air Pb concentrations are
10                   provided, to represent each study.  These were calculated by evaluating each regression
11                   function at ± 0.01 (ig/m3 from the  central estimate of the air Pb concentration. Air Pb
12                   concentration ranges and central estimates varied across studies, making it difficult to
13                   interpret comparisons based  solely on the central estimates of the slope factors.
14                   Therefore, Figure 4-23 depicts the relationship between the blood Pb-air Pb slope factor
15                   as a function of air Pb concentration for the range of air Pb concentrations evaluated in
16                   those studies that provided the regression equation (the central estimate is also shown).
17                   Figure 4-23 provides a more informative picture of the extent to  which slope estimates
18                   vary (and overlap) within and between studies. The Ranft et al. (2008) study includes a
19                   separate term for soil Pb, so the blood Pb-air Pb slope factor presented for that study
20                   underestimates the slope factor that would reflect all air-related pathways, since soil Pb
21                   encompasses deposited ambient air Pb. A few studies used a log-log model that predict
22                   an increase in the  blood Pb-air Pb  slope factor with decreasing air Pb concentration, and
23                   the remainder of the studies used linear models that predict a constant blood Pb-air Pb
24                   slope factor across all air Pb concentrations.
      November 2012                              4-117                   Draft - Do Not Cite or Quote

-------
Table 4-12   Summary of estimated slopes for blood Pb to air Pb slope factors
                 in humans.
Reference
Study Methods
Model Description
Blood Pb-
  AirPb
  Slope3
Children Populations-Air
                   Location: Various countries
                   Years: 1974-1983
Brunekreef (1984)    Subjects: Children (varying age ranges; n>190,000)
                   Analysis: Meta analysis of 96 child populations from
                   18 study locations
                                                Model: Log-Log
                                                Blood Pb: 5-76 ug/dL
                                                (mean range for study
                                                populations)
                                                Air Pb: 0.1-24 ug/m3
                                                (mean range for study locations)
                               All children:
                                 4.6(1.5)b
                                 Children
                                <20 ug/dL:
                                4.8 (0.54)°
                   Location: Chicago, IL
                   Years: 1974-1988
Hayes et al. (1994)   Subjects: 0.5-5 yr (n = 9,604)
                   Analysis: Regression of quarterly median blood Pb
                   and quarterly mean air Pb
                                                Model: Log-Log
                                                Blood Pb: 10-28 ug/dL
                                                (quarterly median range)
                                                Air Pb: 0.05-1.2 ug/m3
                                                (quarterly mean range)
                                8.2 (0.62)d
                   Location: Trail, BC
                   Years: 1989-2001
                   Subjects: 0.5-6 yr (Estimated n = 220-460, based on
                   292-536 blood Pb measurements/yr with 75-85%
                   participation)
                   Analysis: Regression of blood Pb screening and
                   community air Pb following upgrading of a local
                   smelter
                                                Model: Linear
                                                Blood Pb: 4.7-11.5 ug/dL
                                                (annual geometric mean range)
                                                AirPb: 0.03-1.1 ug/m3
                                                (annual geometric mean range)
                                7.0 (0.48)e
Schwartz and
Pitcher (1989). U.S.
EPA (1986a)
Location: Chicago, IL
Years: 1976-1980
Subjects: Black children, 0-5 yr (n = 5,476)
Analysis: Multivariate regression of blood Pb with
mass of Pb in gasoline (derived from gasoline
consumption data and Pb concentrations in gasoline
for the U.S.)
Model: Linear
Blood Pb: 18-27 ug/dL(mean
range)'
Air Pb: 0.36-1.22 ug/m3
(annual maximum quarterly
mean)h
 8.6 (0.75)9
Tripathi et al.
(2001)
Location: Mumbai, India (multiple residential
locations)
Years: 1984-1996
Subjects: 6-10 yr (n = 544)
Analysis: Regression of residential location-specific
average blood Pb and air Pb data
Model: Linear
Blood Pb: 8.6-14.4 ug/dL
(GM range for residential
locations)
Air Pb: 0.10-1.18 ug/m3
(GM range for residential
locations)
 3.6 (0.45)'
November 2012
                             4-118
       Draft - Do Not Cite or Quote

-------
Reference
Study Methods
Model Description
Blood Pb-
  AirPb
  Slope3
Children Populations-Air and Soilj
Ranft et al. (2008)
Location: Germany
Years: 1983-2000 (blood Pb and air Pb), 2000-2001
(soil Pb)
Subjects: 6-11 yr (n = 843)
Analysis: Pooled multivariate regression of 5 cross-
sectional studies
Model: Log-Linear
Blood Pb: 2.2-13.6 ug/dL
(5th-95th percentile)
Air Pb: 0.03-0.47 ug/m3
(5th-95th percentile)
  3.2, 6.4R
Mixed Child-Adult Populations
Schwartz and
Pitcher (1989). U.S.
EPA (1986a)
Location: U.S.
Years: 1976-1980
Subjects: NHANES II, 0.5-74 yr, whites (n = 9,987)
Analysis: Multivariate regression of blood Pb with
mass of Pb in gasoline (derived from gasoline
consumption data and Pb concentrations  in gasoline
for the U.S.)
Model: Linear
Blood Pb: 11-18 ug/dL9
(mean range)'
Air Pb: 0.36-1.22 ug/m3
(annual maximum quarterly
mean)h
 9.3(0.75)'
a Slope is predicted change in blood Pb (ug/dL per ug/m3) evaluated at ± 0.01 ug/m3 from central estimate of air Pb for the study
(shown in parentheses), with the exception of Ranft et al. (2008) in which the slope from the paper is provided because the
regression equation was not available. The central estimate for Brunekreef (1984) is the median of air Pb concentrations since it
was a meta-analysis; for all other studies the mean is presented. For multiple regression models, this is derived based only on air
Pb coefficient and intercept. Depending on extent to which other variables modeled also represent air Pb, this method may
underestimate the slope attributable to air pathways. In single regression models, the  extent to which non-modeled factors,
unrelated to air Pb exposures, exert an impact on blood Pb that covaries with air Pb may lead to the slope presented here to over
represent the role of air Pb.
b In(PbB) = In(PbA) x 0.3485 + 2.853
0 In(PbB) = In(PbA) x 0.2159 + 2.620
d In(PbB) = In(PbA) x 0.24 + 3.17
6PbB = PbAx7.0
f Observed blood Pb values not provided; data are for regressed adjusted blood Pb.
9PbB = PbAx8.6
h Based on air Pb data for U.S. (1986 PbAQCD) as a surrogate for Chicago.
!PbB = PbAx3.6
J Study that considered air Pb and soil Pb where the air Pb-blood Pb relationship was adjusted for soil Pb.
k Slope provided in paper with background blood Pb level of 1.5 and 3 ug/dL, respectively, and GMR of 2.55 for ambient air.
'PbB = PbAx9.63
GM, geometric mean; GSD, geometric standard deviation; PbB, blood  Pb concentration (ug/dL); PbA, air Pb concentration (ug/m3)
November 2012
                              4-119
       Draft - Do Not Cite or Quote

-------
                       0
   10         15

   AirPb (|jg/m3)
    20
25
Note: The regression model is: (ln[|jg/dL blood Pb] = 0.3485-ln[|jg/m air Pb] + 2.85) for all children (n=96 subject groups) and
(ln[ug/dL blood Pb] = 0.2159-ln[ug/m3air Pb] + 2.62) when the sample was restricted to populations that had blood Pb
concentrations <20 ug/dL (n=44 subject groups).
Data provided from Brunekreef (1984).


Figure 4-22      Predicted relationship between air Pb and blood Pb based on a
                   meta analysis of 18 studies.
November 2012
4-120
Draft - Do Not Cite or Quote

-------
                  40
                  35 -
               CO^
               0)30
               o>
                 -25
               O)
               a 20
               o>
               Q.
               O
               w 15
               .a
               CL
               E 10
               00
                   5 -
                    0.0
0.5
        1.0
AirPb (|jg/m3)
1.5
                                         ABrunekreef84<20

                                         AHayes94

                                         »Hilts03

                                         oSchwartz89(Chi)

                                         CSchwartz89(US)

                                         DTripathiOl
2.0
      Note: Slopes are calculated for a change in air Pb (±0.01 ug/m ) over ranges of air Pb concentrations reported in each study (lines).
      The air Pb axis is truncated at 2 ug/m3; the actual range for the Brunkreef et al. (1984) study was 0.1-6.4 ug/dL per ug/m3. The slope
      axis has been truncated at 40; the actual range for the Hayes et al. (1994) study was 5-56 ug/dL per ug/m3 (the high end of the
      range was estimated for the minimum annual average air Pb of 0.05 ug/m3). The two estimates for Schwartz and Pitcher (1989)
      represent data for U.S. and Chicago. Models are log-log (solid lines) and linear (dotted lines). Symbols show the slope at the central
      estimate of air Pb (e.g., median for Brunereef and mean for the other studies).

      Figure 4-23      Blood Pb - air Pb slopes (ug/dL per ug/m3) predicted from
                         epidemiologic studies.

 1                   Hilts et al. (2003) reported child blood Pb and air Pb trends for the city of Trail, British
 2                   Columbia, over a period preceding and following installation of a new smelter process in
 3                   1997 which resulted in lower air Pb concentrations. Blood Pb data were obtained from
 4                   annual (1989-2001) surveys of children 6-60 months of age who lived within 4 km from
 5                   the smelter (n: 292-536 eligible per year,  75-85% participation). Air Pb concentrations
 6                   were obtained from high volume suspended particulate samplers placed within  2 km of
 7                   the smelter that operated 24 hours every 6th day. Data on Pb levels in air, residential soil,
 8                   interior dust, and blood for three sampling periods are  summarized in Table 4-13. Based
 9                   on these data, blood Pb decreased 6.5 (ig/dL per  1 (ig/m3 air Pb and by 0.068 (ig/dL per
10                   mg/kg soil Pb (based on linear regression with air or soil Pb as the sole independent
11                   variable) for the entire period. When considering a 9-month weighted mean of
12                   0.13 (ig/m3) for 2001 (3 months when the smelter was  closed, 0.03 (ig/m3; 6 months
13                   when it was open, 0.18 (ig/m3), the slope is 7.0. Several uncertainties apply to these
      November 2012
       4-121
                                     Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
estimates. Potential mismatching of air Pb concentrations (often termed misclassification)
with individual blood Pb levels may have occurred as a result of air Pb being measured
within 2 km of the smelter, whereas, the blood Pb data included children who resided
>2 km from the smelter. The regression estimates were based on group mean estimates
for three sampling dates, rather than on the individual blood Pb estimates, which included
repeated measures on an unreported fraction of the sample. The limited number of data
pairs (three) constrained parameter estimates to simple regression coefficients. Other
important factors probably contributed to  blood Pb declines in this population that may
have been correlated with  air, soil and dust Pb levels. These factors include aggressive
public education and exposure intervention programs (Hilts et al.. 1998; Hilts.  1996).
Therefore, the coefficients shown in Table 4-13 are likely to overestimate the influence of
air, dust, or soil Pb on blood Pb concentrations at this site.
      Table 4-13   Environmental Pb levels and blood Pb levels in children in Trail,
                     British Columbia.
Date
Blood Pb (ug/dL)
Air Pb (|jg/m3)
Soil Pb (mg/kg)
Interior Dust Pb (mg/kg)
1996 a
11.5
1.1
844
758
1999
5.9
0.3
756
583
2001
4.7
0.13b
750°
580°
Regression Coefficient
NA
7.01 ± 0.009 (R2=1.00, p=0.001)
0.069 ± 0.008 (R2=0.99, p=0.069)
0.035 ± 0.005 (R2=0.98, p=0.097)
      A new smelter process began operation in 1997. Values for air, soil and dust Pb are annual geometric means; values for blood Pb
      are annual geometric means. Regression coefficients are for simple linear regression of each exposure variable on blood Pb.
      a Values for air Pb, soil Pb, and interior dust Pb are actually for period of 1994-1996.
      bNine month time-weighted average of 0.03 ug Pb/m3for3 months and 0.18 ug Pb/m3for6 months.
      0 Values assumed by study authors.
      Source: Data from Hilts et al. (2003).
13
14
15
16
17
18
19
20
21
22
23
24
Ranft et al. (2008) reported a meta-analysis of five cross-sectional surveys of air and soil
Pb levels and blood Pb concentrations in children living in Duisburg, Germany. The
analysis included observations on 843 children (6-11 years of age) made during the
period 1983-2000. Children recruited in 1983 were an average of 9.1 yrs of age, whereas
children recruited in later years of the study averaged 6.3 to 6.4 yrs of age. The 1983 air
Pb concentrations were based on two monitoring stations, while a combination of
dispersion modeling and monitoring data was used in the later years to estimate Pb in PM
in a 200 meter by 200 meter grid that encompassed the city. Pb in surface soil (0-10 cm)
was measured at 145 locations in the city in 2000 and 2001. Air and soil Pb
concentrations were assigned to each participant by spatial interpolation from the
sampling grid data to each home residence. The 5th-95th percentile ranges were
0.025-0.465 (ig Pb/m3 for air and 72-877 mg  Pb/kg for soil. The results of multivariate
      November 2012
                            4-122
Draft - Do Not Cite or Quote

-------
 1                   regression analyses were reported in terms of the relative increase (the geometric mean
 2                   blood Pb ratio, GMR) for an increase in air or soil Pb from the 5th to 95th percentile
 3                   value. In a multivariate regression model (equation not provided) that included air and
 4                   soil Pb in the same model and adjusted for covariates, the GMR values were: 2.55 per
 5                   0.44 (ig/m3 increase in air Pb (95% CI: 2.40, 2.71, R2=0.484, p <0.001) and 1.30 per
 6                   800 mg/kg soil Pb (95% CI: 1.19, 1.43, R2 = 0.017, p <0.001). Based on the values for R2,
 7                   the regression model accounted for approximately 59% of the total variance in blood Pb
 8                   and, of this, 83% was attributed to air Pb. Values for GMR for soil Pb ranged from 1.41
 9                   to 2.89, with most recent blood Pb data (from the year 2000) yielding a value of 1.63 per
10                   800 mg/kg increase in soil Pb.  The GMR values can be converted to regression slopes
11                   (slope = [starting blood Pbxln(GMR)]/[95th - 5th percentile air or soil Pb]) for
12                   calculating equivalent airblood Pb ratios. The model predicts an increase of 3.2 (ig/dL
13                   blood Pb per 1 (ig/m3 increase  in air Pb at the median air Pb concentration for the study
14                   (0.1 (ig/m3) and assuming a background blood Pb concentration of 1.5 (ig/dL. Based on
15                   the GMR estimate of 1.63 for soil Pb, a 1,000 mg/kg increase in soil Pb would be
16                   associated with an increase in blood Pb of 0.9 (ig/dL per mg/kg soil at the median soil Pb
17                   concentration of 206 mg/kg and assuming a background blood Pb concentration of
18                   1.5 (ig/dL. The degree of confounding of the GMR and estimates resulting  from the air
19                   and soil Pb correlation was not reported, although the correlation coefficient for the two
20                   variables was  0.136 for the whole data set and 0.703 when data collected in 1983 was
21                   omitted. Because the model also included Pb levels  in soil, the blood Pb-air Pb ratio may
22                   be underestimated since some of the Pb in soil was likely derived from air.  The blood
23                   Pb-air Pb slope does not include the portion of the soil/dust Pb ingestion pathway that
24                   derives from air Pb, such as recently airborne Pb deposited to soil and dust which remains
25                   available for inhalation and ingestion.

26                   To estimate the blood Pb-air Pb ratio that included all air-related pathways, data for
27                   median of blood Pb and air Pb among the cohort of children studied were extracted from
28                   Table 2 in Ranft et al. (2008) for each of the five study years. The median blood Pb and
29                   air Pb were used in regressions employing linear, log-log, and log-linear (i.e., similar to
30                   authors' approach with ln[blood Pb] against air Pb)  fits. The linear model obtained was:
31                   PbB = 12.2>
-------
 1                   Schnaas et al. (2004) analyzed data on blood Pb and air Pb concentrations during and
 2                   after the phase out of leaded gasoline use in Mexico (1986-1997) in children as part of a
 3                   prospective study conducted in Mexico City. The sample included 321 children born
 4                   during the period 1987 through 1992. Repeated blood Pb measurements were made on
 5                   each child at 6-month intervals up to age 10 years. Air Pb measurements (annual average
 6                   of quarterly means) were derived from three area monitors which represented distinct
 7                   study zones. Children were assigned to study zones based on their current address.
 8                   Associations between lifetime (across the first 10 years of life) blood Pb concentration,
 9                   air Pb concentration (mean annual for each calendar year of study) and other variables
10                   (e-g-, age, year of birth, family use of glazed pottery) were evaluated using multivariate
11                   regression models. The largest air Pb coefficient occurred in the cohort born in 1987, who
12                   experienced the largest decline in air Pb (from 2.8 to about 0.25 (ig/m3); the air Pb
13                   coefficient for this group of children was 0.213 (95% CI: 0.114-0.312) In [(ig/dL blood]
14                   per ln[(ig/m3 air]. The smallest, statistically significant air Pb coefficient occurred for the
15                   1990 birth year cohort, who experienced a decline in air Pb from 1.5 to about 0.1 (ig/m3.
16                   The air Pb coefficient for the 1990 cohort was 0.116 (95% CI: 0.035-0.196). Based on
17                   these air Pb coefficients,  children in the 1987 and 1990 cohorts were estimated to have
18                   24% and 12% decreases in lifetime (across the first 10 years of life) blood Pb levels,
19                   respectively, per natural log decrease in air Pb. Table 4-14 provides predicted blood Pb
20                   and blood Pb-air Pb slopes as a function of age for the 1987 and 1990 cohorts. The values
21                   in Table 4-14 are for children having complete datasets that lived in the Merced (study
22                   region having medium air Pb concentrations) of Mexico City in medium SES families
23                   (for the study population) that did not use clay pottery. Higher estimated blood Pb  and
24                   blood Pb-air Pb slopes than those in Table 4-14. would be predicted for low SES families
25                   living in Xalostoc (study region having highest air Pb) that use clay pottery (e.g., 2-yr-
26                   olds predicted blood Pb of 11-14 ug/dL and blood Pb-air Pb slope of 5.1-12 ug/dL
27                   per ug/m3).  Conversely, lower estimates would be predicted for high SES families  living
28                   in Pedregal (study region having lowest air Pb) that did not use clay pottery (e.g., 2-yr-
29                   olds predicted blood Pb of 7-9 ug/dL and blood Pb-air Pb slope of 3.8-4.8 ug/dL
30                   per ug/m3).  The effect of air Pb on blood Pb may have been underestimated in this study
31                   due to inclusion of location and SES terms in the regression model. It was specifically
32                   noted by the authors that air Pb differed significantly between the locations and the
33                   poorer residential areas were usually the more industrialized areas with higher pollution.
34                   Hence, the inclusion of these terms may have accounted for some of the variance in blood
35                   Pb attributable to air Pb.
      November 2012                             4-124                  Draft - Do Not Cite or Quote

-------
      Table 4-14   Predicted blood Pb levels and blood-air slopes for Mexico City
                     children (1987 and 1990 cohorts).
Age (in years)
1
2
3
4
5
6
7
8
9
10
Blood Pb
(|ig/dL)
7.4-
8.9-
8.4-
7.9-
7.7-
6.9-
6.8-
6.1 -
5.8-
5.6-
8.5 a
10.2
9.7
9.1
8.8
7.9
7.8
7.1
6.7
6.5
Blood Pb-Air Pb Slope (pg/dL
per |jg/m3)
2.1 -4.5a
2.6-
2.4-
2.3-
2.2-
2.0-
2.0-
1.8-
1.7-
1.6-
5.5
5.2
4.9
4.7
4.2
4.2
3.8
3.6
3.5
      'Values are for 1990 and 1987 cohorts, respectively, at an air Pb concentration of 0.4 ug/m3 which is the median and geometric mean of the
      annual air Pb concentrations over the course of the study based on data in Figure 1 of Schnaas et al. (2004)
      Source: Based on Table 4 of Schnaas et al. (2004)

 1                  For an approach that considers all the Schnaas et al. (2004) cohorts simultaneously, data
 2                  for annual geometric mean of blood Pb and air Pb were extracted from Figure 1 in
 3                  Schnaas et al.  (2004). However, in employing this approach, blood Pb is confounded by
 4                  age and year because in the early years of the study, only younger children were available
 5                  and in the later years of the study, only older children  contributed data. The extracted
 6                  values of the geometric mean of blood Pb and mean air Pb were used in regressions
 7                  employing linear and log-log models for comparison to other studies. The linear model
 8                  obtained was:  PbB = 2.50xPbA + 5.61 (R2 = 0.84), i.e., the linear model produced a
 9                  constant slope of 2.50 (ig/dL per (ig/m3. However, inspection of the graph (not shown
10                  here) suggested a bi-linear fit. Regression of the data over the interval 0.1-0.4 (ig/m3
11                  produced a slope  of 9.0 (ig/dL per (ig/m3 (R2 = 0.83), and regression of the data over the
12                  interval 0.4-2.8 (ig/m3 produced a slope of 1.52 (ig/dL per  (ig/m3 (R2 = 0.83). The log-log
13                  model was: In(PbB) = 0.26xm(PbA) + 2.20  (R2 = 0.94), resulting in an inverse curve for
14                  dPbB/dPb, versus PbA, with a slope of 4.5 (ig/dL per  (ig/m3 at PbA = 0.4 (ig/m3.

15                  Schwartz and Pitcher (1989) reported a multivariate regression analysis of associations
16                  between U.S. gasoline  Pb consumption (i.e., sales) and blood Pb concentrations in the
17                  U.S. population during the period 1976-1980 when use of Pb in gasoline was being
18                  phased out. Although this analysis did not directly derive a slope for the air Pb-blood Pb
19                  relationships, other analyses have shown a strong correlation between U.S. gasoline  Pb
20                  consumption and ambient air Pb levels during this same period (U.S. EPA. 1986a).
21                  Therefore, it is possible to infer an air Pb-blood Pb relationship from these data. Two
      November 2012                             4-125                  Draft - Do Not Cite or Quote

-------
 1                  sources of blood Pb data were used in Schwartz and Pitcher (1989): NHANES II
 2                  provided measurements for U.S. individuals 6 months to 74 years of age (n = 9,987
 3                  subjects) between February 1976 and February 1980, and the Chicago blood Pb  screening
 4                  program provided measurements in black children aged birth to 5 years (n = 5476
 5                  subjects) for the period from 1976 to mid-1980. Observed blood Pb levels were not
 6                  provided. Gasoline Pb consumption for the U.S. was estimated as the product of monthly
 7                  gasoline sales and quarterly estimates of Pb concentrations in gasoline reported to U.S.
 8                  EPA. Based on the NHANES blood Pb data for whites, the regression coefficient for
 9                  blood Pb on the previous month's gasoline Pb usage,  adjusted for age, race, sex, income,
10                  degree of urbanization, nutrient intake, smoking, alcohol consumption, occupational
11                  exposure, and other significant covariates was 2.14 (ig/dL blood per 100 metric tons of
12                  gasoline Pb/day (SE=0.19, p=0.0000);  the authors reported that the results for blacks
13                  were essentially identical. Based on the Chicago blood Pb data, the age-adjusted
14                  regression coefficient was  16.12 ((ig/dL per 1,000 metric tons gasoline Pb/quarter
15                  [SE=1.37, p=0.0001]). When the coefficient was scaled by the ratio of Chicago's
16                  gasoline use to the nation's and converted to units of  100 metric tons per day, the
17                  gasoline Pb coefficient was 1.97 (ig/dL blood per 100 metric tons of gasoline Pb/day),
18                  which is similar to the coefficient reported for the NHANES cohort. U.S. EPA (1986a)
19                  reported data on gasoline Pb consumption (sales) and ambient Pb levels in the U.S.
20                  during the period 1976-1984 (Table 4-15). Based on these data, air Pb concentrations
21                  decreased in association with gasoline  Pb consumption. The linear regression coefficient
22                  for the air Pb decrease was 0.23 (ig/m3  per 100 metric tons gasoline Pb/day (SE = 0.02,
23                  R2~0.95, p <0.0001). If this regression coefficient is used to convert the  blood Pb slopes
24                  from Schwartz and Pitcher (1989). the  corresponding air Pb-blood Pb slopes would be
25                  9.3 and 8.6 (ig/dL per (ig/m3, based on  the NHANES  and Chicago data, respectively
26                  (e.g., 2.14/0.23  = 9.3 and 1.97/0.23=8.6).
      November 2012                            4-126                  Draft - Do Not Cite or Quote

-------
Table
Date
1976
1977
1978
1979
1980
1981
1982
1983
1984
4-15 U.S. gasoline Pb consumption and air Pb levels.
Total Gasoline Pb
(103 metric tons/year)
171.4
168.9
153.0
129.4
78.8
60.7
59.9
52.3
46.0
Total Gasoline Pb
(102 metric tons/day)3
4.70
4.63
4.19
3.53
2.16
1.66
1.64
1.43
1.26
The linear regression coefficient is 0.23 |jg/m3air per 100 metric tons/day (SE= 0.020, R2 = 0.95, p
Conversion factor is 10/365 days/year.
bAnnual mean of per site maximum quarterly means (1 984 Trend Report (Available online:
httD://www.eDa.aov/air/airtrends/Ddfs/Trends Report 1984.pdf
AirPb
(M9/m3)b
1.22
1.20
1.13
0.74
0.66
0.51
0.53
0.40
0.36
O.0001)
      Source: Table 5-5, U.S. EPA 1986 Pb AQCD Q986a).

 1                   Tripathi et al. (2001) reported child blood Pb and air Pb for the city and suburbs of
 2                   Mumbai, India over the period 1984-1996. Pb-free petroleum was introduced in India
 3                   beginning in late 1996, which was outside the period of this study. Blood Pb data were
 4                   obtained from children 6-10 years of age (n = 544) who lived in 13 locations within the
 5                   Mumbai area. Air Pb concentrations were measured from high volume PM samplers
 6                   (with the majority of Pb in the respirable size range) placed at a height of 1.6 meters that
 7                   operated 24 hours. Data on Pb concentrations in air and blood are summarized in Table
 8                   4-16. An additional 16 children from two regions of Mumbai were excluded from the
 9                   analysis because of their high blood Pb levels (geometric means: 69.2 and 20.8 (ig/dL)
10                   and proximity to industrial Pb sources with high air Pb concentrations (geometric means:
11                   41.2 and 6.7 (ig/m3). Based on the data from residential locations presented  in Table 4-16.
12                   blood Pb increased 3.6 (ig/dL per 1 (ig/m3 air Pb (based on linear regression with air Pb
13                   as the sole independent variable). Several uncertainties apply to these estimates,
14                   including potential exposure misclassification since the mean air Pb concentration was
15                   used for each suburb over the entire study period. In addition, the regression estimates
16                   were based on group mean blood Pb estimates for the 13 sampling locations, rather than
17                   on the individual blood Pb estimates. Ingestion of Pb-containing food was estimated in
18                   this study,  but was not considered in the regression equation for estimating blood Pb,
19                   despite the author's conclusion that the ingestion route is important for the intake of Pb
20                   by children in Mumbai.
      November 2012                             4-127                  Draft - Do Not Cite or Quote

-------
      Table 4-16   Air Pb concentrations and blood Pb levels in children in Mumbai,
                    India.
Blood Pb (Hg/dL)
Location
Borivilli
Byculla
Deonar
Goregaon
Govandi
Jogeshwari
Khar
Parel
Sion
Thans (SS)
Vile Parle
Colaba
Vakola
N
12
117
46
21
20
20
17
168
34
37
19
12
21
GM
10.4
11.0
9.5
9.1
8.9
8.6
9.0
10.4
9.6
12.0
9.1
9.2
14.4
GSD
1.67
1.99
2.29
1.30
1.42
1.32
1.53
1.91
1.49
1.86
1.46
1.86
1.64
N
10
30
93
24
10
24
22
37
96
4
7
9
7
Air Pb (|jg/m3)
GM
0.32
0.99
0.11
0.35
0.10
0.11
0.18
0.44
0.39
1.18
0.37
0.14
1.12
GSD
1.51
1.73
3.21
1.77
1.52
2.47
3.15
1.48
1.75
1.04
1.34
1.63
1.12
      The linear regression coefficient is 3.62 ug/dL blood per ug/m3 air (SE= 0.61, R2= 0.76, p <0.001).
      GM, geometric mean; GSD, geometric standard deviation; N, number of subjects.
      Source: Data are from Tripathi et al. (2001).

 1                  Hayes et al. (1994) analyzed data collected as part of the Chicago, IL blood Pb screening
 2                  program for the period 1974-1988, following the phase-out of leaded gasoline. The data
 3                  included 9,604 blood Pb measurements in children (age: 6 months to 5 years) and
 4                  quarterly average air Pb concentrations measured at  12 monitoring stations in Cook
 5                  County, IL. Annual median blood Pb levels declined from 30 ug/dL in 1968 to 12 ug/dL
 6                  in 1988. During most of the years of the study, blood Pb measurements at or below
 7                  10 ug/dL were recorded as 10 ug/dL because of concerns over measurement accuracy of
 8                  the instrument below these levels. Quarterly median blood Pb levels declined in
 9                  association with quarterly mean air Pb concentrations. The regression model predicted a
10                  slope of 0.24 In [ug/dL blood] per ln[ug/m3 air], as illustrated in Figure 4-24. This slope
11                  corresponds to an increase of 8.2  ug/dL blood Pb per ug/m3 at the average annual mean
12                  air Pb concentration of 0.62 ug/m3. As shown in Figure 4-25. with decreasing  air Pb
13                  concentration, the slope increases. The study reports a slope of 5.6 associated with
14                  ambient air Pb levels near 1 ug/m3 and a slope  of 16 for ambient air Pb levels in the range
15                  of 0.25 ug/m3, indicating a pattern of higher ratios with lower ambient air Pb and blood
16                  Pb levels.
     November 2012                            4-128                 Draft - Do Not Cite or Quote

-------
                      30
                      25 -
                      20 -
                      15 -
O)
_Q
Q_
T3
O
=  10 H
                       5 -
                       0
                         0.0     0.2    0.4
                             0.6    0.8    1.0
                              AirPb (|jg/m3)
         1.4    1.6
      Note: The regression model is: (ln[|jg/dL blood Pb] = 0.24-ln[|jg/m air Pb] + 3.17).
      Modified from Hayes et al. (1994).

      Figure 4-24      Predicted relationship between air Pb and blood Pb based on data
                        from Chicago, IL in children age 0-5 years (1974-1988).

 1                  The evidence on the quantitative relationship between air Pb and blood Pb is now, as in
 2                  the past, limited by the circumstances in which the data are collected. These estimates are
 3                  generally developed from studies of populations in various Pb exposure circumstances.
 4                  The 1986 Pb AQCD (U.S. EPA. 1986a) discussed the studies available at that time that
 5                  addressed the relationship between air Pb and blood Pb, recognizing that there is
 6                  significant variability in air-to-blood ratios for different populations exposed to Pb
 7                  through different air-related exposure pathways and at different exposure levels. The
 8                  1986 Pb AQCD noted that ratios derived from studies involving higher blood and air Pb
 9                  levels are generally smaller than ratios from studies involving lower blood and air Pb
10                  levels [see the 1986 Pb AQCD, Chapter 11, pp 99 (U.S. EPA. 1986a)1. In consideration
11                  of this factor, slopes in the range of 3 to 5 for children generally reflected study
12                  populations with blood Pb levels in the range of approximately 10-30 ug/dL [see
13                  Chapter 11, pp 100 of the 1986 Pb AQCD, Table 11-36, from (Brunekreef. 1984)1. much
14                  higher than those common in today's population. The slope of 3.6 from  Tripathi et al.
15                  (2001) is consistent with this observation, given that the blood Pb levels were at the lower
16                  end of this range (i.e., 10-15  ug/dL).
     November 2012
                            4-129
Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
There are fewer studies that evaluate the air Pb-blood Pb relationship in conditions that
are more reflective of the current state. Hilts et al. (2003) is one such study that provides
insight because the blood Pb and air Pb levels were relatively lower than those studies
mentioned above; the slope reported was 6.5, but could be as high as 7.0. Similarly,
Hayes et al. (1994) demonstrates greater slopes observed with decreasing air Pb
concentrations. These studies provide evidence that air-to-blood slopes relevant for
today's population of children would likely extend higher than the 3 to 5 range identified
in the 1986 Pb AQCD (U.S. EPA. 1986a). In the 2008 final rule for the Pb NAAQS  (73
FR 66964), with recognition of uncertainty and variability in the absolute value of an air-
to-blood relationship, the air-to-blood slopes of 5, 7, and 10  (ig/dL per (ig/m3 were
utilized in evaluating air-related IQ loss of children.  Figure 4-25 illustrates the impact of
these air-to-blood slopes on the estimated change in  air-related blood Pb as a function of
change in air Pb.
              M

              .Q
              Q.
              ~O
              O
              _o
              CO
              0)
              M
              re
                   1.5  H
 1 -
                   0.5  •
                                                              Slope
                                                        (ng/dLperng/m3)
                                                                  •10
                                                                   7
                                                                   5
                       0
                  0.05            0.1
             Change in Air Pb (u,g/m3)
 0.15
14
      Figure 4-25      Effect of air-to-blood slope on estimated change in air-related
                        blood Pb with change in air Pb.
     November 2012
                           4-130
Draft - Do Not Cite or Quote

-------
      4.5.2       Air Pb-Blood Pb Relationships in Occupational Cohorts

 1                  At the time of the 1986 Pb AQCD, there was a great deal of information on blood Pb
 2                  responses to air Pb exposures of workers in Pb-related occupations (U.S. EPA, 1986a).
 3                  Almost all such exposures were at air Pb exposures far in excess of typical non-
 4                  occupational exposures and typically did not account for other potential sources of Pb
 5                  exposure. The air Pb-blood Pb slopes in these studies were generally much less
 6                  (i.e., 0.03-0.2; pg 11-106) than those observed in children when considering aggregate air
 7                  Pb contributions  (i.e., 3-5; pg 11-106). In addition, the air Pb concentrations in
 8                  occupational studies are typically collected at much shorter durations (e.g., over an 8-hr
 9                  workday) compared to ambient Pb monitoring, making it difficult to draw comparisons
10                  between occupationally and non-occupationally exposed populations. Therefore, only a
11                  few occupational studies are presented below to demonstrate that more recent air Pb and
12                  blood Pb levels remain much higher in these studies compared to those conducted in the
13                  general population.

14                  Rodrigues et al. (2010) examined factors contributing to variability in blood Pb
15                  concentration in New England bridge painters, who regularly use electric grinders to
16                  prepare surfaces for painting. The study included 84 adults (83 males, 1 female) who
17                  were observed during a 2-week period in 1994 or 1995. The geometric mean air Pb
18                  concentration obtained from personal PM samplers worn over the workday was 58 (ig/m3
19                  (GSD 2.8), with a maximum daily value of 210 (ig/m3. Hand wipe samples were
20                  collected and analyzed for Pb (GM = 793 jig, GSD 3.7). Blood Pb samples were
21                  collected at the beginning of the 2-week period (GM =16.1 (ig/dL, GSD 1.7) and at the
22                  end of the period (GM=18.2 (ig/dL, GD=1.6). Associations between exposure variables
23                  and blood Pb concentrations were explored with multivariate regression models. When
24                  the model excluded hand-wipe data, the regression coefficient for the relationship
25                  between ln[blood Pb concentration (ng/dL)] and ln[air Pb ((ig/m3)] was 0.11  (SE = 0.05,
26                  p = 0.03). This corresponds to a slope of 0.009 (ig/dL per (ig/m3 at the geometric mean
27                  air Pb concentration for the study. A second regression model included hand wipe Pb
28                  (n = 54) and yielded a regression coefficient of 0.05 (SE = 0.07, p = 0.45), which
29                  corresponds to a  slope of 0.02 (ig/dL per (ig/m3 at the geometric mean air Pb
30                  concentration for the study.

31                  Two other studies that examined the air Pb-blood Pb relationship in occupational settings
32                  at higher air Pb concentrations (geometric mean of 82 and  111 (ig/m3) for Pb battery and
33                  crystal workers, respectively (Pierre et al., 2002; Lai etal. 1997). Blood Pb levels for the
34                  Pb battery workers averaged 56.9 (ig/dL (SD 25.3) and for the crystal workers was
35                  21.9 (ig/dL. Both studies employed log-log regression models, resulting in slopes of 0.49
36                  (Pierre et al.. 2002) and 0.08 (Laietal.. 1997V
      November 2012                            4-131                  Draft - Do Not Cite or Quote

-------
      4.5.3        Environmental Pb-Blood Pb Relationships

 1                  Empirically-based relationships between blood Pb levels and Pb intakes and/or Pb
 2                  concentrations in environmental media have provided the basis for what has become
 3                  known as slope factor models. Slope factor models are highly simplified representations
 4                  of empirically based regression models in which the slope parameter represents the
 5                  change in blood Pb concentration projected to occur in association with a change in Pb
 6                  intake or uptake. The slope parameter is factored by exposure parameters (e.g., exposure
 7                  concentrations, environmental media intake rates) that relate exposure to blood Pb
 8                  concentration (Maddaloni etal., 2005; U.S. EPA. 2003c; Abadin and Wheeler. 1997;
 9                  Stern. 1996; Bowers et al.. 1994; Stern.  1994; Carlisle and Wade. 1992). In slope factor
10                  models, Pb biokinetics are represented as a linear function between the blood Pb
11                  concentration and either Pb uptake (uptake slope factor, USF) or Pb intake (intake slope
12                  factor, ISF). The models take the general mathematical forms:


                                                     PbB = E x ISF

                                                                                          Equation 4-2

                                                 PbB = E  x AF x USF

                                                                                          Equation 4-3
13                  where PbB is the blood Pb concentration, E is an expression for exposure (e.g., soil
14                  intake x soil Pb concentration) and AF is the absorption fraction for Pb in the specific
15                  exposure medium of interest. Intake slope  factors are based on ingested rather than
16                  absorbed Pb and, therefore, integrate both  absorption and biokinetics into a single slope
17                  factor, whereas models that utilize an uptake slope factor include a separate absorption
18                  parameter. In contrast to mechanistic models, slope factor models predict quasi-steady
19                  state blood Pb concentrations that correspond to time-averaged daily Pb intakes (or
20                  uptakes) that occur over sufficiently long periods to produce a quasi-steady state
21                  (i.e., >75 days, ~3 times the ti/2 for elimination of Pb in blood).
      November 2012                            4-132                  Draft - Do Not Cite or Quote

-------
 1                  The U.S. EPA Adult Lead Methodology (ALM) is an example of a slope factor model
 2                  that has had extensive regulatory use in the EPA Superfund program for assessing health
 3                  risks to adults associated with non-residential exposures to Pb in contaminated soils
 4                  (Maddaloni et al.. 2005; U.S. EPA. 1996a). The model was developed to predict maternal
 5                  and fetal blood Pb concentrations that might occur in relation to maternal exposures to
 6                  contaminated soils.  The model assumes an uptake slope factor of 0.4 (ig/dL blood
 7                  per (ig/day Pb uptake. Additional discussion of slope factor models that have been used
 8                  or proposed for regulatory use can be found in the 2006 Pb AQCD (U.S. EPA. 2006b).

 9                  Previous studies included in the 2006 Pb AQCD (U.S. EPA. 2006b) explored the
10                  relationship between blood Pb in children and environmental Pb concentrations. In a
11                  pooled analysis of 12 epidemiologic studies, interior dust Pb loading, exterior soil/dust
12                  Pb, age, mouthing behavior, and race were all statistically significant variables included
13                  in the regression model for blood Pb concentration (Lanphear et al., 1998). Significant
14                  interactions were found for age and dust Pb loading, mouthing behavior and exterior
15                  soil/dust level, and SES and water Pb level. In a meta-analysis of 11 epidemiologic
16                  studies, among children the most common exposure pathway influencing blood Pb
17                  concentration in structural equation modeling was exterior soil,  operating through its
18                  effect on interior dust Pb and hand Pb (Succop et al.. 1998). Similar to Lanphear et al.
19                  (1998). in the linear regression model, interior dust Pb loading had the strongest
20                  relationships with blood Pb concentration. Individual studies conducted in Rochester,
21                  NY, Cincinnati, OH, and Baltimore, MD report similar relationships between children's
22                  blood Pb and interior dust concentrations (Lanphear and Roghmann. 1997; U.S. EPA.
23                  1996b: Bornschein et al.. 1985).

24                  Dixon et al. (2009) reported a multivariate analysis of associations between
25                  environmental Pb concentrations and blood Pb concentrations, based on data collected in
26                  the NHANES (1999-2004). The analyses included 2,155 children, age 12-60 months. The
27                  population-weighted geometric mean blood Pb concentration was 2.03 (ig/dL
28                  (GSD  1.03). A linear model applied to these data yielded an R2 of 40% (Table 4-17). The
29                  regression coefficient for the relationship between ln[blood Pb concentration ((ig/dL)]
30                  and ln[floor dust Pb concentration ((ig/ft2)] was 0.386 (SE 0.089) for "not smooth and
31                  cleanable" surfaces  (e.g., high-pile carpets) and 0.205 (SE 0.032) for "smooth and
32                  cleanable" surfaces  (e.g., uncarpeted or low-pile carpets). These coefficients correspond
33                  to a 2.4-fold or 1.6-fold increase in blood Pb concentration, respectively, for a 10-fold
34                  increase in floor dust Pb concentration.
      November 2012                             4-133                  Draft - Do Not Cite or Quote

-------
Table 4-17 Linear model relating environmental Pb
concentration in children.
Variables Overall p-value
Intercept 0.172
Age (in years) <0.001
Year of construction 0.014
PIR <0.001
Race/ethnicity <0.001
Country of birth 0.002
Floor surface/condition x log floor <0.001
PbD
Floor surface/condition
x (log floor PbD)2
Floor surface/condition
x (log floor PbD)3
Log windowsill PbD 0.002
Home-apartment type <0.001
Anyone smoke inside the home 0.015
Log cotinine concentration (ng/dL) 0.004
Window, cabinet, or wall renovation 0.045
in a pre-1 978 home
Levels3

Age
Age2
Age3
Age4
Intercept for missing
1990-present
1978-1989
1960-1977
1950-1959
1940-1949
Before 1940
Intercept for missing
Slope
Non-Hispanic white
Non-Hispanic black
Hispanic
Other
Missing
U.S."
Mexico
Elsewhere
Intercept for missing
Not smooth and cleanable
Smooth and cleanable or carpeted
Not smooth and cleanable
Smooth and cleanable or carpeted
Uncarpeted not smooth and
cleanable
Smooth and cleanable or carpeted
Intercept for missing
Slope
Intercept for missing
Mobile home or trailer
One family house, detached
One family house, attached
Apartment (1-9 units)
Apartment (> 10 units)
Missing
Yes
No
Intercept for missing
Slope
Missing
Yes
No
exposure and blood Pb
Estimate (SE)
-0.517(0.373)
2.620 (0.628)
-1.353(0.354)
0.273 (0.083)
-0.019 (0.007)
-0.121 (0.052)
-0.198(0.058)
-0.196(0.060)
-0.174(0.056)
-0.207 (0.065)
-0.012(0.072)
0.000
0.053 (0.065
-0.053(0.012)
0.000
0.247 (0.035
-0.035 (0.030)
0.128(0.070)
-0.077 (0.219)
0.000
0.353 (0.097)
0.154(0.121
0.178(0.094)
0.386 (0.089)
0.205 (0.032)
0.023(0.015)
0.027 (0.008)
-0.020 (0.014)
-0.009 (0.004)
0.053 (0.040
0.041 (0.011
-0.064 (0.097
0.127(0.067)
-0.025 (0.046)
0.000
0.069 (0.060)
-0.133(0.056)
0.138(0.140)
0.100(0.040)
0.000
-0.150(0.063)
0.039(0.012)
-0.008(0.061)
0.097 (0.047)
0.000
p-Value
0.172
<0.001
<0.001
0.002
0.008
0.024
0.001
0.002
0.003
0.003
0.870
0.420
<0.001
<0.001
0.251
0.073
0.728
<0.001
0.209
0.065
<0.001
<0.001
0.124
0.001
0.159
0.012
0.186
<0.001
0.511
0.066
0.596
0.256
0.022
0.331
0.015
0.023
0.002
0.896
0.045
'Children: n = 2,155 (age 10-60 months); R2 = 40%
'includes the 50 states and the District of Columbia
Source: Dixon et al. (20091.
November 2012
4-134
Draft - Do Not Cite or Quote

-------
 1                  Mielke et al. (2007a) analyzed blood Pb and soil Pb concentration data collected as part
 2                  of a blood Pb screening program in New Orleans, Louisiana (2000-2005). The data set
 3                  included 55,551 blood Pb measurements for children 0-6 years of age and 5,467 soil Pb
 4                  measurements. Blood Pb and soil Pb concentrations were matched at the level of census
 5                  tracts. The association between blood Pb concentration and soil Pb concentration was
 6                  evaluated using non-parametric permutation methods. The resulting best-fit model was:
                                           PbB = 2.038  +(0.172 xPbS05)
                                                                                         Equation 4-4
 7                  where PbB is the median blood Pb concentration and PbS is the median soil Pb
 8                  concentration. Although the overall association between blood Pb and soil Pb was strong
 9                  (R2=0.528), there was considerable scatter in the data. For example, at the median soil Pb
10                  levels of 100 and 500 mg/kg, median blood Pb ranged from 2 to 8 (ig/dL  and 3 to
11                  12 (ig/dL, respectively.  The resulting curvilinear relationship predicts a twofold increase
12                  in blood Pb concentration for an increase in soil Pb concentration from 100 to 1,000 ppm
13                  (Figure 4-26).
     November 2012                             4-135                  Draft - Do Not Cite or Quote

-------
                       O)
                       .Q
                       CL
                       •a
                       o
                       _o
                       m
                                                      Blood=2.038+0.172xSoil°-5
                                       400       800       1200
                                                  SoilPb(ppm)
                      1600
          2000
      Note: The data set included 55,551 blood Pb measurements for children 0-6 years of age and 5,467 soil Pb measurements. Blood
      Pb and soil Pb concentrations were matched at the level of census tracts (Mielke et al.. 2007a).

      Figure 4-26      Predicted relationship between soil Pb concentration and blood
                        Pb concentration in children based on data collected in
                        New Orleans, Louisiana:  2000-2005.

 1                  In a subsequent re-analysis of the New Orleans (2000-2005) data, individual child blood
 2                  Pb observations were matched to census tract soil concentrations (Zahran etal. 2011).
 3                  This analysis confirmed the association between blood Pb and both soil Pb and age
 4                  reported in Mielke et al. (2007a). Regression coefficients for soil Pb (random effects
 5                  generalized least squares regression) ranged from 0.217 to 0.214 (per soil Pb05), which is
 6                  equivalent to approximately a 2-fold increase in blood Pb concentration for an increase in
 7                  soil Pb concentration from  100 to 1,000 ppm.

 8                  Several studies have linked elevated blood Pb levels to residential soil exposures for
 9                  populations living nearby industrial or mining facilities. Gulson et al. (2009) studied the
10                  blood Pb and isotopic Pb ratios of children younger than 5-years old and adults older than
11                  18-years old living in the vicinity of a mine producing Magellan Pb ore in western
12                  Australia. They observed a median blood Pb level of 6.6 (ig/dL for the children, with
13                  isotopic ratios indicating contributions from the mine ranging from 27 to 93%. A weak
14                  but significant linear association between blood Pb level and percent Magellan Pb was
15                  observed (R2 = 0.12, p = 0.018). Among children with blood Pb levels over 9 (ig/dL and
16                  among adults, the isotopic ratios revealed Pb exposures from a variety of sources.
17                  Garavan et al. (2008) measured soil Pb and blood Pb levels among children aged 1-month
18                  to 17.7-years old in an Irish town near a coal mine. The blood Pb measurements were
     November 2012
4-136
Draft - Do Not Cite or Quote

-------
 1                  instituted as part of a screening and community education program given that the
 2                  presence of Pb had been documented in the environment. Garavan et al. (2008) found that
 3                  over 3 years of the screening period, median blood Pb levels reduced by roughly 22%
 4                  from 2.7 to 2.1 ng/dL.

 5                  An extensive discussion of the relationships between environmental Pb levels and blood
 6                  Pb concentrations in children at the Bunker Hill Superfund Site location, a former Pb
 7                  mining and smelting site, was provided in the 2006 Pb AQCD (U.S. EPA. 2006c). In the
 8                  most recent analysis (TerraGraphics Environmental Engineering. 2004) of the data on
 9                  environmental Pb levels and child blood Pb concentrations (1988-2002), blood Pb
10                  concentrations (annual GM) ranged from 2.6 to 9.9 (ig/dL. Environmental Pb levels
11                  (e-g-, dust, soil, paint Pb levels) data were collected at -3,000 residences, with interior
12                  dust Pb concentrations (annual GM) ranging from -400 to 4,200 mg/kg and yard soil Pb
13                  concentration (annual GM) ranging from -150 to 2,300 mg/kg. Several multivariate
14                  regression models relating environmental Pb levels and blood Pb concentration were
15                  explored; the model having the highest R2 (0.26) is shown in Table 4-18. The model
16                  predicts significant associations between blood Pb concentration, age, interior dust, yard
17                  soil, neighborhood soil (geometric mean soil Pb concentration for areas within 200 ft of
18                  the residence), and community soil Pb concentration (community GM). Based on the
19                  standardized regression coefficients, the community soil Pb concentration had the largest
20                  effect on blood Pb concentration, followed by neighborhood soil Pb concentration,
21                  interior dust Pb concentration, and yard soil Pb concentration (Table 4-18). The model
22                  predicted a 1.8 (ig/dL decrease in blood Pb concentration in association with a decrease
23                  in community soil Pb concentration from 2,000 to 1,000 mg/kg. The same decrease in
24                  neighborhood soil Pb concentration, interior dust Pb concentration, or yard soil Pb
25                  concentration was predicted to result in a 0.8, 0.5, or 0.2 (ig/dL decrease in blood Pb
26                  concentration, respectively. Note that the soil Pb component of the model was similar to
27                  that derived by Lewin et al. (1999). in which a model of blood Pb as a function of soil Pb
28                  among 0-6 year old children living near one of four industrial sites was given as
29                  PbB = 0.24381n(PbS} + 0.2758.
      November 2012                            4-137                  Draft - Do Not Cite or Quote

-------
      Table 4-18   General linear model relating blood Pb concentration in children and
                    environmental  Pb levels—Bunker Hill Superfund Site.
Parameter
Intercept
Age (yr)
ln(interiordust Pb); (mg/kg)
ln(yard soil Pb); (mg/kg)
GM soil Pb within 200 ft of residence (mg/kg)
GM community soil Pb (mg/kg)
Coefficient
-0.1801
-0.4075
0.7288
0.2555
0.0008
0.0018
P-value
0.7916
<0.0001
<0.0001
0.0002
<0.0001
<0.0001
Standardized
Coefficient
0.00000
-0.2497
0.1515
0.0777
0.1380
0.2250
      R2 = 0.264; p <0.0001; based on data from Bunker Hill Superfund Site, collected over the period 1988-2002.
      GM: geometric mean; In: natural log.
      Source: TerraGraphics (2004).

 1                  Malcoe et al. (2002) analyzed 1997 data on blood Pb and environmental Pb
 2                  concentrations in a representative sample of Native American and white children
 3                  (n = 224, age 1-6 years) who resided in a former Pb mining region in Ottawa County,
 4                  OK. The data set included measurements of blood Pb, yard soil Pb, residential interior
 5                  dust Pb loading, first-draw water Pb, paint Pb assessment and other behavioral (i.e., hand-
 6                  to-mouth activity, hygiene rating) and demographic variables (i.e., hygiene rating,
 7                  poverty level, caregiver education). A multivariate regression model accounted for 34%
 8                  of the observed variability in blood Pb. Yard soil Pb and interior dust Pb loading
 9                  accounted for 10% and 3% of the blood Pb variability, respectfully. The regression model
10                  predicted a slope of 0.74 (ig/dL blood Pb per ln[(ig/g soil Pb] and a slope of 0.45 (ig/dL
11                  blood Pb per ln[(ig/ft2] dust Pb loading.
      4.6       Biokinetic Models of Pb Exposure-Blood Pb Relationships

12                  An alternative to regression models are mechanistic models, which attempt to specify all
13                  parameters needed to describe the mechanisms (or processes) of transfer of Pb from the
14                  environment to human tissues. Such mechanistic models are more complex than
15                  regression models; this added complexity introduces challenges in terms of their
16                  mathematical solution and empirical verification. However, by incorporating parameters
17                  that can be expected to vary spatially or temporally, or across individuals or populations,
18                  mechanistic models can be extrapolated to a wide range of exposure scenarios, including
19                  those that may be outside of the domain of paired predictor-outcome data used to develop
20                  the model. Exposure-intake models, a type of mechanistic models, are highly simplified
21                  mathematical representations of relationships between levels of Pb in environmental
     November 2012                            4-138                 Draft - Do Not Cite or Quote

-------
 1                  media and human Pb intakes (e.g., (ig Pb ingested per day). These models include
 2                  parameters representing processes of Pb transfer between environmental media (e.g., air
 3                  to surface dust) and to humans, including rates of human contact with the media and
 4                  intakes of the media (e.g., g soil ingested per day). Intake-biokinetic models provide the
 5                  analogous mathematical representation of relationships between Pb intakes and Pb levels
 6                  in body tissues (e.g., blood Pb concentration). Biokinetic models include parameters that
 7                  represent processes of Pb transfer (a) from portals of entry into the body and (b) from
 8                  blood to tissues and excreta. Linked together, exposure-intake and intake-biokinetics
 9                  models (i.e., integrated exposure-intake-biokinetics models) provide an approach for
10                  predicting blood Pb concentrations (or Pb concentrations in other tissues) that
11                  corresponds to a specified exposure (medium, concentration, and duration). Detailed
12                  information on exposure and internal dose can be obtained from controlled experiments,
13                  but almost never from epidemiological observations or from public health monitoring
14                  programs. Exposure intake-biokinetics models can provide these predictions in the
15                  absence of complete information on the exposure history and blood Pb concentrations for
16                  an individual (or population) of interest. Therefore, these models are critical to applying
17                  epidemiologic-based information on blood Pb-response relationships to the quantification
18                  and characterization of human health risk. They are also critical for assessing the
19                  potential impacts of public health programs directed at mitigation of Pb exposure or of
20                  remediation of contaminated sites.

21                  However, they are not without their limitations. Human exposure-biokinetics models
22                  include large numbers of parameters, which are required to describe the many processes
23                  that contribute to Pb intake, absorption, distribution, and elimination. The large number
24                  of parameters complicates the assessment of confidence in parameter values, many of
25                  which cannot be directly measured.  Statistical procedures can be used to evaluate the
26                  degree to which model outputs conform to "real-world" observations and values of
27                  influential parameters can be statistically estimated to achieve good agreement with
28                  observations. Still, large uncertainty can be expected to remain about many, or even
29                  most, parameters in complex exposure-biokinetic models. Such uncertainties need  to be
30                  identified and their impacts on model predictions quantified (i.e., sensitivity analysis or
31                  probabilistic methods).

32                  Modeling of human Pb exposures and biokinetics has  advanced considerably during the
33                  past several decades, although there have been relatively few developments since the
34                  2006 Pb AQCD was published. Still in use is the Integrated Exposure Uptake Biokinetic
35                  (IEUBK) Model for Lead in Children (U.S. EPA. 1994) and models that simulate Pb
36                  biokinetics in humans from birth through adulthood (O'Flahertv, 1995; Leggett 1993;
37                  O'Flahertv.  1993). The EPA AALM is still in development. A complete and extensive
38                  discussion of these models can be found in the 2006 Pb AQCD (U.S. EPA. 2006b).
      November 2012                             4-139                  Draft - Do Not Cite or Quote

-------
      4.7        Summary and Conclusions
      4.7.1       Exposure

 1                  Exposure data considered in this assessment build upon the conclusions of the
 2                  2006 Pb AQCD (2006b), which found that air Pb concentrations in the U.S. and
 3                  associated biomarkers of exposure to Pb have decreased substantially following
 4                  reductions in industrial point sources of Pb, and restrictions on Pb in gasoline, house-hold
 5                  paints, and solder. Pb exposure is difficult to assess because Pb has multiple sources in
 6                  the environment and passes through various media. The atmosphere is the main
 7                  environmental transport pathway for Pb, and, on a global scale, atmospheric Pb is
 8                  primarily associated with fine particulate matter, which can deposit to soil and water. In
 9                  addition to primary emission of particle-bearing or gaseous Pb to the atmosphere, Pb can
10                  be suspended to the air from soil or dust. Air-related pathways of Pb exposure are the
11                  focus of this assessment. In addition to inhalation of Pb from ambient air, air-related Pb
12                  exposure pathways include inhalation and ingestion of Pb from indoor dust and/or
13                  outdoor soil that originated from recent or historic ambient air (e.g., air Pb that has
14                  penetrated into the residence either via the air or tracking of soil), ingestion of Pb in
15                  drinking water contaminated from atmospheric deposition  onto surface waters or from
16                  indirect surface runoff of deposition of ambient Pb, and ingestion of Pb in dietary sources
17                  after uptake by plants or grazing animals. Non-air-related Pb exposures may include
18                  occupational exposures, hand-to-mouth contact with Pb-containing consumer goods,
19                  hand-to-mouth contact with dust or chips of peeling Pb-containing paint, or ingestion of
20                  Pb in drinking water conveyed through Pb pipes. Pb can cycle through multiple media
21                  prior to human exposure. Given the multitude of possible air-related exposure scenarios
22                  and the related difficulty of constructing Pb exposure histories, most studies of Pb
23                  exposure through air, water, and soil can be informative to this review. Other exposures,
24                  such as occupational exposures, contact with consumer goods in which Pb has been used,
25                  or ingestion of Pb in drinking water  conveyed through Pb pipes may also contribute to Pb
26                  body burden.

27                  A number of monitoring and modeling techniques have been employed for ambient Pb
28                  exposure assessment. Environmental Pb concentration data can be collected from
29                  ambient air Pb monitors, soil Pb samples, dust Pb samples, and dietary Pb samples to
30                  estimate human exposure. Exposure estimation error depends in part on the collection
31                  efficiency of these methods; collection efficiency for ambient air Pb FRM samplers is
32                  described in Section 3.4. Additionally, high spatial variability of the Pb concentrations in
33                  various media also can contribute to exposure error, as described in  the 2009 PM ISA
34                  (U.S. EPA. 2009a). Models, such as the Integrated Exposure Uptake Biokinetic (IEUBK)
      November 2012                            4-140                  Draft - Do Not Cite or Quote

-------
 1                  model, simulate human exposure to Pb from multiple sources and through various routes
 2                  including inhalation, ingestion, and dermal exposure. IEUBK model inputs include soil
 3                  Pb concentration, air Pb concentration, dietary Pb intake including drinking water, Pb
 4                  dust ingestion, human activity, and biokinetic factors. Measurements and/or assumptions
 5                  can be utilized when formulating the model inputs; errors in measurements and
 6                  assumptions thus have the potential to propagate through the exposure models.

 7                  Section 4.1 presents data illustrating potential exposure pathways. Soil can act as a
 8                  reservoir for deposited Pb emissions, and exposure to soil contaminated with deposited
 9                  Pb can occur through resuspended PM as well as shoe tracking and hand-to-mouth
10                  contact, which is the main pathway of childhood exposure to Pb. Airborne particles
11                  containing Pb tend to be small (much of the distribution <10 nm) compared with Pb in
12                  soil or dust particles (~50 pirn to several hundred nm); Pb deposition to soil is described
13                  in Section 3.3. Hence, hand-to-mouth contact with Pb-bearing soil or dust and/or tracking
14                  Pb contaminated soil or dust into homes are more common means for human exposure to
15                  Pb. Infiltration of Pb dust into indoor environments has been observed, and Pb dust has
16                  been shown to persist in indoor environments even after repeated cleanings.
17                  Measurements of particle-bound Pb exposures reported in this assessment have shown
18                  that personal exposure measurements for Pb concentration are typically higher than
19                  indoor or outdoor ambient Pb concentrations.
      4.7.2        Toxicokinetics

20                  The majority of Pb in the body is found in bone (roughly 90% in adults, 70% in children);
21                  only about 1% of Pb is found in the blood. Pb in blood is primarily (-99%) bound to red
22                  blood cells (RBCs). It has been suggested that the small fraction of Pb in plasma (<1%)
23                  may be the more biologically labile and lexicologically active fraction of the circulating
24                  Pb. The relationship between Pb in blood and plasma is pseudo-linear at relatively low
25                  daily Pb intakes (i.e., <10 ug/day/kg) and at blood Pb concentrations <25 (ig/dL, and
26                  becomes curvilinear at higher blood Pb concentrations  due to saturable binding to RBC
27                  proteins. As blood Pb level increases and the higher affinity binding sites for Pb in RBCs
28                  become saturated, a larger fraction of the blood Pb is available in plasma to distribute to
29                  brain and other Pb-responsive tissues.

30                  The burden of Pb in the body may  be viewed as divided between a dominant slow
31                  (i.e., uptake and elimination) compartment (bone) and smaller fast compartment(s) (soft
32                  tissues). Pb uptake  and elimination in soft tissues is much faster than in bone. Pb
33                  accumulates in bone regions undergoing the most active calcification at the time of
34                  exposure. During infancy and childhood, bone calcification is most active in trabecular
      November 2012                            4-141                 Draft - Do Not Cite or Quote

-------
 1                  bone (e.g., patella); whereas, in adulthood, calcification occurs at sites of remodeling in
 2                  cortical (e.g., tibia) and trabecular bone (Aufderheide and Wittmers. 1992). A high bone
 3                  formation rate in early childhood results in the rapid uptake of circulating Pb into
 4                  mineralizing bone; however, in early childhood bone Pb is also recycled to other tissue
 5                  compartments or excreted in accordance with a high bone resorption rate (O'Flaherty.
 6                  1995). Thus, much of the Pb acquired early in life is not permanently fixed in the bone.

 7                  The exchange of Pb from plasma to the bone surface is a relatively rapid process. Pb in
 8                  bone becomes distributed in trabecular and the more dense cortical bone. The proportion
 9                  of cortical to trabecular bone in the human body varies by age, but on average is about
10                  80% cortical to 20% trabecular. Of the bone types, trabecular bone is more reflective of
11                  recent exposures than is cortical bone due to the slow turnover rate and lower blood
12                  perfusion of cortical bone. Some Pb diffuses to deeper bone regions where it is relatively
13                  inert, particularly in adults. These bone compartments are much more labile in infants
14                  and children than in adults as reflected by half-times for movement of Pb from bone into
15                  to the plasma (e.g., cortical half-time = 0.23 years at birth, 3.7 years at 15 years of age,
16                  and 23 years in adults; trabecular half-time = 0.23 years at birth, 2.0 years at 15 years of
17                  age, and 3.8 years in adults) (Leggett 1993).

18                  Evidence for maternal-to-fetal transfer of Pb in humans is derived from cord blood to
19                  maternal blood Pb ratios. Group mean ratios range from about 0.7 to 1.0 at the time of
20                  delivery for mean maternal blood Pb levels ranging from 1.7 to 8.6 (ig/dL. Transplacental
21                  transfer of Pb may be  facilitated by an increase in the plasma/blood Pb concentration
22                  ratio during pregnancy. Maternal-to-fetal transfer of Pb appears to be related partly to the
23                  mobilization of Pb from the maternal skeleton.

24                  The dominant elimination phase of Pb kinetics in the blood, exhibited shortly after a
25                  change in exposure occurs, has a half-life of-20-30 days. An abrupt change in Pb uptake
26                  gives rise to a relatively rapid change in blood Pb, to a new quasi-steady state, achieved
27                  in -75-100 days (i.e.,  3-4 times the blood elimination half-life). A slower phase of Pb
28                  clearance from the blood may become evident with longer observation periods following
29                  a decrease  in exposure due to the gradual redistribution of Pb among bone and other
30                  compartments.
      4.7.3        Pb Biomarkers

31                   Overall, trends in blood Pb levels have been decreasing among U.S. children and adults
32                   over the past 20 years (Section 4.4). The median blood Pb level for the entire U.S.
33                   population is 1.2 ug/dL and the 95th percentile blood Pb level was 3.7 ug/dL, based on
34                   the 2007-2008 NHANES data (NCHS. 2010). Among children aged 1-5 years, the

      November 2012                             4-142                  Draft - Do Not Cite or Quote

-------
 1                  median and 95th percentiles were slightly higher at 1.4 ug/dL and 4.1 ug/dL,
 2                  respectively.

 3                  Blood Pb is dependent on both the recent exposure history of the individual, as well as
 4                  the long-term exposure history that determines body burden and Pb in bone. The
 5                  contribution of bone Pb to blood Pb changes depending on the duration and intensity of
 6                  the exposure, age, and various other physiological stressors that may affect bone
 7                  remodeling (e.g., nutritional status, pregnancy, menopause, extended bed rest,
 8                  hyperparathyroidism) beyond that which normally and continuously occurs. In children,
 9                  largely due to faster exchange of Pb to and from bone, blood Pb is both an index of recent
10                  exposure and potentially an index of body burden. In adults and children, where exposure
11                  to Pb has effectively ceased or greatly decreased, a slow decline in blood Pb
12                  concentrations over the period of years is most likely due to the gradual release of Pb
13                  from bone. Bone Pb is an index of cumulative exposure and body burden. Even bone
14                  compartments should be recognized as reflective of differing exposure periods with Pb in
15                  trabecular bone exchanging more rapidly than Pb in  cortical bone with the blood. This
16                  difference in the compartments makes Pb in cortical bone a better marker of cumulative
17                  exposure and Pb in trabecular bone more likely to be correlated with blood Pb, even in
18                  adults.

19                  Sampling frequency is an important consideration when evaluating blood Pb and bone Pb
20                  levels in epidemiologic studies, particularly when the exposure is not well characterized.
21                  It is difficult to determine what blood Pb is reflecting in cross-sectional studies that
22                  sample blood Pb once, whether recent exposure or movement of Pb from bone into blood
23                  from historical exposures. In contrast, cross-sectional studies of bone Pb and longitudinal
24                  samples of blood Pb concentrations overtime provide more of an index of cumulative
25                  exposure and are more reflective of average Pb body burdens overtime. The degree to
26                  which repeated sampling will reflect the actual long-term time-weighted average blood
27                  Pb concentration depends on the sampling frequency in relation to variability in
28                  exposure. High variability in Pb exposures can produce episodic (or periodic)  oscillations
29                  in blood Pb concentration that may not be captured with low  sampling frequencies.
30                  Furthermore, similar blood Pb concentrations in two individuals (or populations),
31                  regardless of their age, do not necessarily translate to similar body burdens or similar
32                  exposure histories.

33                  The concentration of Pb in urine follows blood Pb concentration, in that it mainly reflects
34                  the exposure history of the previous few months and therefore, is likely a relatively poor
35                  index of Pb body burden. There is added complexity with Pb in urine because
36                  concentration is also dependent upon  urine flow rate, which requires timed urine samples
      November 2012                            4-143                  Draft - Do Not Cite or Quote

-------
 1                  that is often not feasible in epidemiologic studies. Other biomarkers have been utilized to
 2                  a lesser extent (e.g., Pb in teeth).
      4.7.4       Air Lead-Blood Lead Relationships

 3                  The 1986 Pb AQCD (U.S. EPA. 1986a) described epidemiological studies of
 4                  relationships between air Pb and blood Pb. Much of the pertinent earlier literature for
 5                  child populations described in the 1986 Pb AQCD was also included in a meta-analysis
 6                  by Brunekreef (1984). Based on the studies available at that time, the  1986 Pb AQCD
 7                  concluded that "the blood Pb versus air Pb slope (3 is much smaller at high blood and air
 8                  levels." This is to say that the slope (3 was much smaller for occupational exposures
 9                  where high blood Pb levels (>40 ug/dL) and high air Pb levels (much greater than
10                  10 ug/m3) prevailed relative to lower environmental exposures which showed lower
11                  blood Pb and air Pb concentrations (<30 ug/dL and  <3 ug/m3). For those environmental
12                  exposures, it was concluded that the relationship between blood Pb and air Pb "... for
13                  direct inhalation appears to be approximately linear in the range of normal ambient
14                  exposures (0.1-2.0 ug/m3)" (Chapter 1, pp  98 of the 1986 Pb AQCD). In addition to the
15                  meta-analysis of Brunekreef (1984). more recent studies have provided data from which
16                  estimates of the blood Pb-air Pb slope can be derived for children (Table 4-12).  The range
17                  of estimates from these studies is 4-9 ug/dL per ug/m3, which encompasses the  estimate
18                  from the Brunekreef (1984) meta-analysis. Most studies have described the blood Pb-air
19                  Pb relationship as either log-log (Schnaas et al.. 2004; Hayes et al.. 1994; Brunekreef,
20                  1984). which predicts an increase in the blood Pb-air Pb slope with decreasing air Pb
21                  concentration or linear (Hilts. 2003; Tripathi et al. 2001; Schwartz and Pitcher. 1989).
22                  which predicts a constant blood Pb-air Pb slope regardless of air Pb concentrations. These
23                  differences may simply reflect model selection by the  investigators; alternative models
24                  are not reported in these studies. The blood Pb-air Pb slope may also be affected in some
25                  studies by the inclusion of parameters (e.g., soil Pb) that may account for some of the
26                  variance in blood Pb attributable to air Pb. Other factors that likely contribute to the
27                  derived blood Pb-air Pb slope include  differences in the populations examined and Pb
28                  sources, which varied among individual studies.
      November 2012                            4-144                 Draft - Do Not Cite or Quote

-------
References for Chapter 4
    Abadin. HG: Wheeler. JS. (1997). Guidance for risk assessment of exposure to lead: A site-specific,
        multi-media approach. In JS Andrews; H Frumkin; BL Johnson; MA Mehlman; C Xintaras; JA
        Bucsela (Eds.), Hazardous waste and public health: International congress on the health effects of
        hazardous waste (pp. 477-485). Princeton, NJ: Princeton Scientific Publishing.

    Abrahams. PW; Follansbee. MH; Hunt. A; Smith. B; Wragg. J. (2006). Iron nutrition and possible lead
        toxicity: An appraisal of geophagy undertaken by pregnant women of U.K. Asian communities. Appl
        Geochem 21: 98-108. http://dx.doi.Org/10.1016/i.apgeochem.2005.09.015

    ACGIH (American Conference of Governmental Industrial Hygienists). (1985). Particle size-selective
        sampling in the workplace: Report of the ACGIH Technical Committee on Air Sampling Procedures.
        Cincinnati, OH.

    Adgate. JL; Mongin. SJ; Pratt. GC; Zhang. J; Field. MP; Ramachandran. G; Sexton. K. (2007).
        Relationships between personal, indoor, and outdoor exposures to trace elements in PM2.5. Sci Total
        Environ 386: 21-32. http://dx.doi.Org/10.1016/i.scitotenv.2007.07.007

    Al-Modhefer. AJ; Bradbury. MW; Simons. TJ. (1991). Observations on the chemical nature of lead in
        human blood serum. Clin Sci (Lond) 81: 823-829.

    Alexander. FW; Clayton. BE; Delves. HT. (1974). Mineral and trace-metal balances in children receiving
        normal and synthetic diets. QJM 43: 89-111.

    Alves. C; Goncalves. C; Fernandes. AP; Tarelho. L; Pio. C. (2011). Fireplace and woodstove fine particle
        emissions from combustion of western Mediterranean wood types. Atmos Res 101: 692-700.
        http://dx.doi.0rg/10.1016/i.atmosres.2011.04.015

    Amaral. JH; Rezende. VB; Quintana. SM; Gerlach. RF; Barbosa. F. Jr; Tanus-Santos. JE. (2010). The
        relationship between blood and serum lead levels in peripartum women and their respective umbilical
        cords. Basic Clin Pharmacol Toxicol 107: 971-975.
        http://dx.doi.0rg/10.llll/i.1742-7843.2010.00616.x

    Araki. S; Aono. H; Yokoyama. K; Murata. K. (1986). Filterable plasma concentration, glomerular
        filtration, tubular balance, and renal clearance of heavy metals and organic substances in metal
        workers. Arch Environ Occup Health 41: 216-221. http://dx.doi.org/10.1080/00039896.1986.9938336

    Araki. S; Sata. F; Murata. K. (1990). Adjustment for urinary flow rate: an improved approach to
        biological monitoring. Int Arch Occup Environ Health 62: 471-477.
        http://dx.doi.org/10.1007/BF00379066

    Armstrong. R; Chettle. PR; Scott. MC; Somervaille. LJ; Pendlington. M. (1992). Repeated measurements
        of tibia lead concentrations by in vivo X-ray fluorescence in occupational exposure. Occup Environ
        Med49:  14-16.

    Aro. A; Amarasiriwardena. C; Lee. ML; Kim. R; Hu. H. (2000). Validation of K x-ray fluorescence bone
        lead measurements by inductively coupled plasma mass spectrometry in cadaver legs. Med Phys 27:
        119-123. http://dx.doi.0rg/10.1118/l.598863

    Arora. M; Ettinger. AS; Peterson. KE; Schwartz. J; Hu. H; Hernandez-Avila. M; Tellez-Rojo. MM;
        Wright. RO. (2008). Maternal dietary intake of polyunsaturated fatty acids modifies the relationship
        between lead levels in bone and breast milk. J Nutr 138: 73-79.

    Arruda-Neto. JDT; de Oliveira. MCC; Sarkis. JES; Sordini. P;  Manso-Guevara. MV; Garcia. F; Prado.
        GR; Krug. FJ; Mesa. J; Bittencourt-Oliveira. MC; Garcia. C; Rodrigues. TE; Shtejer. K; Genofre. GC.
        (2009). Study of environmental burden of lead in children using teeth as bioindicator. Environ Int 35:
        614-618. http://dx.doi.0rg/10.1016/i.envint.2008.12.005
November 2012                              4-145                   Draft - Do Not Cite or Quote

-------
     Arruda-Neto. JD; Geraldo. LP; Prado. GR; Garcia. F; Bittencourt-Oliveira. MC: Sarkis. JE; Martinez-
        Lusardo. F; Lima-Cazorla. L; Rosa-Medero. D; Rodrigues. TE; Genofre. GC. (2010). Study of metals
        transfer from environment using teeth as biomonitor. Environ Int 36: 243-246.
        http://dx.doi.0rg/10.1016/i.envint.2009.12.003

     Astrin. KH; Bishop. DF; Wetmur. JG: Kaul. B; Davidow. B; Desnick. RJ. (1987). 5-aminolevulinic acid
        dehydratase isozymes and lead toxicity. Ann N Y Acad Sci 514: 23-29.
        http://dx.doi.org/10.1111/i. 1749-6632.1987.tb48757.x

     Aufderheide. AC: Wittmers. LE. Jr. (1992). Selected aspects of the spatial distribution of lead in bone
        [Review]. Neurotoxicology 13: 809-819.

     Aungst. BJ; Dolce. JA; Fung. H. (1981). The effect of dose on the disposition of lead in rats after
        intravenous and oral administration. Toxicol Appl Pharmacol 61: 48-57.
        http://dx.doi.org/10.1016/0041-008X(81)90006-5

     Aungst. BJ; Fung. H. (1985). The effects of dietary calcium on lead absorption, distribution, and
        elimination kinetics in rats. J Toxicol Environ Health 16: 147-159.
        http://dx.doi.org/10.1080/15287398509530726

     Avula. B; Wang. YH; Smillie. TJ; Duzgoren-Aydin. NS: Khan. IA. (2010). Quantitative determination of
        multiple elements in botanicals and dietary supplements using ICP-MS. J Agric Food Chem 58:
        8887-8894. http://dx.doi.org/10.1021/ifl01598g

     Azar. A; Trochimowicz. HJ; Maxfield. ME. (1973). Review of lead studies in animals carried out at
        Haskell Laboratory: Two year feeding study and response to hemorrhage study. In D Earth; A Berlin;
        R Engel; P Recht; J Smeets (Eds.), Environmental health aspects of lead: Proceedings of an
        international symposium (pp. 199-210). Luxembourg: Commission of the European Communities.

     Bailey. MR; Roy. M. (1994). Clearance of particles from the respiratory tract. In Human respiratory tract
        model for radiological protection: A report of a task group of the International Commission on
        Radiological Protection (pp.  301-413). Oxford, U.K.: Pergamon.

     Bandeen-Roche. K; Glass. TA; Bolla. KI; Todd. AC; Schwartz. BS. (2009). Cumulative lead dose and
        cognitive function in older adults. Epidemiology 20: 831-839.
        http://dx.doi.org/10.1097/EDE.Ob013e3181b5flOO

     Bannon. PI; Abounader. R; Lees. PSJ; Bressler. JP (2003). Effect of DMT1  knockdown on iron,
        cadmium, and lead uptake in Caco-2 cells. Am J Physiol Cell Physiol 284: C44-C50.

     Bannon. PI; Drexler. JW; Pent.  GM; Casteel. SW; Hunter.  PJ; Brattin. WJ; Major. MA. (2009).
        Evaluation of small arms range soils for metal contamination and lead bioavailability. Environ Sci
        Technol 43: 9071-9076. http://dx.doi.org/10.1021/es901834h

     Barbosa. F. Jr; Heloisa. M; Rodrigues. C; Buzalaf. MR; Krug. FJ; Gerlach. RF; Tanus-Santos. JE.
        (2006a). Evaluation of the use of salivary lead levels as a surrogate of blood lead or plasma lead levels
        in lead exposed subjects. Arch Toxicol 80: 633-637. http://dx.doi.org/10.1007/s00204-006-0096-y

     Barbosa. F. Jr; Ramires. I; Rodrigues. MHC; Saint' Pierre. TD; Curtius. AJ; Buzalaf. MR; Gerlach. RF;
        Tanus-Santos. JE. (2006b). Contrasting effects of age on the plasma/whole blood lead ratio in men and
        women with a history of lead exposure. Environ Res 102: 90-95.
        http://dx.doi.0rg/10.1016/i.envres.2006.03.007

     Barltrop. D; Meek. F. (1979). Effect of particle size on lead absorption from the gut. Arch Environ Occup
        Health 34:  280-285.

     Barrett. J; Taylor. K; Hudson-Edwards. K; Charnock. J.  (2010). Solid-phase speciation of Pb in urban
        road dust sediment: A XANES and EXAFS study. Environ Sci Technol 44: 2940-2946.
        http://dx.doi.org/10.1021/es903737k

     Barry. PSI. (1975). A comparison of concentrations of lead in human tissues. Occup Environ Med 32:
        119-139.

     Barry. PSI. (1981). Concentrations of lead in the tissues of children. Occup Environ Med 38: 61-71.


November 2012                             4-146                  Draft - Do Not Cite or Quote

-------
     Barry. PSI; Connolly. R. (1981). Lead concentrations in mediaeval bones. Int Arch Occup Environ Health
        48: 173-177. http://dx.doi.org/10.1007/BF00378438

     Barton. JC: Conrad. ME; Harrison. L; Nubv. S. (1978a). Effects of calcium on the absorption and
        retention of lead.  Transl Res 91: 366-376.

     Barton. JC: Conrad. ME; Nubv. S: Harrison. L. (1978b). Effects of iron on the absorption and retention of
        lead. Transl Res 92: 536-547.

     Beauchemin. S: Maclean. LC: Rasmussen. PE. (2011). Lead speciation in indoor dust: A case study to
        assess old paint contribution in a Canadian urban house. Environ Geochem Health 33: 343-352.
        http://dx.doi.org/10.1007/sl0653-011-9380-8

     Behinaein. S: Chettle. PR; Atanackovic. J: Egden. LM; Fleming. DE; Nie. LH; Richard. N; Stever. S.
        (2011). In vivo measurement of lead in the bones of smelter workers using the four-element 'clover-
        leaf geometry detector system. Phys Med  Biol 56: 653-665.
        http://dx.doi.Org/10.1088/0031-9155/56/3/008

     Bellinger. D; Hu. H;  Titlebaum. L; Needleman. HL. (1994a). Attentional correlates of dentin and bone
        lead levels in adolescents. Arch Environ Health 49: 98-105.
        http://dx.doi.org/10.1080/00039896.1994.9937461

     Bergdahl. IA; Grubb. A; Schiitz. A; Desnick. RJ; Wetmur. JG: Sassa. S: Skerfving. S. (1997a). Lead
        binding to delta-aminolevulinic acid dehydratase (ALAD) in human erythrocytes. Basic Clin
        Pharmacol Toxicol 81: 153-158. http://dx.doi.org/10.1111/i.1600-0773.1997.tb02061.x

     Bergdahl. IA; Schutz. A; Gerhardsson. L; Jensen. A; Skerfving. S. (1997b). Lead concentrations in human
        plasma, urine and whole blood. Scand J Work Environ Health 23: 359-363.

     Bergdahl. IA; Schutz. A; Grubb. A. (1996). Application of liquid chromatography-inductively coupled
        plasma mass spectrometry to the study of protein-bound lead in human erythrocytes. J Anal At
        Spectrom 11: 735-738. http://dx.doi.org/10.1039/JA9961100735

     Bergdahl. IA; Sheveleva. M; Schutz. A; Artamonova. VG: Skerfving. S. (1998). Plasma and blood lead in
        humans: Capacity-limited binding to delta-aminolevulinic acid dehydratase and other lead-binding
        components. Toxicol Sci 46: 247-253. http://dx.doi.Org/10.1093/toxsci/46.2.247

     Bergdahl. IA; Vahter. M; Counter. SA; Schutz. A; Buchanan. LH; Ortega. F; Laurell. G: Skerfving. S.
        (1999). Lead in plasma and whole blood from lead-exposed children. Environ Res 80: 25-33.
        http://dx.doi.org/10.1006/enrs.1998.3880

     Berger. OG: Gregg. DJ: Succop. PA. (1990). Using unstimulated urinary lead excretion to assess the need
        for chelation in the treatment of lead poisoning. J Pediatr 116: 46-51.
        http://dx.doi.org/10.1016/S0022-3476(05)81643-9

     Berkowitz. GS: Wolff. MS: Lapinski. RH; Todd. AC. (2004). Prospective study of blood and tibia lead in
        women undergoing surgical menopause. Environ Health Perspect 112: 1673-1678.

     Besser. JM; Brumbaugh. WG: Ivev. CD; Ingersoll.  CG: Moran. PW. (2008). Biological and chemical
        characterization of metal bioavailability in sediments from Lake Roosevelt, Columbia River,
        Washington, USA. Arch Environ Contam Toxicol 54: 557-570.
        http://dx.doi.org/10.1007/s00244-007-9074-5

     Blake. KCH; Barbezat. GO: Mann. M. (1983). Effect of dietary constituents on the gastrointestinal
        absorption of 203Pb in man. Environ Res 30: 182-187.
        http://dx.doi.org/10.1016/0013-9351(83)90178-0

     Blake. KCH: Mann. M. (1983). Effect of calcium and phosphorus on the gastrointestinal absorption of
        203Pb in man. Environ Res 30: 188-194. http://dx.doi.org/10.1016/0013-9351(83)90179-2

     Bleecker. ML; Ford.  DP; Vaughan. CG: Walsh. KS: Lindgren. KN. (2007b). The association of lead
        exposure and motor performance mediated by cerebral white matter change. Neurotoxicology 28:
        318-323. http://dx.doi.0rg/10.1016/i.neuro.2006.04.008
November 2012                             4-147                   Draft - Do Not Cite or Quote

-------
     Bleecker. ML; Lindgren. KN; Ford. DP. (1997). Differential contribution of current and cumulative
        indices of lead dose to neuropsychological performance by age. Neurology 48: 639-645.

     Bolanowska. W: Piotrowski. J; Garczynski. H. (1967). Triethyllead in the biological material in cases of
        acute tetraethyllead poisoning. Arch Toxicol 22: 278-282. http://dx.doi.org/10.1007/BF00577718

     Bonnard. R; Mckone. TE. (2009). Integration of the predictions of two models with dose measurements in
        a case study of children exposed to the emissions from a lead smelter. Hum Ecol Risk Assess 15:
        1203-1226. http://dx.doi.org/10.1080/10807030903304849

     Booker. DV; Chamberlain. AC: Newton. D; Stott. ANB. (1969). Uptake of radioactive lead following
        inhalation and injection. Br J Radiol 42: 457-466. http://dx.doi.org/10.1259/0007-1285-42-498-457

     Bornschein. RL; Succop. P; Dietrich. KN; Clark. CS: Que Hee SHammond. PB. (1985). The influence of
        social and environmental factors on dust lead, hand lead, and blood lead levels in young children.
        Environ Res 38: 108-118. http://dx.doi.org/10.1016/0013-9351(85)90076-3

     Bos. AJJ; Van der Stap. CCA. H; Valkovic. V: Vis. RD; Verheul. H. (1985). Incorporation routes of
        elements into human hair: Implications for hair analysis used for monitoring. Sci Total Environ 42:
        157-169. http://dx.doi.org/10.1016/0048-9697(85)90015-4

     Bowers. TS: Beck. BD; Karam. HS. (1994). Assessing the relationship between environmental lead
        concentrations and adult blood lead levels. Risk Anal 14:  183-189.
        httrj://dx.doi.org/10.1111/i.l539-6924.1994.tb00043.x

     Brattin. W: Griffin. S. (2011). Evaluation of the contribution of lead in soil to lead in dust at superfund
        sites. Hum Ecol Risk Assess 17: 236-244.  http://dx.doi.org/10.1080/10807039.2011.538638

     Bres. EF; Voegel. JC: Barry. JC: Waddington. WG: Frank. RM. (1986). Feasibility study for the detection
        of lead substitution sites in the hydroxyapatite crystal structure using high-resolution electron
        microscopy (HREM) at optimum focus. Journal of Applied Crystallography 19: 168-173.
        http://dx.doi.org/10.1107/S0021889886089719

     Brodeur. J: Lacasse. Y; Talbot. D. (1983). Influence of removal from occupational lead exposure on blood
        and saliva lead concentrations. Toxicol Lett 19: 195-199.
        http://dx.doi.org/10.1016/0378-4274(83)90282-5

     Brodv. DJ: Pirkle. JL; Kramer. RA; Flegal. KM; Matte. TD; Gunter. EW: Paschal. DC. (1994). Blood
        lead levels in the US population: Phase 1 of the third National Health and Nutrition Examination
        Survey (NHANES III, 1988 to 1991). JAMA 272: 277-283.
        http://dx.doi.org/10.1001/iama.1994.03520040039038

     Bronner. F; Pansu. D: Stein. W. (1986). An analysis of intestinal calcium transport across the rat intestine
        [Review]. Am J Physiol 250: G561- G569.

     Brunekreef. B. (1984). The relationship between air lead and blood lead in children: A critical review
        [Review]. Sci Total Environ 38:  79-123. http://dx.doi.org/10.1016/0048-9697(84)90210-9

     Burke. JM; Rea. AW: Suggs. J: Williams. R; Xue. JP; Ozkavak. H. (2002). Ambient paniculate matter
        exposures: A comparison of sheds-PM exposure model predictions and estimations derived from
        measurements collected during NERL's RTF PM panel study. Epidemiology 13: S83.

     Butterweck. G: Schuler. C: Vezzu. G: Miiller. R; Marsh. J: Thrift.  S: Birchall. A. (2002). Experimental
        determination of the absorption rate of unattached radon progeny from  respiratory  tract to blood.
        Radiat Prot Dosimetry 102: 343- 348.

     Caffo. B; Chen. S: Stewart. W: Bolla. K; Yousem. D; Davatzikos. C:  Schwartz. BS. (2008). Are brain
        volumes based on magnetic resonance imaging mediators of the associations of cumulative lead dose
        with cognitive function? Am J Epidemiol 167: 429-437. http://dx.doi.org/10.1093/aje/kwm326

     Cai. QY; Mo. CH; Wu. QT; Zeng. QY; Katsoyiannis. A. (2007). Concentration and speciation of heavy
        metals in six different sewage sludge-composts. J Hazard Mater 147: 1063-1072.
        http://dx.doi.0rg/10.1016/i.ihazmat.2007.01.142
November 2012                             4-148                   Draft - Do Not Cite or Quote

-------
     Campbell. JR; Toribara. TY. (2001). Hair-root lead to screen for lead toxicity. J Trace Elem Exp Med 14:
        69-72. http://dx.doi.org/10.1002/1520-670X(2001)14:l<69::AID-JTRA1009>3.0.CO:2-A

     Caravanos. J: Weiss. AL; Jaeger. RJ. (2006b). An exterior and interior leaded dust deposition survey in
        New York City: Results of a 2-year study. Environ Res 100: 159-164.
        http://dx.doi.0rg/10.1016/i.envres.2005.08.005

     Carbone. R; Laforgia. N; Crollo. E; Mautone. A; lolascon. A. (1998). Maternal and neonatal lead
        exposure in southern Italy. Neonatology 73: 362-366.

     Carlisle. JC: Wade. MJ. (1992). Predicting blood lead concentrations from environmental concentrations.
        Regul Toxicol Pharmacol 16: 280-289. httrj://dx.doi.org/10.1016/0273-2300(92)90008-W

     Carr. E; Lee. M; Marin. K; Holder. C: Hover. M; Pedde. M; Cook. R; Touma. J. (2011). Development and
        evaluation of an air quality modeling approach to assess near-field impacts of lead emissions from
        piston-engine aircraft operating on leaded aviation gasoline. Atmos Environ 45: 5795-5804.
        http://dx.doi.0rg/10.1016/i.atmosenv.2011.07.017

     Carrizales. L; Razo. I; Tellez-Hernandez. JI; Torres-Nerio. R; Torres. A; Batres. LE; Cubillas. AC: Diaz-
        Barriga. F. (2006). Exposure to arsenic and lead of children living near a copper-smelter in San Luis
        Potosi, Mexico: Importance of soil contamination for exposure of children.  Environ Res 101: 1-10.
        http://dx.doi.0rg/10.1016/i.envres.2005.07.010

     Casteel. SW: Cowart. RP; Weis. CP; Henningsen.  GM;  Hoffman. E; Brattin. WJ: Guzman. RE;  Starost.
        MF; Payne. JT; Stockham. SL; Becker. SV: Drexler. JW: Turk. JR. (1997). Bioavailability of lead to
        juvenile swine  dosed with soil from the Smuggler Mountain NPL site of Aspen, Colorado. Toxicol Sci
        36: 177-187. http://dx.doi.org/10.1006/faat. 1997.2296

     Casteel. SW: Weis. CP: Henningsen. GM; Brattin. WL.  (2006). Estimation of relative bioavailability of
        lead in soil and soil-like materials using young swine.  Environ Health Perspect 114: 1162-1171.
        http://dx.doi.org/10.1289/ehp.8852

     CDC (Centers for Disease Control and Prevention). (2002). Childhood lead poisoning associated with
        tamarind candy and folk remedies-California, 1999-2000. MMWR Recomm Rep 51: 684-686.

     CDC (Centers for Disease Control and Prevention). (2011 a). Fourth national report on human exposure to
        environmental chemicals: Updated tables, February  2011. Atlanta, GA.
        http://www.cdc.gov/exposurereport/pdf/FourthReport.pdf

     Chamberlain. AC.  (1983). Effect of airborne lead on blood lead. Atmos Environ 17: 677-692.
        http://dx.doi.org/10.1016/0004-6981(83)90415-8

     Chamberlain. AC:  Clough. WS: Heard. MJ: Newton. D; Stott. ANB; Wells. AC. (1975). Uptake of lead by
        inhalation of motor exhaust. Proc Biol Sci 192: 77-110.

     Chamberlain. AC:  Heard. MJ: Little. P; Newton. D; Wells. AC: Wiffin. RD. (1978). Investigations into
        lead from motor vehicles. (AERE-R9198). Berkshire, England: Transportation and Road Research
        Laboratory.

     Chen. WP; Krage.  N; Wu. LS: Page. AL; Chang. AC. (2008a). Fertilizer applications and trace elements
        in vegetable production soils of California. Water Air Soil Pollut 190: 209-219.
        http://dx.doi.org/10.1007/sll270-007-9594-7

     Chen. WP: Krage.  N; Wu. LS: Pan. GX; Khosrivafard. M; Chang. AC. (2008b). Arsenic, cadmium, and
        lead in California cropland soils: Role of phosphate  and micronutrient fertilizers.  J Environ Qual 37:
        689-695. http://dx.doi.org/10.2134/iea2007.0444

     Cheng. Y; Schwartz. J: Sparrow. D: Aro. A; Weiss. ST;  Hu. H. (2001). Bone lead and blood lead levels in
        relation to baseline blood pressure and the prospective development of hypertension: The Normative
        Aging Study. Am J Epidemiol 153:  164-171. http://dx.doi.Org/10.1093/aje/153.2.164
November 2012                             4-149                   Draft - Do Not Cite or Quote

-------
     Chia. SE; Zhou. HJ; Yap. E; Tham. MT; Dong. NV; Hong Tu. NT; Chia. KS. (2006). Association of renal
        function and delta-aminolevulinic acid dehydratase polymorphism among Vietnamese and Singapore
        workers exposed to inorganic lead. Occup Environ Med 63: 180-186.
        http://dx.doi.org/10.1136/oem.2005.021154

     Chiang. WF: Yang. HJ: Lung. SCC: Huang. S: Chiu. CY: Liu. IL: Tsai. CL: Kuo. CY. (2008). A
        comparison of elementary schoolchildren's exposure to arsenic and lead. J Environ Sci Health C
        Environ Carcinog Ecotoxicol Rev 26: 237-255. http://dx.doi.org/10.1080/10590500802343958

     Christoffersson. JO; Ahlgren. L; Schwartz. A; Skerfving.  S: Mattsson. S. (1986). Decrease of skeletal lead
        levels in man after end of occupational exposure. Arch Environ Occup Health 41: 312-318.
        http://dx.doi.org/10.1080/00039896.1986.9936703

     Chuang. HY; Schwartz. J; Tsai. SY; M-LT. L; Wang. JD;  Hu. H. (2000). Vibration perception thresholds
        in workers with long term exposure to lead. Occup Environ Med 57: 588-594.

     Cimorelli. AJ; Perry. SG: Venkatram. A; Weil. JC: Paine.  R; Wilson. RB; Lee. RF; Peters. WD; Erode.
        RW. (2005). AERMOD: A dispersion model for industrial source applications. Part I: General model
        formulation and boundary layer characterization. J Appl Meteorol 44: 682-693.
        http://dx.doi.Org/10.1175/JAM2227.l

     Clark. HF; Brabander. DJ; Erdil. RM. (2006). Sources, sinks, and exposure pathways of lead in urban
        garden soil. J Environ Qual 35: 2066-2074. http://dx.doi.org/10.2134/ieq2005.0464

     Clayton. CA; Pellizzari. ED; Whitmore. RW: Perritt RL Quackenboss. JJ. (1999). National human
        exposure assessment survey (NHEXAS): Distributions and associations of lead, arsenic, and volatile
        organic compounds in EPA Region 5. J Expo Sci Environ Epidemiol 9: 381-392.
        http://dx.doi.org/10.1038/sj.jea.7500055

     Coon. S: Stark. A; Peterson. E; Gloi. A; Kortsha. G: Pounds. J: Chettle. D; Gorell. J. (2006). Whole-body
        lifetime occupational lead exposure and risk of Parkinson's disease. Environ Health Perspect 114:
        1872-1876. http://dx.doi.org/10.1289/ehp.9102

     Cornelis. C: Berghmans. P; van Sprundel. M; Van der Auwera. JC. (2006). Use of the IEUBK model for
        determination of exposure routes in view of site remediation. Hum Ecol Risk Assess 12: 963-982.
        http://dx.doi.org/10.1080/10807030600826979

     Costa de Almeida. GR; de Freitas Tavares. CF; de Souza. AM; Sampaio de Sousa. T; Rodrigues
        Funavama. CA; Barbosa. F. Jr; Tanus-Santos. JE; Gerlach. RF. (2010). Whole blood, serum, and saliva
        lead concentrations in 6- to 8-year-old children. Sci Total Environ 408: 1551-1556.
        http://dx.doi.0rg/10.1016/i.scitotenv.2009.12.034

     Costa de Almeida. GR: de Sousa Guerra. C: de Angelo Souza Leite. G: Antonio. RC: Barbosa. F; Tanus-
        Santos. JE; Gerlach. RF. (2011).  Lead contents in the surface enamel of primary and permanent teeth,
        whole blood, serum, and saliva of 6-  to 8-year-old children. Sci Total Environ 409: 1799-1805.
        http://dx.doi.0rg/10.1016/i.scitotenv.2011.01.004

     Costa de Almeida. GR: Pereira Saraiva Mda. C: Barbosa F. J..r; Krug. FJ; Cury. JA; Rosario de Sousa
        Mda. L; Rabelo Buzalaf. MA; Gerlach. RF. (2007). Lead contents in the surface enamel of deciduous
        teeth sampled in vivo from children in uncontaminated and in lead-contaminated areas. Environ Res
        104: 337-345. http://dx.doi.Org/10.1016/i.envres.2007.03.007

     Costa de Almeida. GR: Umbelino de Freitas. C: Barbosa. F. Jr; Tanus-Santos. JE; Gerlach. RF. (2009).
        Lead in saliva from lead-exposed and unexposed children. Sci Total Environ 407: 1547-1550.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.10.058

     Davies. BE: White. HM. (1981). ENVIRONMENTAL-POLLUTION BY WIND BLOWN LEAD MINE
        WASTE - A CASE-STUDY IN WALES, U.K.. Sci Total Environ 20: 57-74.
        http://dx.doi.org/10.1016/0048-9697(81)90036-X
November 2012                              4-150                  Draft - Do Not Cite or Quote

-------
     de Almeida. GRC: de Souza Guerra. C: Tanus-Santos. JE; Barbosa. F. Jr; Gerlach. RF. (2008). A plateau
        detected in lead accumulation in subsurface deciduous enamel from individuals exposed to lead may
        be useful to identify children and regions exposed to higher levels of lead. Environ Res 107: 264-270.
        http://dx.doi.0rg/10.1016/i.envres.2008.01.002

     Del Rio-Celestino. M; Font. R; Moreno-Roj as. R; De Haro-Bailon. A. (2006). Uptake of lead and zinc by
        wild plants growing on contaminated soils. Ind Crop Prod 24: 230-237.
        http://dx.doi.0rg/10.1016/i.indcrop.2006.06.013

     DeSilva. PE. (1981). Determination of lead in plasma and studies on its relationship to lead in
        erythrocytes. Br J Ind Med 38: 209-217.

     Diamond. GL. (1992). Review of default value for lead plasma-to-urine transfer coefficient (TPLUR) in
        the U.S. EPAuptake/biokinetic model.  (SRC TR-92-135). Syracuse, NY: Syracuse Research
        Corporation.

     Diamond. GL. (1988). Biological monitoring of urine for exposure to toxic metals. In TW Clarkson; L
        Friberg; GF Nordberg; PR Sager (Eds.), Biological monitoring of toxic metals (pp. 515-529). New
        York, NY: Plenum Press.

     Dietrich. KN; Berger. OG: Succop. PA. (1993 a). Lead exposure and the motor developmental status of
        urban six-year-old children in the Cincinnati Prospective Study. Pediatrics 91: 301-307.

     Dixon.  SL; Gaitens. JM; Jacobs. DE; Strauss. W: Nagaraia. J: Pivetz. T; Wilson. JW: Ashley. PJ. (2009).
        Exposure of US children to residential  dust lead, 1999-2004: II. The contribution of lead-contaminated
        dust to children's blood lead levels. Environ Health Perspect 117: 468-474.
        http://dx.doi.org/10.1289/ehp.11918

     Dorsev. CD: Lee. BK: Bolla. KI: Weaver. VM: Lee. SS: Lee. GS: Todd. AC: Shi. W: Schwartz. BS.
        (2006). Comparison of patella lead with blood lead and tibia lead and their associations with
        neurobehavioral test scores. J Occup Environ Med 48: 489-496.
        http://dx.doi.org/10.1097/01.iom.0000199678.86629.3b

     Drasch. G: Wanghofer. E; Roider. G. (1997). Are blood, urine, hair, and muscle valid biomonitors for the
        internal burden of men with the heavy metals mercury, lead and cadmium? An investigation on 150
        deceased. Trace Elem Electroly 14: 116-123.

     Drasch. GA; Bohm. J: Baur. C. (1987). Lead in human bones. Investigations on an occupationally non-
        exposed population in southern Bavaria (F.R.G.): I. Adults. Sci Total Environ 64: 303-315.
        http://dx.doi.org/10.1016/0048-9697(87)90252-X

     Drexler. JW: Brattin. WJ. (2007). An in vitro procedure for estimation of lead relative bioavailability:
        With validation. Hum Ecol Risk Assess 13: 383-401. http://dx.doi.org/10.1080/10807030701226350

     Duggan. MJ; Inskip. MJ; Rundle. SA; Moorcroft. JS. (1985). Lead in playground dust and on the hands
        of schoolchildren. Sci Total Environ 44: 65-79. http://dx.doi.org/10.1016/0048-9697(85)90051-8

     DuVal.  G: Fowler. BA. (1989). Preliminary purification and characterization studies of a low molecular
        weight, high affinity cytosolic lead-binding protein in rat brain. Biochem Biophys Res Commun 159:
        177-184. http://dx.doi.org/10.1016/0006-291X(89)92420-0

     Eaton. PL; Stacev. NH; Wong. KL; Klaassen. CD. (1980). Dose-response effects of various metal ions on
        rat liver metallothionein, glutathione, heme oxygenase, and cytochrome P-450. Toxicol Appl
        Pharmacol 55: 393-402.

     Edwards. M; Dudi. A. (2004). Role of chlorine and chloramine in corrosion of lead-bearing plumbing
        materials. J Am Water Works Assoc 96: 69-81.

     Egan. SK; SS-H. T; Pennington. JAT; Bolger. PM. (2002). US Food and Drug Administration's Total Diet
        Study: intake of nutritional and toxic elements, 1991-96. Food Addit Contam 19: 103-125.

     Egeghy. PP; Quackenboss. JJ: Catlin. S: Ryan. PB. (2005). Determinants of temporal variability in
        NHEXAS-Maryland environmental concentrations, exposures, and biomarkers. J Expo Sci Environ
        Epidemiol 15: 388-397.  http://dx.doi.org/10.1038/sj.jea.7500415
November 2012                             4-151                   Draft - Do Not Cite or Quote

-------
     Eisner. AD; Rosati. J: Wiener. R. (2010). Experimental and theoretical investigation of particle-laden
        airflow under a prosthetic mechanical foot in motion. Build Environ 45: 878-886.
        http://dx.doi.0rg/10.1016/i.buildenv.2009.09.007

     Elias. SM; Hashim. Z; Marian. ZM; Abdullah. AS: Hashim. JH. (2007). Relationship between blood lead
        concentration and nutritional status among Malay primary school children In Kuala Lumpur,
        Malaysia. Asia Pac J Public Health 19: 29-37. http://dx.doi.org/10.1177/101053950701900306

     Elmarsafawv. SF; Jain. NB; Schwartz. J; Sparrow. D; Nie. HL; Hu. H. (2006). Dietary calcium as a
        potential modifier of the relationship of lead burden to blood pressure. Epidemiology 17: 531-537.
        http://dx.doi.org/10.1097/01.ede.0000231285.86968.2b
     EPRI (Electric Power Research Institute). (1997). Results of the independent evaluation of ISCST3 and
        ISC-PRIME. (TR-2460026 WO3527-02). Palo Alto, CA.

     ESRI (Environmental Systems Research Institute Inc.). (2011). Updated Demographics Data.
        http://www.esri.com/data/esri data/demographic.html

     Esteban. E; Rubin. CH; Jones. RL; Noonan.  G. (1999). Hair and blood substrates for screening children
        for lead poisoning. Arch Environ Occup Health 54: 436-440.

     Ettinger. AS: Lamadrid-Figueroa. H; Tellez-Rojo. MM; Mercado-Garcia. A; Peterson. KE; Schwartz. J:
        Hu. H; Hernandez-Avila. M. (2009). Effect of calcium supplementation on blood lead levels in
        pregnancy: A randomized placebo-controlled trial. Environ Health Perspect 117: 26-31.
        http://dx.doi.org/10.1289/ehp.11868
     Ettinger. AS: Tellez-Rojo. MM; Amarasiriwardena. C: Bellinger. D; Peterson. K; Schwartz. J: Hu. H;
        Hernandez-Avila. M. (2004a). Effect of breast milk lead on infant blood lead levels at 1 month of age.
        Environ Health Perspect 112: 1381-1385.

     Ettinger. AS: Tellez-Rojo. MM; Amarasiriwardena. C: Gonzalez-Cossio. T; Peterson. KE; Aro. A; Hu. H;
        Hernandez-Avila. M. (2004b). Levels of lead in breast milk and their relation to maternal blood and
        bone lead levels at one month postpartum. Environ Health Perspect 112: 926-931.

     Ettinger. AS: Tellez-Rojo. MM; Amarasiriwardena. C: Peterson. KE: Schwartz. J: Aro. A; Hu. H;
        Hernandez-Avila. M. (2006). Influence of maternal bone lead burden and calcium intake on levels of
        lead in breast milk over the course of lactation. Am J Epidemiol  163: 48-56.
        http://dx.doi.org/10.1093/aie/kwj010
     FAA (U.S. Federal Aviation Administration). (2010). General aviation and part 135 activity surveys - CY
        2010. http://www.faa.gov/data research/aviation data statistics/general aviation/CY2010/

     Falta. T; Limbeck. A; Koellensperger. G: Hann. S. (2008). Bioaccessibility of selected trace metals in
        urban PM2.5 and PM10 samples: A model study. Anal Bioanal Chem 390: 1149-1157.
        http://dx.doi.org/10.1007/s00216-007-1762-5

     FDA (U.S. Food and Drug Administration). (2008). Total diet study.
        http://www.fda.gov/Food/FoodSafetv/FoodContaminantsAdulteration/TotalDietStudv/default.htm

     Fernandez. C: Labanowski. J: Cambier. P; Jongmans. AG: Van Port. F. (2007). Fate of airborne metal
        pollution in soils  as related to agricultural management. 1. Zn and Pb distributions in soil profiles. Eur
        J Soil Sci 58: 547-559. http://dx.doi.Org/10.llll/j.1365-2389.2006.00827.x

     Fernandez. C: Labanowski. J: Jongmans. T; Bermond. A; Cambier. P; Lamy. I; Van Port. F. (2010). Fate
        of airborne metal pollution in soils as related to agricultural management: 2. Assessing the role of
        biological activity in micro-scale Zn and  Pb distributions in A, B and C horizons. Eur J Soil Sci 61:
        514-524. http://dx.doi.0rg/10.llll/j.1365-2389.2010.01256.x

     Fernandez. C: Monna. F; Labanowski. J: Loubet. M; van Port. F. (2008). Anthropogenic lead distribution
        in soils under arable land and permanent grassland estimated by Pb isotopic compositions. Environ
        Pollut 156: 1083-1091. http://dx.doi.Org/10.1016/i.envpol.2008.04.014
November 2012                             4-152                   Draft - Do Not Cite or Quote

-------
     Flegal. AR; Smith. DR. (1995). Measurements of environmental lead contamination and human exposure.
        In GW Ware (Ed.), Reviews of environmental contamination and toxicology, continuation of residue
        reviews, v 143 (pp. 1-45). New York, NY: Springer.

     Fleming. DEB: Boulav. D: Richard. NS: Robin. JP: Gordon. CL: Webber. CE: Chettle. DR. (1997).
        Accumulated body burden and endogenous release of lead in employees of a lead smelter. Environ
        Health Perspect 105: 224-233.

     Forbes. GB; Reina. JC. (1972). Effect of age on gastrointestinal absorption (Fe, Sr, Pb) in the rat. J Nutr
        102: 647-652.

     Fosse. G: Wesenberg. GR; Tvinnereim HM Eide. R; Kristoffersen. O; Nag. OH; Wierzbicka. M; Banoczv.
        J; De Oliveira. AA; Srisopak. C: Zamudio. A. (1995). Lead in deciduous teeth from larger cities of
        some countries. Int J Environ Stud 47: 203-210.

     Fowler. BA. (1989). Biological roles of high affinity metal-binding proteins in mediating cell injury.
        Comments Toxicol 3: 27-46.

     Fowler. BA; DuVal. G. (1991). Effects of lead on the kidney: Roles of high-affinity lead-binding proteins.
        Environ Health Perspect 91: 77-80. http://dx.doi.org/10.1289/ehp.919177

     Franklin. CA; Inskip. MJ; Baccanale. CL; Edwards. CM; Manton.  WI; Edwards. E; O'Flahertv. EJ.
        (1997). Use of sequentially administered stable lead isotopes to investigate changes in blood lead
        during pregnancy in a nonhuman primate (Macaca fascicularis). Toxicol Sci 39: 109-119.

     Freeman. GB; Dill. JA; Johnson. JD; Kurtz. PJ; Parham. F; Matthews. HB. (1996). Comparative
        absorption of lead from contaminated soil and lead salts by weanling Fischer 344 rats. Toxicol Sci 33:
        109-119.

     Freeman. GB: Johnson. JD; Killinger. JM;  Liao. SC: Feder. PI; Davis. AO; Ruby. MV; Chanev. RL;
        Lovre. SC: Bergstrom. PP. (1992). Relative bioavailability of lead from mining waste soil in rats.
        Toxicol Sci 19: 388-398.

     Freeman. GB: Johnson. JD: Liao. SC: Feder. PI; Davis. AO: Ruby. MV: Schoof. RA; Chanev. RL;
        Bergstrom. PP. (1994). Absolute bioavailability of lead acetate and mining waste lead in rats.
        Toxicology 91: 151-163.

     Fukui. Y; Miki. M; Ukai. H; Okamoto. S: Takada. S: Higashikawa. K; Ikeda. M. (1999). Urinary lead as a
        possible surrogate of blood lead  among workers occupationally exposed to lead. Int Arch Occup
        Environ Health 72: 516-520. http://dx.doi.org/10.1007/s004200050409

     Fullmer. CS:  Rosen. JF. (1990). Effect of dietary calcium and lead  status on intestinal calcium absorption.
        Environ Res 51: 91-99.

     Gaitens. JM;  Pixon. SL; Jacobs. PE; Nagaraia. J: Strauss. W: Wilson. JW: Ashley. PJ. (2009). Exposure
        of US children to residential dust lead,  1999-2004:1. Housing and demographic factors. Environ
        Health Perspect 117: 461-467. http://dx.doi.org/10.1289/ehp.11917

     Garavan. C: Breen. J: Moles. R; ORegan. B. (2008). A case study of the health impacts in an abandoned
        lead mining area, using children's blood lead levels. International Journal of Mining, Reclamation and
        Environment 22: 265-284. http://dx.doi.org/10.1080/17480930802109885

     Garrido Latorre. F; Hernandez-Avila. M; Orozco. JT; Medina. CAA; Aro. A; Palazuelos. E; Hu. H.
        (2003). Relationship of blood and bone lead to menopause and  bone mineral density among middle-
        age women in Mexico City. Environ Health Perspect 111: 631-636.

     Georgopoulos. PG: Lioy. PJ. (2006). From theoretical aspects of human exposure and dose assessment to
        computational model implementation: the Modeling ENvironment for TOtal Risk studies (MENTOR).
        J Toxicol Environ Health B Crit  Rev 9:  457-483.

     Gerhardsson. L; Attewell. R: Chettle. PR; Englyst. V: Lundstrom.  NG: Nordberg. GF; Nyhlin. H; Scott.
        MC: Todd. AC. (1993). In vivo measurements of lead in bone in long-term exposed lead smelter
        workers. Arch Environ  Occup Health 48: 147-156.
November 2012                              4-153                  Draft - Do Not Cite or Quote

-------
     Gerhardsson. L; Brune. D; Nordberg. GF; Wester. PO. (1986). Distribution of cadmium, lead and zinc in
        lung, liver and kidney in long-term exposed smelter workers. Sci Total Environ 50: 65-85.

     Gerhardsson. L; Englyst. V: Lundstrom. NG: Nordberg. G: Sandberg. S: Steinvall. F. (1995). Lead in
        tissues of deceased lead smelter worker. J Trace Elem Med Biol 9: 136-143.

     Gerke. TL; Scheckel. KG: Schock. MR. (2009). Identification and Distribution of Vanadinite
        (Pb-5(V5+O4)3Cl) in Lead Pipe Corrosion By-Products. Environ Sci Technol 43: 4412-4418.
        http://dx.doi.org/10.1021/es900501t

     Ghosh. L; Adhikari. S: Ayyappan. S. (2007). Assessment of toxic interactions of heavy metals and their
        effects on accumulation in tissues of freshwater fish. Research Journal of Environmental Toxicology
        1: 37-44. http://dx.doi.org/10.3923/riet.2007.37.44

     Glass. TA; Bandeen-Roche. K; McAtee. M; Bolla. K; Todd. AC: Schwartz. BS. (2009). Neighborhood
        psychosocial hazards and the association of cumulative lead dose with cognitive function in older
        adults. Am J Epidemiol 169: 683-692. http://dx.doi.org/10.1093/aie/kwn390

     Glenn. BS: Bandeen-Roche. K; Lee. BK; Weaver. VM; Todd. AC: Schwartz. BS. (2006). Changes in
        systolic blood pressure associated with lead in blood and bone. Epidemiology 17: 538-544.
        http://dx.doi.org/10.1097/01.ede.0000231284.19078.4b

     Glenn. BS: Stewart. WF; Links. JM; Todd. AC: Schwartz. BS. (2003). The longitudinal association of
        lead with blood pressure. Epidemiology 14: 30-36.

     Godoi. Q: Santos. D; Nunes. LC: Leme. FO; Rufini. IA; Agnelli. JAM; Trevizan. LC: Krug. FJ. (2009).
        Preliminary studies of laser-induced breakdown spectrometry for the determination of Ba, Cd, Cr and
        Pb in toys. Spectrochimica Acta Part B: Atomic Spectroscopy 64: 573-581.
        http://dx.doi.0rg/10.1016/i.sab.2009.05.003

     Gover. RA. (1990). Transplacental transport of lead. Environ Health Perspect 89: 101-105.

     Granev. JR; Hallidav. AN; Keeler. GJ: Nriagu. JO: Robbins. JA; Norton. SA. (1995). Isotopic record of
        lead pollution in lake sediments from the northeastern United States. Geochim Cosmo Act 59:
        1715-1728.

     Graziano. JH; Popovac. D: Factor-Litvak. P; Shrout P; Kline. J: Murphy. MJ; Zhao. YH; Mehmeti. A;
        Ahmedi. X; Raiovic. B; Zvicer. Z; Nenezic. DU: Lolacono. NJ; Stein. Z. (1990).  Determinants of
        elevated blood lead during pregnancy in a population surrounding a lead smelter  in Kosovo,
        Yugoslavia. Environ Health Perspect 89: 95-100. http://dx.doi.org/10.1289/ehp.908995

     Greenwav. JA;  Gerstenberger. S. (2010). An evaluation of lead contamination in plastic toys collected
        from day care centers in the Las Vegas Valley, Nevada, USA. Bull Environ Contam Toxicol 85:
        363-366. http://dx.doi.org/10.1007/s00128-010-0100-3

     Griffin. TB; Coulston. F; Wills. H; Russell. JC. (1975). Clinical studies on men continuously exposed to
        airborne paniculate lead. In TB Griffin; JH Knelson (Eds.), Lead (pp. 221-240). Stuttgart, Germany:
        Georg Thieme Publishers.

     Gross. SB; Pfitzer. EA; Yeager. DW: Kehoe. RA. (1975). Lead in human tissues. Toxicol Appl Pharmacol
        32: 638-651.

     Gulson. B; Jameson. CW:  Mahaffev. KR; Mizon. KJ; Patison. N; Law. AJ; Korsch. MJ: Salter. MA.
        (1998a).  Relationships  of lead in breast milk to lead in blood, urine, and diet of the infant and mother.
        Environ Health Perspect 106: 667-674. http://dx.doi.org/10.1289/ehp.98106667

     Gulson. B; Korsch. M; Matisons. M; Douglas. C: Gillam. L; McLaughlin. V (2009). Windblown lead
        carbonate as the main source of lead in blood of children from a seaside community: An example of
        local birds as "canaries in the mine". Environ Health Perspect 117: 148-154.
        http://dx.doi.org/10.1289/ehp.11577

     Gulson. B; Mahaffev. KR; Jameson. CW: Mizon. KJ: Korsch. MJ: Cameron. MA; Eisman. JA. (1998b).
        Mobilization of lead from the skeleton during the postnatal period is larger than during pregnancy.
        TranslRes 131: 324-329.
November 2012                             4-154                  Draft - Do Not Cite or Quote

-------
     Gulson. B; Mahaffev. KR: Jameson. CW: Patison. N; Law. AJ; Mizon. KJ; Korsch. MJ; Pederson. D.
        (1999). Impact of diet on lead in blood and urine in female adults and relevance to mobilization of
        lead from bone stores. Environ Health Perspect 107: 257-263. http://dx.doi.org/10.1289/ehp.99107257

     Gulson. B; Mahaffev. KR; Mizon. KJ: Korsch. MJ: Cameron. MA; Vimpani. G. (1995a). Contribution of
        tissue lead to blood lead in adult female subjects based on stable lead isotope methods. Transl Res
        125: 703-712.

     Gulson. B; Mizon. K; Korsch. M; Taylor. A. (2006a). Changes in the lead isotopic composition of blood,
        diet and air in Australia over a decade: Globalization and implications  for future isotopic studies.
        Environ Res 100: 130-138. http://dx.doi.Org/10.1016/j.envres.2005.03.006

     Gulson. B; Mizon. K; Smith. H; Eisman. J; Palmer. J; Korsch. M; Donnelly. J: Waite. K. (2002). Skeletal
        lead release during bone resorption: Effect of bisphosphonate treatment in a pilot study. Environ
        Health Perspect 110: 1017-1023.

     Gulson. B; Mizon. K; Taylor. A; Korsch. M; Stauber. J: Davis. JM; Louie. H; Wu. M; Antin. L. (2008).
        Longitudinal monitoring of selected elements in blood of healthy young children. J Trace Elem Med
        Biol 22: 206-214. http://dx.doi.Org/10.1016/i.itemb.2008.04.001

     Gulson. B; Mizon. KJ: Korsch. MJ: Palmer. JM; Donnelly. JB. (2003). Mobilization of lead from human
        bone tissue during pregnancy and lactation: A summary of long-term research. Sci Total Environ 303:
        79-104. http://dx.doi.org/10.1016/S0048-9697(02)00355-8

     Gulson. B; Mizon. KJ: Korsch. MJ: Taylor. AJ. (2006b). Low blood lead levels do not appear to be
        further reduced by dietary supplements. Environ Health Perspect 114:  1186-1192.
        http://dx.doi.org/10.1289/ehp.8605

     Gulson. B; Mizon. KJ: Palmer. JM: Korsch. MJ: Taylor. AJ:  Mahaffev. KR. (2004a). Blood lead changes
        during pregnancy and postpartum with calcium supplementation. Environ Health Perspect  112:
        1499-1507. http://dx.doi.org/10.1289/ehp.6548

     Gulson. B; Wilson. D. (1994). History of lead exposure in children revealed from isotopic analyses of
        teeth. Arch Environ Occup Health 49: 279-283. http://dx.doi.org/10.1080/00039896.1994.9937480

     Gulson. BL; Davis. JJ: Mizon. KJ: Korsch. MJ: Bawden-Smith. J. (1995b). Sources of lead in soil and
        dust and the use of dust fallout as a sampling medium. Sci Total Environ 166: 245-262.
        http://dx.doi.org/10.1016/0048-9697(95)04505-U

     Gulson. BL; Jameson. CW: Mahaffev. KR: Mizon. KJ: Korsch. MJ: Vimpani. G. (1997). Pregnancy
        increases mobilization of lead from maternal skeleton. Transl Res 130: 51-62.
        http://dx.doi.org/10.1016/S0022-2143(97)90058-5

     Gulson. BL; Mizon. KJ: Davis. JD; Palmer. JM: Vimpani. G. (2004b). Identification of sources of lead in
        children in a primary zinc-lead smelter environment. Environ Health Perspect 112: 52-60.
        http://dx.doi.org/10.1289/ehp.6465

     Gultepe. I: Pearson. G: Milbrandt. JA; Hansen. B; Platnick. S: Taylor. P; Gordon. M; Oakley. JP; Cober.
        SG. (2009). The fog remote sensing and modeling field project. Bull Am Meteorol Soc 90: 341-359.
        httrj://dx.doi.org/10.1175/2008bams2354.1

     Haley. VB; Talbot TO. (2004). Seasonality and trend in blood lead levels  of New York State children.
        BMC Pediatr 4:8. http://dx.doi.org/10.1186/1471-2431-4-8

     Hanninen. H; Aitio. A; Kovala. T; Luukkonen. R; Matikainen. E; Mannelin. T; Erkkila. J: Riihimaki. V.
        (1998). Occupational exposure to lead and neuropsychological dysfunction. Occup Environ Med 55:
        202-209.

     Hansen.  S: Nieboer. E; Sandanger. TM; Wilsgaard. T; Thomassen. Y; Veyhe. AS: Odland. JO.  (20lib).
        Changes in maternal blood concentrations of selected essential and toxic elements during and after
        pregnancy. J Environ Monit 13: 2143-2152. http://dx.doi.org/10.1039/cleml0051c

     Havlena. J: Kanarek. MS: Coons. M. (2009).  Factors associated with the seasonality of blood lead levels
        among preschool Wisconsin children. WMJ 108: 151-155.
November 2012                             4-155                   Draft - Do Not Cite or Quote

-------
     Haves. EB; McElvaine. MD; Orbach. HG: Fernandez. AM; Lyne. S: Matte. TD. (1994). Long-term trends
        in blood lead levels among children in Chicago: Relationship to air lead levels. Pediatrics 93: 195-200.

     Healev. N; Chettle. PR; Mcneill. FE; Fleming. DE. (2008). Uncertainties in the relationship between tibia
        lead and cumulative blood lead index [Letter]. Environ Health Perspect 116: A109.
        http://dx.doi.org/10.1289/ehp. 10778

     Healv. MA; Harrison. PG: Aslam. M; Davis. SS: Wilson. CG. (1992). Lead sulphide and traditional
        preparations: Routes for ingestion, and solubility and reactions in gastric fluid. J Clin Hosp Pharm 7:
        169-173. http://dx.doi.0rg/10.llll/j.1365-2710.1982.tb01019.x

     Heard. MJ; Chamberlain. AC. (1982). Effect of minerals and food on uptake of lead from the
        gastrointestinal tract in humans. Hum Exp Toxicol 1: 411-415.
        http://dx.doi.org/10.1177/096032718200100407

     Heard. MJ: Wells. AC: Newton. D; Chamberlain. AC. (1979). Human uptake and metabolism of terra
        ethyl and tetra methyl lead vapour labelled with 203Pb. In International Conference on Management
        and Control of Heavy Metals in the Environment, London, England, September. Edinburgh, United
        Kingdom: CEP Consultants, Ltd.

     Hepp. NM; Mindak. WR; Cheng. J. (2009). Determination of total lead in lipstick: Development and
        validation of a microwave-assisted digestion, inductively coupled plasma-mass spectrometric method.
        J Cosmet Sci 60: 405-414. http://dx.doi.Org/10.llll/i.1468-2494.2010.00577 2.x

     Hernandez-Avila. M; Gonzalez-Cossio. T; Palazuelos. E; Romieu. I; Aro. A; Fishbein. E; Peterson. KE;
        Hu. H.  (1996). Dietary and environmental determinants of blood and bone lead levels in lactating
        postpartum women living in Mexico City. Environ Health Perspect 104: 1076-1082.
        http://dx.doi.org/10.1289/ehp.961041076

     Hernandez-Avila. M; Smith. D; Meneses. F; Sanin. LH; Hu. H. (1998). The influence of bone and blood
        lead on plasma lead levels in environmentally exposed adults. Environ Health Perspect 106: 473-477.
        http://dx.doi.org/10.1289/ehp.98106473

     Hernberg. S: Nikkanen. J: Mellin. G: Lilius. H. (1970). Delta-aminolevulinic acid dehydrase as a measure
        of lead  exposure. Arch Environ Occup Health 21: 140-145.

     Hertz-Picciotto. I; Schramm. M; Watt-Morse. M; Chantala. K; Anderson. J: Osterloh. J.  (2000). Patterns
        and determinants of blood lead during pregnancy. Am J Epidemiol 152: 829-837.
        http://dx.doi.0rg/10.1093/aie/152.9.829

     Hicsonmez. U; Erees. FS: Ozdemir. C: Ozdemir. A; Cam. S. (2009). Determination of major and minor
        elements in the Malva sylvestris L. from Turkey using ICP-OES techniques. Biol Trace Elem Res 128:
        248-257. http://dx.doi.org/10.1007/sl2011-008-8270-0

     Hilts. SR. (1996). A co-operative approach to risk management in an active lead/zinc smelter community.
        Environ Geochem Health 18: 17-24. http://dx.doi.org/10.1007/BF01757215

     Hilts. SR. (2003). Effect of smelter emission reductions on children's blood lead levels. Sci Total Environ
        303: 51-58. http://dx.doi.org/10.1016/S0048-9697(02)00357-l

     Hilts. SR: Bock. SE; Oke. TL; Yates. CL; Copes. RA. (1998). Effect of interventions on children's blood
        lead levels. Environ Health Perspect 106: 79-83. http://dx.doi.org/10.1289/ehp.9810679

     Hopkins. MR; Ettinger. AS: Hernandez-Avila. M; Schwartz. J: Tellez-Rojo. MM; Lamadrid-Figueroa. H;
        Bellinger. D; Hu. H; Wright. RO. (2008). Variants in iron metabolism genes predict higher blood lead
        levels in young children. Environ Health Perspect 116: 1261-1266.
        http://dx.doi.org/10.1289/ehp.11233

     Hsieh. TJ:  Chen. YC: Li.  CW: Liu. GC: Chiu. YW: Chuang. HY. (2009b). A proton magnetic resonance
        spectroscopy study of the chronic lead effect  on the basal ganglion and frontal and occipital lobes in
        middle-age adults. Environ Health Perspect 117: 941-945. http://dx.doi.org/10.1289/ehp.0800187
November 2012                             4-156                   Draft - Do Not Cite or Quote

-------
     Hu. H; Aro. A; Pavton. M; Korrick. S: Sparrow. (1996a). The relationship of bone and blood lead to
        hypertension: The Normative Aging Study. JAMA 275: 1171-1176.
        http://dx.doi.org/10.1001/iama.1996.03530390037031

     Hu. H; Aro. A; Rotnitzky. A. (1995). Bone lead measured by X-ray fluorescence: Epidemiologic methods.
        Environ Health Perspect 103: 105-110. http://dx.doi.org/10.1289/ehp.95103sll05

     Hu. H; Rabinowitz. M; Smith. D. (1998). Bone lead as a biological marker in epidemiologic studies of
        chronic toxicity: Conceptual paradigms [Review]. Environ Health Perspect 106: 1-8.
        http://dx.doi.org/10.1289/ehp.981061

     Hu. H; Shih. R; Rothenberg. S: Schwartz. BS. (2007a). The epidemiology of lead toxicity in adults:
        Measuring dose and consideration of other methodologic issues [Review]. Environ Health Perspect
        115: 455-462. http://dx.doi.org/10.1289/ehrj.9783

     Hunt. A; Johnson. PL.  (2012). Suspension and resuspension of dry soil indoors following track-in on
        footwear. Environ Geochem  Health 34: 355-363. http://dx.doi.org/10.1007/sl0653-011-9400-8

     Hunt. A; Johnson. PL;  Brooks. J: Griffith. DA. (2008). Risk remaining from fine particle contaminants
        after vacuum cleaning of hard floor surfaces. Environ Geochem Health 30: 597-611.
        http://dx.doi.org/10.1007/sl0653-008-9183-8

     Hunt. WG: Burnham. W: Parish. CN; Burnham. KK; Mutch. B; Oaks. JL. (2006). Bullet fragments in
        deer remains: Implications for lead exposure in avian scavengers. Wildlife Society Bulletin 34:
        167-170. http://dx.doi.org/10.2193/0091-7648(2006)34ri67:BFIDPJ12.0.CO:2

     Hunt. WG: Watson. RT: Oaks. JL: Parish. CN: Burnham. KK: Tucker. RL: Belthoff. JR: Hart. G. (2009).
        Lead bullet fragments in venison from rifle-killed deer: Potential for human dietary exposure. PLoS
        ONE 4: e5330. http://dx.doi.org/10.1371/iournal.pone.0005330

     Hursh. JB; Schraub. A; Sattler. EL; Hofmann. HP. (1969). Fate of 212Pb inhaled by human subjects.
        Health Phys 16: 257-267.

     Hursh. JB; Suomela. J.  (1968). Absorption of 212Pb from the gastrointestinal tract of man. Acta Radiol 7:
        108-120. http://dx.doi.org/10.3109/02841866809133184

     ICRP (International Commission on Radiological Protection). (1973). Alkaline earth metabolism in adult
        man. Oxford, U.K.:  Pergamon Press.

     ICRP (International Commission on Radiological Protection). (1993). Age-specific biokinetic models for
        the alkaline earth elements and lead. In Age-dependent doses to members of the public from intake of
        radionuclides: Part 2: Ingestion dose coefficients (pp. 95-120). Elsevier: New York.

     ICRP (International Commission on Radiological Protection). (1994). Human respiratory tract model for
        radiological protection: A report of a task group of the International Commission on Radiological
        Protection. ICRP Publication 66 [Review]. Ann ICRP 24:  1-482.

     Inskip. MJ; Franklin. CA; Baccanale. CL; Manton. WI; O'Flahertv. EJ; Edwards. CMH; Blenkinsop. JB:
        Edwards. EB. (1996). Measurement of the flux of lead from bone to blood in a nonhuman primate
        (Macaca fascicularis) by sequential administration of stable lead isotopes. Toxicol Sci 33: 235-245.
        http://dx.doi.org/10.1006/faat.1996.0161

     Jackson. LW: Cromer. BA; Panneerselvam. A. (2010). Association between bone turnover, micronutrient
        intake and blood lead levels among pre- and post-menopausal women, NHANES 1999-2002. Environ
        Health Perspect 118: 1590-1596. http://dx.doi.org/10.1289/ehp.1002158

     Jaffe. E; Volin. M; Bronson-Mullins. C: Dunbrack. R; Kervinen.  J: Martins. J:  Quinlan. J: Sazinsky. M;
        Steinhouse. E; Yeung. A. (2000). An artificial gene for human porphobilinogen synthase allows
        comparison of an allelic variation implicated in susceptibility to lead poisoning. J Biol Chem 275:
        2619- 2626. http://dx.doi.Org/10.1074/ibc.275.4.2619

     Jain. NB; Potula. V: Schwartz. J: Vokonas. PS: Sparrow. D; Wright. RO: Nie. H; Hu. H. (2007). Lead
        levels and ischemic heart disease in a prospective study of middle-aged and elderly  men: The VA
        Normative Aging Study. Environ Health Perspect 115: 871-875. http://dx.doi.org/10.1289/ehp.9629
November 2012                             4-157                   Draft - Do Not Cite or Quote

-------
     James. HM; Hilburn. ME; Blair. JA. (1985). Effects of meals and meal times on uptake of lead from the
        gastrointestinal tract of humans. Hum Exp Toxicol 4: 401-407.

     Jin. C: Li. Y; Li. YL; Zou. Y; Zhang. GL; Normura. M; Zhu. GY. (2008). Blood lead: Its effect on trace
        element levels and iron structure in hemoglobin. Nucl Instrum Methods Phys Res B 266: 3607-3613.
        http://dx.doi.0rg/10.1016/i.nimb.2008.05.087

     Jin. CW: Zhang. SJ: He. YF; Zhou. GD; Zhou. ZX. (2005). Lead contamination in tea garden soils and
        factors affecting its bioavailability. Chemosphere 59: 1151-1159.
        http://dx.doi.0rg/10.1016/i.chemosphere.2004.ll.058

     Johansen. P; Pedersen. HS: Asmund. G: Riget. F. (2006). Lead shot from hunting as a source of lead in
        human blood. Environ Pollut 142: 93-97. http://dx.doi.Org/10.1016/i.envpol.2005.09.015

     Johnson. D; Bretsch. J. (2002). Soil lead and children's blood lead levels in Syracuse, NY, USA. Environ
        Geochem Health 24: 375-385. http://dx.doi.Org/10.1023/A:1020500504167

     Johnson. PL; McDade. K; Griffith. D. (1996). Seasonal variation in paediatric blood lead levels in
        Syracuse, NY, USA. Environ Geochem Health 18: 81-88. http://dx.doi.org/10.1007/BF01771136

     Jones. RL; Homa. DM; Meyer. PA; Brodv. DJ: Caldwell. KL; Pirkle. JL; Brown. MJ. (2009a). Trends in
        blood lead levels and blood lead testing  among US children aged 1 to 5 Years, 1988-2004. Pediatrics
        123: e376-e385. http://dx.doi.org/10.1542/peds.2007-3608

     Juhasz. AL; Weber. J; Smith. E. (2011). Impact of soil particle size and bioaccessibility on children and
        adult lead exposure in peri-urban contaminated soil. J Hazard Mater 186: 1870-1879.
        http://dx.doi.0rg/10.1016/i.jhazmat.2010.12.095

     Juhasz. AL; Weber. J: Smith. E; Naidu. R; Marschner. B; Rees. M; Rofe. A; Kuchel. T; Sansom. L.
        (2009). Evaluation of SBRC-gastric and SBRC-intestinal methods for the prediction of in vivo relative
        lead bioavailability in contaminated soils. Environ Sci Technol 43: 4503-4509.
        http://dx.doi.org/10.1021/es803238u

     Kamel. F; Umbach. DM; Hu. H; Munsat. TL;  Shefner. JM; Taylor. JA; Sandier. DP. (2005). Lead
        exposure as a risk factor for amyotrophic lateral sclerosis. Neurodegener Dis 2: 195-201.
        http://dx.doi.org/10.1159/000089625

     Kamel. F; Umbach. DM; Munsat. TL; Shefner. JM; Hu. H; Sandier. DP. (2002). Lead exposure and
        amyotrophic lateral sclerosis. Epidemiology 13: 311-319.

     Kamel. F; Umbach. DM; Stallone. L; Richards. M; Hu. H; Sandier. DP. (2008). Association of lead
        exposure with survival in amyotrophic lateral sclerosis. Environ Health Perspect 116: 943-947.
        http://dx.doi.org/10.1289/ehp.11193

     Kang. HG: Jeong. SH; Cho. MR; Cho. JH; Bischoff. K. (2009). Time-dependent changes in lead and
        delta-aminolevulinic acid after  subchronic lead exposure in rats. Hum Exp Toxicol 28: 647-654.
        http://dx.doi.org/10.1177/0960327109107046

     Kehoe. RA. (196la). The Harben Lectures,  1960: The metabolism of lead in man in health and disease. 1.
        The normal metabolism of lead. J R Inst Public Health 24: 81-97.

     Kehoe. RA. (1961b). The Harben Lectures,  1960: The metabolism of lead in man in health and disease.
        2(2). The metabolism of lead under abnormal conditions. J R Inst Public Health 24:  129-143.

     Kehoe. RA. (1961c). The Harben Lectures,  1960: The metabolism of lead in man in health and disease. 3.
        Present hygienic problems relating to the absorption of lead. J R Inst Public Health 24: 177-203.

     Kehoe. RA. (1987). Studies of lead administration and elimination in adult volunteers under natural and
        experimentally induced conditions over extended periods of time. Food Chem Toxicol 25: 425-493.

     Kemp. FW: Neti. PVS. V: Howell. RW: Wenger. P; Louria. DB; Bogden. JD. (2007). Elevated blood lead
        concentrations and vitamin D deficiency in winter and summer in young urban children. Environ
        Health Perspect 115: 630-635. http://dx.doi.org/10.1289/ehp.9389
November 2012                             4-158                  Draft - Do Not Cite or Quote

-------
     Kessler. M; Durand. PY; Huu. TC: Royer-Morot MJ; Chanliau. J: Netter. P; Due. M. (1999).
        Mobilization of lead from bone in end-stage renal failure patients with secondary
        hyperparathyroidism. Nephrol Dial Transplant 14: 2731-2733.
        http://dx.doi.org/10.1093/ndt/14.ll.2731

     Khalil. N; Morrow. LA; Needleman. H; Talbott. EO; Wilson. JW: Caulev. JA. (2009a). Association of
        cumulative lead and neurocognitive function in an occupational cohort. Neuropsychology 23: 10-19.
        http://dx.doi.org/10.1037/a0013757

     Khoder. MI; Hassan. SK; El-Abssawv. AA. (2010). An evaluation of loading rate of dust, Pb, Cd, and Ni
        and metals mass concentration in the settled surface dust in domestic houses and factors affecting
        them. Indoor Built Environ  19: 391-399. http://dx.doi.org/10.1177/1420326X10367284

     Khourv. GA; Diamond. GL.  (2003). Risks to children from exposure to lead in air during remedial or
        removal activities at Superfund sites: a case study of the RSR lead smelter superfund site. J Expo Sci
        Environ Epidemiol 13: 51-65. http://dx.doi.org/10.1038/sj.jea.7500254

     Kim. EJ; Herrera. JE.  (2010). Characteristics of lead corrosion scales formed during drinking water
        distribution and their potential influence on the release of lead and other contaminants. Environ Sci
        Technol 44: 6054-6061. http://dx.doi.org/10.1021/esl01328u

     Kim. HK; Yoon. EK; Jang. J: Hwang. M; Kim. J: Ha. JH; Jang. DP; Yoo. TM; Park. KL. (2009a).
        Assessment of heavy metal  exposure via the intake of oriental medicines in Korea. J Toxicol Environ
        HealthA72:  1336-1342.  http://dx.doi.org/10.1080/15287390903212485

     Kim. HS: Lee. SS: Lee. GS:  Hwangbo. Y; Ahn. KD; Lee. BK. (2004). The protective effect of delta-
        aminolevulinic acid dehydratase 1-2 and 2-2 isozymes against blood lead with higher hematologic
        parameters. Environ Health Perspect 112: 538-541. http://dx.doi.org/10.1289/ehp.6464

     Kim. R; Aro. A;  Rotnitzky. A; Amarasiriwardena.  C: Hu. H. (1995). K x-ray fluorescence measurements
        of bone lead concentration:  the analysis of low-level data. Phys Med Biol 40: 1475-1485.
        http://dx.doi.Org/10.1088/0031-9155/40/9/007

     Kim. R; Landrigan. C: Mossmann. P; Sparrow. D; Hu. H. (1997). Age and secular trends in bone lead
        levels in middle-aged and elderly men: Three-year longitudinal follow-up in the Normative Aging
        Study. Am J Epidemiol 146: 586-591.

     Kissel. JC: Richter. KY; Fenske. RA. (1996). Field measurement of dermal soil loading attributable to
        various activities: implications for exposure assessment. Risk Anal 16:  115-125.

     Klotzback. JM; Follansbee. MH; Diamond. GL. (2003). Evaluation of the ICRP lead biokinetics model:
        Empirical comparisons with observations of plasma-blood lead concentration relationships in humans
        (draft final). Washington, DC: U.S. Environmental Protection Agency.

     Kordas. K; Ettinger. AS: Lamadrid-Figueroa. H; Tellez-Rojo. MM; Hernandez-Avila. M; Hu. H; Wright.
        RO. (2009). Methylenetetrahydrofolate reductase (MTHFR) C677T, A1298C and Gl 793 A genotypes,
        and the relationship between maternal folate intake, tibia lead and infant size at birth. Br J Nutr 102:
        907-914. http://dx.doi.org/10.1017/s0007114509318280

     Korrick. SA; Hunter. DJ; Rotnitzky. A; Hu. H; Speizer. FE. (1999). Lead and hypertension in a sample of
        middle-aged women. Am J Public Health 89: 330-335. http://dx.doi.Org/10.2105/AJPH.89.3.330

     Korrick. SA; Schwartz. J: Tsaih. SW: Hunter. DJ: Aro. A; Rosner. B; Speizer. FE; Hu. H. (2002).
        Correlates of bone and blood lead levels among middle-aged and elderly women. Am J Epidemiol
        156: 335-343. http://dx.doi.org/10.1093/aje/kwf042

     Koster. J: Erhardt. A; Stoeppler. M; Mohl. C: Ritz. E. (1989). Mobilizable lead in patients with chronic
        renal failure. Eur J Clin Invest 19: 228-233. http://dx.doi.Org/10.llll/i.1365-2362.1989.tb00222.x

     Kostial. K; Kello. D; Jugo. S: Rabar. I; Maljkovic. T. (1978). Influence  of age on metal metabolism and
        toxicity. Environ Health Perspect 25: 81-86. http://dx.doi.org/10.1289/ehp.782581
November 2012                             4-159                   Draft - Do Not Cite or Quote

-------
     Koyashiki. GA; Paoliello. MM; Matsuo. T; de Oliveira. MM; Mezzaroba. L; de Fatima Carvalho. M;
        Momoyo Sakuma. A; Turini. C: Terezinha Oliveira Vannuchi. M; Barbosa. CS. (2010). Lead levels in
        milk and blood from donors to the breast milk bank in Southern Brazil. Environ Res 110: 265-271.
        http://dx.doi.0rg/10.1016/i.envres.2009.12.001

     Krell. U: Roeckner. E. (1988). Model simulation of the atmospheric input of lead and cadmium into the
        North Sea. Atmos Environ 22: 375-381. http://dx.doi.org/10.1016/0004-6981(88)90043-l

     Krieg. EF. Jr; Butler. MA; Chang. MH; Liu. TB; Yesupriya. A; Lindegren. ML; Dowling. N. (2009). Lead
        and cognitive function in ALAD genotypes in the Third National Health and Nutrition Examination
        Survey. Neurotoxicol Teratol 31: 364-371. http://dx.doi.Org/10.1016/i.ntt.2009.08.003

     Kumar. A; Pastore. P. (2007). Lead and cadmium in soft plastic toys. Curr Sci 93: 818-822.

     Lacev. RF; Moore. MR; Richards. WN. (1985). Lead in water,  infant diet and blood: The Glasgow
        Duplicate Diet Study. Sci Total Environ 41: 235-257.

     Lagerkvist BJ; Ekesrydh. S: Englyst. V: Nordberg. GF; Soderberg. HA; Wiklund. DE. (1996). Increased
        blood lead and decreased calcium levels during pregnancy:  A prospective study of Swedish women
        living near a smelter. Am J Public Health 86: 1247-1252.

     Lai. JS: Wu. TN: Liou. SH: Shen. CY: Guu. CF: Ko. KN: Chi.  HY: Chang. PY. (1997). A study of the
        relationship between ambient lead and blood lead among lead battery workers. Int Arch Occup
        Environ Health 69: 295-300.

     Laidlaw. MAS: Mielke. HW: Filippelli. GM; Johnson. PL; Gonzales. CR. (2005). Seasonality and
        children's blood lead levels: Developing a predictive model using climatic variables and blood lead
        data from Indianapolis, Indiana, Syracuse, New York, and New Orleans, Louisiana (USA). Environ
        Health Perspect 793-800: 113. http://dx.doi.org/10.1289/ehp.7759

     Laidlaw. MAS: Zahran. S: Mielke. HW: Taylor.  MP; Filippelli. GM. (2012). Re-suspension of lead
        contaminated urban soil as a dominant source of atmospheric lead in Birmingham, Chicago, Detroit
        and Pittsburgh, USA. Atmos Environ 49: 302-310. http://dx.doi.Org/10.1016/i.atmosenv.2011.ll.030

     Lamadrid-Figueroa. H; Tellez-Rojo. MM; Hernandez-Cadena.  L; Mercado-Garcia. A; Smith. D;  Solano-
        Gonzalez. M; Hernandez-Avila. M; Hu. H. (2006). Biological markers of fetal lead exposure at each
        stage of pregnancy. J Toxicol Environ Health A 69: 1781-1796.
        http://dx.doi.org/10.1080/15287390600630195

     Lanphear. BP; Hornung. R; Ho. M; Howard. CR: Eberly. S: Knauf. K. (2002). Environmental lead
        exposure during early childhood. J Pediatr 140: 40-47.

     Lanphear. BP; Matte. TD; Rogers. J: Clickner. RP; Dietz. B; Bornschein. RL; Succop. P; Mahaffev. KR;
        Dixon. S: Galke. W: Rabinowitz. M; Farfel M; Rohde. C: Schwartz. J: Ashley. P; Jacobs. DE. (1998).
        The contribution of lead-contaminated house dust and residential soil to children's blood lead levels: A
        pooled analysis of 12 epidemiologic studies.  Environ Res 79: 51-68.
        http://dx.doi.org/10.1006/enrs.1998.3859

     Lanphear. BP: Roghmann. KJ. (1997). Pathways of lead exposure in urban children. Environ  Res 74:
        67-73.

     Lasheen. MR; Sharabv. CM; El-Kholy. NG: Elsherif. IY; El-Wakeel. ST. (2008). Factors influencing lead
        and iron release from some Egyptian drinking water pipes. J Hazard Mater 160: 675-680.
        http://dx.doi.0rg/10.1016/i.jhazmat.2008.03.040

     Lavton. DW: Beamer. PI. (2009). Migration of contaminated soil and airborne particulates to indoor dust.
        Environ Sci Technol 43: 8199-8205. http://dx.doi.org/10.1021/es9003735

     Lee. BK; Lee. GS: Stewart. WF; Ahn. KD; Simon. D; Kelsev. KT; Todd. AC: Schwartz. BS. (200la).
        Associations of blood pressure and hypertension with lead dose measures and polymorphisms in the
        vitamin D receptor and delta-aminolevulinic  acid dehydratase genes. Environ Health Perspect 109:
        383-389.
November 2012                             4-160                  Draft - Do Not Cite or Quote

-------
     Leggett. RW. (1993). An age-specific kinetic model of lead metabolism in humans. Environ Health
        Perspect 101: 598-616.

     Levesque. B; Duchesne. JF; Gariepv. C: Rhainds. M; Dumas. P; Scheuhammer. AM; Proulx. JF; Dery. S:
        Muckle. G: Dallaire. F; Dewailly. E. (2003). Monitoring of umbilical cord blood lead levels and
        sources assessment among the Inuit. Occup Environ Med 60: 693-695.

     Levin. R; Brown. MJ; Kashtock. ME; Jacobs. DE; Whelan. EA; Rodman. J; Schock. MR; Padilla. A;
        Sinks. T. (2008). Lead exposures in US children, 2008: Implications for prevention. Environ Health
        Perspect 116: 1285-1293. http://dx.doi.org/10.1289/ehp.11241

     Lewin. MD; Sarasua. S: Jones. PA. (1999). A multivariate linear regression model for predicting
        children's blood lead levels based on soil lead levels: A study at four superfund sites. Environ Res 81:
        52-61. http://dx.doi.org/10.1006/enrs.1998.3952

     Liang. F; Zhang. G: Tan. M; Yan. C: Li. X; Li. Y; Zhang. Y; Shan. Z. (2010). Lead in children's blood is
        mainly caused by coal-fired ash after phasing out of leaded gasoline in Shanghai. Environ Sci Technol
        44: 4760-4765. http://dx.doi.org/10.1021/es9039665

     Lima. FD; do Nascimento. CWA; da Silva. FBV; de Carvalho. VGB; Ribeiro. MR. (2009).  Lead
        concentration and allocation in vegetable crops grown in a soil contaminated by battery  residues.
        Horticulture Brasileira 27: 362-365. http://dx.doi.org/10.1590/S0102-05362009000300019

     Liu. J; McCaulev. L; Compher. C: Yan. C: Shen. X; Needleman. H; Pinto-Martin. JA. (2011 a). Regular
        breakfast and blood lead levels among preschool children. Environ Health Global Access Sci Source
        10: 28. http://dx.doi.org/10.1186/1476-069X-10-28

     Ljung. K; Selinus. O: Otabbong. E; Berglund. M. (2006). Metal and arsenic distribution in soil particle
        sizes relevant to soil ingestion by children. Appl Geochem 21: 1613-1624.
        http://dx.doi.0rg/10.1016/i.apgeochem.2006.05.005

     Long. GJ: Rosen. JF: Pounds. JG. (1990). Cellular lead toxicity and metabolism in primary and clonal
        osteoblastic bone cells. Toxicol Appl Pharmacol 102: 346-361.

     Loos M. SA. (2010). Receptor-oriented approaches in wildlife and human exposure modelling: A
        comparative  study. Environ Modell Softw 25: 369-382.

     Loos, M; Schipper, AM; Schlink, U; Strebel, K; Ragas, AJM. (2010). Receptor-oriented approaches in
        wildlife and human exposure modelling: A comparative study. Environ Modell Softw 25: 369-382.

     Lorenzana. RM; Troast R: Klotzbach. JM; Follansbee. MH; Diamond. GL. (2005). Issues related to time
        averaging of exposure in modeling risks associated with intermittent exposures to lead. Risk Anal 25:
        169-178. http://dx.doi.0rg/10.llll/i.0272-4332.2005.00576.x

     Lu. Y; Yin. W: Huang. L; Zhang. G: Zhao. Y. (2011). Assessment of bioaccessibility and exposure risk of
        arsenic and lead in urban soils of Guangzhou City, China. Environ Geochem Health 33:  93-102.
        http://dx.doi.org/10.1007/sl0653-010-9324-8

     Lynch.  RA; Boatright. DT; Moss. SK. (2000). Lead-contaminated imported tamarind candy and children's
        blood lead levels. Public Health Rep 115: 537-543. http://dx.doi.Org/10.1093/phr/115.6.537

     Lytle. DA; Schock. MR; Scheckel. K. (2009). The inhibition of Pb(IV) oxide formation in chlorinated
        water by orthophosphate. Environ Sci Technol 43: 6624-6631. http://dx.doi.org/10.1021/es900399m

     Maas. RP; Patch. SC: Christian. AM; Coplan. MJ. (2007). Effects of fluoridation and disinfection agent
        combinations on lead leaching from leaded-brass parts. Neurotoxicology 28: 1023-1031.
        http://dx.doi.0rg/10.1016/i.neuro.2007.06.006

     Macek. MD; Matte. TD; Sinks. T; Malvitz. DM. (2006). Blood lead concentrations in children and
        method of water fluoridation in the United States, 1988-1994. Environ Health Perspect 114: 130-134.
        http://dx.doi.org/10.1289/ehp.8319
November 2012                              4-161                   Draft - Do Not Cite or Quote

-------
     Maddaloni. M; Ballew. M; Diamond. G: Follansbee. M; Gefell. D; Goodrum. P; Johnson. M; Koporec. K;
        Khoury. G: Luev. J: Odin. M; Troast. R; Van Leeuwen. P; Zaragoza. L. (2005). Assessing lead risks at
        non-residential hazardous waste sites. Hum Ecol Risk Assess 11: 967-1003.
        http://dx.doi.org/10.1080/10807030500257838

     Maddaloni. M; Lolacono. N; Manton. W: Blum. C: Drexler. J; Graziano. J. (1998). Bioavailability of
        soilborne lead in adults, by stable isotope dilution. Environ Health Perspect 106: 1589-1594.
        http://dx.doi.org/10.1289/ehp.98106s61589

     Mahaffev. KR. (1977). Quantities of lead producing health effects in humans: Sources and bioavailability.
        Environ Health Perspect 19: 285-295. http://dx.doi.org/10.1289/ehp.7719285

     Mahaffev. KR; Annest. JL. (1986). Association of erythrocyte protoporphyrin with blood lead level and
        iron status in the second national health and nutrition examination survey, 1976-1980. Environ Res 41:
        327-338. http://dx.doi.org/10.1016/80013-9351(86)80194-3

     Mahaffev. KR: Gartside. PS: Glueck. CJ. (1986). Blood lead levels and dietary calcium intake in 1-11
        year-old children: The second national health and nutrition examination survey, 1976-1980. Pediatrics
        78: 257-262.

     Malcoe. LH; Lynch. RA; Kegler. MC: Skaggs. VJ. (2002). Lead sources, behaviors, and socioeconomic
        factors in relation to blood lead of Native American and white children: A community-based
        assessment of a former mining area. Environ Health Perspect 110: 221-231.
        http://dx.doi.org/10.1289/ehp.02110s2221

     Manton. WI. (1985). Total contribution of airborne lead to blood lead. Occup Environ Med 42: 168-172.
        http://dx.doi.0rg/10.1136/oem.42.3.168

     Manton. WI: Angle. CR; Krogstrand. KLS. (2005). Origin of lead in the United States diet. Environ Sci
        Technol 39: 8995-9000. http://dx.doi.org/10.1021/es051145e

     Manton. WI: Angle. CR: Stanek. KL; Kuntzelman. D; Reese. YR; Kuehnemann. TJ. (2003). Release of
        lead from bone in pregnancy and lactation. Environ Res 92: 139-151.
        http://dx.doi.org/10.1016/80013-9351(03)00020-3

     Manton. WI: Angle. CR: Stanek. KL; Reese. YR; Kuehnemann. TJ. (2000). Acquisition and retention of
        lead by young children. Environ Res 82: 60-80. http://dx.doi.org/10.1006/enrs.1999.4003

     Manton. WI: Cook. JD.  (1984). High accuracy (stable isotope dilution) measurements of lead in serum
        and cerebrospinal fluid. Occup Environ Med 41: 313-319. http://dx.doi.Org/10.1136/oem.41.3.313

     Manton. WI: Mallov. CR. (1983). Distribution of lead in body fluids after ingestion of soft solder. Occup
        Environ Med 40: 51-57. http://dx.doi.Org/10.1136/oem.40.l.51

     Manton. WI: Rothenberg. SJ: Manalo. M. (2001). The lead content of blood serum. Environ Res 86:
        263-273. http://dx.doi.org/10.1006/enrs.2001.4271

     Marcus. AH. (1985). Multicompartment kinetic model for lead, III: Lead in blood plasma and
        erythrocytes. Environ Res 36: 473-489. http://dx.doi.org/10.1016/0013-9351(85)90039-8

     Marcus. AH; Schwartz.  J. (1987). Dose-response curves for erythrocyte protoporphyrin vs blood lead:
        Effects of iron status. Environ Res 44: 221-227. http://dx.doi.org/10.1016/S0013-9351(87)80230-X

     Markowitz. ME; Weinberger. HL. (1990). Immobilization-related lead toxicity in previously lead-
        poisoned children. Pediatrics 86: 455-457.

     Marschner. B; Welge. P: Hack. A; Wittsiepe. J: Wilhelm. M. (2006). Comparison of soil Pb in vitro
        bioaccessibility and in vivo bioavailability with Pb pools from a sequential soil extraction. Environ Sci
        Technol 40: 2812-2818. http://dx.doi.org/10.1021/es051617p

     Marsh. J: Birchall. A. (1999). Determination of lung-to-blood absorption rates for lead and bismuth which
        are appropriate for radon progeny. Radiat Prot Dosimetry 83: 331- 337.
November 2012                              4-162                  Draft - Do Not Cite or Quote

-------
     Martin. D; Glass. TA; Bandeen-Roche. K; Todd. AC: Shi. WP; Schwartz. BS. (2006). Association of
        blood lead and tibia lead with blood pressure and hypertension in a community sample of older adults.
        Am J Epidemiol 163: 467-478. http://dx.doi.org/10.1093/aje/kwj060

     Martinez-Haro. M; Taggart. MA; Mateo. R. (2010). Pb-Al relationships in waterfowl feces discriminate
        between sources of Pb exposure. Environ Pollut 158: 2485-2489.
        http://dx.doi.0rg/10.1016/i.envpol.2010.03.018

     Mateo. R; Baos. AR; Vidal. D; Camarero. PR; Martinez-Haro. M; Taggart. MA. (2011). Bioaccessibility
        of Pb from ammunition in game meat is affected by cooking treatment. PLoS ONE 6:  e!5892.
        http://dx.doi.org/10.1371/iournal.pone.0015892

     McMichael. AJ; Baghurst. PA; Wigg. NR; Vimpani. GV: Robertson. EF; Roberts. RJ. (1988). Port Pirie
        cohort study: Environmental exposure to lead and children's abilities at the age of four years. N Engl J
        Med 319: 468-475. http://dx.doi.org/10.1056/NEJM19880825319080

     McNeill. A; Bedi. R: Islam. S: Alkhatib. MN; West. R. (2006). Levels of toxins in oral tobacco products
        in the U.K.. Tob Control 15:  64-67. http://dx.doi.org/10.1136/tc.2005.013011

     McNeill. FE; Stokes. L; Brito. JA; Chettle. PR; Kave.  WE. (2000).  109Cd K x-ray fluorescence
        measurements of tibial lead content in young adults exposed to lead in early childhood. Occup
        Environ Med 57: 465-471. http://dx.doi.Org/10.1136/oem.57.7.465

     Meirer. F; Pemmer. B; Pepponi. G: Zoeger. N; Wobrauschek. P; Sprio. S: Tampieri. A; Goettlicher. J:
        Steininger. R; Mangold. S: Roschger. P; Berzlanovich. A; Hofstaetter. JG: Streli. C. (2011).
        Assessment of chemical species of lead accumulated in tidemarks of human articular cartilage by X-
        ray absorption near-edge structure analysis. J Synchrotron Radiat 18: 238-244.
        http://dx.doi.org/10.1107/S0909049510052040

     Meltzer. HM; Brantsaeter. AL; Borch-Iohnsen. B; Ellingsen. DG: Alexander. J: Thomassen. Y; Stigum.
        H; Ydersbond. TA. (2010). Low iron stores are related to higher blood concentrations  of manganese,
        cobalt and cadmium in non-smoking, Norwegian women in the HUNT 2 study. Environ Res 110:
        497-504. http://dx.doi.0rg/10.1016/i.envres.2010.03.006

     Meyer. PA; Brown. MJ; Falk. H. (2008). Global approach to reducing lead exposure and poisoning
        [Review]. Mutat Res Rev Mutat Res 659: 166-175. http://dx.doi.Org/10.1016/i.mrrev.2008.03.003

     Mielke. HW: Covington. TP: Mielke. PW: Wolman. FJ: Powell. ET: Gonzales. CR. (2011 a). Soil
        intervention as a strategy for lead exposure prevention: The New Orleans lead-safe childcare
        playground project. Environ  Pollut 159: 2071-2077. http://dx.doi.Org/10.1016/i.envpol.2010.ll.008

     Mielke. HW: Gonzales. C. (2008). Mercury (Hg) and lead (Pb) in interior and exterior New Orleans
        house paint films. Chemosphere 72: 882-885. http://dx.doi.Org/10.1016/i.chemosphere.2008.03.061

     Mielke. HW: Gonzales. CR: Mielke. PW. (20 lib). The continuing impact of lead dust on children's blood
        lead: Comparison of public and private properties in New Orleans. Environ Res 111:  1164-1172.
        http://dx.doi.0rg/10.1016/i.envres.2011.06.010

     Mielke. HW: Gonzales. CR: Powell. E; Jartun. M; Mielke. PW. Jr. (2007a). Nonlinear association
        between soil lead and blood lead of children in metropolitan New Orleans, Louisiana: 2000-2005. Sci
        Total Environ 388: 43-53.  http://dx.doi.Org/10.1016/i.scitotenv.2007.08.012

     Mielke. HW: Powell. ET; Shah. A; Gonzales. CR: Mielke. PW.  (2001). Multiple metal contamination
        from house paints: Consequences of power sanding and paint scraping in New Orleans. Environ
        Health Perspect 109: 973-978. http://dx.doi.org/10.1289/ehp.01109973

     Miller. A; Drake. P: Hintz. P;  Habian. M. (2010). Characterizing exposures to airborne metals and
        nanoparticle emissions in a refinery. Ann Occup Hyg 54: 504-513.
        http://dx.doi.org/10.1093/annhyg/meq032

     Mindak. WR; Cheng.  J: Canas. BJ; Bolger. PM. (2008). Lead in women's and children's vitamins. J Agric
        Food Chem 56: 6892-6896. http://dx.doi.org/10.1021/if801236w
November 2012                             4-163                   Draft - Do Not Cite or Quote

-------
     Miodovnik. A; Landrigan. PJ. (2009). The U.S. Food and Drug Administration risk assessment on lead in
        women's and children's vitamins is based on outdated assumptions. Environ Health Perspect 117:
        1021-1022. http://dx.doi.org/10.1289/ehp.0900573

     Miranda. ML; Anthopolos. R; Hastings. D. (2011). Ageospatial analysis of the effects of aviation
        gasoline on childhood blood lead levels. Environ Health Perspect 119: 1513-1516.
        http://dx.doi.org/10.1289/ehp. 1003231

     Miranda. ML; Edwards. SE; Swamy. GK; Paul. CJ: Neelon. B. (2010). Blood lead levels among pregnant
        women: Historical versus contemporaneous exposures. Int J Environ Res Public Health 7: 1508-1519.
        http://dx.doi.org/10.3390/iierph7041508

     Miranda. ML; Kim. D; Hull. AP; Paul. CJ: Galeano. MAO. (2007b). Changes in blood lead levels
        associated with use of chloramines in water treatment systems. Environ Health Perspect 115: 221-225.
        http://dx.doi.org/10.1289/ehp.9432

     Mitchell. RA; Drake. JE; Wittlin. LA; Rejent. TA. (1977). Erythrocyte porphobilinogen synthase (delta-
        aminolaevulinate dehydratase) activity: A reliable and quantitative indicator of lead exposure in
        humans. Clin Chem 23:  105-111.

     Miyake. M. (1986). Structure refinements of Pb2+ ion-exchanged apatites by x-ray powder pattern-
        fitting. J Solid State Chem 61: 230-235. http://dx.doi.org/10.1016/0022-4596(86)90026-5

     Miyaki. K; Lwin. H; Masaki. K; Song. YX; Takahashi. Y; Muramatsu. M; Nakavama. T. (2009).
        Association between a polymorphism of aminolevulinate dehydrogenase (ALAD) gene and blood lead
        levels in Japanese subjects. Int J Environ Res Public Health 6: 999-1009.
        http://dx.doi.org/10.3390/ijerph6030999

     Molnar. P; Bellander. T; Sallsten. G: Boman. J. (2007). Indoor and outdoor concentrations of PM2.5 trace
        elements at homes, preschools and schools in  Stockholm, Sweden. J Environ Monit 9: 348-357.
        http://dx.doi.org/10.1039/b616858b

     Molnar. P; Gustafson. P; Johannesson. S: Boman. J: Barregard.  L; Sallsten. G. (2005). Domestic wood
        burning and PM2.5 trace elements: Personal exposures, indoor and outdoor levels. Atmos Environ  39:
        2643-2653. http://dx.doi.Org/10.1016/i.atmosenv.2005.01.016

     Montenegro.  MF; Barbosa.  F. Jr; Sandrim. VC: Gerlach. RF; Tanus-Santos. JE. (2006). A polymorphism
        in the delta-aminolevulinic acid dehydratase gene modifies plasma/whole blood lead ratio. Arch
        Toxicol 80: 394-398. http://dx.doi.org/10.1007/s00204-005-0056-v

     Montenegro.  MF; Barbosa.  F. Jr; Tanus-Santos. JE. (2008). Assessment of how pregnancy modifies
        plasma lead and plasma/whole blood lead ratio in ALAD 1-1 genotype women. Basic Clin Pharmacol
        Toxicol 102: 347-351. http://dx.doi.Org/10.llll/i.1742-7843.2007.00205.x

     Morrison. JN; Quarterman.  J. (1987). The relationship between iron status and lead absorption in rats.
        Biol Trace Elem Res 14: 115-126. http://dx.doi.org/10.1007/BF02795602

     Morrow. PE;  Beiter. H; Amato. F; Gibb. FR. (1980). Pulmonary retention of lead: An experimental study
        in man. Environ Res 21: 373-384. http://dx.doi.org/10.1016/0013-9351(80)90040-7

     Mushak. P. (1991). Gastro-intestinal absorption of lead in children and adults: Overview of biological  and
        biophysico-chemical aspects. Chem Speciation Bioavailability 3: 87-104.

     Mushak. P. (1998). Uses and limits of empirical data in measuring and modeling human lead exposure
        [Review]. Environ Health Perspect 106: 1467-1484. http://dx.doi.org/10.1289/ehp.98106s61467

     Naithani. V; Kakkar. P. (2006). Effect of ecological variation on heavy metal content of some medicinal
        plants used as herbal tea ingredients in India. Bull Environ Contam Toxicol 76: 285-292.
        http://dx.doi.org/10.1007/s00128-006-0919-9

     Navas-Acien. A; Schwartz.  BS: Rothenberg. SJ: Hu. H; Silbergeld. EK; Guallar. E. (2008). Bone lead
        levels and blood pressure endpoints: Ameta-analysis. Epidemiology 19: 496-504.
        http://dx.doi.org/10.1097/EDE.Ob013e31816a2400
November 2012                             4-164                  Draft - Do Not Cite or Quote

-------
     NCHS (National Center for Health Statistics). (2010). National health and nutrition examination survey:
        Questionnaires, datasets, and related documentation. Available online at
        http://www.cdc.gov/nchs/nhanes/nhanes questionnaires.htm (accessed November 19, 2010).

     Needleman. HL; McFarland. C: Ness. RB; Fienberg. SB; Tobin. MJ. (2002). Bone lead levels in
        adjudicated delinquents: A case control study. Neurotoxicol Teratol 24: 711-717.
        http://dx.doi.org/10.1016/S0892-0362(02)00269-6

     Nie. H; Sanchez. BN; Wilker. E; Weisskopf. MG: Schwartz. J; Sparrow. D; Hu. H. (2009). Bone lead and
        endogenous exposure in an environmentally exposed elderly population: The Normative Aging Study.
        J Occup Environ Med 51: 848-857. http://dx.doi.org/10.1097/JOM.Ob013e3181aa0106

     Nie. L; Sanchez. S: Newton. K; Grodzins. L; Cleveland. R; Weisskopf. M. (2011 a). In vivo quantification
        of lead in bone with a portable x-ray fluorescence system: Methodology and feasibility.  Phys Med
        Biol 56: N39-N51. http://dx.doi.Org/10.1088/0031-9155/56/3/N01

     Nie. LH; Wright. RO: Bellinger. DC: Hussain. J: Amarasiriwardena. C: Chettle. PR; Pejovic-Milic. A;
        Woolf. A; Shannon. M. (20lib). Blood lead levels and cumulative blood lead index (CBLI) as
        predictors of late neurodevelopment in lead poisoned children. Biomarkers 16: 517-524.
        http://dx.doi.org/10.3109/1354750X.2011.604133

     Nielsen. T; Jensen. KA; Grandjean. P. (1978). Organic lead in normal human brains. Nature 274:
        602-603. http://dx.doi.org/10.1038/274602aO

     Nilsson. U; Attewell. R; Christoffersson. JO; Schutz. A; Ahlgren. L; Skerfving. S: Mattsson. S. (1991).
        Kinetics of lead in bone and blood after end of occupational exposure. Basic Clin Pharmacol Toxicol
        68: 477-484.

     NIOSH (National Institute for Occupational Safety and Health). (1994). NIOSH manual of analytical
        methods:  Method no. 7105: Lead by GFAAS. (NIOSH 94-113).  Cincinnati, OH: U.S. Department of
        Health and Human Services, National Institute for Occupational Safety and Health.

     Niu. J; Rasmussen. P; Hassan. N; Vincent. R. (2010). Concentration distribution and bioaccessibility of
        trace elements in nano and fine urban airborne paniculate matter: Influence of particle size. Water Air
        Soil Pollut 213: 211-225. http://dx.doi.org/10.1007/sll270-010-0379-z

     Nriagu. J: Burt. B; Linder. A; Ismail. A; Sohn. W. (2006). Lead levels in blood and saliva in a low-income
        population of Detroit, Michigan. Int J Hyg Environ Health 209:  109-121.
        http://dx.doi.0rg/10.1016/i.iiheh.2005.ll.005

     Nwude. DO: Okoye. PAC: Babavemi. JO. (2010). Blood heavy metal levels in cows at slaughter at awka
        abattoir, Nigeria. International Journal of Dairy Science 5: 264-270.
        http://dx.doi.org/10.3923/iids.2010.264.270

     Nziguheba. G: Smolders. E.  (2008). Inputs of trace elements in agricultural soils via phosphate fertilizers
        in European countries. Sci Total Environ 390: 53-57. http://dx.doi.Org/10.1016/j.scitotenv.2007.09.031

     O'Flahertv. EJ. (1993). Physiologically based models for bone-seeking elements: IV. Kinetics of lead
        disposition in humans. Toxicol Appl Pharmacol 118: 16-29. http://dx.doi.org/10.1006/taap.1993.1004

     O'Flahertv. EJ. (1995). Physiologically based models for bone-seeking elements: V. Lead absorption and
        disposition in childhood [Review]. Toxicol Appl Pharmacol 131: 297-308.
        http://dx.doi.org/10.1006/taap.1995.1072

     O'Flahertv. EJ. (1998). A physiologically based kinetic model for lead in children and adults [Review].
        Environ Health Perspect  106: 1495-1503. http://dx.doi.org/10.1289/ehp.98106s61495

     O'Flahertv. EJ: Hammond. PB; Lerner. SI.  (1982). Dependence of apparent blood lead half-life on the
        length of previous lead exposure in humans. Toxicol Sci 2: 49-54.
        http://dx.doi.org/10.1093/toxsci/2.1.49

     O'Rourke. MK; Rogan. SP; Jin. S: Robertson. GL. (1999). Spatial distributions of arsenic exposure and
        mining communities from NHEXAS Arizona. National Human Exposure Assessment Survey. J Expo
        Anal Environ Epidemiol  9:  446-455. http://dx.doi.org/10.1038/sj.jea.7500050
November 2012                              4-165                   Draft - Do Not Cite or Quote

-------
     Obi. E; Akunyili. DN; Ekpo. B; Orisakwe. OE. (2006). Heavy metal hazards of Nigerian herbal remedies.
        Sci Total Environ 369: 35-41. http://dx.doi.Org/10.1016/i.scitotenv.2006.04.024

     Oldereid. NB; Thomassen. Y; Attramadal. A; Olaisen. B; Purvis. K. (1993). Concentrations of lead,
        cadmium and zinc in the tissues of reproductive organs of men. Reprod Fertil Dev 99: 421-425.
        http://dx.doi.org/10.1530/irf.0.0990421

     Omokhodion.  FO; Crockford. GW. (1991). Lead in sweat and its relationship to salivary and urinary
        levels in normal healthy subjects. Sci Total Environ 103: 113-122.
        http://dx.doi.org/10.1016/0048-9697(91)90137-4

     Omolaove. JA; Uzairu. A; Gimba. CE. (2010a). Heavy metal assessment of some eye shadow products
        imported into Nigeria from  China. Arch Appl Sci Res 2: 76-84.

     Omolaove. JA; Uzairu. A; Gimba. CE. (201 Ob). Heavy metal assessment of some soft plastic toys
        imported into Nigeria from  China. Journal of Environmental Chemistry and Ecotoxicology 2:
        126-130.

     Ong. CN; Lee. WR. (1980a). Distribution of lead-203 in human peripheral blood in vitro. Br J Ind Med
        37: 78-84.  http://dx.doi.Org/10.1136/oem.37.l.78

     Opler. MGA; Brown. AS: Graziano. J; Desai. M; Zheng. W: Schaefer. C: Factor-Litvak. P; Susser. ES.
        (2004). Prenatal lead exposure, delta-aminolevulinic acid, and schizophrenia. Environ Health Perspect
        112: 548-552. http://dx.doi.org/10.1289/ehp.10464

     Opler. MGA: Buka. SL; Groeger. J: McKeague. I; Wei. C: Factor-Litvak. P; Bresnahan. M; Graziano. J:
        Goldstein.  JM; Seidman. LJ; Brown. AS: Susser. ES. (2008). Prenatal exposure to lead, delta-
        aminolevulinic acid, and schizophrenia: Further evidence. Environ Health Perspect 116: 1586-1590.
        http://dx.doi.org/10.1289/ehp.10464

     Osterberg. K;  Borjesson. J: Gerhardsson. L; Schutz. A; Skerfving. S. (1997). A neurobehavioural study of
        long-term occupational inorganic lead exposure. Sci Total Environ 201: 39-51.
        http://dx.doi.org/10.1016/80048-9697(97)84051-X

     Otto. DA; Robinson. G: Baumann. S: Schroeder. S: Mushak. P; Kleinbaum. D; Boone. L. (1985). 5-year
        follow-up study of children with low-to-moderate lead absorption:  Electrophysiological evaluation.
        Environ Res 38: 168-186. http://dx.doi.org/10.1016/0013-9351(85)90082-9

     Oulhote. Y; Bot. BL; Poupon. J: Lucas. JP; Mandin. C: Etchevers. A; Zmirou-Navier. D; Glorennec. P.
        (2011). Identification of sources of lead exposure in French children by lead isotope analysis: A cross-
        sectional study. Environ Health Global Access  Sci Source 10: 75.
        http://dx.doi.org/10.1186/1476-069X-10-75

     P'an. AYS. (1981). Lead levels in saliva and in blood. J Toxicol Environ Health 7: 273-  280.
        http://dx.doi.org/10.1080/15287398109529978

     Pan. TL; Wang. PW: Al-Suwaveh. SA; Chen. CC: Fang. JY. (2010). Skin toxicology  of lead species
        evaluated by their permeability and proteomic profiles: A comparison of organic and inorganic lead.
        Toxicol Lett 197: 19-28. http://dx.doi.Org/10.1016/i.toxlet.2010.04.019

     Park. SK; Elmarsafawv. S: Mukherjee. B; Spiro. A. Ill; Vokonas. PS: Nie. H; Weisskopf MG: Schwartz.
        J; Hu. H. (2010). Cumulative lead exposure and age-related hearing loss: The VA  Normative Aging
        Study. Hear Res 269: 48-55. http://dx.doi.Org/10.1016/i.heares.2010.07.004

     Park. SK; Hu. H; Wright. RO: Schwartz. J: Cheng. Y; Sparrow. D: Vokonas. PS: Weisskopf. MG.
        (2009b). Iron metabolism genes, low-level lead exposure, and QT interval. Environ Health Perspect
        117: 80-85. http://dx.doi.org/10.1289/ehp.11559

     Park. SK: Mukherjee. B; Xia. X; Sparrow. D: Weisskopf. MG: Nie. H; Hu. H. (2009c). Bone lead level
        prediction  models and their application to examine the relationship of lead exposure and hypertension
        in the third National Health and Nutrition Examination Survey. J Occup Environ Med 51: 1422-1436.
        http://dx.doi.org/10.1097/JOM.Ob013e3181bf6c8d
November 2012                             4-166                  Draft - Do Not Cite or Quote

-------
     Park. SK; Schwartz. J; Weisskopf. M; Sparrow. D; Vokonas. PS: Wright. RO: Coull. B; Nie. H; Hu. H.
        (2006). Low-level lead exposure, metabolic syndrome, and heart rate variability: The VA Normative
        Aging Study. Environ Health Perspect 114: 1718-1724.

     Patel. AB; Prabhu. AS. (2009). Determinants of lead level in umbilical cord blood. Indian Pediatr 46:
        791-793.

     Pavton. M; Puggs. KM; Spiro. A. Ill; Weiss. ST; Hu. H. (1998). Relations of bone and blood lead to
        cognitive function: The VA Normative Aging Study. Neurotoxicol Teratol 20: 19-27.

     Pekev. B; Bozkurt. ZB; Pekev. H; Dogan. G: Zararsiz. A; Efe. N; Tuncel. G. (2010). Indoor/outdoor
        concentrations and elemental composition of PM10/PM2.5 in urban/industrial areas of Kocaeli City,
        Turkey. Indoor Air 20: 112-125. http://dx.doi.org/10. Illl/i.l600-0668.2009.00628.x

     Perez-Bravo. F; Ruz. M; Moran-Jimenez. MJ; Olivares. M; Rebolledo. A; Codoceo. J: Sepulveda. V:
        Jenkin. A; Santos. JL; Fontanellas. A. (2004). Association between aminolevulinate dehydrase
        genotypes and blood lead levels in children from a lead-contaminated area in Antofagasta, Chile. Arch
        Environ Contam Toxicol 47: 276-280. http://dx.doi.org/10.1007/s00244-004-2215-l

     Perry. SG: Cimorelli. AJ; Paine. RJ: Erode. RW: Weil. JC: Venkatram. A; Wilson. RB: Lee. RF; Peters.
        WD. (2005). AERMOD: A dispersion model for industrial source applications. Part II: Model
        performance against 17 field study databases. J Appl Meteorol 44: 694-708.
        http://dx.doi.Org/10.1175/JAM2228.l

     Peters. JL; Kubzansky. L; McNeely. E; Schwartz. J: Spiro. A. Ill; Sparrow. D; Wright. RO: Nie. H; Hu. H.
        (2007). Stress as a potential modifier of the impact of lead levels on blood pressure: The Normative
        Aging Study. Environ Health Perspect 115: 1154-1159. http://dx.doi.org/10.1289/ehp.10002

     Pierre. F: Vallaver. C: Baruthio. F: Peltier. A; Pale.  S: Rouver. J: Goutet. P; Aubrege. B; Lecossois. C:
        Guillemin. C: Elcabache. JM; Verelle. B; Fabries. JF. (2002). Specific relationship between blood lead
        and air lead in the crystal industry. Int Arch Occup Environ Health 75: 217-223.

     Pirkle. JL: Brodv. DJ: Gunter. EW: Kramer. RA; Paschal. DC: Flegal. KM; Matte. TD. (1994). The
        decline in blood lead levels in the United States: The National Health and Nutrition Examination
        Surveys (NHANES). JAMA 272: 284-291. http://dx.doi.org/10.1001/jama.1994.03520040046039

     Pirkle. JL; Kaufmann. RB; Brodv. DJ: Hickman. T; Gunter. EW: Paschal. DC. (1998). Exposure of the
        U.S. population to lead, 1991-1994. Environ Health Perspect 106: 745-750.

     Pocock. SJ: Shaper. AG: Walker. M; Wale. CJ: Clayton. B; Delves. T; Lacev. RF: Packham. RF; Powell.
        P (1983). Effects of tap water lead, water hardness,  alcohol, and cigarettes on blood lead
        concentrations. J Epidemiol Community Health 37:  1-7. http://dx.doi.Org/10.1136/jech.37.l.l

     Popovic. M; McNeill. FE; Chettle. PR; Webber. CE; Lee. CV: Kave. WE. (2005). Impact of occupational
        exposure on lead levels in women. Environ Health Perspect 113: 478-484.
        http://dx.doi.org/10.1289/ehp.7386

     Pounds. JG: Leggett. RW. (1998). The ICRP age-specific biokinetic model for lead: Validations, empirical
        comparisons, and explorations [Review]. Environ Health Perspect 106: 1505-1511.

     Pounds. JG: Long. GJ: Rosen. JF. (1991). Cellular and molecular toxicity of lead in bone [Review].
        Environ Health Perspect 91: 17-32. http://dx.doi.org/10.1289/ehp.919117

     Pounds. JG: Marlar. RJ: Allen. JR. (1978). Metabolism  of lead-210 in juvenile and adult rhesus monkeys
        (Macaca mulatta). Bull Environ Contam Toxicol 19: 684-691.

     Pounds. JG: Rosen. JF. (1986). Cellular metabolism of lead: Akinetic analysis in cultured osteoclastic
        bone cells. Toxicol Appl Pharmacol 83:  531-545. http://dx.doi.org/10.1016/0041-008X(86)90236-X

     Pruvot. C: Douav. F; Herve. F; Waterlot. C. (2006). Heavy metals in soil, crops and grass as a source of
        human exposure in the former mining areas. Journal of Soils and Sediments 6: 215-220.
        http://dx.doi.org/10.1065/iss2006.10.186
November 2012                             4-167                   Draft - Do Not Cite or Quote

-------
     Qian. J; Ferro. AR. (2008). Resuspension of dust particles in a chamber and associated environmental
        factors. Aerosol Sci Technol 42: 566-578. http://dx.doi.org/10.1080/02786820802220274

     Qian. J; Ferro. AR; Fowler. KR. (2008). Estimating the resuspension rate and residence time of indoor
        particles. J Air Waste Manag Assoc 58: 502-516. http://dx.doi.Org/10.3155/1047-3289.58.4.502

     Que Hee. SS: Peace. B; Clark. CS: Boyle. JR; Bornschein. RL; Hammond. PB. (1985). Evolution of
        efficient methods to sample lead sources, such as house dust and hand dust, in the homes of children.
        Environ Res 38:  77-95. http://dx.doi.org/10.1016/0013-9351(85)90074-X

     R.. ZA; A.. SL; S.. EA. (2011). Metal sources and exposures in the homes of young children living near a
        mining-impacted Superfund site. J Expo Sci Environ Epidemiol 21: 495-505.
        http://dx.doi.org/10.1038/ies.2011.21

     Raab. GM; Laxen. D; Lindsay. F; Anderson. N; Fulton. M. (1991). The influence of pH and household
        plumbing on water lead levels in Edinburgh. In JG Farmer (Ed.), International conference: Heavy
        metals in the environment, v 1; September; Edinburgh, United Kingdom (v. 1 ed., pp. 127-130).
        Edinburgh, United Kingdom: CEP Consultants, Ltd.
     Rabinowitz. M; Leviton. A; Needleman. H; Bellinger. D; Waternaux. C. (1985). Environmental correlates
        of infant blood lead levels in Boston. Environ Res 38: 96-107.
        http://dx.doi.org/10.1016/0013-9351(85)90075-1

     Rabinowitz. M; Needleman. H; Burlev. M; Finch. H; Rees. J. (1984). Lead in umbilical blood, indoor air,
        tap water, and gasoline in Boston. Arch Environ Health 39: 299-301.
     Rabinowitz. MB.  (1991). Toxicokinetics of bone lead [Review]. Environ Health Perspect 91: 33-37.
        httrj://dx.doi.org/10.1289/ehp.919133

     Rabinowitz. MB.  (1995). Relating tooth and blood lead levels in children. Bull Environ Contam Toxicol
        55: 853-857. http://dx.doi.org/10.1007/BF00209464

     Rabinowitz. MB;  Kopple. JD; Wetherill. GW. (1980). Effect of food intake and fasting on gastrointestinal
        lead absorption in humans. Am J Clin Nutr 33:  1784-1788.

     Rabinowitz. MB;  Leviton. A; Bellinger. D. (1993). Relationships between serial blood lead levels and
        exfoliated tooth dentin lead levels: Models of tooth lead kinetics. Calcif Tissue Int 53: 338-341.
        http://dx.doi.org/10.1007/BF01351840

     Rabinowitz. MB;  Leviton. A; Bellinger. DC. (1989). Blood lead - tooth lead relationship among Boston
        children. Bull  Environ Contam Toxicol 43: 485-492. http://dx.doi.org/10.1007/BF01701924

     Rabinowitz. MB;  Wetherill. GW: Kopple. JD. (1973). Lead metabolism in the normal human: Stable
        isotope studies. Science 182: 725-727. http://dx.doi.org/10.1126/science.182.4113.725
     Rabinowitz. MB;  Wetherill. GW: Kopple. JD. (1976). Kinetic analysis of lead metabolism in healthy
        humans. J Clin Invest 58: 260-270. http://dx.doi.org/10.1172/JCI108467

     Rabinowitz. MB;  Wetherill. GW: Kopple. JD. (1977). Magnitude of lead intake from respiration by
        normal man. Transl Res 90: 238-248.

     Rabstein. S: Unfried. K; Ranft. U: Illig. T; Kolz. M; Mambetova. C: Vlad. M; Roman. C: Weiss. T;
        Becker. D; Bruning. T; Pesch. B. (2008). Lack of association of delta-aminolevulinate dehydratase
        polymorphisms with blood lead levels and hemoglobin in Romanian women from a lead-contaminated
        region. J Toxicol Environ Health A 71: 716-724. http://dx.doi.org/10.1080/15287390801985190

     Rajan. P; Kelsev.  KT; Schwartz. JD: Bellinger. DC: Weuve. J: Sparrow. D: Spiro. A. Ill; Smith. TJ; Nie.
        H; Hu. H; Wright. RO. (2007). Lead burden and psychiatric symptoms and the modifying influence of
        the delta-aminolevulinic acid dehydratase (ALAD) polymorphism: The VA Normative Aging Study.
        Am J Epidemiol 166:  1400-1408. http://dx.doi.org/10.1093/aie/kwm220
November 2012                             4-168                  Draft - Do Not Cite or Quote

-------
     Raian. P; Kelsev. KT; Schwartz. JD; Bellinger. DC: Weuve. J: Spiro. A. Ill; Sparrow. D; Smith. TJ; Nie.
        H; Weisskopf. MG: Hu. H; Wright. RO. (2008). Interaction of the delta-aminolevulinic acid
        dehydratase polymorphism and lead burden on cognitive function: The VA Normative Aging Study. J
        Occup Environ Med 50: 1053-1061. http://dx.doi.org/10.1097/JOM.Ob013e3181792463

     Ranft. U: Delschen. T; Machtolf. M; Sugiri. D; Wilhelm. M. (2006). Lead concentration in the blood of
        children and its association with lead in soil and ambient air: Trends between 1983 and 2000 in a
        German industrial city [Abstract]. Epidemiology 17: S122-S122.

     Ranft. U: Delschen. T; Machtolf. M; Sugiri. D; Wilhelm. M. (2008). Lead concentration in the blood of
        children and its association with lead in soil and ambient air: Trends between 1983 and 2000 in
        Duisburg. J Toxicol Environ Health A 71: 710-715. http://dx.doi.org/10.1080/15287390801985117

     Rankin. C: Nriagu. JO: Aggarwal. JK; Arowolo. TA; Adebavo. K; Flegal. AR. (2005). Lead
        contamination in cocoa and cocoa products: Isotopic evidence of global contamination. Environ
        Health Perspect 113: 1344-1348. http://dx.doi.org/10.1289/ehp.8009

     Rasmussen. PE; Wheeler. AJ; Hassan. NM; Filiatreault. A; Lanouette. M. (2007). Monitoring personal,
        indoor, and outdoor exposures to metals in airborne paniculate matter: Risk of contamination during
        sampling, handling and analysis. Atmos Environ 41: 5897-5907.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.03.018

     Reglero. MM; Taggart. MA; Monsalve-Gonzalez. L; Mateo. R. (2009b). Heavy metal exposure in large
        game from a lead mining area: Effects on oxidative stress and fatty acid composition in liver. Environ
        Pollut 157: 1388-1395. http://dx.doi.Org/10.1016/i.envpol.2008.ll.036

     Rentschler. G: Broberg. K; Lundh. T; Skerfving.  S. (2012). Long-term lead elimination from plasma and
        whole blood after poisoning. Int Arch Occup Environ Health 85: 311-316.
        http://dx.doi.org/10.1007/s00420-011-0673-0

     Rhodes. D: Spiro. A. Ill; Aro. A; Hu. H. (2003). Relationship of bone and blood lead levels to psychiatric
        symptoms: The Normative Aging Study. J Occup Environ Med 45: 1144-1151.
        http://dx.doi.org/10.1097/01.iom.0000094995.23808.7b

     Riedt. CS: Buckley. BT; Brolin. RE; Ambia-Sobhan. H; Rhoads. GG: Shapses. SA. (2009). Blood lead
        levels and bone turnover with weight reduction  in women. J Expo Sci Environ Epidemiol 19: 90-96.
        http://dx.doi.0rg/10.1038/ies.2008.5

     Robbins. N; Zhang. ZF; Sun. J: Ketterer. ME; Lalumandier. JA; Shulze. RA. (2010).  Childhood lead
        exposure and uptake in teeth in the Cleveland area during the era of leaded gasoline. Sci Total Environ
        408: 4118-4127. http://dx.doi.Org/10.1016/i.scitotenv.2010.04.060

     Rodrigues. EG: Virji.  MA; McClean. MD; Weinberg. J: Woskie. S: Pepper. LD. (2010). Personal
        exposure, behavior, and work site conditions as  determinants of blood lead among bridge painters. J
        Occup Environ Hyg 7: 80-87. http://dx.doi.org/10.1080/15459620903418316

     Rodrigues. JL; Batista. BL; Nunes. JA; Passos. CJS: Barbosa. F. (2008). Evaluation of the use of human
        hair for biomonitoring the deficiency of essential and exposure to toxic elements.  Sci Total Environ
        405: 370-376. http://dx.doi.Org/10.1016/i.scitotenv.2008.06.002

     Roels. H; Konings. J:  Green. S: Bradley. D: Chettle. D: Lauwerys. R. (1995). Time-integrated blood lead
        concentration is a valid surrogate for estimating the cumulative lead dose assessed by tibial lead
        measurement. Environ Res 69: 75-82.

     Roels. H; Lauwervs. R: Konings. J: Buchet JP; Bernard. A; Green. S: Bradley. D: Morgan. W: Chettle.
        EX (1994).  Renal function and hyperfiltration capacity in lead smelter workers with high bone lead.
        Occup Environ Med 51: 505-512. http://dx.doi.Org/10.1136/oem.51.8.505

     Rosen. JF.  (1983). The metabolism of lead in isolated bone cell populations: Interactions between lead
        and calcium. Toxicol Appl Pharmacol 71: 101-112. http://dx.doi.org/10.1016/0041-008X(83)90049-2
November 2012                             4-169                   Draft - Do Not Cite or Quote

-------
     Rosen. JF; Pounds. JG. (1998). Severe chronic lead insult that maintains body burdens of lead related to
        those in the skeleton: Observations by Dr Clair Patterson conclusively demonstrated [Review].
        Environ Res 78: 140-151. http://dx.doi.org/10.1006/enrs.1997.3830

     Rothenberg. SJ: Karchmer. S: Schnaas. L; Perroni. E; Zea. F; Alba. JF. (1994a). Changes in serial blood
        lead levels during pregnancy. Environ Health Perspect 102: 876-880.
        http://dx.doi.org/10.1289/ehp.94102876

     Rothenberg. SJ: Kondrashov. V; Manalo. M; Jiang. J; Cuellar. R; Garcia. M; Reynoso. B; Reyes. S: Diaz.
        M; Todd. AC. (2002a). Increases in hypertension and blood pressure during pregnancy with increased
        bone lead levels. Am J Epidemiol 156: 1079-1087. http://dx.doi.org/10.1093/aie/kwfl63

     Rothenberg. SJ: Schnaas. L; Perroni. E; Hernandez. RM; Karchmer. S. (1998). Secular trend in blood lead
        levels  in a cohort of Mexico City children. Arch Environ Health 53: 231-235.
        http://dx.doi.org/10.1080/00039899809605700

     Roussel. H; Waterlot. C: Pelfrene. A; Pruvot. C: Mazzuca. M; Douav. F. (2010). Cd, Pb and Zn oral
        bioaccessibility of urban soils contaminated in the past by atmospheric emissions from two lead and
        zinc smelters. Arch Environ Contam Toxicol 58: 945-954.
        http://dx.doi.org/10.1007/s00244-009-9425-5

     Rubio-Andrade. M; Valdes-Perezgasga. F; Alonso. J: Rosado. JL; Cebrian. ME; Garcia-Vargas. GG.
        (2011). Follow-up study on lead exposure in children living in a smelter community in northern
        Mexico. Environ Health Global Access Sci Source 10: 66.  http://dx.doi.org/10.1186/1476-069X-10-66

     Ryu. JE; Ziegler.  EE; Nelson. SE; Fomon. SJ. (1983). Dietary intake of lead and blood lead concentration
        in early infancy. Am J Dis Child 137: 886-891.

     Saikat. S: Barnes. B; Westwood. D. (2007). A review of laboratory results for bioaccessibility values of
        arsenic, lead and nickel in contaminated U.K. soils. J Environ Sci Health A Tox Hazard Subst Environ
        Eng42: 1213-1221. http://dx.doi.org/10.1080/10934520701435486

     Sakai. T; Yanagihara. S: Kunugi. Y; Ushio. K. (1982). Relationships between distribution of lead in
        erythrocytes in vivo and in vitro and inhibition of ALA-D. Br J Ind Med 39:  382-387.

     Sammut. ML; Noack. Y; Rose. J: Hazemann. JL; Proux. O; Depoux. M; Ziebel. A; Fiani. E. (2010).
        Speciation of  Cd and Pb in dust emitted from sinter plant. Chemosphere 78:  445-450.
        http://dx.doi.0rg/10.1016/i.chemosphere.2009.10.039

     Sayre. JW: Charnev. E; Vostal. J: Pless. IB. (1974). House and hand dust  as a potential source of
        childhood lead exposure. Am J Dis Child 127: 167-170.

     Schell. LM; Denham. M; Stark. AD; Ravenscroft. J: Parsons. P; Schulte.  E. (2004). Relationship between
        blood  lead concentration and dietary intakes of infants from 3 to 12 months of age. Environ Res 96:
        264-273. http://dx.doi.0rg/10.1016/i.envres.2004.02.008

     Schmitt. CJ: Brumbaugh. WG: May. TW. (2007a). Accumulation of metals in fish from lead-zinc mining
        areas of southeastern Missouri, USA. Ecotoxicol Environ Saf 67: 14-30.
        http://dx.doi.0rg/10.1016/i.ecoenv.2006.ll.002

     Schmitt. CJ: Brumbaugh. WG: May. TW. (2009).  Concentrations of cadmium, cobalt, lead, nickel, and
        zinc in blood and fillets of northern hog sucker (Hypentelium nigricans) from streams contaminated
        by lead-zinc mining: Implications for monitoring. Arch Environ Contam Toxicol 56: 509-524.
        http://dx.doi.org/10.1007/s00244-009-9288-9

     Schmitt. CJ: Whyte. JJ: Brumbaugh. WG: Tillitt. DE. (2005). Biochemical effects of lead, zinc, and
        cadmium from mining on fish in the Tri-States District of northeastern Oklahoma, USA. Environ
        Toxicol Chem 24: 1483-1495. http://dx.doi.Org/10.1897/04-332R.l

     Schnaas. L; Rothenberg. SJ: Flores. MF; Martinez. S: Hernandez. C: Osorio. E; Perroni. E. (2004). Blood
        lead secular trend in a cohort of children in Mexico City (1987-2002). Environ Health Perspect 112:
        1110-1115.  http://dx.doi.org/10.1289/ehp.6636
November 2012                              4-170                  Draft - Do Not Cite or Quote

-------
     Schnaas. L; Rotfaenberg. SJ: Perroni. E; Martinez. S: Hernandez. C: Hernandez. RM. (2000). Temporal
        pattern in the effect of postnatal blood lead level on intellectual development of young children.
        Neurotoxicol Teratol 22: 805-810. http://dx.doi.org/10.1016/S0892-0362(00)00101-X

     Schock. MR; Hyland. RN; Welch. MM. (2008). Occurrence of contaminant accumulation in lead pipe
        scales from domestic drinking-water distribution systems. Environ Sci Technol 42: 4285-4291.
        http://dx.doi.org/10.1021/es702488v

     Schroeder. HA; Tipton. IH. (1968). The human body burden of lead [Review]. Arch Environ Health 17:
        965-978.

     Schuhmacher. M; Hernandez. M; Domingo. JL; Fernandez-Ballart. JD; Llobet. JM; Corbella. J. (1996). A
        longitudinal study of lead mobilization during pregnancy: Concentrations in maternal and umbilical
        cord blood. Trace Elem Electroly 13: 177-181.

     Schutz. A; Bergdahl. IA; Ekholm. A; Skerfving. S. (1996). Measurement by ICP-MS of lead in plasma
        and whole blood of lead workers and controls. Occup Environ Med 53: 736-740.
        http://dx.doi.org/10.1136/oem.53.ll.736

     Schwartz. BS: Lee. BK; Bandeen-Roche. K; Stewart. W: Bolla. K; Links. J: Weaver. V; Todd. A. (2005).
        Occupational lead exposure and longitudinal decline in neurobehavioral test scores. Epidemiology 16:
        106-113. http://dx.doi.org/10.1097/01.ede.0000147109.62324.51

     Schwartz. BS: Lee. BK: Lee. GS: Stewart. WF: Lee. SS: Hwang. KY: Ahn. KD: Kim. YB: Bolla. KI:
        Simon. D; Parsons. PJ: Todd. AC.  (2001). Associations of blood lead, dimercaptosuccinic acid-
        chelatable lead, and tibia lead with neurobehavioral test scores in South Korean lead workers. Am J
        Epidemiol 153: 453-464. http://dx.doi.Org/10.1093/aie/153.5.453

     Schwartz. BS: Stewart. WF; Bolla. KI; Simon. MS: Bandeen-Roche. K; Gordon. B; Links. JM; Todd. AC.
        (2000b). Past adult lead exposure is associated with longitudinal decline in cognitive function.
        Neurology 55: 1144-1150.

     Schwartz. BS: Stewart. WF: Todd. AC: Simon. D; Links. JM. (2000c). Different associations of blood
        lead, meso 2,3-dimercaptosuccinic acid (DMSA)-chelatable lead, and tibial lead levels with blood
        pressure in 543 former organolead manufacturing workers. Arch Environ Occup Health 55: 85-92.
        http://dx.doi.org/10.1080/00039890009603392

     Schwartz. J: Pitcher. H. (1989). The relationship between gasoline lead and blood lead in the United
        States. J Offic Stat 5: 421-431.

     Scinicariello. F; Murray. HE; Moffett DB; Abadin. HG: Sexton. MJ; Fowler. BA. (2007). Lead and delta-
        aminolevulinic acid dehydratase polymorphism: Where does it lead? Ameta-analysis. Environ Health
        Perspect 115: 35-41. http://dx.doi.org/10.1289/ehp.9448

     Scinicariello. F; Yesupriya. A; Chang. MH; Fowler. BA. (2010). Modification by ALAD of the
        association between blood lead and blood pressure in the U.S.  population: Results from the Third
        National Health and Nutrition Examination Survey. Environ Health Perspect 118: 259-264.
        http://dx.doi.org/10.1289/ehp.0900866

     Sesli. E; Tuzen. M; Soylak. M. (2008). Evaluation of trace metal contents of some wild edible
        mushrooms from Black Sea region, Turkey. J Hazard Mater 160: 462-467.
        http://dx.doi.0rg/10.1016/i.jhazmat.2008.03.020

     Shah. F; Kazi. TG: Afridi. HI; Khan. S: Kolachi. NF; Arain. MB; Baig. JA. (2011). The influence of
        environmental exposure on lead concentrations in scalp hair of children in Pakistan. Ecotoxicol
        Environ  Saf 74: 727-732. http://dx.doi.Org/10.1016/j.ecoenv.2010.10.036

     Shaik. AP; Jamil. K. (2009). Individual susceptibility and genotoxicity in workers exposed to hazardous
        materials like lead. J Hazard Mater 168: 918-924. http://dx.doi.Org/10.1016/i.jhazmat.2009.02.129

     Sharma. A;  ChY. R; Tiwari. RK; Kumar Tyagi. L; Kori. ML; Singh. V; Gaur. K; Shankar. K. (2009). Role
        of Aloe barbadensis Linn in the removal of toxic heavy metal of Kukkutandatwak (shell of hen's egg):
        A drug used in Indian system of medicine (Ayurveda). Adv Biol Res 3: 79-83.
November 2012                             4-171                  Draft - Do Not Cite or Quote

-------
     Sheets. RG: Bergquist. BA. (1999). Laboratory treatability testing of soils contaminated with lead and
        PCBs using particle-size separation and soil washing. J Hazard Mater 66: 137-150.

     Sheppard. SC. (1995). Parameter values to model the soil ingestion pathway. Environ Monit Assess 34:
        27-44. http://dx.doi.org/10.1007/BF00546244

     Sherlock. J; Smart. G: Forbes. GI; Moore. MR; Patterson. WJ: Richards. WN; Wilson. TS. (1982).
        Assessment of lead intakes and dose-response for a population in Ayr exposed to a plumbosolvent
        water supply. Hum Exp Toxicol 1: 115-122. http://dx.doi.org/10.1177/096032718200100203

     Sherlock. JC: Ashbv. D; Delves. HT; Forbes. GI: Moore. MR; Patterson. WJ: Pocock. SJ: Quinn. MJ;
        Puchards. WN; Wilson. TS. (1984). Reduction in exposure to lead from drinking water and its effect
        on blood lead concentrations. Hum Exp Toxicol 3: 383-392.
        http://dx.doi.org/10.1177/096032718400300503

     Sherlock. JC: Quinn. MJ. (1986). Relationship between blood and lead concentrations and dietary lead
        intake in infants: The Glasgow Duplicate Diet Study 1979-1980. Food Addit Contam 3:  167-176.
        http://dx.doi.org/10.1080/02652038609373579

     Shih. RA; Glass. TA; Bandeen-Roche. K; Carlson. MC:  Bolla. KI; Todd. AC: Schwartz. BS. (2006).
        Environmental lead exposure and cognitive function  in community-dwelling older adults. Neurology
        67: 1556-1562. http://dx.doi.org/10.1212/01.wnl.0000239836.26142.c5

     Shih. RA; Hu. H; Weisskopf. MG: Schwartz. BS. (2007). Cumulative lead dose and cognitive function in
        adults: A review of studies that measured both blood lead and bone lead [Review]. Environ Health
        Perspect 115: 483-492. http://dx.doi.org/10.1289/ehp.9786

     Shotyk. W: Krachler. M. (2009). Determination of trace  element concentrations in natural freshwaters:
        How low is "low," and how low do we need to go? J  Environ Monit 11:  1747-1753.
        http://dx.doi.org/10.1039/b917090c

     Siciliano. SD; James. K; Zhang. GY; Schafer. AN; Peak. JD. (2009). Adhesion and Enrichment of Metals
        on Human Hands from Contaminated Soil at an Arctic Urban Brownfield. Environ Sci Technol 43:
        6385-6390. http://dx.doi.org/10.1021/es901090w

     Silbergeld. EK. (1991). Lead in bone: Implications for toxicology during pregnancy and lactation
        [Review]. Environ Health Perspect 91: 63-70.

     Silbergeld. EK; Schwartz. J: Mahaffev. K. (1988). Lead  and osteoporosis: Mobilization of lead from bone
        in postmenopausal women. Environ Res 47: 79-94. http://dx.doi.org/10.1016/80013-9351(88)80023-9

     Simon. PL; Maynard. EJ; Thomas. KD. (2007). Living in a sea of lead changes in blood- and hand-lead
        of infants living near a smelter. J Expo Sci Environ Epidemiol 17: 248-259.
        http://dx.doi.org/10.1038/sj.jes.7500512

     Smith. CM; Hu. H; Wang. X; Kelsev. KT. (1995a). ALA-D genotype is not associated with HT or HB
        levels among workers exposed to low levels of lead.  Med Lav 86: 229-235.

     Smith. D; Hernandez-Avila. M; Tellez-Rojo. MM; Mercado. A; Hu. H.  (2002). The relationship between
        lead in plasma and whole blood in women. Environ Health Perspect 110: 263-268.
        http://dx.doi.org/10.1289/ehp.02110263

     Smith. DM; Mielke. HW: Heneghan. JB.  (2008). Subchronic lead feeding study in male rats. Arch
        Environ Contam Toxicol 55:  518-528. http://dx.doi.org/10.1007/s00244-008-9138-l

     Smith. DM; Mielke. HW: Heneghan. JB.  (2009a). Subchronic lead feeding study in male rats and
        micropigs. Environ Toxicol 24: 453-461. http://dx.doi.org/10.1002/tox.20448

     Smith. PR; Kahng. MW: Quintanilla-Vega. B; Fowler. BA. (1998). High-affinity renal lead-binding
        proteins in environmentally-exposed humans. Chem  Biol Interact 115: 39-52.
        http://dx.doi.org/10.1016/S0009-2797(98)00060-X
November 2012                             4-172                  Draft - Do Not Cite or Quote

-------
     Smith. PR; Osterloh. JD; Flegal. AR. (1996). Use of endogenous, stable lead isotopes to determine
        release of lead from the skeleton. Environ Health Perspect 104: 60-66.
        http://dx.doi.org/10.1289/ehp.9610460

     Smith. E; Weber. J; Naidu. R; McLaren. RG: Juhasz. AL. (2011). Assessment of lead bioaccessibility in
        peri-urban contaminated soil. J Hazard Mater 186: 300-305.
        http://dx.doi.0rg/10.1016/i.jhazmat.2010.10.lll

     Sobin. C: Gutierrez. M; Alterio. H. (2009). Polymorphisms of delta-aminolevulinic acid dehydratase
        (ALAD) and peptide transporter 2 (PEPT2) genes in children with low-level lead exposure.
        Neurotoxicology 30: 881-887. http://dx.doi.Org/10.1016/i.neuro.2009.08.006

     Soderholm. SC. (1989). Proposed international conventions for particle size-selective sampling. Ann
        Occup Hyg 33: 301-320. http://dx.doi.Org/10.1093/annhyg/33.3.301

     Spalinger. SM; von Braun. MC: Petrosvan. V; von Lindern. IH. (2007). Northern Idaho house dust and
        soil lead levels compared to the Bunker Hill superfund site. Environ Monit Assess 130: 57-72.
        http://dx.doi.org/10.1007/sl0661-006-9450-z

     SRC (Syracuse Research Corporation). (2002). Baseline human health risk assessment: Eureka Mills  -
        Eureka, Utah. Denver, CO: U.S. Environmental Protection Agency.
        http://www.epa.gov/region8/r8risk/pdf/hhra_eureka.pdf

     Srivastava. SK; Rai. V: Srivastava. M; Rawat. AKS: Mehrotra. S. (2006). Estimation of heavy metals in
        different Berberis species and its market  samples. Environ Monit Assess 116: 315-320.
        http://dx.doi.org/10.1007/sl0661-006-7395-x

     Sterling. DA; Roegner. KG: Lewis. RD; Luke. DA; Wilder. LC: Burchette. SM. (1999). Evaluation of
        four sampling methods for determining exposure of children to lead-contaminated household dust.
        Environ Res 81: 130-141.  http://dx.doi.org/10.1006/enrs.1999.3962

     Stern. AH. (1994). Derivation of a target level of lead in soil at residential sites corresponding to a de
        minimis contribution to blood lead concentration. Risk Anal 14: 1049-1056.
        http://dx.doi.0rg/10.llll/i.1539-6924.1994.tb00075.x

     Stern. AH. (1996). Derivation of a target concentration of Pb  in soil based on elevation of adult blood
        pressure. Risk Anal 16: 201-210. http://dx.doi.Org/10.llll/i.1539-6924.1996.tb01450.x

     Stewart. WF; Schwartz. BS: Davatzikos. C:  Shen. D; Liu. D; Wu. X; Todd. AC: Shi. W: Bassett. S:
        Youssem. D. (2006).  Past adult lead exposure is linked to  neurodegeneration measured by brain MRI.
        Neurology 66: 1476-1484. http://dx.doi.org/10.1212/01.wnl.0000216138.69777.15

     Stewart. WF; Schwartz. BS: Simon. D; Bolla. KI; Todd. AC:  Links. J. (1999). Neurobehavioral function
        and tibial and chelatable lead levels in 543 former organolead workers. Neurology 52: 1610-1617.

     Stokes. L; Letz. R; Gerr. F; Kolczak. M; McNeill. FE; Chettle. PR; Kave. WE. (1998). Neurotoxicity in
        young adults 20 years after childhood exposure to lead: the Bunker Hill experience. Occup Environ
        Med 55: 507-516.

     Succop. P; Bornschein. R; Brown. K; Tseng. CY. (1998). An empirical comparison of lead exposure
        pathway models. Environ Health Perspect 1066:  1577-1583.

     Tahir. SN; Alaamer. AS. (2008). PB-210 concentrations in cigarettes tobaccos and radiation doses to the
        smokers. Radiat Prot Dosimetry 130: 389-391. http://dx.doi.org/10.1093/rpd/ncn097

     Tellez-Rojo. MM; Hernandez-Avila. M; Gonzalez-Cossio. T; Romieu. I; Aro. A; Palazuelos. E; Schwartz.
        J; Hu. H. (2002). Impact of breastfeeding on the  mobilization of lead from bone. Am J Epidemiol  155:
        420-428. http://dx.doi.0rg/10.1093/aje/155.5.420

     TerraGraphics Environmental Engineering (TerraGraphics Environmental Engineering Inc). (2004).
        Human health remedial evaluation report for the Bunker Hill Superfund site box. Moscow, ID.
November 2012                              4-173                   Draft - Do Not Cite or Quote

-------
     Theppeang. K; Glass. TA; Bandeen-Roche. K; Todd. AC: Rohde. CA; Links. JM; Schwartz. BS. (2008a).
        Associations of bone mineral density and lead levels in blood, tibia, and patella in urban-dwelling
        women. Environ Health Perspect 116: 784-790. http://dx.doi.org/10.1289/ehp.10977

     Todd. AC: Buchanan. R; Carroll. S: Moshier. EL; Popovac. D; Slavkovich. V; Graziano. JH. (2001). Tibia
        lead levels and methodological uncertainty in 12-year-old children. Environ Res 86: 60-65.
        http://dx.doi.org/10.1006/enrs.2001.4251

     Todd. AC: Carroll S: Geraghtv. C: Khan. FA; Moshier. EL; Tang. S: Parsons. PJ. (2002). L-shell x-ray
        fluorescence measurements of lead in bone: accuracy and precision. Phys Med Biol 47:  1399-1419.
        http://dx.doi.org/10.1088/0031-9155/47/8/312

     Todd. AC: Carroll. S: Godbold. JH; Moshier. EL; Khan. FA. (2000). Variability in XRF-measured tibia
        lead levels. Phys Med Biol 45: 3737-3748.  http://dx.doi.org/10.1088/0031-9155/45/12/316

     Tovalin-Ahumada. H; Whitehead. L; Blanco. S. (2007). Personal exposure to PM2.5 and element
        composition—A comparison between outdoor and indoor workers from two Mexican cities. Atmos
        Environ 41: 7401-7413. http://dx.doi.Org/10.1016/i.atmosenv.2007.05.059

     Treble. RG: Thompson. TS. (1997). Preliminary results of a survey of lead levels in human liver tissue.
        Bull Environ Contam Toxicol 59: 688-695. http://dx.doi.org/10.1007/s001289900535

     Tripathi. RM; Raghunath. R; Kumar. AV; Sastry. VN; Sadasivan. S.  (2001). Atmospheric and children's
        blood lead as indicators of vehicular traffic and other emission sources in Mumbai, India. Sci Total
        Environ 267: 101-108. htrp://dx.doi.org/10.1016/S0048-9697(00)00770-l

     Tsuji. LJS: Wainman. BC: Javasinghe. RK; VanSpronsen. EP; Liberda. EN. (2009). Determining tissue-
        lead levels in large game mammals harvested with lead bullets: Human health concerns. Bull Environ
        Contam Toxicol 82: 435-439. http://dx.doi.org/10.1007/s00128-009-9647-2

     Tsuji. LJS: Wainman. BC: Martin. ID; Sutherland. C: Weber. JP; Dumas. P; Nieboer. E. (2008). The
        identification of lead ammunition as a source of lead exposure in First Nations: The use of lead
        isotope ratios. Sci Total Environ 393: 291-298. http://dx.doi.Org/10.1016/i.scitotenv.2008.01.022

     Tu. C: Zheng. CR; Chen. HM. (2000). Effect of applying chemical fertilizers on forms of lead and
        cadmium in red soil. Chemosphere 41: 133-138. http://dx.doi.org/10.1016/S0045-6535(99)00400-2

     Turlakiewicz. Z; Chmielnicka. J. (1985). Diethyllead as a specific indicator of occupational exposure to
        tetraethyllead. Occup Environ Med 42: 682-685. http://dx.doi.org/10.1136/oem.42.10.682

     Turner. A; Simmonds. L. (2006). Elemental concentrations and metal bioaccessibility in U.K. household
        dust.  Sci Total Environ 371: 74-81. http://dx.doi.Org/10.1016/i.scitotenv.2006.08.011

     Tvinnereim. HM; Eide. R: Riise. T; Wesenberg. GR; Fosse. G: Steinnes. E. (1997). Lead in primary teeth
        from Norway: Changes in lead levels from  the 1970s to the 1990s. Sci Total Environ 207: 165-177.
        htrp://dx.doi.org/10.1016/S0048-9697(97)00263-5

     U.S. EPA (U.S. Environmental Protection Agency). (1986a). Air quality criteria for lead [EPA Report].
        (EPA/600/8-83/028aF-dF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1990b). Field evaluation of a high volume surface
        sampler for pesticides in floor dust [EPA Report]. (EPA/600/S3-90/030). Research Triangle Park.

     U.S. EPA (U.S. Environmental Protection Agency). (1994). Guidance  manual for the integrated exposure
        uptake biokinetic model for lead in children [EPA Report]. (EPA/540/R-93/081). Washington, D.C.
        http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=2000WN4R.txt

     U.S. EPA (U.S. Environmental Protection Agency). (1996a).  Recommendations of the Technical Review
        Workgroup for Lead for an interim approach to assessing risks associated with adult exposures to lead
        in soil [EPA Report]. Washington, DC: U.S. Environmental Protection Agency, Technical Review
        Workgroup for Lead.
November 2012                             4-174                   Draft - Do Not Cite or Quote

-------
     U.S. EPA (U.S. Environmental Protection Agency). (1996b). Urban soil lead abatement demonstration
        project: Volume I: EPA integrated report [EPA Report]. (EPA/600/P-93/001aF). Washington, D.C.
        http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockev=30002663.txt

     U.S. EPA (U.S. Environmental Protection Agency). (2001). Baseline human health risk assessment:
        Vasquez Boulevard and 1-70 Superfund site, Denver, CO [EPA Report]. Denver, CO.
        http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=P1006STM.txt

     U.S. EPA (U.S. Environmental Protection Agency). (2003c). Recommendations of the Technical Review
        Workgroup for Lead for an approach to assessing risks associated with adult exposures to lead in soil
        [EPA Report]. (EPA-540-R-03-001). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2005a). All ages lead model (AALM) (Version Draft
        1.05) [Computer Program]. Research Triangle Park, NC: U.S. Environmental Protection Agency,
        National Center for Environmental Assessment.

     U.S. EPA (U.S. Environmental Protection Agency). (2005f). Re-evaluation of the lEUBK-based clean-up
        level for lead in soil in East Helena, Montana [EPA Report]. Denver, CO.

     U.S. EPA (U.S. Environmental Protection Agency). (2006b). Air quality criteria for lead: Volume I of II
        [EPA Report]. (EPA/600/R-05/144aF). Research Triangle Park, NC.
        http ://cfpub. epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823

     U.S. EPA (U.S. Environmental Protection Agency). (2006c). Air quality criteria for lead: Volume II of II
        [EPA Report]. (EPA/600/R-05/144bF). Research Triangle Park, NC.
        http ://cfpub .epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823

     U.S. EPA (U.S. Environmental Protection Agency). (2007b). Estimation of relative bioavailability of lead
        in soil and soil-like materials using in vivo and in vitro methods [EPA Report]. (OSWER 9285.7-77).
        Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2007e). Guidance for evaluating the oral
        bioavailability of metals in soils for use in human health risk assessment [EPA Report]. (OSWER
        9285.7-80). Washington, D.C.
        http://www.epa.gov/superfund/health/contaminants/bioavailabilitv/guidance.htm

     U.S. EPA (U.S. Environmental Protection Agency). (20071). Lead: Human exposure and health risk
        assessments for selected case studies. Volume II: Appendices [EPA Report]. (452R07014b). Research
        Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2007g). Lead: Human exposure and health risk
        assessments for selected case studies: Volume 1: Human exposure and health risk assessments - full-
        scale [EPA Report]. (EPA-452/R-07-014a).  Research Triangle Park, NC.
        http://www.ntis.gov/search/product.aspx?ABBR=PB2008102573

     U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for paniculate
        matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546

     Uzu. G: Sobanska. S: Sarret.  G: Munoz. M; Dumat. C. (2010). Foliar lead uptake by lettuce exposed to
        atmospheric fallouts. Environ Sci Technol 44: 1036-1042. http://dx.doi.org/10.1021/es902190u

     Van de Wiele. TR; Oomen. AG: Wragg. J: Cave. M; Minekus. M; Hack. A; Cornells. C: Rompelberg.
        CJM; De Zwart. LL; Klinck. B; Van Wijnen. J; Verstraete. W: Sips. A. (2007). Comparison of five in
        vitro digestion models to in vivo experimental results: Lead bioaccessibility in the human
        gastrointestinal tract. J Environ Sci Health A Tox Hazard Subst Environ Eng 42: 1203-1211.
        htrp://dx.doi.org/10.1080/10934520701434919

     van Wijngaarden. E; Campbell. JR; Cory-Slechta. DA. (2009). Bone lead levels are associated with
        measures of memory impairment in older adults. Neurotoxicology 30: 572-580.
        http://dx.doi.0rg/10.1016/i.neuro.2009.05.007
November 2012                             4-175                  Draft - Do Not Cite or Quote

-------
     Vandenhove. H; Olyslaegers. G: Sanzharova. N; Shubina. O: Reed. E; Shang. Z; Velasco. H. (2009).
        Proposal for new best estimates of the soil-to-plant transfer factor of U, Th, Ra, Pb and Po. J Environ
        Radioact 100: 721-732. http://dx.doi.Org/10.1016/i.jenvrad.2008.10.014

     Vander. AJ; Taylor. PL; Kalitis. K; Mouw. PR; Victery. W. (1977). Renal handling of lead in dogs:
        Clearance studies. Am J Physiol 233: F532-F538.

     Verbeeck. RMH; Lassuvt. CJ: Heijligers. HJM; Driessens. FCM; Vrolijk. JWG. A. (1981). Lattice
        parameters and cation distribution of solid solutions of calcium and lead hydroxyapatite. Calcif Tissue
        Int 33: 243-247. http://dx.doi.org/10.1007/BF02409444

     Victery. W: Vander. AJ: Mouw. DR. (1979). Effect of acid-base status on renal excretion and
        accumulation of lead in dogs and rats. Am J Physiol 237: F398-F407.

     Villalobos. M; Merino-Sanchez. C: Hall. C: Grieshop. J: Gutierrez-Ruiz. ME; Handlev. MA. (2009). Lead
        (II) detection and contamination routes in environmental sources, cookware and home-prepared foods
        from Zimatlan, Oaxaca, Mexico. Sci Total Environ 407: 2836-2844.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.12.059

     Vural. N: Duvdu. Y. (1995). Biological monitoring of lead in workers exposed to tetraethyllead. Sci Total
        Environ 171: 183-187. http://dx.doi.org/10.1016/0048-9697(95)04676-6

     Waalkes. MP; Klaassen. CD. (1985). Concentration of metallothionein in major organs of rats after
        administration of various metals. Toxicol Sci 5: 473-477. http://dx.doi.Org/10.1093/toxsci/5.3.473

     Wananukul. W: Sura. T; Salaitanawatwong. P.  (2006). Polymorphism of delta-aminolevulinic acid
        dehydratase and its  effect on blood lead levels  in Thai workers. Arch Environ Occup Health 61: 67-72.
        http://dx.doi.Org/10.3200/AEOH.61.2.67-72

     Wasserman. GA; Factor-Litvak. P; Liu. X; Todd. AC: Kline. JK; Slavkovich. V; Popovac. D; Graziano.
        JH. (2003). The relationship between blood lead, bone lead and child intelligence. Child Neuropsychol
        9: 22-34. http://dx.doi.Org/10.1076/chin.9.l.22.14497

     Watson. WS: Morrison. J: Bethel. MIF; Baldwin. NM; Lyon. DTB; Dobson. H; Moore. MR; Hume. R.
        (1986). Food iron and lead absorption in humans. Am J Clin Nutr 44: 248-256.

     Weaver. VM; Ellis. LR; Lee. BK; Todd. AC: Shi. W: Ahn. KD; Schwartz. BS. (2008). Associations
        between patella lead and blood pressure in lead workers. Am J Ind Med 51: 336-343.
        http://dx.doi.org/10.1002/aiim.20573

     Webber. CE; Chettle. PR; Bowins. RJ: Beaumont. LF; Gordon. CL; Song. X; Blake. JM; McNutt. RH.
        (1995). Hormone replacement therapy may reduce the return of endogenous lead from bone to the
        circulation. Environ Health Perspect 103: 1150-1153.

     Weis. CP; Lavelle. JM. (1991). Characteristics to consider when choosing an animal model for the study
        of lead bioavailability. Chem Speciation Bioavailability 3: 113-119.

     Weisskopf. MG: Hu. H; Sparrow. D; Lenkinski. RE; Wright. RO. (2007a). Proton magnetic resonance
        spectroscopic evidence of glial effects of cumulative lead exposure in the adult human hippocampus.
        Environ Health Perspect 115:  519-523. http://dx.doi.org/10.1289/ehp.9645

     Weisskopf. MG: Jain. N;  Nie. HL; Sparrow. D: Vokonas. P; Schwartz. J: Hu. H. (2009). A prospective
        study of bone lead concentration and death from  all causes, cardiovascular diseases, and cancer in the
        department of veterans affairs normative aging study. Circulation 120: 1056-1064.
        http://dx.doi.org/10.1161/circulationaha.108.827121

     Weisskopf. MG: Proctor.  SP; Wright. RO: Schwartz. J: Spiro. A. Ill; Sparrow. D: Nie. HL; Hu. H.
        (2007b). Cumulative lead exposure and cognitive performance among elderly men. Epidemiology 18:
        59-66. http://dx.doi.org/10.1097/01.ede.0000248237.35363.29

     Weisskopf. MG: Weuve. J: Nie. H; Saint-Hilaire. MH; Sudarsky. L; Simon. DK; Hersh. B; Schwartz. J:
        Wright. RO: Hu. H. (2010). Association of cumulative lead exposure with Parkinson's Disease.
        Environ Health Perspect 118:  1609-1613. http://dx.doi.org/10.1289/ehp.1002339
November 2012                             4-176                   Draft - Do Not Cite or Quote

-------
     Weisskopf MG: Wright. RO: Schwartz. J; Spiro. A. Ill: Sparrow. D; Aro. A; Hu. H. (2004). Cumulative
        lead exposure and prospective change in cognition among elderly men: The VA Normative Aging
        Study. Am J Epidemiol 160: 1184-1193. http://dx.doi.org/10.1093/aje/kwh333

     Welt. M: Mielke. HW: Gonzales. C: Cooper. KM: Batiste. CG: Cresswell. LH. Ill: Mielke. PW. (2003).
        Metal contamination of sediments and soils of bayou Saint John: A potential health impact to local
        fishermen? Environ Geochem Health 25: 387-396.

     Wetmur. JG. (1994). Influence of the common human delta-aminolevulinate dehydratase polymorphism
        on lead body burden [Review]. Environ Health Perspect 102: 215-219.

     Wetmur. JG: Kava. AH; Plewinska. M; Desnick. RJ. (1991a). Molecular characterization of the human 5-
        aminolevulinate dehydratase-2 (ALAD2) allele: Implications for molecular screening of individuals
        for genetic susceptibility to lead poisoning. Am J Hum Genet 49: 757-763.

     Wetmur. JG: Lehnert. G: Desnick. RJ. (1991b). The delta-aminolevulinate dehydratase polymorphism:
        Higher blood lead levels in lead workers and environmentally exposed children with the 1-2 and 2-2
        isozymes. Environ Res 56: 109-119. http://dx.doi.org/10.1016/80013-9351(05)80001-5

     Weuve. J: Kelsev. KT; Schwartz. J: Bellinger. D; Wright. RO: Raian. P; Spiro. A. Ill; Sparrow. D; Aro. A;
        Hu. H. (2006). Delta-aminolevulinic acid dehydratase polymorphism and the relation between low
        level lead exposure and the Mini-Mental Status Examination in older men: The Normative Aging
        Study. Occup Environ Med 63: 746-753. http://dx.doi.org/10.1136/oem.2006.027417

     Weuve. J: Korrick. SA; Weisskopf. MA; Ryan. LM; Schwartz. J: Nie. HL; Grodstein. F; Hu. H. (2009).
        Cumulative exposure to lead in relation to cognitive function in older women. Environ Health
        Perspect 117: 574-580. http://dx.doi.org/10.1289/ehp.11846

     Wigginton. NS: Haus. KL; Hochella. MF. (2007). Aquatic environmental nanoparticles [Review]. J
        Environ Monit 9: 1306-1316. http://dx.doi.org/10.1039/b712709i

     Wilhelm. M; Lombeck. I; Hafner. D; Ohnesorge. FK. (1989). Hair lead levels in young children from the
        FRG. J Trace Elem Med Biol 3: 165-170.

     Wilhelm. M; Pesch. A; Rostek. U; Begerow. J: Schmitz. N; Idel. H; Ranft. U. (2002). Concentrations of
        lead in blood, hair and saliva of German children living in three  different areas of traffic density. Sci
        Total Environ 297: 109-118.

     Wilker. E; Korrick. S: Nie. LH; Sparrow. D: Vokonas. P; Coull. B; Wright. RO: Schwartz. J: Hu. H.
        (2011). Longitudinal changes in bone lead levels: The VA Normative Aging Study. J Occup Environ
        Med 53:  850-855. http://dx.doi.org/10.1097/JOM.Ob013e31822589a9

     Wilson. J; Dixon. S: Galke. W: McLaine. P. (2007). An investigation of dust lead sampling locations and
        children's blood lead levels. J Expo Sci Environ Epidemiol 17: 2-12.
        http://dx.doi.org/10.1038/sj.jes.7500514

     Wright. RO: Silverman. EK; Schwartz. J: Tsaih.  SW: Senter. J: Sparrow. D: Weiss. ST; Aro. A; Hu. H.
        (2004). Association between hemochromatosis genotype and lead exposure among elderly men: The
        Normative Aging Study. Environ Health Perspect 112: 746-750.

     Wright. RO: Tsaih. SW: Schwartz. J: Spiro. A; McDonald. K; Weiss. ST;  Hu. H. (2003). Lead exposure
        biomarkers and mini-mental status exam scores in older men. Epidemiology 14: 713-718.
        http://dx.doi.org/10.1097/01.EDE.0000081988.85964.db

     Xie. Y; Chiba. M; Shinohara. A; Watanabe. H; Inaba. Y. (1998). Studies on lead-binding protein and
        interaction between  lead and selenium in the  human erythrocytes. Ind  Health 36: 234-239.
        http://dx.doi.org/10.2486/indhealth.36.234

     Xie. Y; Wang. Y; Giammar. DE. (2010). Impact  of chlorine  disinfectants on dissolution of the lead
        corrosion product PbO2. Environ Sci Technol 44: 7082-7088.  http://dx.doi.org/10.1021/esl016763

     Yamamoto. N; Takahashi. Y; Yoshinaga. J: Tanaka. A; Shibata. Y. (2006). Size  distributions of soil
        particles adhered to children's hands. Arch Environ Contain Toxicol 51: 157-163.
        http://dx.doi.org/10.1007/s00244-005-7012-y
November 2012                              4-177                  Draft - Do Not Cite or Quote

-------
     Yiin. LM; Rhoads. GG: Lioy. PJ. (2000). Seasonal influences on childhood lead exposure. Environ Health
        Perspect 108: 177-182.

     Yu. CH; Yiin. LM; Lioy. PJ. (2006). The bioaccessibility of lead (Pb) from vacuumed house dust on
        carpets in urban residences. Risk Anal 26: 125-134.
        http:/7dx.doi.org/10.1111/i.l539-6924.2006.00710.x

     Zahran. S: Mielke. HW: Weiler. S: Gonzales. CR. (2011). Nonlinear associations between blood lead in
        children, age of child, and quantity of soil lead in metropolitan New Orleans. Sci Total Environ 409:
        1211-1218. http://dx.doi.0rg/10.1016/j.scitotenv.2010.ll.036

     Zhang. W: Zhang.  GG: He. HZ; Bolt. HM. (1994). Early health effects and biological monitoring in
        persons occupationally exposed to tetraethyl lead. Int Arch Occup Environ Health 65: 395-399.
        http://dx.doi.org/10.1007/BF00383250

     Zhang. X; Ahmadi. G: Qian. J; Ferro. A. (2008). Particle detachment, resuspension and transport due to
        human walking in indoor environments. J Adhes Sci Tech 22: 591-621.
        http://dx.doi.org/10.1163/156856108X305624

     Zhang. Y; Griffin. A; Rahman. M; Camper. A; Baribeau. H; Edwards. M. (2009b). Lead contamination of
        potable water due to nitrification. Environ Sci Technol 43:  1890-1895.
        http://dx.doi.org/10.1021/es802482s

     Zhao. Y; Wang. L; Shen. HB; Wang. ZX; Wei. QY; Chen. F. (2007). Association between delta-
        aminolevulinic acid dehydratase (ALAD) polymorphism and blood lead levels: A meta-regression
        analysis. J Toxicol Environ Health 70: 1986-1994. http://dx.doi.org/10.1080/15287390701550946

     Ziegler. EE; Edwards. BB; Jensen. RL; Mahaffev. KR; Fomon. SJ. (1978). Absorption and retention of
        lead by infants. Pediatr Res 12: 29-34. http://dx.doi.org/10.1203/00006450-197801000-00008
November 2012                             4-178                   Draft - Do Not Cite or Quote

-------
       5 INTEGRATED  HEALTH  EFFECTS   OF  LEAD
          EXPOSURE
      5.1        Introduction

 1                   This chapter summarizes, integrates, and evaluates the evidence for the broad spectrum of
 2                   health effects associated with exposure to Pb. The chapter begins (Section 5.2) with a
 3                   discussion of the evidence for the modes of action that mediate the health effects of Pb,
 4                   including those modes of action that are shared by all of the health effects evaluated in
 5                   this ISA and those modes of action that are specific to particular endpoints. Subsequent
 6                   sections comprise evaluations of the epidemiologic  and toxicological evidence for the
 7                   effects of Pb exposure on health outcomes related to nervous system effects (Section 5.3).
 8                   cardiovascular effects (Section 5.4). renal effects (Section 5.5). immune effects
 9                   (Section 5.6). hematological effects (Section 5.7). and reproductive and developmental
10                   effects (Section  5.8). Section 5.9 reviews the evidence for the effects  of Pb on other
11                   noncancer health outcomes, for which the cumulative bodies of evidence are smaller,
12                   including those related to the hepatic system (Section 5.9.1). gastrointestinal  system
13                   (Section 5.9.2). endocrine system (Section 5.9.3). bone and teeth (Section 5.9.4). ocular
14                   health (Section 5.9.5). and respiratory  system (Section 5.9.6). Chapter 5 concludes with a
15                   discussion of the evidence for Pb effects on cancer (Section 5.10).

16                   Individual sections for major outcome categories (e.g., nervous system, cardiovascular,
17                   renal) begin with a brief summary of conclusions from the 2006 Pb AQCD (U.S. EPA.
18                   2006c) followed by an evaluation of recent (i.e., published since the completion of the
19                   2006 Pb AQCD) studies that is intended to build upon evidence from previous reviews.
20                   Within each of these sections, results are organized  by endpoint (e.g., cognitive function,
21                   behavior, neurodegenerative diseases) then by specific scientific discipline
22                   (i.e., epidemiology, toxicology). This chapter evaluates evidence for both short- and long-
23                   term Pb exposures, which are defined as less than four weeks and greater than
24                   four weeks, respectively, in animal toxicological studies and less than one year and
25                   greater than one year, respectively, in epidemiologic studies (Section 2.1).

26                   Sections for each of the major outcome categories (e.g., nervous system, cardiovascular,
27                   renal effects) conclude with an integrated summary of the evaluation  of evidence and a
28                   conclusion regarding causality. Based  upon the framework (described in the Preamble to
29                   this ISA), a determination of causality was made for a group of related endpoints within a
30                   major outcome category (e.g., cognitive function, attention-related behavioral problems).
31                   In judgments regarding causality, emphasis was placed on studies with relevant Pb
32                   exposure routes  and concentrations in toxicological studies and internal dose measures in

      November 2012                              5-1                   Draft - Do Not Cite or Quote

-------
 1                   epidemiologic studies (generally one order of magnitude above blood Pb levels in the
 2                   current U.S. population as described in Section 2.1). Studies that examined higher Pb
 3                   concentrations were evaluated particularly to inform mode of action. Further, evidence
 4                   was evaluated for consistency of findings across multiple studies and the extent to which
 5                   chance, confounding (i.e., bias due to a correlation with Pb biomarker level and causal
 6                   association with the outcome), and other biases could be ruled out with reasonable
 7                   confidence. Such evidence included high quality epidemiologic studies with
 8                   representative population-based groups or samples, prospective versus cross-sectional or
 9                   ecologic design; rigorous statistical analysis (i.e., multivariate regression) with
10                   assessment of potential confounding factors; information on the concentration-response
11                   relationship; and supporting toxicological evidence. The extent of consideration for
12                   potential confounding varied among epidemiologic studies. Because no single study
13                   considered all potential confounding factors, and not all potential confounding factors
14                   were examined in the collective body of evidence, residual confounding by unmeasured
15                   factors is possible. However, the examination of factors well documented in the literature
16                   to be associated with Pb exposure and health outcomes and supporting toxicological
17                   evidence help to minimize the undue influence of confounding bias  on the observed
18                   epidemiologic associations. The biological plausibility provided by the coherence of
19                   evidence between toxicology and epidemiology and across a spectrum of related
20                   endpoints, including evidence for modes of action, was used as support to address
21                   uncertainties in the epidemiologic evidence due to biases from factors such as selective
22                   publication, recruitment or participation of subjects; reverse causality;  or confounding.
      5.2       Modes of Action


      5.2.1        Introduction

23                   The diverse health effects associated with Pb exposure are dependent on multiple factors,
24                   including the concentration and duration of exposure, the particular Pb compounds
25                   constituting the exposure, and which tissues are affected. Pb exposure is linked to
26                   downstream health effects by various modes of action. A mode of action (MOA) is the
27                   common set of biochemical, physiological, or behavioral responses (i.e., empirically
28                   observable precursor steps) that can cumulatively result in the formation of negative
29                   health outcomes. Although the effects of Pb exposure appear to be  mediated through
30                   multiple modes of action, alteration of cellular ion status (including disruption of calcium
31                   homeostasis, altered ion transport mechanisms, and perturbed protein function through
32                   displacement of metal cofactors) seems to be the major unifying mode of action
33                   underlying all subsequent modes of action (Figure 5-1). This section draws information

      November 2012                              5-2                   Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
from all of the subsequent health effects sections in Chapter 5, and identifies the major
modes of action operating at the molecular, cellular, and tissue/organ level. In turn, the
individual health effect sections bridge these MOA effects to those observed on the
organismal level. Each of the individual health effect sections includes a more detailed
description of the mechanisms specific to the individual health effect. Accordingly, this
section differs in structure and content from other health effects sections as it does not
primarily focus on the literature published since the 2006 Pb AQCD, but rather
incorporates recent information with earlier studies (which together represent the current
state of the science) on the possible modes of action of Pb. Higher concentrations of Pb
are often utilized in mode of action studies. This section includes some studies that are
conducted at concentrations greater than one order of magnitude above the upper end of
the blood Pb distribution of the general U.S. population when it is likely that the mode of
action does not differ at higher concentrations.
Oxidative
Stress
(5.2.4)


Cell Death
Genotoxicity
(5.2.7)
Inflammation
(5.2.5)

Endocrine
Disruption
(5.2.6)
      Note: The subsections where these MOAs are discussed are indicated in parentheses.
      (Section 5.2.2: Section 5.2.3: Section 5.2.4: Section 5.2.5: Section 5.2.6: and Section 5.2.7).
      Figure 5-1       Schematic representation of the relationships between the
                        various  MOAs by which Pb exposure exerts its health effects.
      November 2012
                            5-3
Draft - Do Not Cite or Quote

-------
      5.2.2        Altered Ion Status

 1                   Physiologically-relevant metal ions (e.g., Ca2+, Mg, Zn, Fe) are known to have a
 2                   multitude of functions in biological systems, including roles as charge carriers,
 3                   intermediates in enzymatically-catalyzed reactions, and structural elements in the proper
 4                   maintenance of tertiary protein conformations (Garza et al.. 2006). It is through
 5                   disruption of these biological functions that Pb exerts its negative actions, ultimately
 6                   interfering with such tightly regulated processes as cell signaling, intracellular ion
 7                   homeostasis, ion transport, energy metabolism, and enzymatic function.
                     5.2.2.1      Disruption of Ca2+ Homeostasis

 8                   Calcium (Ca2+) is one of the most important carriers of cell signals and regulates virtually
 9                   all aspects of cell function, including energy metabolism, signal transduction, hormonal
10                   regulation, cellular motility, and apoptosis (Carafoli. 2005). Ca2+ homeostasis is
11                   maintained through a tightly regulated balance of cellular transport and intracellular
12                   storage (Pentvala et al.. 2010). Disruption of Ca2+ homeostasis by Pb has been observed
13                   in a number of different cell types and cell-free environments, indicating that this is a
14                   major mode of action for Pb-induced toxicity on a cellular level.

15                   Ca2+ homeostasis is particularly important in bone cells, as the skeletal system serves as
16                   the major dynamic reservoir of Ca2+ in the body (Wiemann et al.,  1999; Long et al.,
17                   1992). Bone  cells also are unique in that they exist in a microenvironment that is high in
18                   Ca2+, and potentially high in Pb concentrations. This may increase their relative exposure
19                   to Pb and thus Pb-induced effects (Long et al.. 1992). A series of studies from the
20                   laboratory of Long, Dowd, and Rosen have indicated that exposure of cultured
21                   osteoblastic bone cells to Pb alters intracellular Ca2+ levels ([Ca2+]0.  Exposure of
22                   osteoblasts to 1, 5, or 25 (iM Pb for 40-300 minutes resulted in prolonged increases in
23                   [Ca2+]j of 36, 50 and 120% over baseline,  respectively (Schanne etal.. 1997; Schanne et
24                   al.. 1989). Long et al. (1992) observed that exposure of osteoblasts to either 400 ng
25                   parathyroid hormone (PTH)/mL culture medium for  1 hour or 25 (iM Pb for 20 hours
26                   increased [Ca2+]j. Pb-exposed cells pretreated with PTH increased [Ca2+]j above
27                   concentrations observed in either single exposure (Pb alone or PTH alone), indicating
28                   that Pb may disrupt the ability of bone cells to respond to normal hormonal control. A
29                   similar increase  in [Ca2+]j was also observed when bone cells were co-treated with
30                   epidermal growth factor (EGF, 50 ng/mL) plus Pb (5 (iM), versus EGF alone (Long and
31                   Rosen. 1992). Pb-induced increases in [Ca2+]j were blocked by a protein kinase C (PKC)
32                   inhibitor, indicating that PKC activation may serve as one mechanism by which Pb
33                   perturbs [Ca2+]j (Schanne etal.. 1997). Schirrmacher et al. (1998)  also observed
      November 2012                              5-4                   Draft - Do Not Cite or Quote

-------
 1                   alterations in Ca2+ homeostasis in osteoblasts exposed to 5 (iM Pb for 50 minutes due to
 2                   potential disruption of Ca2+ATPases. However, Wiemann et al. (1999) demonstrated that
 3                   exposure to 5 or 12.5 (iM Pb inhibited the Ca2+-release-activated calcium influx of Ca2+
 4                   independently of any inhibitory effect on Ca2+ATPases.

 5                   Ca2+ homeostasis has also been shown to be disturbed in erythrocytes exposed to Pb
 6                   (Quintanar-Escorza et al., 2010; Quintanar-Escorza et al., 2007; Shin et al.. 2007). In
 7                   blood samples taken from Pb-exposed workers (mean [SD] blood Pb level: 74.4
 8                   [21.9] (ig/dL), the [Ca2+]j was approximately 2.5-fold higher than that seen in nonexposed
 9                   workers (mean [SD] blood Pb level: 9.9 [2.0] (ig/dL) (Ouintanar-Escorza et al.. 2007).
10                   The increase in  [Ca2+]j was associated with higher osmotic fragility and modifications in
11                   erythrocyte shape. In a separate investigation, when erythrocytes from 10 healthy
12                   volunteers were exposed (in vitro) at concentrations of 0.2 to 6.0 (iM Pb for 24 or 120
13                   hours, concentration-related increases in [Ca2+]j were observed across all concentrations
14                   for both durations of exposure (Quintanar-Escorza et al.. 2010). Subsequent exposures of
15                   erythrocytes to either 0.4 or 4.0 (iM Pb [corresponding to 10 or 80 (ig/dL in exposed
16                   workers (Quintanar-Escorza et al.. 2007)]  for 12-120 hours resulted in duration-related
17                   increases with durations >12 hours. Osmotic fragility (measured as percent hemolysis)
18                   was increased in erythrocytes exposed to 0.4 (iM Pb for 24 hours. Co-incubation with a
19                   vitamin E analog mitigated these effects, indicating that the increase in [Ca2+]j is
20                   dependent on the oxidative state of the erythrocytes. Shin et al. (2007) observed that
21                   incubation of human erythrocytes with 5 (iM Pb for 1 hour resulted in a 30-fold increase
22                   in [Ca2+]j in vitro, inducing the pro-coagulant activity of exposed erythrocytes. Induction
23                   of pro-coagulant activity in erythrocytes could lead to thrombus formation and negatively
24                   contribute to overall cardiovascular health; whereas increased osmotic fragility could
25                   substantially reduce erythrocyte life span and ultimately lead to anemic conditions.

26                   Similar to effects seen in erythrocytes, Pb has been observed to interfere with Ca2+
27                   homeostasis in platelets and white blood cells. Dowd and Gupta (1991) observed that
28                   1 (iM Pb (for 3.5 hours) was the lowest exposure concentration to result in increases in
29                   [Ca2+]j in human platelets (in vitro). The observed increase in  [Ca2+]j levels was attributed
30                   to the increased influx of external Ca2+, possibly through ligand-gated Ca2+ channels. In
31                   mouse  splenic lymphocytes,  1 (iM Pb was the lowest exposure concentration found to
32                   increase [Ca2+]j  with incubation periods of 10 minutes or greater (Li et al., 2008c). These
33                   increases in [Ca2+]j appeared to be reversible as [Ca2+]j returned to baseline after one
34                   hour. Pretreatment with a calmodulin antagonist slightly mitigated the  effects of Pb
35                   exposure, indicating a role for calmodulin in disruption of Ca2+ homeostasis by Pb
36                   exposure in lymphocytes. In rat tail arteries exposed to  1.2 (iM Pb acetate for 1 hour,
37                   [Ca2+]j increased over controls, possibly through increased transmembrane influx of
38                   external Ca2+ (Piccinini et al.. 1977).
      November 2012                              5-5                   Draft - Do Not Cite or Quote

-------
 1                  Exposure of the microsomal fraction (prepared from rat brain cells) to as little as 0.25
 2                  Pb for 2 minutes resulted in increased release of Ca2+ into the culture medium (Pentvala
 3                  et al., 2010). Further, Pb exposure also decreased the activity of microsomal Ca2+ATPase,
 4                  thus decreasing the sequestration of Ca2+ into microsomes. The results of this study
 5                  suggest that disruption of microsomal release and re-uptake of Ca2+ may alter Ca2+
 6                  homeostasis, ultimately leading to altered signal transduction and neuronal dysfunction.
 7                  However, Ferguson et al. (2000) observed that [Ca2+]j was decreased in rat hippocampal
 8                  neurons in response to exposure to 0.1 (iM Pb for 1-48 hours; although the observed
 9                  decreases were not time-dependent. The decrease in [Ca2+]j was shown to be due to
10                  increased efflux of Ca2+ out of the neuron via a calmodulin-regulated mechanism,
11                  possibly through stimulated Ca2+ efflux via Ca2+ATPase.

12                  Pb exposure has been shown to disrupt [Ca2+]j levels in multiple cell types including
13                  osteoblasts, erythrocytes, platelets, and neuronal cells. This alteration in Ca2+ homeostasis
14                  could potentially affect cell signaling and disrupt the normal physiological function of
15                  these cells.
                    5.2.2.2     Disruption of Ion Transport Mechanisms

16                  As described above, deregulation of Ca2+ homeostasis can result in negative effects in
17                  multiple organ systems. Under normal conditions in the life cycle of most cells, cytosolic
18                  concentrations of free Ca2+ fluctuate between approximately 100 to 200 nM and Ca2+that
19                  has entered the cell must be removed in order to maintain normal homeostatic
20                  concentrations (Carafoli, 2005). An important component in the maintenance of Ca2+
21                  homeostasis is transmembrane transport of Ca ions via Ca2+ATPase and voltage-gated
22                  Ca2+ channels (Carafoli. 2005). Pb has been shown to disrupt the normal movement of
23                  Ca2+ ions, as well as other physiologically important ions through interactions with these
24                  transport mechanisms.

25                  Multiple studies have reported alterations in the activity of Na+/K+ATPase, Ca2+ATPase,
26                  and Mg2+ATPases after Pb exposure in animal models. Decreases in the activity of all
27                  three ATPases were observed in the kidneys and livers of rats exposed to 750 ppm Pb in
28                  drinking water for 11 weeks (mean [SD] blood Pb level: 55.6 [6.3] (ig/dL) (Kharoubi et
29                  al., 2008a) and in erythrocytes from rats exposed to 2,000 ppm Pb in drinking water for
30                  5 weeks (mean [SD] blood Pb level: 97.56 [11.8] jig/dL) (Sivaprasad et al.. 2003).
31                  Increases in lipid peroxidation were seen in both studies, and the decrements in ATPase
32                  activities may be explained by generation of free radicals in Pb-exposed animals. A
33                  decrease in the activity of Na+/K+ATPase was observed in rabbit kidney membranes
34                  exposed to 0.01 to 10 (iM Pb, possibly due to Pb inhibiting the hydrolytic cleavage of
      November 2012                             5-6                  Draft - Do Not Cite or Quote

-------
 1                  phosphorylated intermediates in the K-related branch of the pump (Gramigni et al.,
 2                  2009). Similar decreases in Na+/K+ATPase activity were observed in brain synaptosomes
 3                  isolated from rats that were exposed to 200 ppm Pb in drinking water for 3 months (blood
 4                  Pb level: 37.8 ng/dL) (Rafalowska et al.. 1996) or 15 mg Pb/kg injected (i.p.) for 7 days
 5                  (blood Pb level: 112.5 jig/dL) (Struzynska et al..  1997a). Inhibition of Na+/K+ATPase
 6                  activity was also observed in primary cerebellar granule neuronal cultures obtained from
 7                  rat pups that were pre- and post-natally (to PND8) exposed to Pb (1,000 ppm Pb acetate
 8                  in dams' drinking water, resulting in blood Pb level of 4 (ig/dL) (Baranowska-Bosiacka
 9                  et al.. 20 lib). The activity of Ca2+ATPase in the sarcoplasmic reticulum of rabbits
10                  exposed to 0.01 (iM Pb was similarly decreased (Hechtenberg and Beyersmann. 1991).
11                  The inhibitory effect of Pb was diminished in the presence of high Mg-ATP
12                  concentrations. The activity of generic ATPase was reported to be altered in the testes of
13                  rat pups exposed to 300 ppm (mg/L) Pb acetate, both during lactation and in drinking
14                  water after weaning to the age of 6, 8, 10, or 12 weeks (Liu et al., 2008). In pregnant rats
15                  fed a Pb-depleted (20 ± 5 ng/kg) or control (1 mg/kg) diet during gestation and lactation,
16                  no difference was observed in the activity of Na+/K+ATPase and Ca2+/Mg2+ATPase in the
17                  parental generation (Eder et al..  1990). However, the offspring (exposed via placental and
18                  lactational transfer of Pb) of Pb-depleted rats displayed  decreased activities in both
19                  enzymes compared with offspring of rats with higher Pb exposures. An increase in the
20                  Na+/K+ATPase activity was observed in rats treated (i.p.) with 20 mg/kg Pb for
21                  14 consecutive days (Jehan and Motlag. 1995). Co-exposure of Pb with Zn and Cu
22                  greatly attenuated the increase in ATPase activity. Although the precise mechanism was
23                  not investigated, Navarro-Moreno et al. (2009) reported that Ca2+ uptake was diminished
24                  in proximal renal tubule cells in rats chronically exposed to 500 ppm Pb in drinking water
25                  for 7 months (mean [SD] blood Pb level: 43.0 [7.6] (ig/dL).

26                  In vitro studies of ATPase activities in human erythrocyte ghosts have also shown that Pb
27                  affects the transport of metal ions across membranes. Calderon-Salinas et al.  (1999a)
28                  observed that 1-5 x  103 (iM Pb and Ca2+ were capable of inhibiting the passive transport
29                  of each other in human erythrocyte ghosts incubated with both cations.  Subsequent
30                  inhibition experiments indicated that both cations share  the same electrogenic transport
31                  pathway (Sakuma et al., 1984). Further study by this group (Calderon-Salinas et al.,
32                  1999b) demonstrated that Pb can noncompetitively block the transport of Ca2+ by
33                  inhibiting the activity of Ca2+/Mg2+ATPase at concentrations of 1-5 x 103 (iM. Mas-Oliva
34                  (1989) demonstrated that the activity of Ca2+/Mg2+ATPase in human erythrocyte ghosts
35                  was inhibited by incubation with 0.1-100 (iM Pb. The inhibitory action was most likely
36                  due to direct reaction with sulfhydryl groups on the ATPase enzyme at Pb concentrations
37                  greater than 1 \iM, but due to the action of Pb on calmodulin at lower concentrations.
38                  Grabowska and Guminska (1996) observed that 10 (ig/dL was the lowest
39                  Pb concentration to decrease the activity of Na+/K+ATPase in erythrocyte ghosts; activity

      November 2012                             5-7                  Draft - Do Not Cite or Quote

-------
 1                  of Ca2+/Mg2+ATPase was less sensitive to Pb exposure, and Mg2+ATPase activity was not
 2                  affected.

 3                  Effects on ATPase activity are also observed in association with blood Pb levels in
 4                  human populations. In a study investigating ATPase activities in Pb-exposed workers in
 5                  Nigeria, Abam et al. (2008) observed that the activity of erythrocyte membrane-bound
 6                  Ca2+/Mg2+ATPase was decreased by roughly 50% in all occupational groups (range of
 7                  mean [SD] blood Pb level across nine occupational groups: 28.75  [11.31] to 42.07
 8                  [12.01] (ig/dL) compared to nonexposed controls (mean [SD] blood Pb level: 12.34
 9                  [2.44] in males and 16.85 [6.01] (ig/dL in females). Higher membrane concentrations of
10                  Ca2+ and Mg2+ were also observed, indicating that Pb prevented the efflux of those
11                  cations from the cell, most likely by substituting for those metals in the active site of the
12                  ATPase. In a study of 247 mother-newborn pairs, Campagna et al. (2000) observed that
13                  newborn (cord) blood Pb (geometric mean [5th-95th percentile]: 4.8 [2.8-9.2] (ig/dL) was
14                  negatively and significantly associated with maternal blood Ca2+ pump activities;
15                  however, newborn (cord) blood Pb was not significantly associated with newborn (cord)
16                  blood Ca2+ pump activities. Newborn hair Pb (geometric mean [5th-95th percentile]: 1.1
17                  [0.1-8.0] (ig/g) was negatively and significantly associated with both maternal and
18                  newborn (cord) blood Ca2+ pump activities. In a population of 81 newborns, Huel et al.
19                  (2008) found that newborn hair and newborn (cord) blood Pb levels (mean [SD] newborn
20                  hair Pb and newborn [cord] blood Pb levels: 1.22 [1.41] (ig/g and  3.54 [1.72] (ig/dL)
21                  were negatively associated with Ca2+ATPase activity in plasma membranes of
22                  erythrocytes isolated from newborn (cord) blood; newborn hair Pb levels were more
23                  strongly associated with newborn (cord) Ca2+ pump activity than were newborn (cord)
24                  blood Pb levels.

25                  Pb has also been shown to disrupt cation transport mechanisms through direct action on
26                  voltage-gated cation channels. Audesirk and Audesirk (1993. 1991) demonstrated that
27                  extracellular free Pb inhibits the action of multiple voltage-gated Ca2+ channels, with free
28                  Pb IC50 (half maximal inhibitory concentration) values of 0.7 (iM  for L-type channels and
29                  1.3 (iM for T-type channels in neuroblastoma cells maintained in culture media, and IC50
30                  values as low as 0.03 (iM for L-type channels in cultured hippocampal neurons.  Sun and
31                  Suszkiw (1995) corroborated the inhibitory action of extracellular Pb on voltage-gated
32                  Ca2+ channels, demonstrating an IC50 value of 0.3 (iM in bovine adrenal chromaffin cells.
33                  The observed disruption of the voltage-gated Ca2+ channels most likely reflects
34                  competition between Pb and Ca2+ for the  extracellular Ca2+ binding domain of the
35                  channel.  Research by other laboratories supported these findings:  Pb inhibited the action
36                  of multiple Ca2+ channels in human embryonic kidney cells transfected with L-, N-, and
37                  R-type channels (IC50 values of 0.38 (JVI, 1.31  (iM, and 0.10 (iM,  respectively) (Peng et
38                  al.. 2002) and P-type channels in cultured hippocampal neurons at concentrations up to
      November 2012                             5-8                  Draft - Do Not Cite or Quote

-------
 1                  3 (iM (Ujihara et al., 1995). However, in bovine adrenal chromaffin cells, intracellular Pb
 2                  was observed to enhance Ca2+ currents through attenuation of the Ca2+ dependent
 3                  deactivation of Ca2+ channels at an EC50 value of 200 (iM, possibly through blocking the
 4                  intracellular Ca2+ binding domain, or through Ca2+ dependent dephosphorylation of the
 5                  channel (Sun and Suszkiw.  1995). Recently, Pb has also been shown to enter cells
 6                  (HEK293, HeLa, and PC 12 cell lines) through store-operated Ca2+ channels (Chiu et al..
 7                  2009; Chang et al., 2008b).  In particular, the Orail-STIMl complex was shown to be
 8                  critical in the entry of Pb ions into cells, and increased Pb permeation was directly related
 9                  to decreased [Ca2+]j concentrations at exposure concentrations as low as 0.1
10                  Pb also has been found to disrupt the action of Ca2+-dependent K+ channels. Alvarez et al.
1 1                  (1986) observed that Pb promoted the efflux of K+ from inside-out erythrocyte vesicles in
12                  a concentration-dependent manner at concentrations of 1-300 (iM, either through action
13                  on a Mg2+ modulatory site or through direct interaction with the Ca2+ binding site. Fehlau
14                  et al. (1989) also demonstrated Pb-induced activation of the K+ channel in erythrocytes.
15                  However, Pb only activated the K+ channels at concentrations below 10 (iM; higher
16                  concentrations of Pb completely inhibited channel activity, indicating the modulation of
17                  K+ permeability is due to concentration dependent alterations in channel gating. Silken et
18                  al. (2001) observed that Pb activated K+ channels in erythrocytes from the marine teleost
19                  Scorpaena porcus in a concentration-dependent manner after a 20-minute incubation;
20                  minor loss of K+ was seen at Pb concentrations of 1-2 (iM, whereas exposure to
21                  20-50 (iM Pb resulted in approximately 70% K+ loss. Competitive and inhibitory binding
22                  assays suggest that Pb directly activates K+ channels in S. porcus.


                    Disruption  of Neurotransmitter Release

23                  Pb has been shown to inhibit the evoked release of neurotransmitters by inhibiting Ca2+
24                  transport through voltage-gated channels in in vitro experiments (Cooper and Manalis.
25                  1984; Suszkiw et al.. 1984). However, in these same experiments, concentrations of Pb
26                  >  5 (iM were also observed to actually increase the spontaneous release of
27                  neurotransmitters. Subsequent research by other groups affirmed that Pb demonstrates
28                  Ca2+-mimetic properties in enhancing neurotransmitter release from cells in the absence
29                  of Ca2+ and Ca2+-induced depolarization. Tomsig and Suszkiw (1993) reported that Pb
30                  exposure induced the release of norepinephrine (NE) from bovine adrenal chromaffin
3 1                  cells, and was considerably more potent (as measured by half-maximal metal -dependent
32                  release [K0 5]) than was Ca2+ (K0 5 of 4.6 x 10'3 jiM for Pb versus 2.4 jiM for Ca2+).
33                  Activation of PKC was observed to enhance the Pb-induced release of NE (Tomsig and
34                  Suszkiw. 1995). Westerink and Vijverberg (2002) observed that Pb acted as a high
35                  affinity substitute for Ca2+, and triggered enhanced catecholamine release from PC 12
36                  cells at 10 (iM in intact cells and 0.03 (iM in permeabilized cells. The suppression of

      November 20 12                              5-9                   Draft - Do Not Cite or Quote

-------
 1                  Ca2+-evoked release of neurotransmitters combined with the ability of Pb to enhance
 2                  spontaneous releases could result in higher noise observed in the synaptic transmission of
 3                  nerve impulses in Pb-exposed animals.

 4                  In rats exposed to Pb at concentrations of 1,000-10,000 ppm in drinking water beginning
 5                  at gestational days GDI5-GDI6 and continuing to postnatal days PND120, decreases in
 6                  total K+-stimulated hippocampal gamma aminobutyric acid (GABA) release were seen at
 7                  exposure levels of 1,000-5,000 ppm (range of mean [SD] blood Pb levels: 26.8 [1.3] -
 8                  61.8 [2.9] (ig/dL) (Lasley and Gilbert. 2002). Maximal effects were observed at
 9                  2,000 ppm Pb in drinking water, but effects were less evident at 5,000 ppm, and were
10                  absent at 10,000 ppm. In the absence of Ca2+, K+-induced GABA release  was increased
11                  with the two highest Pb exposure concentrations, suggesting a Pb-induced enhancement
12                  of K+-evoked release of GABA. The authors suggest that this pattern of response
13                  indicates that Pb is a potent suppressor of K+-evoked release at low concentrations, but a
14                  Ca2+ mimic in regard to independently inducing exocytosis and evoking neurotransmitter
15                  release at higher concentrations (Lasley and Gilbert. 2002). Suszkiw (2004) reports that
16                  augmentation of spontaneous release of neurotransmitters may involve Pb-induced
17                  activation of CaMKII-dependent phosphorylation of synapsin I or direct activation of
18                  synaptotagmin I. Further, Suszkiw (2004) suggests that unlike the intracellularly
19                  mediated effects of Pb on spontaneous release of neurotransmitters, Pb-induced inhibition
20                  of evoked transmitter releases is largely due to extracellular blockage of the voltage-gated
21                  Ca2+ channels.

22                  In summary, Pb has been shown to disrupt ion transport mechanisms in toxicological and
23                  epidemiologic studies. Specific mechanisms disrupted include various cation-specific
24                  ATPases and voltage-gated cation channels. Alterations in ion transport functions have
25                  also been shown to disrupt neurotransmitter release in both in vivo and in vitro
26                  experiments.
                    5.2.2.3      Displacement of Metal Ions and Perturbed Protein
                                 Function

27                  The binding of metal ions to proteins causes specific changes in protein shape, and these
28                  conformational changes may alter specific cellular function of many proteins (Kirberger
29                  and Yang. 2008). Metal binding sites on proteins are generally ion-specific and are
30                  influenced by multiple factors, including binding geometries, ligand preferences, ionic
31                  radius, and metal coordination numbers (Kirberger and Yang. 2008; GarzaetaL 2006).
32                  The coordination chemistry that normally regulates metal-protein binding makes many
33                  proteins particularly susceptible to perturbation from Pb, as it is able to function with
      November 2012                             5-10                  Draft - Do Not Cite or Quote

-------
 1                   flexible coordination numbers and can bind multiple ligands (Kirberger and Yang. 2008;
 2                   Garza et al.. 2006). However, due to differences in its physical properties, Pb induces
 3                   abnormal conformational changes when it binds to proteins (Kirberger and Yang, 2008;
 4                   Bitto et al.. 2006; Garza et al.. 2006; Magyar et al.. 2005). and these structural changes
 5                   elicit altered protein function. It is known that [Ca2+]j is an important second messenger
 6                   in cell signaling pathways, and operates by binding directly to and activating proteins
 7                   such as calmodulin and PKC (Goldstein. 1993). Alterations in the functions of both of
 8                   these proteins due to direct interaction with Pb have been well documented in the
 9                   literature.

10                   PKC is a family of serine/threonine protein kinases critical for cell signaling and
11                   important for cellular processes, including growth and differentiation (Goldstein. 1993).
12                   PKC contains a "C2" Ca2+-binding domain and requires binding of the cation, as well as
13                   the presence of diacylglycerol and phospholipids, for proper cellular activity (Garza et
14                   al.. 2006).  Markovac and Goldstein (1988b) observed that, in the absence of Ca2+,
15                   exposure to 10"6 (iM concentrations of Pb for 5 minutes directly activated PKC purified
16                   from rat brains. The activation of PKC by Pb was more potent than was Ca2+-dependent
17                   activation by five orders of magnitude.  Long et al. (1994)  affirmed these findings,
18                   reporting that Pb had a Kact 4,800 times smaller than that of Ca2+ (5.5 x 10"5 (iM versus
19                   25 (iM, following  a 3 minute exposure). However, Ca2+ had a higher maximal activation
20                   of PKC than did Pb. This possibly indicates the presence of multiple Ca2+-binding sites
21                   on the protein, and that Pb may bind the first site more efficiently than does Ca2+, but not
22                   subsequent sites. Tomsig and Suszkiw (1995) further demonstrated the ability of Pb to
23                   activate PKC in bovine adrenal chromaffin cells incubated with 10"6 (iM concentrations
24                   of Pb for 10 minutes but also reported that activation of PKC by Pb was only partial
25                   (approximately 40% of the maximum activity induced by  Ca2+) and tended to decrease at
26                   concentrations >1  x 10"3 (iM.

27                   Contrary to the above findings, Markovac and Goldstein (1988a) observed that Pb and
28                   Ca2+ activated PKC at equivalent concentrations and efficacies when broken cell
29                   preparations of rat brain microvessels were  incubated with either cation for 45 minutes.
30                   However, when PKC activation was investigated in whole vessel preparations, no
31                   activation was observed, but PKC did become redistributed from the cytosolic to the
32                   particulate fraction after centrifugation. This suggests that Pb redistributes PKC at (iM
33                   concentrations, but does not activate the protein in brain microvessels. In human
34                   erythrocytes exposed to Pb acetate for 60 minutes, the amount of PKC found in
35                   erythrocyte membranes and  total PKC activity was increased at concentrations greater
36                   than 0.1 (iM (Belloni-Olivi et al., 1996). The observation that neither Ca2+ nor
37                   diacylglycerol concentrations were increased due to Pb exposure,  indicates that
38                   Pb-induced activation of PKC is due to direct interaction with the protein. Pb-induced
      November 2012                             5-11                  Draft - Do Not Cite or Quote

-------
 1                  alterations in PKC have also been observed in other tissues, including increased activity
 2                  in rabbit mesenteric arteries at 10~6 (iM concentrations of Pb (Watts et al.. 1995; Chai and
 3                  Webb, 1988) and human erythrocytes from Pb-exposed workers (range of blood Pb
 4                  levels: 5.4 to 69.3 (ig/dL) (Hwang et al.. 2002). and decreased activity in mouse
 5                  macrophages and the rat brain cortex at (iM concentrations (Murakami et al., 1993; Lison
 6                  etal.. 1990).

 7                  Calmodulin is another important protein essential for proper Ca2+-dependent cell
 8                  signaling. Calmodulin contains an "EF-hand" Ca2+ binding domain, which is dependent
 9                  on the cation for proper activity (GarzaetaL 2006). Calmodulin regulates events as
10                  diverse as cellular structural integrity, gene expression, and maintenance of membrane
11                  potential (Vetter and Leclerc, 2003; Saimi and Kung. 2002). Habermann et al. (1983)
12                  observed that exposure to Pb altered numerous cellular functions of calmodulin,
13                  including activation of calmodulin-dependent phosphodiesterase activity after 10 minutes
14                  incubation (minimal activation at 0.1 (iM, EC50 = 0.5-1.0 (JVI), stimulation of brain
15                  membrane phosphorylation at Pb concentrations greater than 0.4 (iM after 1  minute
16                  incubation, and increased binding of calmodulin to brain membranes at Pb concentrations
17                  greater than 1 (iM after 10 minutes incubation. Habermann et al. (1983) reported that the
18                  affinity of Pb for Ca2+-binding sites on calmodulin was approximate to that of Ca2+ itself
19                  (Kd ~20 (JVI), whereas Richardt et al. (1986) observed that Pb was slightly more potent
20                  than Ca2+ was at binding calmodulin (IC50 =11 and 26 (iM, respectively). Both studies
21                  indicated that Pb was much more effective at binding to calmodulin than was any other
22                  metal cation investigated (e.g., Hg, Cd, Fe). Kern et al. (2000) observed that Pb was more
23                  potent in binding to, and affecting conformational changes in, calmodulin compared to
24                  Ca2+ (EC50 values of 4-5.5 x  10'4 jiM [threshold = 1 x  10'4 jiM] and 0.45-0.5 jiM
25                  [threshold = 0.1 |iM], respectively). Pb, in the absence of Ca2+, was also observed to
26                  activate calmodulin-dependent cyclic nucleotide phosphodiesterase activity at much
27                  lower concentrations compared to Ca2+ (EC50 value 4.3 x 10"4 (iM  [threshold = 3  x
28                  10"4 (iM] versus EC50 1.2 x 10"3 \iM [threshold = 0.2 (iM; 50 minute incubation]). When
29                  incubated with physiological concentrations of Ca2+, Pb induced phosphodiesterase
30                  activity at concentrations as low as 5 x 10"5 (iM. Pb activated calcineurin, a Ca2+-
31                  dependent phosphatase with widespread distribution in the brain and immune system, at
32                  threshold concentrations as low as 2 x  10"5 (iM in the presence of Ca2+ (incubation
33                  time = 30 minutes), but inhibited its activity at concentrations greater than 2 x 10"4  (iM
34                  (Kern and Audesirk. 2000). Thus, 10"6 (iM concentrations of intracellular Pb appear to
35                  amplify the  activity of calmodulin and thus can be expected to alter intracellular Ca2+
36                  signaling in exposed cells (Kern et al.. 2000). Mas-Oliva (1989) observed that
37                  low-exposure (<1 (iM,  20 minute incubation) stimulatory effects of Pb exposure on the
38                  activity of Ca2+/Mg2+ATPase was due to Pb binding to calmodulin and subsequent
39                  activation of the ion pore. Ferguson et al. (2000) observed that exposure of rat

      November 2012                             5-12                  Draft - Do Not Cite or Quote

-------
 1                  hippocampal neurons to Pb for 1 to 48 hours resulted in increased activation of a
 2                  calmodulin-dependent Ca2+ extrusion mechanism.

 3                  Pb has also been observed to alter the activity of other proteins that rely on Ca2+ binding
 4                  for normal cellular function. Osteocalcin is a matrix protein important in bone resorption,
 5                  osteoclast differentiation, and bone growth; and has three Ca2+-binding sites  (Dowd et al..
 6                  2001). Incubation of osteocalcin in solution with Ca2+ and Pb resulted in the  competitive
 7                  displacement of Ca2+ by Pb (Dowdetal.. 1994). Pb was found to bind to osteocalcin
 8                  more than 1,000 times more tightly than was Ca2+ (Kd = 1.6 x  10"2 (iM versus 7.0 (iM,
 9                  respectively), and analysis with nuclear magnetic resonance (NMR) indicated that Pb
10                  induced similar, though slightly different, secondary structures in osteocalcin, compared
11                  to Ca2+. The authors hypothesized that the observed difference in Pb-bound osteocalcin
12                  structure may explain previous findings in the literature that Pb exposure reduced
13                  osteocalcin adsorption to hydroxyapatite  (Dowd etal. 1994). Further research by this
14                  group also found that Pb binded osteocalcin approximately 10,000-times more tightly
15                  than did Ca2+ (Kd = 8.5 x 1Q"2 jiM versus 1.25 x 1Q3 jiM, respectively) (Dowd et al..
16                  2001). However, the authors reported that Pb exposure actually caused increased
17                  hydroxyapatite adsorption at concentrations 2-3 orders of magnitude lower than that seen
18                  with Ca2+. Additionally, Pb can displace Ca2+ in numerous other Ca2+-binding proteins,
19                  such as proteins important in muscle contractions, renal Ca2+ transport and
20                  neurotransmission, including troponin C, parvalbumin, CaBP I and II, phospholipase A2,
21                  and synaptotagmin I, at concentrations as low as the 10"3 (iM range (Bouton  etal.. 2001;
22                  Osterode and Ulberth. 2000: Richardt et al.. 1986).

23                  Pb can displace metal cations  other than Ca2+that are requisite for protein function. One
24                  of the most researched targets for molecular toxicity of Pb is the second enzyme in the
25                  heme synthetic pathway, aminolevulinic acid dehydratase (ALAD). ALAD contains four
26                  Zn-binding sites and all four need to be occupied to confer full enzymatic activity
27                  (Simons. 1995). ALAD has been identified as the major protein binding target for Pb in
28                  human erythrocytes (Bergdahl et al., 1997a). and blood Pb levels are associated with
29                  inhibition of the enzyme in the erythrocytes of Pb-exposed workers and adolescents
30                  (blood Pb levels > 10 (ig/dL) (Ahamed et al.. 2006; Ademuyiwa et al.. 2005b). in lysed
31                  human erythrocytes exposed to Pb in vitro for 60 minutes (K; = 7 x 10"6 (iM) (Simons.
32                  1995). and in rats exposed to 25 mg/kg Pb once a week for 4 weeks (mean [SD] blood Pb
33                  level: 6.56 [0.98] (ig/dL) (Lee et al.. 2005). Additional experiments indicated that lower
34                  concentrations of Zn result in  greater inhibition of enzyme activity by Pb, suggesting a
35                  competitive inhibition between Zn and Pb at a single site (Simons. 1995).

36                  Zn-binding domains are also found in transcription factors and proteins necessary for
37                  gene expression, including GATA proteins and transcription factors TFIIIA, Spl, and
      November 2012                             5-13                   Draft - Do Not Cite or Quote

-------
 1                  Erg-1 (Gheringetal.. 2005; Huang et al.. 2004; Reddv and Zawia. 2000; Hanas et al..
 2                  1999; Zawia et al.. 1998). Pb was found to form tight complexes with the cysteine
 3                  residues in GATA proteins (Pb stoichiometric stability constant (CF(31pb) = 6.4 (± 2.0)x
 4                  109 M'1 for single C-terminal GATA Zn finger from chicken and DFp2Pb2 = 6.3 (± 6.3) x
 5                  1019 M"2 for double-GATA Zn finger from human), and was able to displace bound Zn
 6                  from the protein under physiologically relevant conditions (Ghering et al.. 2005). Once
 7                  Pb was bound to GATA proteins, they displayed decreased ability to bind to DNA (Pb
 8                  concentrations > 1.25 (iM) and activate transcription. Pb at a minimum concentration of
 9                  10 (iM also binds to the Zn domain of TFIIIA, inhibiting its ability to bind DNA at
10                  concentrations (Huang et al.. 2004; Hanas et al.. 1999). Huang et al. (2004) also reported
11                  that exposure to Pb caused the dissociation of TFIIIA-DNA adducts and using NMR
12                  spectroscopy,  found that altered TFIIIA activity was the result of a Pb-induced abnormal
13                  protein conformation.

14                  Pb exposure modulated the DNA-binding profiles of the transcription factors Spl and
15                  Erg-1 in rat pups exposed to 2,000 ppm Pb acetate via lactation, resulting in a shift in
16                  DNA-binding toward early development (i.e., the first week following birth) (Reddy and
17                  Zawia. 2000; Zawia et al.. 1998). The shifts in Spl DNA-binding profiles were shown to
18                  be associated with abnormal expression of genes related to myelin formation
19                  (Section 5.2.7.5). Further mechanistic research utilizing a synthetic peptide containing a
20                  Zn finger motif demonstrated that Pb can bind the histidine and cysteine residues of the
21                  Zn finger motif, thus displacing Zn and resulting in an increase in the DNA-binding
22                  efficiency of the synthetic peptide (Razmiafshari et al..  2001; Razmiafshari and Zawia.
23                  2000). However, in DNA-binding assays utilizing recombinant Spl (which has three Zn
24                  finger motifs,  opposed to only one in the synthetic peptide), 37 (iM Pb was  the lowest
25                  concentration  observed to abolish the DNA-binding capabilities of Spl (Razmiafshari
26                  and Zawia. 2000).

27                  Pb has also been reported to competitively inhibit Mg binding and thus inhibit the
28                  activities of adenine and hypoxanthine/guanine phosphoribosyltransferase in erythrocyte
29                  lysates from rats exposed to 1,000 ppm Pb in drinking water for 9 months (mean [SD]
30                  blood Pb level: 7.01 [1.64] (ig/dL) and in in vitro human erythrocyte lysates exposed to
31                  0.1 (iM Pb for as little as 5 minutes (Baranowska-Bosiacka et al.. 2009). and cGMP
32                  phosphodiesterase at 10"6 (iM concentrations in homogenized bovine retinas (Srivastava
33                  et al.. 1995). Pb was also reported to inhibit pyrimidine 5'-nucleotidase through
34                  competitive inhibition of Mg binding, resulting in conformational changes and improper
35                  amino acid positioning in the active site (Bitto et al.. 2006).
      November 2012                             5-14                  Draft - Do Not Cite or Quote

-------
 1                  In summary, Pb has been shown to displace metal cations from the active sites of
 2                  multiple enzymes and proteins, and thus to alter the functions of those proteins in
 3                  occupationally-exposed humans with blood Pb levels of 5.4-69.3 (ig/dL, in adult rodents
 4                  with blood Pb levels of 6.5 (ig/dL (exposure 4 weeks), in suckling rats exposed to
 5                  2,000 ppm Pb via lactation, and in cell-free  and cellular in vitro experiments conducted at
 6                  exposure concentrations ranging from 10"6 (iM to 1  (iM. These alterations in protein
 7                  function have  implications for numerous cellular and physiological processes, including
 8                  cell signaling, growth and differentiation, gene expression, energy metabolism, and
 9                  biosynthetic pathways. Table 5-1 provides a list of enzymes and proteins whose function
10                  may be perturbed by Pb exposure.
      November 2012                             5-15                  Draft - Do Not Cite or Quote

-------
Table 5-1    Enzymes and proteins potentially affected by exposure to Pb and the
            metal cation cofactors necessary for their proper physiological
            activity.

Enzymes
Ion Channels/
Transport
Signal
Transduction
Pb Binding
DMA Binding
Metalloprotein/Enzyme
Aminolevulinic acid
dehydratase
Ferrochelatase
Superoxide dismutase
Catalase
Glutathione peroxidase
Guanylate cyclase
cGMP phosphodiesterase
NAD synthase
NAD(P)H oxidase
Pyrimidine 5'-nucleotidase
Erythrocyte
phosphoribosyltransferase
ATPase
Mitochondrial
transmembrane pore
Calcium-dependent
potassium channel
Protein kinase C
Calmodulin
Metallothionein
GATAtranscriptional factors
a^ indicates increased activity; J, indicates decreased
Direction
1
1
IT
IT
IT
1
1
1
T
1
1
IT
T
T
IT
T
T
1
activity; J,f
of Action3 Metal Cation; Reference
Zn; Simons (1995)
Fe (2Fe-2S Cluster);
Crooks et al. (2010)
Mn, Cu, Zn, Fe;
Antonyuk et al. (2009),
Borgstahl et al. (1992)
Fe (Heme); Putnam et al. (2000)
Se; Rotruck et al. (1973)
Fe (Heme);
Boerrigter and Burnett (2009)
Mg, Zn; Ke (2004)
Mg; Hara et al. (2003)
Ca2+; Leseney et al. (1999)
Mg, Ca2+;
Bitto et al. (2006).
Amici et al. (1997).
Paglia and Valentine (1975)
Mg (Mn, Ca2+, Co, Ni, Zn);
Dengetal. (2010),
Arnold and Kelley (1978)
Ca2+, Mg, Na/K; Technische
Universitat Braunschweig (2011)
Ca2+; He et al. (2000)
Ca2+;
Silkin et al. (2001),
Alvarez et al. (1986)
Ca2+; Garza et al. (2006)
Ca2+; Garza et al. (2006)
Zn, Cu; Yu et al. (2009)
Zn;
Hanas et al. (1999),
Huang et al. (2004)
indicates activity can be alternatively increased or
decreased.
November 2012
5-16
Draft - Do Not Cite or Quote

-------
                    5.2.2.4     Mitochondrial Abnormality

 1                  Alterations in mitochondrial function, including disruptions in ion transport,
 2                  ultrastructural changes, altered energy metabolism, and perturbed enzyme activities due
 3                  to Pb exposure are well documented in the scientific literature. Exposure of rats to Pb in
 4                  feed (10,000 ppm Pb for 4, 6, 8, 10, 12, or 20 weeks) or drinking water (300 ppm for
 5                  8 weeks, 500 ppm for 7 months, or 10,000 ppm Pb for 9 months) resulted in gross
 6                  ultrastructural changes in renal tubule mitochondria and epididymal mitochondria
 7                  characterized as a general swollen appearance with frequent rupture of the outer
 8                  membrane, distorted cristae, loss of cristae, frequent inner compartment vacuolization,
 9                  observation of small inclusion bodies, and fusion with adjacent mitochondria (Wang et
10                  al..2010d: Marchlewicz et al.. 2009; Navarre-Moreno et al.. 2009; Gover. 1968; Gover et
11                  al.. 1968).

12                  Transmembrane mitochondrial ion transport mechanisms have been found to be
13                  perturbed by exposure to Pb. Pb inhibits the uptake of Ca2+ into mitochondria (Parr and
14                  Harris. 1976). while simultaneously stimulating the efflux of Ca2+ out of the organelle
15                  (Simons. 1993a). thus disrupting intracellular/mitochondrial Ca2+ homeostasis. Pb
16                  exposure has also been shown to decrease the mitochondrial transmembrane potential in
17                  primary cerebellar granule neuronal cultures from rats exposed to 1,000 ppm Pb in
18                  drinking water throughout gestation and lactation (Baranowska-Bosiacka et al., 201 Ib).
19                  astroglia incubated with 0.1 or 1.0 (iM Pb for 14 days (Legare etal..  1993). proximal
20                  tubule cells exposed to 0.25, 0.5, and 1.0  (iM for 12 hours (Wang et al..  2009c). and
21                  retinal rod photoreceptor cells incubated with 0.01 to 10 (iM for 15 minutes (He et al..
22                  2000). Further research indicated that Pb-induced mitochondrial swelling and decreased
23                  membrane potential is the result of the opening of a mitochondrial transmembrane pore
24                  (MTP), possibly by directly binding to the metal (Ca2+)-binding  site on the matrix side of
25                  the pore (Bragadin et al.. 2007;  He et al.. 2000). Opening of the MTP is  the first step of
26                  the mitochondrial-regulated apoptotic cascade pathway in many cells (Rana. 2008;
27                  Lidsky and Schneider. 2003). He et al. (2000) additionally observed other indicators of
28                  apoptosis including, cytochrome c release from mitochondria, and caspase-9 and -3
29                  activation following exposure of retinal rod cells to Pb. Induction of mitochondrially-
30                  regulated apoptosis via stimulation of the  caspase cascade following exposure to Pb has
31                  also been observed in rat hepatic oval cells (Agarwal et al.. 2009).


                    Altered Energy Metabolism

32                  Pb has been reported to alter normal cellular bioenergetics. In mitochondria isolated from
33                  the kidneys of rats exposed to 10,000 ppm Pb in feed for 6 weeks, the rate of oxygen
34                  uptake during ADP-activated (state 3) respiration was lower compared to controls (Gover

      November 2012                            5-17                  Draft - Do Not Cite or Quote

-------
 1                  et al.. 1968). The rate of ATP formation from exposed mitochondria was observed to be
 2                  approximately 50% that of control mitochondria. A decrease in state 3 respiration and
 3                  respiratory control ratios (state 3/state 4 [succinate or pyruvate/malate-activated]) was
 4                  also observed in kidney mitochondria from rats exposed continuously from conception to
 5                  six or nine months of age (i.e., gestationally, lactationally, and via drinking water after
 6                  weaning) to 50 or 250 ppm Pb (Fowler et al.. 1980). Pb-induced decreases in ATP and
 7                  adenylate energy charge (AEC) were observed concurrently with increases in ADP,
 8                  AMP, and adenosine in adult rats exposed to 10,000 ppm Pb in drinking water for
 9                  9 months (Marchlewicz et al., 2009). Similarly, ATP and AEC were decreased, and AMP
10                  increased, in primary cerebellar granule neuronal cultures from rats exposed to
1 1                  1,000 ppm Pb in drinking water throughout gestation and lactation (Baranowska-
12                  Bosiacka et al., 201 Ib). One (iM Pb (48 hours) was the lowest concentration observed to
13                  decrease cellular ATP levels in NGF-differentiated PC- 12 cells, and these  changes were
14                  correlated with a Pb-induced decrease in the expression of the mitochondrial voltage-
15                  dependent anion channel, which maintains cellular ATP levels in neurons (Prins et al..
16                  2010). Dowd et al. (1990) reported that oxidative phosphorylation was decreased up to
17                  74% after exposure of osteoblasts to 10 (iM Pb. Parr and Harris (1976) reported that Pb
18                  inhibited both coupled and uncoupled respiratory oxygen use in mitochondria, and that
19                  Pb prevented pyruvate, but not malate, uptake. Mitochondrial levels of ATP were
20                  diminished after Pb exposure, and the authors compared the effects of Pb on the energy
21                  supply to the actions of classic respiratory inhibitors, low temperature, and chemical
22                  uncouplers. Bragadin et al. (1998) supported this view by demonstrating that alkylated Pb
23                  compounds acted as a chemical uncoupler of respiration by abolishing the  proton gradient
24                  necessary for oxidative phosphorylation. Further, the enzymatic activities of complex I
25                  and IV of the respiratory chain have been shown to be decreased in the peroneous longus
26                  muscle of rats exposed to 250 ppm Pb (or  5 ppm thallium) in drinking water for 90 days
27                  (Mendez-Armenta et al.. 201 1). Contrary to the above findings, Rafalowska et al. (1996)
28                  reported that, although ATP levels did decrease in the forebrain synaptosomes prepared
29                  from rats exposed to 200 ppm Pb in water for 3 months, this chronic exposure to Pb did
30                  not inhibit oxidative phosphorylation in the synaptosomal mitochondria. Similar effects
3 1                  with regard to the activity of the mitochondrial oxidative chain were observed in rats
32                  injected with 15 mg Pb/kg (i.p.) daily for seven days, as reported by Struzynksa et al.
33                  (1997a). although ATP levels were reported to increase after exposure to Pb.
34                  Pb has also been shown to decrease glycolysis in osteoblasts exposed to 10 (iM Pb and in
35                  human erythrocytes exposed (in vitro) to 30 (ig/dL Pb (Grabowska and Guminska. 1996;
36                  Dowd et al.. 1990). Contrary to these findings, Antonowicz et al. (1990) observed higher
37                  levels of glycolytic enzymes in erythrocytes obtained from Pb workers directly exposed
38                  to Pb, compared to workers exposed to lower concentrations of Pb (blood Pb levels: 82.1
39                  versus 39.9 (ig/dL), and suggested that Pb activated anaerobic glycolysis. In vitro

      November 2012                             5-18                 Draft - Do Not Cite or Quote

-------
 1                  exposure of human umbilical cord erythrocytes to 100-200 (ig/dL Pb for 20 hours was
 2                  observed to lower the cellular pools of adenine and guanine nucleotide pools, including
 3                  NAD and NADPH (Baranowska-Bosiacka and Hlynczak. 2003). These decreases in
 4                  nucleotide pools were accompanied by an increase in purine degradation products
 5                  (adenosine, etc.). Similar decreases in cellular nucleotide pools were observed when rats
 6                  were exposed to 10,000 ppm Pb in drinking water for four weeks (Baranowska-Bosiacka
 7                  and Hlynczak. 2004). In erythrocytes, nucleotides are synthesized via salvage pathways
 8                  such as the adenine pathway, which requires adenine phosphoribosyltransferase. The
 9                  activity of this enzyme is inhibited by exposure to Pb in human and rat erythrocytes (see
10                  above for concentration and duration) (Baranowska-Bosiacka et al.. 2009).

11                  Disruptions in erythrocyte energy metabolism have been observed in adults
12                  occupationally exposed to Pb. Nikolova and Kavaldzhieva (1991) reported higher ratios
13                  of ATP/ADP in Pb-exposed workers with an average duration of exposure of 8.4 years
14                  (blood Pb not reported) than in unexposed controls. Morita et al. (1997) evaluated the
15                  effect of Pb on NAD synthetase in the erythrocytes of Pb-exposed workers (mean [SD]
16                  blood Pb level: 34.6 [20.7] (ig/dL) and observed an apparent concentration-dependent
17                  decrease in NAD synthetase activity with increased blood Pb level. The blood Pb level
18                  associated with 50% inhibition of NAD synthetase, which requires a Mg2+ cation for
19                  activity (Hara et al., 2003), was 43 (ig/dL.


                    Altered Heme Synthesis

20                  Exposure to Pb is demonstrated to inhibit two key steps in the  synthesis of heme:
21                  porphobilinogen synthase (i.e., 5-aminolevulinic acid dehydratase), a cytoplasmic
22                  enzyme requiring Zn for enzymatic activity that condenses two molecules of
23                  aminolevulinic acid into porphobilinogen, and ferrochelatase, a mitochondrial iron-sulfur
24                  containing enzyme that incorporates Fe2+ into protoporphyrin IX to create heme. Farant
25                  and Wigfield (1990. 1987) observed that Pb  inhibits the activity of porphobilinogen
26                  synthase in rabbit and human erythrocytes, and that the effect on the enzyme was
27                  dependent on the affinity for thiol groups at its active site. Taketani et al. (1985)
28                  examined the activity of Pb on  ferrochelatase in rat liver mitochondria and observed that
29                  10 (iM Pb (30  minute incubation) reduced NAD(P)H-dependent heme synthesis by half
30                  when ferric, but not ferrous, iron was used. Pb inhibits  the insertion of Fe2+ into the
31                  protoporphyrin ring and instead, Zn is inserted into the ring creating Zn-protoporphyrin
32                  (ZPP). While not directly measuring the activity of ferrochelatase, numerous studies have
33                  shown that blood Pb levels are  associated with increased erythrocyte ZPP levels in
34                  humans (mean blood Pb levels  ranging from 21.92 to 53.63 (ig/dL) (Mohammad et al.,
35                  2008; Counter et al.. 2007; Patil et al.. 2006b: Ademuyiwa et al.. 2005b) and in animals
36                  (blood Pb level: 24.7 (ig/dL) (Rendon-Ramirez et al.. 2007).

      November 2012                             5-19                   Draft - Do Not Cite or Quote

-------
 1                  In summary, Pb has been shown to disrupt mitochondrial function including
 2                  transmembrane mitochondrial ion transport mechanisms and has been shown to perturb
 3                  normal cell bioenergetics. These effects have not only been demonstrated in in vitro
 4                  toxicological studies but also exposed worker populations.
      5.2.3        Protein Binding

 5                  Evidence indicates that Pb binds to proteins within cells through interactions with side
 6                  group moieties (e.g., thiol residues) to form inclusion bodies and can thereby potentially
 7                  disrupt cellular function (Sections 5.2.2.3 and 5.2.2.4). However, some proteins are also
 8                  able to bind Pb and protect against its negative effects through sequestration. The ability
 9                  of Pb to bind proteins was first reported by Blackman (1936). In children exposed to high
10                  levels of Pb, formation of intranuclear inclusion bodies in the liver and kidney was
11                  observed. Since that time, further research has been conducted into characterizing the
12                  composition of intranuclear inclusion bodies and identifying specific Pb-binding proteins.
                    5.2.3.1      Intranuclear and Cytoplasmic Inclusion Bodies

13                  Goyer (1968) and Goyer et al. (1968) observed the formation of intranuclear inclusion
14                  bodies in the renal tubules of rats fed 10,000 ppm Pb in food for up to 20 weeks. The
15                  observation of inclusion bodies was accompanied by altered mitochondrial structure and
16                  reduced rates of oxidative phosphorylation. Pb has further been observed to form
17                  cytoplasmic inclusion bodies preceding the formation of the intranuclear bodies, and to
18                  be concentrated within the subsequently induced intranuclear inclusion bodies following
19                  i.p. injection, drinking water, and dietary exposures (Navarre-Moreno et al., 2009;
20                  Oskarsson and Fowler. 1985: Fowler etal.. 1980: McLachlin et al.. 1980: Choie and
21                  Richter. 1972: Goyer etal.. 1970b: Goyer etal.. 1970a). Inclusion bodies have also been
22                  observed in the mitochondria of kidneys and the perinuclear space in the neurons of rats
23                  exposed to 500 ppm Pb acetate in drinking water for 60 days or 7 months (Navarro-
24                  Moreno et al., 2009: Deveci, 2006). Intranuclear and cytoplasmic inclusions have also
25                  been found in organs other than the kidney, including liver, lung,  and glial cells (Singh et
26                  al.. 1999:  Gover and Rhvne. 1973). Pb found within nuclei has also been shown to bind
27                  to the nuclear membrane and histone fractions (Sabbioni and Marafante. 1976).

28                  Upon denaturing intranuclear inclusion bodies with strong denaturing agents, Moore et
29                  al. (1973) observed that proteins included in the bodies were rich  in aspartic and glutamic
30                  acid, glycine, and cysteine. Further work by Moore and Goyer (1974) characterized the
31                  protein as a 27.5 kilodalton (kDa) protein that migrates as a single band on
      November 2012                             5-20                   Draft - Do Not Cite or Quote

-------
 1                  polyacrylamide gel electrophoresis. In contrast with the findings of Moore and Goyer,
 2                  Shelton and Egle (1982) identified a 32 kDa protein with an isoelectric point of 6.3 from
 3                  the kidneys of rats exposed to 10,000 ppm Pb acetate in feed or 7,500 ppm in drinking
 4                  water. This protein, dubbed p32/6.3, was not found in control rats, indicating that the
 5                  protein was induced by Pb exposure. This finding was in agreement with studies that
 6                  indicated the formation of intranuclear inclusion bodies required protein synthesis
 7                  (McLachlin et al.,  1980; Choie et al.,  1975). In addition to its presence in kidneys of
 8                  Pb-exposed animals, p32/6.3 has been observed to be present and highly conserved in the
 9                  brains of rats, mice, dogs, chickens, and humans (Egle and Shelton. 1986). Exposure of
10                  neuroblastoma cells to 50 or 100 (iM Pb glutamate for 1  or 3 days increased the
11                  abundance of p32/6.3 (Klann and Shelton. 1989).  Shelton et al. (1990) determined that
12                  p32/6.3 was enriched in the basal ganglia, diencephalon, hippocampus, cerebellum,
13                  brainstem, spinal cord, and cerebral cortex, and that it contained a high percentage of
14                  glycine, aspartic, and glutamic acid residues. Selvin-Testa et al. (1991) and Harry et al.
15                  (1996) reported that pre- and post-natal exposure of rats to 2,000-10,000 ppm Pb in
16                  drinking water increased the levels of another brain protein, glial fibrillary acidic protein,
17                  in developing astrocytes; and that this increase may be indicative of a demand for
18                  astrocytes to sequester Pb.
                    5.2.3.2     Cytosolic Pb Binding Proteins

19                  Numerous studies have also identified cytosolic Pb-binding proteins. Two binding
20                  proteins, with molecular weights (MW) of 11.5 and 63 kDa, were identified by Oskarsson
21                  et al. (1982) in the kidney postmitochondrial cytosolic fraction prepared from adult male
22                  rats after i.p. injection with 50 mg Pb acetate/kg, followed by i.p. injection of 50 (iCi
23                  203Pb acetate 6 days later. The two proteins were  also found in the brain, but not the liver
24                  or lung. Mistry et al. (1985) identified three proteins (MW = 11.5, 63, and >200 kDa) in
25                  rat kidney cytosol, two of which, the 11.5 and 63 kDa proteins, were able to translocate
26                  into the nucleus. The 11.5 kDa kidney protein was also able to reverse Pb binding to
27                  ALAD through chelation of Pb and donation of a Zn cation to ALAD (Goering and
28                  Fowler. 1985. 1984). Cd and Zn, but not Ca2+ or  Fe, prevented the binding of Pb to the 63
29                  and 11.5 kDa cytosolic proteins, which agrees with previous observations that Cd is able
30                  to reduce total kidney Pb and prevent the formation of intranuclear inclusion bodies
31                  (Mistry et al.. 1986; Mahaffev et al.. 1981; Mahaffev and Fowler. 1977). Additional
32                  cytosolic Pb-binding proteins have been identified in the kidneys of Pb-exposed rats and
33                  humans, including the cleavage product of a2-microglobulin, acyl-CoA binding protein
34                  (MW = 9 kDa), and thymosin |34 (MW = 5 kDa)  (Smith etal.. 1998:  Fowler and DuVal.
35                  1991).
      November 2012                             5-21                   Draft - Do Not Cite or Quote

-------
 1                  Cytosolic Pb-binding proteins distinct from kidney proteins have also been identified in
 2                  the brain of exposed rats, and in human brain homogenates exposed to Pb in vitro
 3                  (Quintanilla-Vega et al.. 1995; DuVal and Fowler. 1989; Goering et al.. 1986). One
 4                  protein (MW =12 kDa) was shown to alleviate hepatic ALAD inhibition due to Pb
 5                  exposure through competitive binding with Pb and donation of Zn to ALAD. Cytosolic
 6                  Pb-binding proteins have been shown to be high in glutamic acid, aspartic acid, and
 7                  cysteine residues (Fowler et al..  1993; DuVal and Fowler.  1989). Some evidence exists
 8                  that cytosolic Pb-binding proteins directly target Pb and compartmentalize intracellular
 9                  Pb as a protective measure against toxicity (Oian etal. 2005; Oian et al., 2000).
                    5.2.3.3      Erythrocytic Pb Binding Proteins

10                  The majority (94%) of Pb in whole blood is found in erythrocytes (Ong and Lee. 1980a).
11                  Originally, the major Pb-binding protein in erythrocytes was identified as hemoglobin
12                  (Cohen et al.. 2000; Lolin and O'Gorman.  1988; Ong and Lee. 1980a. b; Raghavan and
13                  Gonick. 1977). However, Bergdahl et al. (1997b) observed the principal Pb-binding
14                  protein to be 240 kDa and identified it as ALAD. Two smaller Pb-binding proteins were
15                  observed, but not identified (MW = 45 and <10 kDa). ALAD levels are inducible by Pb
16                  exposure; the total concentration of the enzyme, but not the activity, is higher in both Pb-
17                  exposed humans (blood Pb = 30-75 (ig/dL) and rats (Pb exposure = 2.5 x 10"4 (iM in
18                  drinking water) (Boudene et al.. 1984; Fujitaetal.. 1982; Fujitaetal.. 1981).

19                  ALAD is a polymorphic gene with three isoforms: ALAD 1-1, ALAD 1-2, or ALAD 2-2.
20                  Carriers of the ALAD-2 allele have been shown to have higher blood  Pb levels than
21                  carriers of the homozygous ALAD-1 allele (Scinicariello et al.. 2007;  Zhao et al.. 2007;
22                  Kim et al., 2004; Perez-Bravo et al., 2004; Smith etal.. 1995a; Wetmur. 1994; Wetmur et
23                  al.. 1991b; Astrin et al.. 1987). Some recent studies, however, either observed lower
24                  blood Pb levels in carriers of the ALAD-2 allele or no difference in Pb levels among the
25                  different allele carriers (Scinicariello et al.. 2010; Krieg et al.. 2009; Chen et al.. 2008c;
26                  Chia et al.. 2007; Chia et al.. 2006; Wananukul et al.. 2006).

27                  The ALAD-2 protein binds Pb more tightly than the ALAD-lform: in Pb-exposed
28                  workers carrying the ALAD-2 gene, 84% of blood Pb was bound to ALAD versus 81%
29                  in carriers of the ALAD-1 gene (p = 0.03) (Bergdahl et al.. 1997a). This higher affinity
30                  for Pb in ALAD-2 carriers may sequester Pb and prevent its bioavailability for reaction
31                  with other enzymes or cellular components. This is supported by the observation that
32                  carriers of the ALAD-2 gene have higher levels of hemoglobin (Scinicariello et al..
33                  2007). decreased plasma levulinic acid (Schwartz et al.. 1997b). decreased levels of Zn
34                  protoporphyrin (Scinicariello et al.. 2007; Kim et al.. 2004). lower cortical bone Pb
      November 2012                             5-22                  Draft - Do Not Cite or Quote

-------
 1                  (Smith et al.. 1995b). and lower amounts of DMSA-chelatable Pb (Scinicariello et al..
 2                  2007; Schwartz et al. 2000a: Schwartz et al.. 1997a). However, the findings, that
 3                  ALAD-2 polymorphisms reduced the bioavailability of Pb, are  somewhat equivocal. Wu
 4                  et al. (2003a) observed that ALAD-2 carriers had lower blood Pb level (5.8 ± 4.2 (ig/dL)
 5                  than carriers of the ALAD-1 gene (blood Pb level = 6.2 ± 4.1 (ig/dL), and that ALAD-2
 6                  carriers demonstrated decreased renal function at lower patellar Pb concentrations than
 7                  those associated with decreased renal function in ALAD-1 carriers. This potentially
 8                  indicates that ALAD-2 carriers have enhanced Pb bioavailability. Weaver et al. (2003b)
 9                  observed that ALAD-2 polymorphisms were  associated with higher DMSA-chelatable Pb
10                  concentrations, when normalized to creatinine levels. Further, Montenegro et al. (2006)
11                  observed that compared with individuals with the ALAD 1-1 genotype, individuals with
12                  the ALAD 1-2/2-2 genotypes had a higher amount of Pb in the  plasma (0.44 (ig/L versus
13                  0.89 (ig/L, respectively) and in the percent plasma/blood ratio (0.48% versus 1.45%,
14                  respectively). This potentially suggests that individuals with the ALAD 1-2/2-2 genotype
15                  are at increased risk of Pb-induced health effects due to lower amounts of Pb
16                  sequestration by erythrocyte ALAD, although this  study did not specifically investigate
17                  the clinical implications of ALAD polymorphisms.

18                  ALAD has the estimated capacity to bind Pb  at 85  (ig/dL in erythrocytes and 40 (ig/dL in
19                  whole blood (Bergdahl et al.. 1998). The 45 kDa and <10 kDa Pb-binding proteins bound
20                  approximately 12-26% and <1% of the blood Pb, respectively. At blood Pb
21                  concentrations greater than 40 (ig/dL, greater binding to these components would likely
22                  be observed. Bergdahl et al. (1998) tentatively identified the 45 kDa protein as
23                  pyrimidine-5'-nucleotidase and the <10 kDa protein as acyl-CoA binding protein. Smith
24                  et al. (1998) previously identified acyl-CoA binding protein as a Pb-binding protein
25                  found in the kidney.
                    5.2.3.4      Metallothionein

26                  In adults occupationally exposed to Pb, the presence of an inducible, low-molecular
27                  weight (approximately 10 kDa) Pb-binding protein was identified in multiple early
28                  studies (Gonick et al.. 1985; Raghavan et al.. 1981. 1980; Raghavan and Gonick. 1977).
29                  The presence of this low molecular weight protein seemed to have a protective effect, as
30                  workers who exhibited toxicity at low blood Pb concentrations were observed to have
31                  lowered expression of this protein or low levels of Pb bound to it (Raghavan et al.. 1981.
32                  1980). The presence of low molecular weight Pb-binding proteins in exposed workers
33                  was corroborated by Lolin and O'Gorman (1988) and Church et al. (1993a. b). Further
34                  Lolin and O'Gorman (1988) reported that the observed protein was only present when
35                  blood Pb levels were greater than 39 (ig/dL, in agreement with the Pb-binding capacity of


      November 2012                            5-23                  Draft - Do Not Cite or Quote

-------
 1                  ALAD, identified by Bergdahl et al. (1998). Xie et al. (1998) confirmed this, observing
 2                  the presence of a second low molecular weight protein with greater affinity than ALAD,
 3                  only at higher blood Pb levels. Church et al. Q993a, b) observed the presence of a 6-7
 4                  kDa protein in the blood of two Pb workers (blood Pb >160 (ig/dL); approximately 67%
 5                  of Pb was bound to the protein in the blood of the asymptomatic worker, whereas only
 6                  22% of the Pb was bound to it in the symptomatic (tremor, ataxia, extremity numbness)
 7                  worker. The reported protein was rich in cysteine residues and tentatively identified as
 8                  metallothionein.

 9                  Metallothionein is a low-MW metal-binding protein, most often binding Zn or Cu, that is
10                  rich in cysteine residues and plays an important role in protecting against heavy metal
11                  toxicity, maintaining trace element homeostasis, and scavenging free radicals (Yu et al..
12                  2009). Exposure to Pb acetate induced the production of Pb- and Zn-metallothionein in
13                  mice treated via i.p. or intravenous (i.v.) injection at 30 mg/kg (Maitani et al.. 1986). in
14                  mice treated via i.p. injection at 300 (imol/kg (Yu et al.. 2009). or in rats treated via i.p.
15                  injection at 24 (imol/lOOg  (Ikebuchi et al..  1986). The induced Pb-metallothionein
16                  consisted of 28% half-cysteine and reacted with an antibody for Zn-metallothionein II
17                  (Ikebuchi et al..  1986). In contrast, exposure of rats to Pb via drinking water (200 or
18                  300 ppm) failed to induce  metallothionein in the kidneys or intestines (Wang et al..
19                  2009b; Jamieson et al.. 2007). Goering and Fowler (1987a. b) observed that pretreatment
20                  of rats with Zn before injection with Pb resulted in Pb and Zn co-eluting with Zn-
21                  thionein, and that Zn-thionein I and II were able to bind Pb in vitro  (Goering and Fowler.
22                  1987a. b). Further, Goering and Fowler (1987a) found that kidney and liver Zn-thionein
23                  decreased binding of Pb to liver ALAD and was able to donate Zn to ALAD, thus
24                  attenuating the inhibition of ALAD due to Pb exposure. These findings are  in agreement
25                  with Goering et al. (1986)  and DuVal and Fowler (1989) who demonstrated that rat brain
26                  Pb-binding proteins attenuated Pb-induced inhibition of ALAD.

27                  Metallothionein has been reported to be important in the amelioration of Pb-induced
28                  toxicity effects. Liu et al. (1991) reported that Zn-metallothionein reduced Pb-induced
29                  membrane leakage and loss of K+ in cultured hepatocytes incubated with 600-3,600 (iM
30                  Pb. Metallothionein-null mice exposed to  1,000, 2,000, or 4,000 ppm Pb for 20 weeks
31                  suffered renal toxicity described as nephromegaly and decreased renal function compared
32                  to Pb-treated wild-type mice (Qu et al.. 2002). Interestingly, metallothionein-null mice
33                  were unable to form intranuclear inclusion bodies and accumulated less renal Pb than did
34                  the wild-type mice (Qu et  al.. 2002). Increased metallothionein levels were  induced by Pb
35                  exposure in non-null mice. Exposure to Pb (1,000, 2,000, or 4,000 ppm), both for
36                  104 weeks as adults and from GD8 to early adulthood, resulted in increased preneoplastic
37                  lesions and carcinogenicity in the testes, bladder, and kidneys of metallothionein-null rats
38                  compared to wild type mice (Tokaretal.. 2010; Waalkes et al.. 2004).  Inclusion bodies
      November 2012                             5-24                  Draft - Do Not Cite or Quote

-------
 1                  were not observed in null mice. The authors concluded that metallothionein is important
 2                  in the formation of inclusion bodies and in the mitigation of Pb-induced toxic effects, and
 3                  that those with polymorphisms in metallothionein coding genes that are associated with
 4                  reduced inducibility may be at greater susceptibility to Pb.  In support of this theory, Chen
 5                  et al. (2010a) observed that Pb-exposed workers with a mutant metallothionein allele had
 6                  higher blood Pb levels than did carriers of the normal allele (24.17 and 21.27 versus
 7                  17.03 (ig/dL), and had larger blood Pb-associated changes  in systolic BP and serum renal
 8                  function parameters.

 9                  In summary, a number of proteins have been identified that can bind and sequester Pb
10                  including ALAD and metallothionein. Additionally, evidence suggests that certain
11                  polymorphisms that alter the binding capability or inducibility of these proteins can
12                  increase the risk of Pb induced health effects.
      5.2.4        Oxidative Stress

13                  Oxidative stress occurs when free radicals or reactive oxygen species (ROS) exceed the
14                  capacity of antioxidant defense mechanisms. This Oxidative imbalance results in
15                  uncontained ROS, such as superoxide (O2~), hydroxyl radical (OH), and hydrogen
16                  peroxide (H2O2), which can attack and denature functional/structural molecules and,
17                  thereby, promote tissue damage, cytotoxicity, and dysfunction. Pb exposure has been
18                  shown to cause Oxidative damage to the heart, liver, kidney, reproductive organs, brain,
19                  and erythrocytes, which may be responsible for a number of Pb-induced health effects
20                  (Salawu et al.. 2009; Shan et al.. 2009; Vaziri. 2008b: Gonicketal.. 1997; Sandhir and
21                  Gill. 1995: Khalil-Manesh et al..  1994: Khalil-Manesh et al.. 1992aV The origin of ROS
22                  (produced after Pb exposure) is likely a multipathway process, resulting from oxidation
23                  of 5-aminolevulinic acid (ALA), membrane and lipid oxidation, NAD(P)H oxidase
24                  activation, and antioxidant enzyme depletion, as discussed below.  Some of these
25                  processes result from the disruption of functional metal ions within oxidative stress
26                  enzymes,  such as superoxide dismutase (SOD), catalase (CAT), and glutathione
27                  peroxidase (GPx).  Interestingly, Pb exposure in many species of plants, invertebrates, and
28                  vertebrates discussed in ChapterJ? (Ecological Effects of Lead) results in upregulation of
29                  antioxidant enzymes and increased lipid peroxidation. Oxidative stress is a common
30                  mode of action for a number of other metals (e.g., Cd, Mn, As, Co, Cr) that are often
31                  found with Pb and by which possible interactions with Pb have been suggested to occur
32                  (Jomova and Valko. 2011; Jomovaet al.. 2011; Matovic et al.. 2011; HaMai and Bondy.
33                  2004). Not all of these co-occurring metals directly produce ROS or redox cycle, but
34                  instead may suppress the free radical scavenging ability of the organism thus leading to
35                  oxidative  stress.
      November 2012                             5-25                  Draft - Do Not Cite or Quote

-------
                    5.2.4.1      5-ALA Oxidation

 1                  The majority of Pb present in the blood accumulates in erythrocytes where it enters
 2                  through passive carrier-mediated mechanisms including a vanadate-sensitive Ca2+ pump.
 3                  Once Pb enters erythrocytes, it is predominantly found in the protein-bound form, with
 4                  hemoglobin and 5-ALAD both identified as targets (Bergdahl et al., 1997a). Through its
 5                  sulfhydryl and metal ion disrupting properties, Pb incorporates with and inhibits a
 6                  number of enzymes in the heme biosynthetic process, including 5-ALA synthetase,
 7                  5-ALAD, and ferrochelatase. Pb has been shown to be able to disrupt the Zn ions
 8                  requisite for the activity of 5-ALAD, the rate limiting step in heme synthesis, leading to
 9                  enzyme inhibition at 10"6 (iM concentrations (Simons. 1995). Additionally, blood Pb
10                  levels (mean:  7 (ig/dL) have been associated with inhibited activity of 5-ALAD in
11                  humans, and the lowest blood Pb level observed to be associated with lower 5-ALAD
12                  activity in these studies was 5 (ig/dL (Ahamed et al., 2005; Sakai and Morita. 1996). A
13                  negative correlation (r = -0.6) was found between blood Pb levels in adolescents (range
14                  of blood Pb levels: 4 to 20 (ig/dL) and blood 5-ALAD activity (Ahamed et al.. 2006).
15                  This inhibition of 5-ALAD results in the accumulation of 5-ALA in blood and urine,
16                  where 5-ALA undergoes tautomerization and autoxidation. Oxidized  5-ALA generates
17                  ROS through reduction of ferricytochrome c and electron transfer from oxyHb, metHb,
18                  and other ferric and ferrous iron complexes (Hermes-Lima et al.. 1991; Monteiro et al..
19                  1991). The autoxidation of 5-ALA produces O2~, OH, H2O2, and an ALA radical
20                  (Monteiro et al.. 1989; Monteiro et al.. 1986).
                    5.2.4.2      Membrane and Lipid Peroxidation

21                  A large number of studies in humans and experimental animals have indicated that
22                  exposure to Pb can lead to membrane and lipid peroxidation. It is possible that ROS
23                  produced from 5-ALA oxidation, as described above, interacts with and disrupts
24                  membrane lipids (Oteizaet al., 1995; Bechara et al.. 1993). Additionally, Pb has the
25                  capacity to stimulate ferrous ion initiated membrane lipid peroxidation serving as a
26                  catalyst for these events (Adonavlo and Oteiza.  1999; Quinlan et al.. 1988). The extent of
27                  peroxidation of lipids varies based on the number of double bonds present in unsaturated
28                  fatty acids, since double bonds weaken the C-H bonds on the adjacent carbon, making
29                  hydrogen (H) removal easier (Yiin and Lin. 1995). After the essential unsaturated fatty
30                  acid solutions were incubated with Pb (4-12 (ig/dL, 24 hours), the production of
31                  malondialdehyde (MDA), a marker of oxidative stress and lipid oxidation end product,
32                  increased relative to the number of double bonds of the fatty acid (Yiin and Lin. 1995). In
33                  the absence of Fe2+, Pb has not been shown to promote lipid peroxidation; however, it
34                  may accelerate peroxidation by H2O2 (Quinlan et al., 1988). This could be due to altering

      November 2012                             5-26                 Draft - Do Not Cite or Quote

-------
 1                  membrane structure, restricting phospholipid movement, and facilitating the propagation
 2                  of peroxidation.

 3                  Pb induces changes in the fatty acid composition of a membrane, which could lead to
 4                  oxidative damage. Exposure to Pb (>62.5 ppm in drinking water, 3 weeks) in chicks
 5                  promoted an increase in arachidonic acid (20:4) as a percentage of total fatty acids, and
 6                  decreased the relative proportion of shorter chain fatty acids (linoleic acid, 18:2) (Lawton
 7                  and Donaldson. 1991). It is possible that Pb depressed the desaturation of saturated fatty
 8                  acids to the corresponding monoenoic fatty acids, while stimulating elongation and
 9                  desaturation of linoleic acid to arachidonic acid. Since fatty acid chain length and
10                  unsaturation are related to the oxidative potential, changes in fatty acid membrane
11                  composition may result in enhanced lipid peroxidation. In addition, changes in fatty
12                  acids, thus membrane composition, can result in altered membrane fluidity (Donaldson
13                  and Knowles.  1993). Changes in membrane fluidity will disturb the conformation of the
14                  active sites of membrane associated enzymes, disrupt metabolic regulation, and alter
15                  membrane permeability and function.

16                  A number of recent studies report increased measures of lipid peroxidation in various
17                  organs, tissues, and species in association with Pb. Occupational Pb exposure resulting in
18                  elevated blood Pb levels (means >8 (ig/dL) reported in various  countries provides
19                  evidence of lipid peroxidation, including increased plasma MDA levels (Ergurhan-Ilhan
20                  et al.. 2008; Khan et al.. 2008; Mohammad et al.. 2008; Quintanar-Escorza et al.. 2007;
21                  Patil et al.. 2006a: Patil et al.. 2006b). One study found a correlation between the MDA
22                  levels and blood Pb levels even in the unexposed workers, although they had blood Pb
23                  levels higher than the mean blood Pb level of the current U.S. population
24                  (i.e., <12 (ig/dL) (Quintanar-Escorza et al., 2007). Other studies found evidence of
25                  increased lipid peroxidation among the general population, including children, with
26                  elevated blood Pb levels (means >10 (ig/dL) (Ahamed et al.. 2008; Ahamed et al.. 2006;
27                  Jin et al.. 2006). In adolescents, Ahamed et al. (2006) found a blood MDA levels to be
28                  positively correlated (r = 0.7) with concurrent blood Pb levels ranging between 4 and
29                  20 (ig/dL. Similar results have been shown after Pb exposure in animal studies (Abdel
30                  Moneimetal. 20lib; Pandvaet al.. 2010; Dogru et al.. 2008; Yu et al.. 2008; Adegbesan
31                  and Adenuga.  2007; Lee et al.. 2005). Enhanced lipid peroxidation has been found in Pb
32                  treated (50 (ig, 1-4 hours) rat brain homogenates (Rehman et al.. 1995). rat proximal
33                  tubule cells (0.5-1 (JVI, 12 hours) (Wang etal. 20lib), and in specific  brain regions,
34                  hippocampus and cerebellum, after Pb exposure (500 ppm, 8 weeks) to rats (Bennet et al..
35                  2007). Overall, there was a correlation between the blood Pb level and measures of lipid
36                  peroxidation often measured by MDA levels.
      November 2012                             5-27                  Draft - Do Not Cite or Quote

-------
 1                  In summary, studies in humans and animals provide evidence for increased lipid and
 2                  membrane oxidation following Pb exposure. Interestingly, many species of plants,
 3                  invertebrates, and other vertebrates also exhibit increased lipid peroxidation with Pb
 4                  exposure (Sections 7.3.12.6 and 7.4.12.6). The increase in lipid peroxidation following
 5                  Pb exposure observed across species and kingdoms demonstrate an evolutionarily
 6                  conserved oxidative response following Pb exposure.
                    5.2.4.3      NAD(P)H Oxidase Activation

 7                  NAD(P)H oxidase is a membrane bound enzyme that requires Ca2+ in order to catalyze
 8                  the production of O2 from NAD(P)H and molecular oxygen (Leseney et al., 1999). Two
 9                  studies provide evidence for increased activation of NAD(P)H oxidase that may
10                  contribute to the production of ROS after Pb exposure (Ni et al., 2004; Vaziri et al.,
11                  2003). Vaziri et al. (2003) found increased protein expression of the NAD(P)H subunit
12                  gp91phox in the brain, heart, and renal cortex of Pb-treated rats (100 ppm in drinking
13                  water, 12 weeks). This upregulation was present in Pb-treated (1-10 ppm) human
14                  coronary artery endothelial cells, but not vascular smooth muscle cells (VSMC), which
15                  do not express the protein (Ni et al., 2004). It is possible that NAD(P)H oxidase serves as
16                  a potential source of ROS in cells that express this protein.
                    5.2.4.4      Antioxidant Enzyme Disruption

17                  Oxidative stress can result not only from the increased production of ROS, but also from
18                  the decreased activity of antioxidant defense enzymes. Pb has been shown to alter the
19                  function of several antioxidant enzymes, including SOD, CAT, glucose-6-phosphate
20                  dehydrogenase (G6PD), and the enzymes involved in glutathione metabolism, GPx,
21                  glutathione-S-transferase (GST), and glutathione reductase (GR). These changes in the
22                  antioxidant defense system could be due to the high affinity of Pb for sulfhydryl groups
23                  contained within proteins and its metal ion mimicry. However changes could also be a
24                  consequence of increased oxidative damage by Pb.

25                  Studies of the effects of Pb exposure on the activities of SOD and CAT give divergent
26                  results. These metalloprotein enzymes require various essential trace elements for proper
27                  structure  and function, making them a target for Pb toxicity. CAT is a heme containing
28                  protein that requires Fe ions to function (Putnam et al.. 2000). SOD exists in multiple
29                  isoforms  that require Cu and Zn (SOD 1 and SOD3) (Antonyuk et al.. 2009) or Mn
30                  (SOD2) (Borgstahl et al.. 1992). A number of studies have found decreased activity of
31                  these enzymes (Pandya et al., 2010; Ergurhan-Ilhan et al., 2008; Mohammad et al., 2008;
      November 2012                            5-28                  Draft - Do Not Cite or Quote

-------
 1                  Yu et al.. 2008; Patil et al.. 2006a: Patil et al.. 2006b: Conterato et al.. In Press), whereas
 2                  others have observed increased activity following Pb exposure (Ahamed et al.. 2008; Lee
 3                  et al.. 2005). The heterogeneity in species examined, (i.e., humans, rodents, boars), and
 4                  Pb exposure metrics reported did not permit evaluation of whether a nonlinear
 5                  concentration-response relationship could explain heterogeneity in findings. Pb exposure
 6                  (500 ppm Pb acetate, 1, 4, and 8 weeks) in adult male rats showed that SOD and CAT
 7                  activity responded differently depending on the brain region analyzed and length of
 8                  exposure (Bennet et al.. 2007). Another study found that the brain had consistently
 9                  decreased SOD activity, irrespective of dose in prenatally-exposed animals (0.3 and
10                  3.0 ppm, blood Pb level 20.4 and 24.5 (ig/dL); however hepatic SOD activity increased at
11                  low level Pb administration and decreased after high level exposure (Uzbekov et al..
12                  2007). It is possible that the increased activity of the SOD and CAT proteins is due to
13                  activation by ROS, while decreased enzyme activity is the result of metal ion substitution
14                  by Pb, causing enzyme inactivation.

15                  Glutathione is a tripeptide antioxidant containing a cysteine with a reactive thiol group
16                  that can act nonenzymatically as a direct antioxidant or as a cofactor in enzymatic
17                  detoxification reactions by GST. Glutathione will donate an electron while in its reduced
18                  state (GSH), which leads to conversion to the oxidized form, glutathione disulfide
19                  (GSSG). Pb binds to the thiol and can both interfere with the antioxidant capacity of
20                  GSH, and can decrease levels of GSH. Short-term administration of Pb in vitro (0.1 (iM)
21                  and observed biomarkers of Pb exposure in humans  (18 (ig/dL mean blood Pb level) have
22                  been associated with decreased blood and organ GSH and cysteine content, which may be
23                  due to increased GSH efflux from tissues (Pandya et al.. 2010; Pillai et al..  2010; Ahamed
24                  et al.. 2009; Ahamed et al.. 2008; Flora et al.. 2007; Ahamed etal. 2005; Chettv et al..
25                  2005; Nakagawa. 1991. 1989). Long-term Pb exposure has elicited a compensatory
26                  upregulation in the biosynthesis of GSH in the attempt to overcome Pb toxicity, thus
27                  often manifesting as an increase in Pb-induced GSH in animals (Daggett et al.. 1998;
28                  Corongiu and Milia. 1982; Hsu. 1981)  and occupationally exposed adults (SRC. 2002;
29                  Conterato et al.. In Press). However, other studies have found that long-term Pb exposure,
30                  resulting in mean blood Pb levels between 6.6 and 22 (ig/dL, causes the depletion of GSH
31                  in animals (Lee et al..  2005; Ercal etal.. 1996) and occupationally exposed adults
32                  (Mohammad et al.. 2008). Thus, the duration of Pb exposure is important to consider
33                  when measuring GSH levels.

34                  Glutathione reductase is able to reduce GSSG back to GSH. Therefore, an increased
35                  GSSG/GSH ratio is considered to be indicative of oxidative stress. Epidemiologic studies
36                  have found higher blood Pb levels to be associated with increases in the GSSG/GSH ratio
37                  (Mohammad et al.. 2008; Ercal etal.. 1996; Sandhir and Gill. 1995). In one study, this
38                  association was observed in a population of children with a mean blood Pb level below
      November 2012                             5-29                   Draft - Do Not Cite or Quote

-------
 1                   10 (ig/dL (Diouf etal.. 2006). Studies have found mixed effects on GR activation. GR
 2                  possesses a disulfide at its active site that is a target for inhibition by Pb. Studies in
 3                  animals and cells have reported decreased (Bokara et al.. 2009; Sandhir and Gill, 1995;
 4                  Sandhiretal.. 1994). increased (Sobekova et al. 2009; Howard. 1974). and no change
 5                  (Hsu. 1981) in GR activity after Pb exposure. This could be because the effect of Pb on
 6                  GR varies depending on sex (Sobekova et al.. 2009). and organ or organ region (Bokara
 7                  et al.. 2009). The heterogeneity in species examined, (i.e., humans, rodents), and Pb
 8                  exposure duration and metrics reported did not permit evaluation of whether a nonlinear
 9                  concentration-response relationship could explain heterogeneity in findings.

10                  GSH is used as a cofactor for peroxide reduction and detoxification of xenobiotics by the
11                  enzymes GPx and GST.  GPx requires Se (selenium) for peroxide decomposition (Rotruck
12                  et al.. 1973). whereas GST functions via a sulfhydryl group. Evidence indicates that by
13                  reducing the uptake  of Se, depleting cellular GSH, and disrupting protein thiols, Pb
14                  decreases the activity of GPx and GST (Pillai etal.. 2010; Yu et al.. 2008; Lee et al..
15                  2005; Nakagawa.  1991;  Schrauzer. 1987). Similar to other antioxidant enzymes,
16                  compensatory upregulation of these enzymes was observed after Pb exposure in animals
17                  and in Pb-exposed workers (painters with a mean blood Pb level of 5.4 (ig/dL) (Bokara et
18                  al.. 2009; Ergurhan-Ilhan et al.. 2008; Conterato et al.. 2007; Daggett et al.. 1998;
19                  Conterato et al.. In Press). However, in another study, these enzymes  were  not able to
20                  compensate for the increased Pb-induced ROS, further contributing to the oxidative
21                  environment (Farmand et al.. 2005).

22                  Recently, y-glutamyltransferase (GGT) within its normal range has been regarded as an
23                  early and sensitive marker of oxidative stress. This may be because cellular GGT
24                  metabolizes extracellular GSH to be used in intracellular GSH synthesis. Thus, cellular
25                  GGT acts as an antioxidant enzyme by increasing the intracellular GSH pool. However,
26                  the reasons for the association between GGT and oxidative stress have not  been fully
27                  realized (Lee et al.. 2004).  In one study, occupational Pb exposure (mean blood Pb level
28                  of 29.1 (ig/dL) was associated with increased serum GGT levels (Khan et al.. 2008).
29                  Interestingly, higher blood Pb level was similarly associated with higher serum GGT
30                  levels in a sample of the U.S. adult population (NHANES III) (Lee et al.. 2006a). In this
31                  study of nonoccupationally-exposed individuals, a concentration-dependent relationship
32                  was observed with blood Pb levels <7 (ig/dL.

33                  In summary, Pb has  been shown to alter the function of several antioxidant enzymes,
34                  including SOD, CAT, G6PD, and the enzymes  involved in glutathione metabolism, GPx,
35                  GST, and GR in human populations and experimental animal models. Alteration of these
36                  enzymes may lead to further oxidative stress  following Pb exposure.
      November 2012                             5-30                  Draft - Do Not Cite or Quote

-------
                    5.2.4.5     Nitric Oxide Signaling

 1                  NO (nitric oxide radical), also known as endothelium-derived relaxing factor, is a potent
 2                  endogenous signaling molecule involved in vasodilation. Short- and long-term Pb
 3                  exposure in animals have been found to decrease the biologically active NO, not through
 4                  reduction in NO-production capacity (Vaziri and Ding. 2001; Vaziri etal.. 1999a). but as
 5                  a result of inactivation and sequestration of NO by ROS (Malvezzi et al.. 2001; Vaziri et
 6                  al.. 1999b). Endogenous NO can interact with ROS, specifically O2~, produced following
 7                  exposure to Pb to form the highly cytotoxic reactive nitrogen species, peroxynitrite
 8                  (ONOO"). This reactive compound can damage cellular DNA and proteins, resulting in
 9                  the formation of nitrotyrosine among other products. Overabundance of nitrotyrosine in
10                  plasma and tissues is present after exposure to Pb (Vaziri et al.. 1999b). NO is also
11                  produced by macrophages in the defense against certain infectious agents, including
12                  bacteria. Studies have indicated that Pb exposure can significantly reduce production of
13                  NO in human (Pineda-Zavaleta et al.. 2004) and animal immune cells (Lee et al.. 200Ib;
14                  Tian and Lawrence. 1995). possibly leading to reduced host resistance (Tian and
15                  Lawrence.  1996).

16                  Production of NO is catalyzed by a family of enzymes called nitric oxide synthases
17                  (NOS), including endothelial NOS (eNOS), neuronal NOS (nNOS), and inducible NOS
18                  (iNOS), which require a heme prosthetic group and a Zn cation for enzymatic activity
19                  (Messerschmidt et al.. 2001). Paradoxically, the reduction in NO availability in vascular
20                  tissue following Pb exposure is accompanied by statistically significant upregulation in
21                  NOS isotypes (Vaziri and Ding. 2001; Vaziri etal.. 1999a; Gonicketal. 1997). A direct
22                  inhibitory action of Pb on NOS enzymatic activity has been rejected (Vaziri et al..
23                  1999a). Instead, the upregulation of NOS occurs as compensation for the decreased NO
24                  resulting from ROS inactivation (Vaziri et al.. 2005; Vaziri and Ding. 2001; Vaziri and
25                  Wang. 1999).


                    Soluble Guanylate Synthase

26                  Many biological actions of NO, such as vasorelaxation, are mediated by cyclic guanosine
27                  monophosphate (cGMP), which is produced by soluble guanylate  cyclase (sGC) from the
28                  substrate guanosine triphosphate. Soluble guanylate cyclase is a heterodimer requiring
29                  one molecule of heme for enzymatic activity (Boerrigter and Burnett. 2009). In VSMC,
30                  sGC serves as the NO receptor. Marked reduction in plasma concentrations and urinary
31                  excretion of cGMP is observed after  Pb exposure to rats [5 ppm Pb in drinking water for
32                  30 days (Marques et al.. 2001)] and [100 ppm Pb acetate in drinking water for 3 months,
33                  resulting in a mean blood Pb level of 29.4 (ig/dL) (Khalil et al.. 2008)1 (Marques et al..
34                  2001; Khalil-Manesh et al.. 1993b). In addition, Pb exposure downregulated the protein

      November 2012                             5-31                  Draft - Do Not Cite or Quote

-------
 1                  abundance of sGC in vascular tissue (Farmand et al., 2005; Courtois etal. 2003;
 2                  Marques et al.. 2001). This downregulation in sGC was prevented by antioxidant therapy
 3                  (ascorbic acid) suggesting that oxidative stress also plays a role in Pb-induced
 4                  downregulation of sGC (no change in blood Pb level was observed after ascorbic acid
 5                  treatment) (Marques et al., 2001).
      5.2.5       Inflammation

 6                  Misregulated inflammation represents one of the major hallmarks of Pb-induced immune
 7                  effects. It is important to note that this can manifest in any tissue where immune cell
 8                  mobilization and tissue insult occurs. Enhanced inflammation and tissue damage occurs
 9                  through the modulation of inflammatory cell function and production of pro-
10                  inflammatory cytokines and metabolites. Overproduction of ROS and an apparent
11                  depletion of antioxidant protective enzymes and factors (e.g., Se) accompany this
12                  immunomodulation (Chetty et al., 2005).

13                  Traditional immune-mediated inflammation can be seen with bronchial
14                  hyperresponsiveness, asthma, and respiratory infections, some of which have been
15                  associated with exposure to Pb. But it is important to recognize that any tissue or organ
16                  can be affected by immune-mediated inflammatory dysfunction given the distribution of
17                  immune cells as both permanent residents and infiltrating cell populations (Mudipalli.
18                  2007; Carmignani et al., 2000). Pb has been associated with multiple indicators of
19                  inflammation in multiple cell types. Pb has also induced renal tubulointerstitial
20                  inflammation (100 ppm exposure for 14 weeks) (Rodriguez-Iturbe et al., 2005)
21                  (24.6 (ig/dL blood Pb level,  150 ppm for 16 weeks) (Roncal et al.. 2007). Renal
22                  tubulointerstitial inflammation has been coupled with activation of the redox sensitive
23                  nuclear transcription factor kappa B (NFKB) and lymphocyte and macrophage infiltration
24                  in rats (100 ppm for 14 weeks resulting in mean blood Pb levels ranging 23.7 (ig/dL)
25                  (Bravo et al.. 2007). These events could be in response to the oxidative environment
26                  arising from Pb exposure, since Pb-induced inflammation and NFxB activation can be
27                  ameliorated by antioxidant therapy (Rodriguez-Iturbe et al.. 2004). Pb spheres implanted
28                  in the brains of rats produced neutrophil-driven inflammation with apoptosis and
29                  indications of neurodegeneration (Nakao et al.. 2010).

30                  Inflammation can be mediated by the production of chemical messengers such as
31                  prostaglandins (PG). Pb exposure has been associated with increased arachidonic acid
32                  (AA) metabolism, thus elevating the production of PGE2, PGF2, and thromboxane in
33                  occupationally-exposed humans (mean blood Pb level 48 (ig/dL) (Cardenas et al., 1993)
34                  and animal and cell models (e.g., 0.01 (JVI, 48 hours) (Chetty et al.. 2005; Flohe et al..
      November 2012                            5-32                  Draft - Do Not Cite or Quote

-------
 1                  2002; Knowles and Donaldson. 1997; Lee and Battles. 1994). Dietary Pb exposure of
 2                  animals (500 ppm, 19 days) can increase the percentage of cell membrane AA, the
 3                  precursor of cyclooxygenase and lipoxygenase metabolism to PGs and leukotrienes
 4                  (Knowles and Donaldson. 1990). Additionally, Pb (1 (iM) may promote the release of
 5                  AA via activation of phospholipase A2, as shown in isolated VSMC (Dorman and
 6                  Freeman. 2002).

 7                  Inflammation may be the result of increased pro-inflammatory signaling or may stimulate
 8                  these signaling pathways. Pb can elevate the expression of the pro-inflammatory
 9                  transcription factors NFxB and activator protein-1 (AP-1), as well as the AP-1
10                  component c-Jun (Korashy and El-Kadi. 2008; Korashy and Ei-Kadi. 2008; Bravo et al..
11                  2007; RameshetaL 1999; Pvatt et al.. 1996). Pb exposure (25 (iM) to dendritic cells
12                  stimulated phosphorylation of the Erk/MAPK pathway, but not p38, STATS or 5, or
13                  CREB (Gao et al..  2007)
                    5.2.5.1      Cytokine Production

14                  There are three modes by which Pb has been shown to affect immune cytokine
15                  production. First, Pb can act on macrophages to elevate the production of pro-
16                  inflammatory cytokines such as TNF-a and interleukin (IL)-6 (Cheng et al.. 2006; Chen
17                  et al.. 1999; Dentener et al.. 1989). This can result in local tissue damage during the
18                  course of immune responses affecting such targets as the liver. Second, Pb can skew the
19                  ratio of IL-12/IL-10 such that T-derived lymphocyte helper (Th)l responses are
20                  suppressed and Th2 responses are promoted (Chen et al., 2004; Miller et al..  1998)
21                  possibly by affecting dendritic cells. Third, when acquired immune responses occur
22                  following exposure to Pb, Thl lymphocyte production of cytokines is suppressed
23                  (e-g-, IFN-y) (Lynes et al.. 2006; Heo et al.. 1996); in contrast, Th2 cytokines such as
24                  IL-4, IL-5, and IL-6 are elevated (Gao et al.. 2007; Kim and Lawrence. 2000). The
25                  combination of these three modes of cytokine changes induced by Pb can create a
26                  hyperinflammatory state among innate immune cells and skew acquired immunity toward
27                  Th2 responses.

28                  lavicoli et al. (2006b) reported that low blood Pb concentrations produced significant
29                  changes in cytokine levels in mice. At a low dietary Pb concentration (0.11 ppm, blood
30                  Pb level of 1.6 (ig/dL), IL-2 and IFN-y were decreased compared to the controls
31                  (0.02 ppm, 0.8 (ig/dL), indicating a suppressed Thl  response. As the dietary and blood Pb
32                  concentrations increased (resulting in blood Pb levels 12-61 (ig/dL), a Th2 phenotype
33                  was observed with suppressed IFN-y and IL-2 and elevated IL-4 production.  These
34                  findings support the notion that the immune system is remarkably sensitive to Pb-induced
      November 2012                             5-33                  Draft - Do Not Cite or Quote

-------
 1                   functional alterations and that nonlinear effects may occur at low Pb exposures. TGF-(3
 2                   production was also altered by Pb exposure to transfected mouse limb bud mesenchymal
 3                   stem cells (1 (JVI, 3 days) (Zuscik et al., 2007). IL-2 is one of the more variable cytokines
 4                   with respect to Pb-induced changes. Depending upon the protocol it can be slightly
 5                   elevated in production or unchanged. Recently, Gao et al. (2007) found that Pb-treated
 6                   dendritic cells (25  (iM) promoted a slight but statistically significant increase in IL-2
 7                   production among lymphocytes. Proinflammatory cytokines have been measured in other
 8                   organs and cell types after Pb exposure. Elevation of IL-1(3 and TNF-a were observed in
 9                   the hippocampus after Pb treatment (15 ppm, i.p., daily for 2 weeks, blood Pb level of
10                   30.8 (ig/dL) and increased IL-6 was found  in the forebrain (Struzvnska et al., 2007).

11                   Consistent with animal studies, epidemiologic studies also found higher concurrent blood
12                   Pb levels in children and occupationally-exposed adults to be associated with a shift
13                   toward production of Th2 cytokines relative to Thl cytokines. The evidence in children
14                   was based on comparisons of serum cytokine levels among groups with different blood
15                   Pb levels without consideration of potential confounding factors. Among children ages
16                   9 months to 6 years in Missouri, Lutz et al. (1999) found that children with concurrent
17                   blood Pb levels 15-19 (ig/dL had higher serum levels of IL-4 and IgE (Section 5.6.3) than
18                   did children with lower blood Pb levels. These results were consistent with the mode of
19                   action for IL-4 to activate B cells to induce B cell class switching to IgE. Concurrent
20                   blood Pb levels did not differ by residence  in old versus new homes or by urban versus
21                   rural residence (means: 3.2-3.8 (ig/dL) but  were higher among  children living near an oil
22                   refinery, in particular, among children with known respiratory allergies (mean:
23                   8.8 (ig/dL). This latter group of children also had the lowest serum levels of IFN-y and
24                   highest levels of IL-4. There was no direct  comparison of cytokine levels between blood
25                   Pb level  groups in the population overall; however, cytokine levels were similar between
26                   healthy and allergy groups in the other Pb source groups that had similar blood Pb levels.
27                   Thus, the differences in cytokine levels between healthy and allergic children living near
28                   the oil refinery may have been influenced by differences in their blood Pb levels or other
29                   factors related to residence near an oil refinery.

30                   Evidence of association between blood Pb  levels and cytokine levels in
31                   nonoccupationally-exposed adults was unclear. Among healthy adult university students
32                   in Incheon, Korea, Kim et al. (2007) found associations of concurrent blood Pb level with
33                   serum levels of TNF-a and IL-6 that were larger among male students with blood Pb
34                   levels 2.51-10.47 (ig/dL. Notably, the relative  contributions of lower recent versus higher
35                   past Pb exposures to these cytokine effects  is not known. In models that adjusted for age,
36                   sex, BMI, and smoking status, a 1 (ig/dL increase in blood Pb level was associated with a
37                   23% increase (95% CI: 4, 55%) in log of TNF-a and a 26% increase in log of IL-6 (95%
38                   CI: 0, 55%). The association between levels of blood Pb and plasma TNF-a was greater
      November 2012                             5-34                  Draft - Do Not Cite or Quote

-------
 1                  among men who were GSTM1 null or had the TNF-a GG genotype. For the association
 2                  between blood Pb level and plasma IL-6, the effect estimate was slightly elevated in
 3                  TNF-a GG genotype but not elevated in the GSTM1 positive group. The effects of Pb on
 4                  several physiological systems have been hypothesized to be mediated by the generation
 5                  of ROS (Daggett et al..  1998). Thus, the null variant of GSTM1, which is associated with
 6                  reduced elimination of ROS, may increase the risk of Pb-associated immune effects. The
 7                  results for the TNF-a polymorphism are difficult to interpret. The GG genotype is
 8                  associated with lower expression of TNF-a, and the literature is mixed with respect to
 9                  which variant increases risk of inflammation-related conditions. A study of adults in Italy
10                  did not provide quantitative results and only reported a lack of statistically significant
11                  correlation between blood Pb levels with Th2 or Thl cytokine levels in men (Boscolo et
12                  al., 1999) and women (Boscolo et al., 2000).

13                  Results from studies of occupationally-exposed adults also suggested that Pb exposure
14                  may be associated with decreases in Thl cytokines and increases in Th2 cytokines;
15                  however, analyses were mostly limited to comparisons of levels among different
16                  occupational groups with different mean blood Pb levels (Pi Lorenzo et al.. 2007;
17                  Valentino et al., 2007; Yiicesoy et al., 1997a) without consideration for potential
18                  confounding factors including other occupational exposures. An exception was a study of
19                  male foundry workers, pottery workers, and unexposed workers by Valentino et al.
20                  (2007). Although quantitative regression results were not provided, higher blood Pb level
21                  was associated with higher IL-10 and TNF-a with adjustment for age, BMI, smoking, and
22                  alcohol consumption. In analyses of blood Pb groups, levels of IL-2, IL-10, and IL-6 also
23                  increased from the lowest to highest blood Pb group. In contrast with most  other studies,
24                  both exposed worker groups had lower IL-4 levels compared with controls. In a similar
25                  analysis, DiLorenzo et al. (2007) separated exposed workers into intermediate
26                  (9.1-29.4 (ig/dL) and high (29.4-81.1 (ig/dL) blood Pb level groups, with unexposed
27                  workers comprising the low exposure group (blood Pb levels 1-11 (ig/dL). Mean TNF-a
28                  levels showed a monotonic increase from the low to high blood Pb group. Levels of
29                  granulocyte colony-stimulating factor (G-CSF) did not differ between the intermediate
30                  and high blood Pb groups among the Pb recyclers; however, G-CSF levels were higher in
31                  the Pb recyclers than in the unexposed controls. Furthermore, among all subjects, blood
32                  Pb  showed a strong, positive correlation with G-CSF. Yucesoy et al. (1997a) found lower
33                  serum levels of the Thl cytokines, IL-lp and IFN-y, in workers (mean blood Pb level of
34                  59.4 (ig/dL) compared with controls (mean blood Pb level of 4.8 (ig/dL); however levels
35                  of the Th2 cytokines, IL-2 and TNF-a levels, were similar between groups. As most
36                  occupationally-exposed cohorts represent populations highly exposed to Pb (with mean
37                  blood Pb levels >22 (ig/dL), effects observed within these cohorts  may not be
38                  generalizable to the population as a whole.
      November 2012                             5-35                  Draft - Do Not Cite or Quote

-------
 1                  In summary, animal, general population, and occupational studies suggest that exposure
 2                  to Pb increases the production of pro-inflammatory cytokines, skews the ratio of Thl and
 3                  Th2 cytokines to favor Th2 responses, and suppresses lymphocyte cytokine production.
      5.2.6       Endocrine Disruption
                    5.2.6.1      Hypothalamic-Pituitary-Gonadal Axis

 4                  Evidence indicates that Pb is a potent endocrine disrupting chemical found to be
 5                  associated with reproductive and developmental effects in both male and female animal
 6                  models (see Section 5.8). Pb may act both at multiple points along the hypothalamic-
 7                  pituitary-gonadal (HPG) axis and directly at gonadal sites. The HPG axis functions in a
 8                  closely regulated manner to produce circulating sex steroids and growth factors required
 9                  for normal growth and development. Long-term Pb exposure in animals has been shown
10                  to alter serum levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH),
11                  testosterone, and estradiol (Biswas and Ghosh. 2006; Rubio et al. 2006). Similar changes
12                  in serum HPG hormones have been observed after high-level Pb exposure in animals,
13                  resulting in blood Pb levels >20 (ig/dL (Dearth et al.. 2002; Ronisetal.. 1998b: Foster.
14                  1992; Sokol and Berman. 1991). Increases in serum LH and FSH have been associated
15                  with increasing concurrent blood Pb levels in adult women from the NHANES cohort
16                  (Krieg. 2007). The change in HPG hormones likely occurs through the inhibition of LH
17                  secretion and the reduction in the expression of the steroidogenic acute regulatory protein
18                  (StAR) (Huang and Liu. 2004; Srivastavaet al.. 2004; Huang et al.. 2002; Ronis et al..
19                  1996). StAR expression is the rate-limiting step essential in maintaining gonadotropin-
20                  stimulated steroidogenesis, which results in the formation of testosterone and estradiol.
21                  Prenatal and lactational Pb exposure (resulting in  3 (ig/dL blood Pb in the female rat
22                  offspring at PND31) was found to decrease basal  StAR synthesis, but not gonadotropin-
23                  stimulated StAR synthesis, suggesting that Pb may not directly  affect ovarian
24                  responsiveness to gonadotropin stimulation (Srivastava et al.. 2004). Instead, Pb may act
25                  at the hypothalamic-pituitary level to alter LH secretion, which is necessary to drive
26                  StAR production and subsequent sex hormone synthesis. Release of LH and FSH from
27                  the pituitary is controlled by gonadotropin-releasing hormone (GnRH). Pb exposure
28                  (10 (iM, 90 min) in rat brain median eminence cells can block GnRH release (Bratton et
29                  al.. 1994). Pb may also interfere with release of pituitary hormones through interference
30                  with cation-dependent secondary messenger systems that mediate hormone release and
31                  storage.
      November 2012                            5-36                  Draft - Do Not Cite or Quote

-------
 1                  Endocrine disruption may also be a result of altered hormone binding to endocrine
 2                  receptors. Prenatal and postnatal Pb exposure (20 ppm in drinking water) to rats was able
 3                  to decrease the number of estrogen, LH, and FSH receptors found in the uterus or ovaries
 4                  and receptor binding affinity ("Wiebe etal.. 1988; Wiebe and Barr. 1988). Altered
 5                  hormone binding ability may be due to the ion binding properties of Pb, resulting in
 6                  changes in receptor tertiary structure that will disrupt ligand binding. In addition,
 7                  Pb-induced changes in hormone levels that act as inducing agents for receptor synthesis
 8                  may affect the number of hormone receptors produced.

 9                  Some of these endocrine disrupting effects of Pb have been related to the generation of
10                  ROS. Treatment with antioxidants has been able to counteract a number of the endocrine
11                  disrupting effects of Pb, including apoptosis and decreased sperm motility and production
12                  (Salawu et al.. 2009: Shan  et al. 2009: Madhavi et al.. 2007: Rubio et al.. 2006: Wang et
13                  al.. 2006a: Hsu etal.. 1998b). Direct generation of ROS in epididymal spermatozoa was
14                  observed after Pb treatment in rats (i.p. 20 or 50 ppm, 6 weeks) (Hsu etal.. 1998a). In
15                  addition, lipid peroxidation has been observed in Pb-treated rats (i.p. 0.025 ppm, 15 days)
16                  (Pandya et al.. 2012). Lipid peroxidation in the seminal plasma was significantly
17                  increased in a group of Pb-exposed workers with high blood Pb levels (>40 (ig/dL) than
18                  in unexposed controls (Kasperczyk et al.. 2008).

19                  The liver is often associated with the HPG axis due in part to its production of insulin-
20                  like growth factor 1 (IGF-1). Children with higher concurrent blood Pb levels (>4 (ig/dL)
21                  (Huseman et al.. 1992) and Pb-exposed animals (blood Pb level of 14 (ig/dL) (Pine etal..
22                  2006: Dearth et al., 2002) and gonadal cells (46 ppm Pb exposure) (Kolesarova et al.,
23                  2010) have shown a decrease in plasma IGF-1, which may be the result of decreased
24                  translation or secretion of IGF-1 (Dearth et al., 2002). IGF-1 also induces LH-releasing
25                  hormone release, such that IGF-1 decrements may explain decreased LH and estradiol
26                  levels. IGF-1 production is stimulated by growth hormone (GH)  secreted from the
27                  pituitary gland and could be the result of GH depletion.

28                  A number of studies have revealed that Pb exposure affects the dynamics of growth (see
29                  Section 5.8.1). Decreased growth after Pb exposure could be the  result of Pb-induced
30                  decreased GH levels (Berry et al.. 2002: Camoratto et al.. 1993: Huseman et al.. 1992:
31                  Huseman et al., 1987). This decrease in GH could be a result of decreased release of GH
32                  releasing hormone  (GHRH) from the hypothalamus or disrupted GHRH binding to its
33                  receptor, which has been reported in vitro after Pb treatment (IC50 free Pb in solution 5.2
34                  x  10"5 (iM, 30 minutes) (Lauet al.. 1991). GH secretion may also be altered from
35                  decreased testosterone, a result of Pb exposure.
      November 2012                             5-37                  Draft - Do Not Cite or Quote

-------
                    5.2.6.2     Hypothalamic-Pituitary-Thyroid Axis

 1                  The evidence for the effects of Pb exposure on the hypothalamic-pituitary-thyroid (HPT)
 2                  axis is mixed. Pb exposure impacts a variety of components in the thyroid hormone
 3                  system. A number of occupational studies (blood Pb levels >7.3 (ig/dL) have shown that
 4                  elevated blood Pb are associated with lower thyroxine (T4) (and free T4 levels) without
 5                  alteration in triiodothyronine (T3), suggesting that long-term Pb exposure may depress
 6                  thyroid function in workers (Dundar et al.. 2006; Tuppurainen et al.. 1988; Robins et al..
 7                  1983). However, animal studies on thyroid hormones have shown mixed results.
 8                  Pb-exposed cows (blood Pb levels >51 (ig/dL) were reported to have an increase in
 9                  plasma T3 and T4 levels (Swarup et al.. 2007). whereas mice and chickens manifested
10                  decreased serum T3 concentrations after Pb exposure, which was accompanied by
11                  increased lipid peroxidation (Chaurasia et al..  1998; Chaurasia and Kar. 1997). Both
12                  decreased serum T3 and increased lipid peroxidation were restored by vitamin E
13                  treatment, suggesting the disruption of thyroid hormone homeostasis could be a result of
14                  altered membrane architecture and oxidative stress; however, no data were provided to
15                  exclude changes in Pb kinetics as the mechanism of protection (Chaurasia and Kar.
16                  1997).

17                  Decreased T4 and T3 may be the result of altered pituitary release of thyroid stimulating
18                  hormone (TSH). However, several studies have reported higher TSH levels in high-level
19                  Pb-exposed workers (blood Pb levels >39 (ig/dL) (Lopez et al.. 2000; Singh et al.. 2000;
20                  Gustafson et al.. 1989). which would result in increased T4 levels. Overall, results on the
21                  effects of Pb on the HPT axis are inconclusive.
      November 2012                             5-38                  Draft - Do Not Cite or Quote

-------
      5.2.7       Cell Death and Genotoxicity

 1                  A number of studies have attempted to characterize the genotoxicity of inorganic Pb in
 2                  human populations, laboratory animals, and cell cultures. Endpoints investigated include
 3                  DNA damage (single- and double-strand breaks, DNA-adduct formation), mutagenicity,
 4                  clastogenicity (sister chromatid exchange, micronucleus formation, chromosomal
 5                  aberrations), and epigenetic changes (changes in gene expression, DNA methylation,
 6                  mitogenesis). It is important to note that numerous studies have utilized exposure to
 7                  Pb chromate to investigate genotoxicity endpoints; some studies have specifically
 8                  attributed the observed increases in DNA damage and clastogenicity to the chromate ion
 9                  while others have not. Due to the uncertainty regarding whether observed genotoxic
10                  effects are due to chromate or Pb in studies using this form of inorganic Pb, only studies
11                  utilizing other forms of inorganic Pb (e.g., Pb nitrate, acetate, chloride, sulfate) are
12                  discussed below. Overall, evidence indicates that in vitro or in vivo exposure to various
13                  Pb compounds can increase risk of genotoxic effects, including DNA  damage,
14                  clastogenicity, and mutagenicity.
                    5.2.7.1      DNA Damage

15                  A number of studies in human populations have observed associations between indicators
16                  of Pb exposure and increased DNA damage, as measured as DNA strand breaks. Most of
17                  these associations have been observed in occupationally-exposed populations (Grover et
18                  al..201Q: Minozzo etal.. 2010: Shaik and Jamil. 2009: Danadevi et al.. 2003: Hengstler
19                  etal.. 2003: Palus etal.. 2003: Fracasso et al.. 2002: de Restrepo et al.. 2000). Evidence
20                  overall was equivocal in regard to how blood Pb levels correlated with DNA damage:
21                  Fracasso et al. (2002) observed that DNA damage increased with increasing blood Pb
22                  levels (blood Pb levels, <25, 25-35, and >35 (ig/dL), whereas Palus et al. (2003) (mean
23                  blood Pb level: 50.4 (ig/dL [range: 28.2 to 65.5 ng/dL]) and Minozzo et al. (2010) (mean
24                  [SD]: 59.43 (ig/dL [28.34]) observed no correlation. Hengstler et al. (2003) examined
25                  workers exposed to Pb, Cd, and Co and observed that neither blood (mean: 4.4 [IQR:
26                  2.84-13.6] (ig/dL) nor air Pb levels (mean:  3.0 [IQR: 1.6-50.0] (ig/m3) were associated
27                  with DNA damage when examined alone, but that blood Pb influenced the occurrence of
28                  single strand DNA breaks when included in a multiple regression model along with Cd in
29                  air and blood and Co in air.

30                  A few studies were found that investigated Pb-induced DNA damage resulting from
31                  nonoccupational exposures. Mendez-Gomez (2008) observed that children attending
32                  grade schools at close and intermediate distances to a Pb smelter had mean (range) blood
33                  Pb levels of 28.6 (11.4 to 47.5) and 19.5 (11.3 to 49.2) ng/dL, respectively, compared to
      November 2012                             5-39                  Draft - Do Not Cite or Quote

-------
 1                  blood Pb level of 4.6 (0.1 to 8.7) (ig/dL for children living distant to the smelter. DNA
 2                  damage in lymphocytes was higher in children living nearest to the smelter, compared to
 3                  the children at the intermediate distance, but was not different from children living
 4                  farthest away from the smelter. Multivariate analysis (which considered children urinary
 5                  As levels, highest in children farthest from the smelter), revealed no statistically
 6                  significant associations between DNA damage and blood Pb level. Further, DNA repair
 7                  ability was also observed to be unrelated to blood Pb levels. Alternatively, Yanez et al.
 8                  (2003) observed that children living close to a mining complex (mean [range] blood Pb
 9                  level: 11.6 [3.0 to 19.5] (ig/dL) did have higher levels of DNA damage compared to
10                  control children who lived further away from the mining facility (mean [range] blood Pb
11                  level: 8.3  [3.0 to 25.0] (ig/dL).

12                  Pb-induced DNA damage was observed in multiple animal studies. In mice exposed to Pb
13                  (blood Pb level of 0.68 ug/dL) via inhalation for up to 4 weeks, differential levels of
14                  DNA damage were observed in different organ systems, with only the lung and the liver
15                  demonstrating statistically greater DNA damage compared to the respective organ
16                  controls after acute exposure (Valverde et al.. 2002). Statistically elevated levels of DNA
17                  damage were observed in the kidneys, lungs, liver, brain,  nasal cavity, bone marrow, and
18                  leukocytes of mice exposed to Pb over a period of 4 weeks, although variability was high
19                  in all groups. The magnitude of the DNA damage was characterized as weak and did not
20                  increase with increasing durations of exposure. In mice given Pb nitrate (0.7 to
21                  89.6 mg/kg) by gavage for 24, 48, or 72 hours, or 1 or 2 weeks, single strand DNA breaks
22                  in white blood cells were  observed but did not increase with increasing concentration
23                  (Devi et al.. 2000). The three highest concentrations had responses that were similar in
24                  magnitude to each other and were  actually lower than the responses to the lower
25                  concentrations tested. Xu et al. (2008) exposed mice to 10-100 mg/kg Pb acetate via
26                  gavage for four weeks and observed a concentration-dependent increase in DNA single
27                  strand breaks in white blood cells that was statistically significant at 50 and 100 mg/kg.
28                  The authors characterized the observed DNA damage as severe. Pb nitrate induced DNA
29                  damage in primary spermatozoa in rats (blood Pb  levels of 19.5 and 21.9 (ig/dL) over that
30                  in control rats (Nava-Hernandez et al.. 2009). The level of DNA damage was not
31                  concentration dependent and was comparable in both exposure groups. Narayana and Al-
32                  Bader (2011) observed no increase in DNA damage in the livers of rats exposed to 5,000
33                  or 10,000 ppm Pb nitrate  in drinking water for 60  days. Interestingly, although the results
34                  were not statistically significant and were  highly variable within exposure groups, DNA
35                  fragmentation appeared to be lower in the exposed animals.

36                  Studies investigating Pb-induced DNA damage in human cell cultures were
37                  contradictory. Pb acetate did not induce DNA strand breaks in human HeLa cells when
38                  exposed in vitro to 500 (iM Pb  acetate for 20-25 hours or 100 (iM for 0.5-4 hours
      November 2012                             5-40                  Draft - Do Not Cite or Quote

-------
 1                  (Hartwig et al., 1990; Snyder and Lachmann. 1989). Pb nitrate, administered to
 2                  lymphoma cells in vitro at 1,000-10,000 (iM for 6 hours, did not result in any
 3                  DNA-protein crosslinks (Costa et al.. 1996). Pb acetate was observed by Wozniak and
 4                  Blasiak (2003) to result in DNA single and double strand breaks in primary human
 5                  lymphocytes exposed in vitro to 1-100 (iM for 1 hour, although the pattern of damage
 6                  was peculiar.  DNA damage was greater in cells exposed to 1 or 10 (iM, compared to
 7                  those exposed to 100 (iM. DNA-protein crosslinks were only observed in the 100 (iM
 8                  exposure group, suggesting that the decreased strand breaks observed in the high
 9                  exposure group may be a result of increased crosslinking in this group. Pasha Shaik et al.
10                  (2006) also observed DNA damage in human lymphocytes exposed in vitro to
11                  2,100-3,300 (iM Pb nitrate for 2 hours. Although there was a concentration-dependent
12                  increase in DNA damage from 2,100-3,300 (iM, no statistics were reported and no
13                  unexposed control group was included, making it difficult to interpret these results.
14                  Gastaldo et al. (2007) observed that in vitro exposure of human endothelial cells to
15                  1-1,000 uM Pb nitrate  for 24 hours resulted in a concentration-dependent increase in
16                  DNA double strand breaks.

17                  Studies in animal cell lines collectively were equally as ambiguous as those using human
18                  cell lines. Zelikoff et al. (1988) and Roy and Rossman (1992) reported that Pb acetate
19                  (concentration not stated and 1,000 uM, respectively) did not induce single or double
20                  DNA strand breaks or  DNA-protein or DNA-DNA crosslinks in CHV79 cells. However,
21                  both Xu et al. (2006) and Kermani  et al. (2008) reported Pb acetate-induced DNA
22                  damage in undifferentiated PC12 cells exposed to 0.1, 1, or 10 (iM for 24 hours; and in
23                  bone marrow mesenchymal stem cells exposed to 60 (iM for 48 hours, respectively.
24                  Wedrychowski et al. (1986)  reported that DNA-protein crosslinks were induced in a
25                  concentration-dependent manner in hepatoma cells exposed to 50-5,000 (iM Pb nitrate
26                  for 4 hours. Pb acetate and Pb nitrate increased the incidence of nick translation in
27                  CHV79 cells when a bacterial DNA polymerase was added.

28                  Pb exposure has also been shown to inhibit DNA repair mechanisms. Pb acetate did not
29                  induce single strand DNA breaks in HeLa cells  exposed to 500 uM for 20-25 hours
30                  (Hartwig etal., 1990).  However, exposure to both Pb acetate and UV light resulted in
31                  increased persistence of UV-induced strand breaks, compared with exposure to UV light
32                  alone. Similar effects were seen in  hamster V79 cells:  UV-induced mutation rates and
33                  SCE frequency was exacerbated by co-incubation with Pb acetate. Taken together, these
34                  data suggest that Pb  acetate interferes with normal DNA repair mechanisms triggered by
35                  UV exposure alone.  Pb nitrate was observed to affect different DNA double strand break
36                  repair pathways in human endothelial cells exposed in vitro to 100 uM for 24 hours.
37                  Exposure to Pb inhibited nonhomologous end joining repair, but increased two other
38                  repair pathways, MRE11 -dependent and Rad51 -related repair (Gastaldo et al.. 2007).
      November 2012                            5-41                  Draft - Do Not Cite or Quote

-------
 1                  Interestingly, in contrast to the above studies, exposure of lung carcinoma cells to 100,
 2                  300, or 500 uM Pb acetate for 24 hours resulted in an increase in nucleotide excision
 3                  repair efficiency (Li et al., 2008a). Roy and Rossman (1992) observed an increase in UV-
 4                  induced mutagenicity when CHV79 cells were co-exposed to 400 uM Pb acetate (a
 5                  nonmutagenic concentration of Pb acetate), indicating an inhibition of DNA repair.
 6                  Treatment of Chinese hamster ovary cells to 0.5-500 uM Pb acetate resulted in a
 7                  concentration-dependent accumulation of apurinic/apyrimidinic site incision activity,
 8                  indicating that DNA repair was diminished (McNeill et al..  2007).
                    5.2.7.2     Mutagenicity

 9                  Only one human study was found that investigated Pb-induced mutagenicity. Van
10                  Larebeke et al. (2004) investigated the frequency of mutations in the hypoxanthine
11                  phosphoribosyltransferase (HPRT) gene in Flemish women without occupational
12                  exposures to Pb or to a number of other heavy metals and organic contaminants. Higher
13                  blood Pb level (1 Oth-90th percentile: 1.6 to 5.2 (ig/dL) was associated with greater HPRT
14                  mutation frequency than found in the total population. Also, women with high blood Pb
15                  levels (i.e., greater than the population median, not reported) demonstrated a greater
16                  mutation frequency compared to women with lower blood Pb levels.

17                  Pb-induced mutagenicity was investigated in a few studies using human cell cultures.
18                  Ye (1993) exposed human keratinocytes to  100 (iM to 1 * 105 (iM Pb acetate for 2-24
19                  hours. This study did not measure HPRT mutations directly, but rather measured the
20                  amount of tritium (3H) incorporated into DNA as an indicator of mutation. In the
21                  presence of 6-thioguanine, tritium incorporation was increased in exposed cells,
22                  indicating weak mutagenicity. Hwua and Yang (1998) reported that Pb acetate was not
23                  mutagenic in human foreskin fibroblasts exposed to 500-2,000 (iM for 24 hours.
24                  Pb acetate remained nonmutagenic in the presence of 3-aminotriazole, a catalase
25                  inhibitor, indicating that oxidative metabolism did not play a part in potential
26                  mutagenicity of Pb. Exposure to Pb acetate  alone did not induce mutagenicity in lung
27                  carcinoma cells (100-500 (iM for 24 hours) or fibroblasts (300-500 (iM for 24 hours) (U
28                  et al., 2008a; Wang et al., 2008c). However, pretreatment with PKC inhibitors before Pb
29                  treatment did result in statistically significant increases in mutagenicity in both cell lines.

30                  Results from investigations into Pb-induced mutagenicity using animal cell lines were as
31                  equivocal as were the findings from human cell line studies, although the mixed findings
32                  may be reflective of specific  Pb compounds used. Pb acetate was observed to be
33                  nonmutagenic (HPRT assay) in CHV79 cells exposed to 1-25 (iM of the  compound for
34                  24 hours (Hartwig et al.. 1990). but elicited a mutagenic response in CHV79 cells (gpt
      November 2012                             5-42                  Draft - Do Not Cite or Quote

-------
 1                  assay) exposed to 1,700 (iM for 5 days (Roy and Rossman. 1992). Pb acetate was
 2                  observed to be nonmutagenic (HPRT assay) in Chinese hamster ovary cells exposed to
 3                  5 (jJVI for 6 hours (McNeill et al.. 2007). The implication of mutagenicity in the latter
 4                  study is complicated by the concurrent observation of severe cytotoxicity at the same
 5                  concentration. Pb nitrate was alternatively found to be nonmutagenic in CHV79 cells (gpt
 6                  assay) exposed to 0.5-2,000 (iM for 5 days (Roy and Rossman. 1992) but mutagenic in
 7                  the same cell line (HPRT assay) exposed to 50-5,000 (iM for 5 days (Zelikoff et al..
 8                  1988). However, mutagenicity was only observed at 500 (JVI, and was higher than that
 9                  observed at higher Pb concentrations. Pb sulfate was also observed to be mutagenic in
10                  CHV79 cells (HPRT assay) exposed to 100-1,000 (iM for 24 hours, but as with
11                  Pb nitrate, it was not concentration-dependent (Zelikoff etal. 1988). Pb chloride was the
12                  only Pb compound tested in animal cell lines that was consistently mutagenic: three
13                  studies from the same laboratory observed concentration-dependent mutagenicity in the
14                  gpt assay in Chinese hamster ovary cells exposed to 0.1-1 (iM Pb chloride for one hour
15                  (Ariza and Williams. 1999: Arizaetal.. 1998: Ariza and Williams. 1996V
                    5.2.7.3      Clastogenicity

16                  Clastogenicity is the ability of a compound to induce chromosomal damage, and is
17                  commonly observed as sister chromatid exchange (SCE), micronuclei formation, or
18                  incidence of chromosomal aberrations (i.e., breaks or gaps in chromosomes). Pb has been
19                  shown to increase sister chromatid exchange, micronuclei formation, and chromosomal
20                  aberrations in human populations, exposed animal models, and in vitro experiments.


                    Sister Chromatid Exchange

21                  An association between blood Pb levels (means: 10.48 - 86.9 (ig/dL) and sister chromatid
22                  exchange (SCE) was observed in a number of occupational studies (Wiwanitkit et al..
23                  2008: Duvdu etal..  2005: Palus etal.. 2003: Duvdu etal.. 2001: Pinto et al.. 2000:
24                  Bilban. 1998: Anwar and Kamal. 1988: Huang etal.. 1988). In most studies that
25                  attempted to investigate the concentration-response relationship in workers, no
26                  association was observed between increasing blood Pb levels and the number of SCE
27                  (Palus etal.. 2003: Duvdu etal.. 2001:  Pinto  et al.. 2000V However, Huang et al. (1988)
28                  did observe increased SCE in exposed workers in the two highest blood Pb groups (52.1
29                  and 86.9 (ig/dL), with a statistically significant association observed in the 86.9 (ig/dL
30                  group. Pinto et al. (2000) did report an  association with duration of exposure (range of
31                  years exposed: 1.6-40). Two  studies reported no correlation between occupational
32                  exposure to Pb and number of SCE (Rajah and Ahuja. 1996: Rajah and Ahuja. 1995).
      November 2012                             5-43                  Draft - Do Not Cite or Quote

-------
 1                  Mielzynska et al. (2006) found no association between blood Pb level and SCEs in
 2                  children in Poland. Children had an average blood Pb level of 7.69 (ig/dL and 7.87
 3                  SCEs/cell.

 4                  Pb exposure has been observed to induce SCEs in multiple laboratory animal studies. In
 5                  mice treated with up to 100 mg/kg Pb acetate i.p., Pb induced SCEs with 50 and
 6                  100 mg/kg (Fahmy. 1999). Pb nitrate, also administered i.p. induced the formation of
 7                  increased SCE levels in a concentration-dependent manner (10-40 mg/kg) in the bone
 8                  marrow of exposed mice (Dhir et al., 1993). Nayak et al. (1989b) treated pregnant mice
 9                  with 100-200 mg/kg Pb nitrate via i.v. injection and observed an increase in the number
10                  of SCE in dams at  150 and 200 mg/kg; no increases in SCE levels were observed in the
11                  fetuses. Tapisso et al. (2009) treated rats with 21.5 mg/kg Pb acetate (l/10th the LD50) via
12                  i.p. injection on alternating days for 11 or 21 days, for a total of 5 or 10 treatments.
13                  Induction of SCE in the bone marrow of exposed rats was increased over controls in a
14                  statistically significant duration-dependent manner. It is important to note that all of these
15                  studies utilized an injection route of exposure that may not be relevant to routes of
16                  exposure in the human population (e.g., air, drinking water exposure).

17                  Few studies were found that investigated SCE formation due to Pb exposure in human
18                  cell lines. Statistically significant, concentration-dependent increases in SCE were
19                  observed in human lymphocytes obtained from a single donor when incubated with 1,5,
20                  10, or 50 (iM Pb nitrate (Ustundag and Duydu. 2007). Melatonin and N-acetylcysteine
21                  were reported to ameliorate these effects, indicating Pb may induce increases in SCE
22                  levels through increased oxidative stress. Pb chloride was also observed to increase SCE
23                  levels in human lymphocytes exposed to 3 or 5 ppm (Turkez et al.. 2011).

24                  Evidence from studies investigating SCE in rodent cells was more equivocal than that in
25                  human cells. Pb sulfate, acetate, and nitrate were found not to induce SCE in CHV79
26                  cells (Hartwig et al.. 1990; Zelikoff et al.. 1988). Both of these studies only examined
27                  25-30 cells per concentration, reducing their power to detect Pb-induced increases in SCE
28                  levels. Cai and Arenaz (1998). on the other hand, used 100 cells per treatment and
29                  observed that exposure to 0.05-1  (iM Pb nitrate for 3-12 hours resulted in a weak,
30                  concentration-dependent increase in SCE levels in Chinese hamster ovary cells. Lin et al.
31                  (1994) also observed a concentration-dependent increase in SCE levels in Chinese
32                  hamster cells exposed to 3-30 (iM Pb nitrate for 2 hours.
      November 2012                             5-44                   Draft - Do Not Cite or Quote

-------
                     Micronucleus Formation

 1                   Pb-induced micronucleus formation was observed in numerous occupational studies
 2                   (Groveretal..201Q: Khanetal.. 2010b: Minozzo etal.. 2010: Shaik and Jamil. 2009:
 3                   Minozzo et al.. 2004: Palus etal.. 2003: Vaglenov etal.. 2001: Pinto et al.. 2000: Bilban.
 4                   1998: Vaglenov et al.. 1998). Pinto et al. (2000) observed increased micronuclei in
 5                   exposed workers with an average blood Pb level of 10.48 (ig/dL compared with
 6                   unexposed controls. In studies investigating the correlation between blood Pb levels and
 7                   micronucleus formation, no association was observed (Minozzo et al.. 2010: Minozzo et
 8                   al.. 2004: Palus etal.. 2003: Pinto et al.. 2000). although Pinto et al. (2000). Grover et al.
 9                   (2010). and Minozzo et al. (2010) did report an association between micronuclei
10                   formation and duration of exposure. Mielzynska et al. (2006) investigated micronucleus
11                   formation in  a nonworker population and reported a statistically significant positive
12                   correlation between blood Pb levels and micronuclei frequency in children in Poland.
13                   Children, with an average blood Pb level of 7.69 (ig/dL, were observed to have 4.44
14                   micronucleated cells per 1,000 cells analyzed. Children with blood Pb levels greater than
15                   10 (ig/dL had significantly more micronucleated cells than did children with blood Pb
16                   levels less than 10 (ig/dL.

17                   Micronucleus formation in response to Pb exposure has been observed in rodent animal
18                   studies. Celik et al. (2005) observed that exposure of female rats to Pb acetate (140, 250,
19                   or 500 mg/kg once per week for 10 weeks) resulted in statistically significant increases in
20                   numbers of micronucleated polychromatic erythrocytes (PCEs) compared to controls.
21                   Similarly, Alghazal et al. (2008b) exposed rats to Pb acetate (100 ppm daily for 125 days)
22                   and observed statistically significant increases in micronucleated PCEs in both sexes.
23                   Tapisso et al. (2009) treated rats with Pb acetate (21.5 mg/kg; l/10th the LD50) via i.p.
24                   injection on alternating days for 11 or 21 days, for a total of 5 or 10 exposures. Formation
25                   of micronuclei in the bone marrow of exposed rats was increased over formation in
26                   controls in a  significant duration-dependent manner.  Two further studies investigated
27                   formation of micronuclei in the bone marrow of exposed mice: Roy et al. (1992) treated
28                   mice with Pb nitrate (10 or 20 mg/kg, i.p.) and observed a concentration-dependent
29                   increase in micronuclei, whereas Jagetia and Aruna (1998) observed an  increase in
30                   micronuclei in mice treated with Pb nitrate (0.625-80 mg/kg, i.p.), though the increase
31                   was not concentration-dependent. Mice exposed to Pb acetate (0.1 (ig/L via drinking
32                   water, a more environmentally relevant route of exposure,  for 90 days) had statistically
33                   significant increases in micronucleated PCEs (Marques et al.. 2006).

34                   A few studies were found that reported increased micronucleus formation in human cell
35                   lines treated with Pb. Concentration-dependent micronucleus formation was observed in
36                   human lymphocytes when exposed in vitro to either 1, 5, 10, or 50 (iM Pb nitrate or 3 or
37                   5 ppm Pb chloride (Turkez etal.. 2011: Ustundag and Duydu. 2007). Gastaldo et al.

      November 2012                             5-45                  Draft - Do Not Cite or Quote

-------
 1                  (2007) also observed a concentration-dependent increase in micronuclei in human
 2                  endothelial cells exposed in vitro to 1-1,000 (iM Pb nitrate for 24 hours. Animal cell
 3                  culture studies investigating micronuclei formation produced contrasting results. One
 4                  study observed that micronuclei were not induced in Chinese hamster cells exposed to
 5                  3-30 (iM Pb nitrate for 2 hours (Lin et al.. 1994). whereas the other observed that
 6                  Pb acetate induced a concentration-dependent increase in Chinese hamster cells when
 7                  administered at 0.03-10 (iM for 18 hours (Bonacker et al., 2005).
                    Chromosomal Aberrations

 8                  Chromosomal aberrations (e.g., chromosome breaks, nucleoplasmic bridges, di- and a-
 9                  centric chromosomes, and rings) were examined in a number of occupational studies
10                  (Groveretal..201Q: Shaik and Jamil. 2009; Pinto et al.. 2000; Bilban. 1998; De et al..
1 1                  1995; Huang etal.. 1988). No correlation was observed between increasing blood Pb
12                  level and the number of chromosomal aberrations, although an association was observed
13                  between duration of exposure and chromosomal damage (Grover et al.. 2010; Pinto et al..
14                  2000). Other studies reported no association between occupational exposure to Pb and
15                  chromosomal aberrations (Anwar and Kamal. 1988; Andreae.  1983). Smejkalova (1990)
16                  observed greater chromosomal damage and aberrations in children living in  a heavily
17                  Pb-contaminated area of Czechoslovakia compared with children living in an area with
18                  less contamination, although the difference between the two areas was not statistically
19                  significant. Blood Pb levels were comparable between children living in the
20                  Pb-contaminated area and children living in the less contaminated area (low 30s versus
21                  high 20s (ig/dL, respectively), indicating there may not be enough of a dose  contrast to
22                  detect a significant difference in aberration rates.

23                  The majority of animal studies investigating Pb-induced  genotoxicity focused on the
24                  capacity of Pb to produce chromosomal damage. Fahmy (1999) treated mice with
25                  Pb acetate (25-400 mg/kg i.p.), either as a single dose or repeatedly for 3,  5, or 7 days.
26                  Chromosomal damage was observed to increase in bone  marrow cells (100-400 mg/kg)
27                  and spermatocytes (50-400 mg/kg) in a concentration-dependent manner after both
28                  dosing regimens. Pb nitrate was also observed to produce concentration-dependent
29                  chromosomal damage in mice treated i.p. to a single dosage of 5, 10, or 20 mg/kg (Dhir
30                  et al..  1992b). In a similar experiment, Dhir et al. (1990)  treated mice with Pb nitrate (10,
3 1                  20, or 40 mg/kg) and saw an increase in chromosomal aberrations, although there was no
32                  concentration-dependent response as the response was similar in all concentrations tested.
33                  Nayak et al. (1989b) treated pregnant mice with Pb nitrate (100, 150, or 200 mg/kg via
34                  i.v. injection) and observed no chromosomal gaps or breaks in dams or fetuses but did
35                  report some karyotypic chromosomal damage and weak aneuploidy at the low dose. In a
36                  similar experiment, low levels of chromosomal  aberrations were observed in dams and

      November 20 12                             5-46                  Draft - Do Not Cite or Quote

-------
 1                  fetuses injected with Pb nitrate (12.5, 50, or 75 mg/kg), but there was no concentration-
 2                  dependent response reported and few cells were analyzed (Navaketal.. 1989a). In rats
 3                  given Pb acetate (2.5 mg/100 g body weight, i.p. daily for 5-15 days or 10-20 mg/100 g
 4                  once and analyzed after 15 days), Pb-induced chromosomal aberrations were observed
 5                  (Chakraborty et al.. 1987). The above studies all are limited by the use of a route of
 6                  exposure that may not be relevant to human environmental exposures. However, studies
 7                  utilizing oral exposures also observed increases in chromosomal damage. Aboul-Ela
 8                  (2002)  exposed mice to Pb acetate (200 or 400 mg/kg by gavage for 5 days) and reported
 9                  that chromosomal damage was present in the bone marrow cells and spermatocytes of
10                  animals exposed to both concentrations. Dhir et al. (1992a) also observed a
11                  concentration-dependent increase in chromosomal damage in mice exposed via gavage,
12                  albeit at much lower concentrations: either 5 or 10 mg/kg. Nehez et al. (2000) observed a
13                  Pb-induced increase in aneuploidy and percent of cells with damage after exposure to
14                  10 mg/kg administered by gavage 5 days a week for 4 weeks. In the only study that
15                  investigated dietary exposure, El-Ashmawy et al. (2006) exposed mice to 5,000 ppm
16                  Pb acetate in feed, and observed an increase in abnormal cells and frequency of
17                  chromosomal damage.

18                  In the few studies that investigated the capacity of Pb to induce chromosomal damage in
19                  human cell lines, Pb exposure did not induce chromosomal damage. Wise et al. (2005;
20                  2004) observed that Pb glutamate was not mutagenic in human lung cells exposed in vitro
21                  to 250-2,000 (iM for 24 hours. Pasha Shaik et al. (2006) observed that Pb nitrate did not
22                  increase chromosomal aberrations in primary lymphocytes (obtained from  healthy
23                  volunteers) when incubated with 1,200 or 2,000 (iM for 2 hours. Studies utilizing animal
24                  cell lines generally supported the finding of no Pb-induced chromosomal damage in
25                  human cell lines. Pb nitrate was found to induce no chromosomal damage in Chinese
26                  hamster ovary cells exposed to 500-2,000 (iM for 24 hours (Wiseetal.. 1994). 3-30 (iM
27                  for 2 hours (Linetal.. 1994). or 0.05 -1 (JVI for 3 -12 hours (Cai and Arenaz. 1998). Wise
28                  et al. (1994) did observe increased chromosomal damage in Chinese hamster ovary cells
29                  exposed to 1,000 (iM Pb glutamate for 24 hours, but did not see any damage in cells
30                  exposed to higher concentrations (up to 2,000
      November 2012                             5-47                  Draft - Do Not Cite or Quote

-------
                     5.2.7.4      Epigenetic Effects

 1                   Epigenetic effects are heritable changes in gene expression resulting without changes in
 2                   the underlying DNA sequence. A prime example of an epigenetic effect is the abnormal
 3                   methylation of DNA, which could lead to altered gene expression and cell proliferation
 4                   and differentiation. Possible indications of Pb-induced epigenetic changes include
 5                   alterations in methylation patterns in exposed rats, and alterations in mitogenesis and cell
 6                   proliferation in exposed humans and animals, as well as human and animal cell cultures.


                     DNA Methylation

 7                   A single i.v. injection of Pb nitrate (75 umol/kg) resulted in global hypomethylation of
 8                   hepatic DNA in rats (Kanduc et al..  1991). The observed hypomethylation in the liver
 9                   was associated with an increase in cell proliferation.  A few additional studies in humans
10                   observed that higher bone Pb levels  were associated with lower global DNA methylation
11                   patterns in adults and cord blood of newborns (Wright et al.. 2010;  Pilsner et al.. 2009).
12                   Hypomethylation specifically is associated with increased gene expression.  Changes in
13                   DNA methylation patterns could potentially lead to dysregulation of gene expression and
14                   altered tissue differentiation.


                     Mitogenesis

15                   Conflicting results have been reported regarding Pb-induced effects on mitogenesis, with
16                   both increased and decreased cell growth and mitogenesis. A discernible pattern of
17                   effects is difficult to detect when analyzing effects across human, in vivo animal, and in
18                   vitro studies. Only a few studies have investigated the mitogenic effects of Pb exposure
19                   in human populations indirectly by examining mitogenesis or the induction  of cell
20                   proliferation, which can be a consequence of epigenetic changes. These studies (Minozzo
21                   et al.. 2010; Minozzo et al.. 2004; Rajah and Ahuja. 1995) reported reduced mitogenesis
22                   in two groups of Pb-exposed workers compared with unexposed  controls (mean blood Pb
23                   levels: 35.4 (ig/dL, 59.4 (ig/dL, and  not reported, respectively). The observation of
24                   decreased cell division in exposed workers may indicate that cells suffered DNA damage
25                   and died during division, or that division was delayed to allow for DNA repair to occur. It
26                   is also possible that Pb exerts an aneugenic effect and arrests the  cell cycle.

27                   Many studies have investigated the ability of Pb to induce mitogenesis in animal models,
28                   and have consistently shown that Pb nitrate can stimulate DNA synthesis and cell
29                   proliferation in the liver of animals treated with  100 (iM Pb per kg body weight, via i.v.
30                   injection (Nakaiima et al.. 1995; Conietal.. 1992; Ledda-Columbano et al.. 1992;
31                   Columbano et al., 1990; Columbano et al., 1987). Shinozuka et al. (1996) observed that

      November 2012                             5-48                   Draft - Do Not Cite or Quote

-------
 1                   Pb-induced hepatocellular proliferation was similar in magnitude to that induced by
 2                   TNF-a at 100 (iM/kg; and Pb was observed to induce TNF-a in glial and nerve cells in
 3                   mice (and NF-KB, TNF-a, and iNOS in rat liver cells) from mice treated with Pb at
 4                   12.5 mg/kg and 100 (imol/kg, respectively (Cheng et al.. 2002; Menegazzi et al. 1997).
 5                   The only study that examined Pb exposure via inhalation (Pb acetate, 10,000 (iM for
 6                   4 weeks) resulted in increased cellular proliferation in murine lungs (Fortoul et al.. 2005).

 7                   Extensive research has been conducted investigating the potential effects of Pb on
 8                   mitogenesis in human and animal cell cultures. In human cell cultures, Pb acetate
 9                   inhibited cell growth in hepatoma cells (0.1-100 (iM for 2-6 days) (Heiman and Tonner.
10                   1995) and primary oligodendrocyte progenitor cells (1  (iM for 24 hours) (Deng and
11                   Poretz. 2002) but had no observable effects on growth  in glioma cells (0.01-10 (iM for
12                   12-72 hours) (Liu et al.. 2000). Pb glutamate had no effect on cell growth in human lung
13                   cells in vitro, but did increase the mitotic index (250-1,000 (iM exposure for 24 hours)
14                   (Wise et al.. 2005). The increase in the mitotic index was attributed to an arrest of the cell
15                   cycle at M-phase, and was not attributed to an actual increase of cell growth and
16                   proliferation. Gastaldo et al. (2007) also reported S and G2 cell cycle arrests  in human
17                   endothelial cells following exposure to 100 uM Pb nitrate for 24 hours. Conflicting
18                   results with regard to DNA synthesis were reported, with a concentration-dependent
19                   inhibition of DNA synthesis reported in hepatoma cells (1-100 (iM for 72 hours) (Heiman
20                   and Tonner. 1995). but an induction of synthesis observed in astrocytoma cells (1-50 (iM
21                   for 24 hours) (Lu et al.. 2002).

22                   In rat fibroblasts and epithelial cells, Pb acetate, Pb chloride, Pb  oxide, and Pb sulfate
23                   were all  observed to inhibit cell growth (10-1,000 (iM for 1-7 days and 0.078-320 (iM for
24                   48 hours, respectively) (lavicoli et al.. 2001;  Apostoli et al.. 2000). lavicoli et al. (2001)
25                   observed that in addition to inhibiting cell growth in rat fibroblasts, Pb acetate caused
26                   GS/M and S-phase arrest. Pb acetate decreased cell proliferation in mouse bone  marrow
27                   mesenchymal stem cells when administered at 20-100  uM for 48 hours (Kermani et al..
28                   2008). Pb nitrate was alternatively reported to increase (Lin et al.. 1994) and decrease
29                   (Cai and Arenaz. 1998) the mitotic index in Chinese hamster ovary cells exposed to 1 uM
30                   Pb nitrate. Lin et al. (1994) did not consider cell cycle arrest when measuring the mitotic
31                   index and did not observe a decrease at higher concentrations; in fact, the highest
32                   concentration tested, 30 uM, had a mitotic index equal to that in the untreated control
33                   cells.
      November 2012                             5-49                   Draft - Do Not Cite or Quote

-------
                    5.2.7.5     Gene Expression

 1                  A few animal studies have investigated the ability of Pb exposure to alter gene expression
 2                  in regard to phase I and II metabolizing enzymes. Suzuki et al. (1996) treated rats with
 3                  Pb acetate or Pb nitrate (100 ug/kg via i.p. injection) and observed an induction of GST-P
 4                  with both Pb compounds. The induction of GST-P by Pb was observed to occur on the
 5                  transcriptional level and to be dependent on the direct activation of the cis-element GPEI
 6                  enhancer. Degawa et al. (1993) reported that Pb nitrate (20, 50, or 100 umol/kg, via i.v.)
 7                  selectively inhibited CYP1A2 levels. Pb was shown not to inhibit CYP1A2 by direct
 8                  enzyme inhibition, but rather to decrease the amount of CYP1A2 mRNA. In contrast,
 9                  Korashy and El Kadi (2004) observed that exposure of murine hepatoma cells to
10                  Pb nitrate (10-100 uM for 24 hours) increased the amount of CYP1A1 mRNA while not
11                  influencing the  activity of the enzyme. NAD(P)H:quinone oxidoreductase and GST Ya
12                  activities and mRNA levels were increased after exposure to Pb. Incubation of primary
13                  human bronchial epithelial cells with Pb acetate (500 ug/L for 72 hours) resulted in the
14                  up-regulation of multiple genes associated with cytochrome P450 activity, glutathione
15                  metabolism, the pentose phosphate pathway, and amino acid metabolism (Glahn et al..
16                  2008).

17                  Additional animal studies provide further evidence that exposure to Pb compounds can
18                  perturb gene expression. Zawia and Harry (1995) investigated whether the observed
19                  Pb-induced disruption of myelin formation in rat pups exposed postnatally was due to
20                  altered gene expression. In pups exposed to 2,000 ppm Pb acetate via lactation from
21                  PND1-PND20, the expression of proteolipid protein, a major structural constituent of
22                  myelin, was elevated (statistically significant) at PND20, compared to controls. The
23                  expression of another structural element of myelin (myelin basic protein) was similarly
24                  elevated in exposed animals,  although not significantly so. The expression of both genes
25                  returned to control levels 5 days following the termination of exposure. These data
26                  suggest that altered gene expression in structural myelin proteins due to Pb exposure may
27                  be responsible for observed alterations in abnormal conduction of nerve impulses. Long
28                  et al. (2011) investigated the Pb-induced increase in ABCC5, an ATP-binding cassette
29                  transporter, in embryonic and adult zebrafish. In the initial in vitro portion of the study,
30                  exposure of zebrafish fibroblasts to 20 (iM Pb nitrate for 24 hours significantly increased
31                  the induction of ABBC5 mRNA 2.68-fold over controls. Similar levels of induction were
32                  observed when  embryonic zebrafish were exposed to 5 (iM for 24 to 96 hours;
33                  specifically, induction of ABCC5 was seen in the livers of developing embryos. In adult
34                  fish, induction of ABCC5 was observed in the brains, intestines, and kidneys of exposed
35                  fish, but a decrease was found in their livers. Induction of ABCC5 in adult fish was
36                  observed to attenuate the toxicity of Cd (but not Hg or As); however, in developing
37                  embryos, the attenuation of Pb-induced toxicity was not investigated. These findings

      November 2012                             5-50                  Draft - Do Not Cite or Quote

-------
 1                  indicate that increased expression of ABCC5 due to heavy metal exposure may play a
 2                  part in cellular defense mechanisms.
                     5.2.7.6     Apoptosis

 3                   Occupational exposure to Pb and induction of apoptosis in various cell types was
 4                   investigated in a few studies. The study that directly measured apoptosis reported that
 5                   exposure to Pb increased apoptosis of lymphocytes compared to nonexposed controls
 6                   (Minozzo et al.. 2010). whereas the others reported that two early indicators of apoptosis,
 7                   karyorrhexis and karyolysis, were elevated in occupationally exposed workers (Grover et
 8                   al.. 2010; Khan et al.. 2010b). Pb nitrate was also observed to induce apoptosis in the
 9                   liver of exposed animals (Columbano et al.. 1996; Nakaiima et al.. 1995). Apoptosis was
10                   observed in rat fibroblasts exposed in vitro to Pb acetate and rat alveolar macrophages
11                   exposed to Pb nitrate davicoli et al.. 2001; Shabani and Rabbani. 2000). Observation of
12                   Pb-induced apoptosis may represent the dysregulation of genetically-controlled cell
13                   processes and tissue homeostasis.
      5.2.8        Summary

14                  The diverse health effects of Pb are mediated through multiple, interconnected modes of
15                  action. Each of the modes of action discussed here has the potential to contribute to the
16                  development of a number of Pb-induced health effects (Table 5-2). While this section
17                  draws from earlier literature as well as newer lines of evidence, the inclusion of recent
18                  evidence does not qualitatively change the previous conclusions regarding individual
19                  modes of action. Rather, the more recent evidence agrees with, and thus strengthens these
20                  conclusions. Evidence for the majority of these modes of action is observed with blood
21                  Pb levels in humans ranging between 2 and 17 (ig/dL, with supporting evidence from
22                  animal and in vitro assays. As many of these  studies examined adults, with likely higher
23                  past than current Pb exposures, uncertainty exists as to the Pb exposure level, duration,
24                  frequency, and timing associated with these modes of action. The blood Pb levels or in
25                  vitro concentrations presented in Table 5-2 reflect the current evidence for these modes
26                  of action and are not intended to convey conclusions regarding specific thresholds. Also,
27                  the data presented in this table do not inform the exposure frequency and duration
28                  required to elicit a particular MOA.
      November 2012                             5-51                  Draft - Do Not Cite or Quote

-------
      Table 5-2     MOAs, their related health effects, and information on concentrations
                       eliciting the MOAs.
            Mode of Action
        [Related Health Effects
            (ISA Section)]
                                                       Concentrations or Doses (Conditions)
                           Blood Pb
                                                                     Dose
      Altered Ion Status
      [All Health Effects of Pb]
             3.5 ug/dL
             (Mean in cord blood; association with cord
             blood Ca2+ATPase pump activity)

             Hueletal. (2008)
                                                                     0.00005 uM free Pb2+
                                                                     (In vitro; 30 minutes; calmodulin activation
                                                                     assay)
                                                                     Kern et al. (2000)
      Protein Binding
      [Renal (5.5). Hematological
      Effects (5.7)1
             17.0ug/dL
             (Concurrent mean in adult workers with
             wildtype metallothionein expression;
             increased BP susceptibility)
             Chen et al. (201 Oa)
                                                                     50 uM Pb glutamate
                                                                     (In vitro; 24 hours; increased nuclear protein
                                                                     in neurological cell)
                                                                     Klann and Shelton (1989)
      Oxidative Stress
      [All Heath Effects of Pb]
             5.4 ug/dL
             (Concurrent mean in adult male workers;
             decreased CAT activity in blood)
             Conterato et al. (In Press)
                                                                     0.1 uM Pb acetate
                                                                     (In vitro; 48 hours; decreased cellular GSH in
                                                                     neuroblastoma cells)
                                                                     Chetty et al. (2005)
      Inflammation
      [Nervous System (5.3).
      Cardiovascular (5.4). Renal
      (5.5). Immune (5.6). Respiratory
      (5.9.6). Hepatic (5.9.1)1
             Among males with concurrent blood Pb
             > 2.5 ug/dL
             (Increased serum TNF-a and blood WBC
             count)
             Kim et al. (2007)
                                                                     0.01 uM Pb acetate
                                                                     (In vitro; 48 hours; increased cellular PGE2 in
                                                                     neuroblastoma cells)
                                                                     Chetty et al. (2005)
      Endocrine Disruption
      [Reproductive and
      Developmental Effects (5.8).
      Endocrine System (5.9.3). Bone
      and Teeth (5.9.4)1
             1.7 ug/dL
             (lowest level at which a relationship could be
             detected in adult women with both ovaries
             removed; increased serum FSH)
             Krieg (2007)
                                                                     10 uM Pb nitrate
                                                                     (In vitro; 30 minutes; displaced GHRH
                                                                     binding to rat pituitary receptors)
                                                                     Lau et al. (1991)
Cell Death/Genotoxicity
[Cancer (5.10). Reproductive
and Developmental Effects
(5.8). Bone and Teeth (5.9.4)1
                                    3.3 ug/dL
                                    (concurrent median in adult women;
                                    increased rate of HPRT mutation frequency)
                                    Van Larebeke et al. (2004)
                                                     0.03 uM Pb acetate
                                                     (In vitro; 18 hours; increased formation of
                                                     micronuclei)
                                                     Bonacker et al. (2005)
      aThis table provides examples of studies that report effects with low Pb dosages or concentrations; they are not the full body of
      evidence used to characterize the weight of the evidence. In addition, the levels cited are reflective of the data and methods
      available and do not imply that these modes of action are not acting at lower Pb exposure or blood Pb levels or that these doses
      represent the threshold of the effect. Additionally, the blood concentrations and doses (indicating Pb concentrations from in vitro
      systems) refer to the concentrations and doses at which these modes of action were observed. While the individual modes of
      action are related back to specific health effects sections (e.g., Nervous System, Cardiovascular), the concentrations and doses
      given should not be interpreted as levels at which those specific health effects occur.
 1

 2

 o
 J

 4

 5

 6

 7

 8

 9

10

11
The alteration of cellular ion status (including disruption of Ca2+ homeostasis, altered ion

transport mechanisms, and perturbed protein function through displacement of metal

cofactors) appears to be the major unifying mode of action underlying all subsequent

modes of action (Figure  5-1). Pb is well characterized to interfere with endogenous Ca2+

homeostasis (necessary as a cell signal carrier mediating normal cellular functions).

[Ca2+]j has been shown to increase after Pb exposure in a number of cell types including

bone, erythrocytes, brain cells, and white blood cells, due to the increased flux of

extracellular Ca2+ into the cell. This disruption of ion transport is due in part to the

alteration of the activity  of transport channels and proteins,  such as Na+/K+ATPase and

voltage-gated Ca2+ channels. Pb can interfere with these proteins through direct

competition between Pb and the native metals present in the protein metal binding
      November 2012
                                5-52
                                                                         Draft - Do Not Cite or Quote

-------
 1                   domain or through disruption of proteins important in Ca2+-dependent cell signaling, such
 2                   as PKC or calmodulin.

 3                   Disruption of ion transport not only leads to altered Ca2+ homeostasis, but it can also
 4                   result in perturbed neurotransmitter function. Pb has been shown to displace metal ions,
 5                   (such as Zn2+, Mg2+, and Ca2+) from proteins due to the flexible coordination number of
 6                   Pb and multiple ligand binding ability, leading to abnormal conformational changes in
 7                   proteins and altered protein function. Evidence for this metal ion displacement and
 8                   protein perturbation has been shown at t 10~6 (iM concentrations of Pb. Additional effects
 9                   of altered cellular ion status are the inhibition of heme synthesis and decreased cellular
10                   energy production due to perturbation of mitochondrial function.

11                   Although Pb can  bind to proteins within cells through interactions with side group
12                   moieties, thus potentially disrupting cellular function, protein binding of Pb may
13                   represent a mechanism by which cells protect themselves against the toxic effects of Pb.
14                   Intranuclear and intracytosolic inclusion body formation has been observed in the kidney,
15                   liver, lung, and brain following Pb exposure. A number of unique Pb binding proteins
16                   have been detected, constituting the observed inclusion bodies. The major Pb binding
17                   protein in blood is ALAD with carriers of the ALAD-2 allele potentially exhibiting
18                   higher Pb binding affinity. Additionally, metallothionein is an important protein in the
19                   formation of inclusion bodies and mitigation of the toxic effects of Pb.

20                   A second major mode of action of Pb is its role in the development of oxidative stress,
21                   due  in many instances to the antagonism of normal metal ion functions. The origin of
22                   oxidative  stress produced after Pb exposure is likely a multipathway process, resulting
23                   from oxidation of 5-ALA, NAD(P)H oxidase activation, membrane and lipid
24                   peroxidation, and antioxidant enzyme depletion. Through the inhibition of 5-ALAD (due
25                   to displacement of Zn by Pb), accumulated 5-ALA goes through an auto-oxidation
26                   process to produce ROS. Additionally, Pb can induce the production of ROS through the
27                   activation of NAD(P)H oxidase. Pb-induced ROS can interact with membrane lipids to
28                   cause a membrane and lipid peroxidation cascade.  Enhanced lipid peroxidation can also
29                   result from Pb potentiation of Fe2+ initiated lipid peroxidation and alteration of membrane
30                   composition after Pb exposure. Increased Pb-induced ROS can also sequester and
31                   inactivate biologically active NO, leading to the  increased production of the toxic product
32                   nitrotyrosine, increased compensatory NOS, and decreased sGC protein. Pb-induced
33                   oxidative  stress not only can result from increased  ROS production but also through the
34                   alteration and reduction in activity of the antioxidant defense enzymes. The biological
35                   actions of a number of these enzymes are antagonized due to the displacement of the
36                   protein functional metal ions by Pb.
      November 2012                             5-53                  Draft - Do Not Cite or Quote

-------
 1                  In a number of organ systems, Pb-induced oxidative stress is accompanied by
 2                  misregulated inflammation. Pb exposure can modulate inflammatory cell function,
 3                  production of pro-inflammatory cytokines and metabolites, inflammatory chemical
 4                  messengers, and pro-inflammatory signaling cascades. Cytokine production is skewed
 5                  toward the production of pro-inflammatory cytokines like TNF-a and IL-6 as well as
 6                  toward the promotion of a Th2 response and suppression of a Thl response accompanied
 7                  by decreased production of related cytokines.

 8                  Evidence indicates that Pb is a potent endocrine disrupting chemical. Pb can disrupt the
 9                  HPG axis evidenced by altered serum hormone levels, such as FSH, LH, testosterone,
10                  and estradiol. Pb can interact with the hypothalamic-pituitary level hormone control
11                  causing a decrease in pituitary hormones, alteration of growth dynamics due to decreased
12                  IGF-1, inhibition of LH secretion, and reduction in StAR protein. Pb has also been shown
13                  to alter hormone receptor binding likely due to interference of metal cations with
14                  secondary messenger systems  and receptor ligand binding and through generation of
15                  ROS. Pb also may disrupt the HPT axis by alteration of a number of thyroid hormones,
16                  possibly due to oxidative stress. However, the results of these studies investigating HPT
17                  are mixed.

18                  The association of Pb with increased genotoxicity and cell death has been investigated in
19                  humans, animals,  and cell models. Occupational Pb exposure in humans has been
20                  associated with increased DNA damage; however, lower blood Pb and exposure levels
21                  have been associated with these effects in experimental animals and cells. While not
22                  entirely consistent, a number of studies reported decreased repair processes following Pb
23                  exposure. There is evidence of mutagenesis and clastogenicity in highly-exposed
24                  humans; however, weak evidence has been shown in animals and cell based systems.
25                  Human occupational studies provide limited evidence for micronucleus formation (blood
26                  Pb levels >10 (ig/dL) and are supported by Pb-induced effects in both animal and  cell
27                  studies at higher exposure levels.  Animal studies have also provided evidence for
28                  Pb-induced chromosomal aberrations. The observed increases in clastogenicity may be
29                  the result of increased oxidative damage to DNA due to Pb exposure, as co-exposures
30                  with antioxidants  ameliorate the observed toxicities. Limited evidence of epigenetic
31                  effects is available, including abnormal DNA methylation, mitogenesis, and gene
32                  expression. Pb may alter gene  expression by displacing Zn from multiple transcriptional
33                  factors, thus perturbing their normal cellular activities. Consistently positive results have
34                  provided evidence of increased apoptosis following Pb exposure.

35                  Similar to Pb, other polyvalent metal ions (e.g.,  Cd, Cr, Be, Ba,  Se, Sr, As, Al, Cu) have
36                  demonstrated molecular mimicry  and displacement of biological cations (Garza et al..
37                  2006). In this manner, these metal ions share with Pb a common central mode of action of
      November 2012                             5-54                  Draft - Do Not Cite or Quote

-------
 1                  disruption of ion status. Specifically, these metals have been shown to disrupt cellular
 2                  processes as diverse as Ca2+ homeostasis, cell signaling, neurotransmitter release, cation
 3                  membrane channel function, protein-DNA binding, and cellular membrane structure
 4                  (Pentyalaetal.. 2010; Huang et al.. 2004; Atchison. 2003; Jehan and Motlag. 1995;
 5                  Richardt et al.. 1986; Cooper and Manalis. 1984; Habermann et al., 1983). Additionally,
 6                  presumably through their shared central mode of action, some of these metal ions also
 7                  display corresponding downstream modes of actions such as oxidative stress, apoptosis,
 8                  and genotoxicity (Jomova and Valko. 2011; Jomovaet al.. 2011; Matovic et al.. 2011;
 9                  Agarwal et al., 2009; Mendez-Gomez et al.. 2008; Rana. 2008; Hengstler et al., 2003).

10                  Overall, Pb-induced health effects can occur through a number of interconnected modes
11                  of action that generally originate with the alteration of ion status.
      5.3        Nervous System Effects
      5.3.1  Introduction

12                  The 2006 Pb AQCD concluded that the "overall weight of the available evidence
13                  provides clear substantiation of neurocognitive decrements being associated in young
14                  children with blood-Pb concentrations..." (U.S. EPA. 2006b). This conclusion was based
15                  on evidence from several prospective and cross-sectional epidemiologic studies
16                  conducted in diverse populations with adjustment for potential confounding by
17                  socioeconomic status (SES), parental intelligence, and parental caregiving quality and
18                  stimulation. The association between blood Pb levels and cognitive function decrements
19                  was substantiated in an international pooled analysis of children, ages 4.8 to 10 years,
20                  participating in seven prospective studies (Boston, MA; Cincinnati, OH; Rochester, NY;
21                  Cleveland, OH; Mexico City, Mexico; Port Pirie, Australia; and Kosovo, Yugoslavia)
22                  (Lanphear et al.. 2005). Across all previously evaluated studies, associations between
23                  blood Pb levels and decrements in full-scale intelligence quotient (FSIQ), infant mental
24                  development, memory, learning, and executive function were found in children ages  2 to
25                  17 years with population mean blood Pb levels (measured at various lifestages and time
26                  periods) 5-10 (ig/dL; however, several results indicated associations in groups of children
27                  (ages 2-10 years) with mean blood Pb levels in the lower range of 3-5 (ig/dL (Bellinger.
28                  2008; Canfield. 2008; Hornung. 2008; Tellez-Roio. 2008). Based on fewer available
29                  studies, the 2006 Pb AQCD described evidence from prospective and cross-sectional
30                  epidemiologic studies for associations of childhood blood Pb levels with attention-related
31                  behavioral problems in children ages 6-13 years and misconduct and delinquent behavior
32                  in children ages 7-17 years and young adults ages 21-22 years (U.S. EPA. 2006b).
      November 2012                             5-55                  Draft - Do Not Cite or Quote

-------
 1                   Biological plausibility for epidemiologic evidence in children was provided by similarly
 2                   consistent toxicological findings for Pb-induced impairments in learning and behavior in
 3                   rodents and monkeys (U.S. EPA. 2006b). Pb exposure was not found consistently to
 4                   affect the memory of animals. In animals, learning impairments were demonstrated
 5                   largely as poorer performance in maze tests, shorter interresponse times on schedule
 6                   controlled behavior tasks, and response perseveration errors in discrimination reversal
 7                   tests. Some results from these tests also indicated Pb-induced increases in inattention.
 8                   Pb-induced impulsivity in animals was demonstrated as increased response rates on the
 9                   Fixed Ratio (FR)/waiting for reward test. These effects on learning and behavioral
10                   problems in animals were found predominately with Pb exposures that resulted in blood
11                   Pb levels 20-50 (ig/dL; however, some studies observed these impairments in rodents
12                   (pre- and/or post-natal Pb exposure) and monkeys (postnatal Pb exposure) with blood Pb
13                   levels 14-25 (ig/dL (Kuhlmann et al.. 1997; Altmann et al.. 1993; Rice and Karpinski.
14                   1988; Gilbert and Rice. 1987). Toxicological studies further provided biological
15                   plausibility for Pb-induced  learning impairments and behavioral problems by
16                   characterizing modes of action.  Evidence for Pb affecting neuronal development and
17                   function at the cellular and  subcellular level (e.g., blood brain barrier integrity, synaptic
18                   architecture during development, neurite outgrowth, glial growth, neurotransmitter
19                   release, oxidative stress), provided biological plausibility for associations observed
20                   between blood Pb levels and deficits in multiple functional domains such as cognitive
21                   function, motor function, memory, mood, and behavioral problems in children.
22                   Additional biological plausibility was provided by associations observed of childhood
23                   blood Pb levels with changes indicative of neuronal damage and altered brain physiology
24                   assessed in small groups of children (Meng etal. 2005; Trope etal.. 2001) and young
25                   adults (Yuan et al.. 2006; Cecil et al.. 2005) using magnetic resonance imaging
26                   techniques.

27                   A common finding across several different populations of children was a supralinear
28                   concentration-response relationship between blood Pb level and cognitive function
29                   decrements, i.e., a larger decrement in cognitive function per unit increase in blood Pb
30                   level in children in the lower range of the population blood Pb level distribution (Kordas
31                   et al.. 2006; Schnaas et al.,  2006; Tellez-Rojo et al.. 2006; Bellinger and Needleman.
32                   2003; Canfield et al.. 2003a). Most of these epidemiologic results were based on the
33                   analysis of concurrent blood Pb levels and a cut-point of 10 (ig/dL to define lower and
34                   higher blood Pb levels. These findings were corroborated in pooled analyses of seven
35                   cohorts, which indicated that a nonlinear relationship fit the data better than a linear
36                   relationship (Lanphear et al.. 2005; Rothenberg and Rothenberg. 2005). Explanations for
37                   the supralinear concentration-response were not well characterized.
      November 2012                             5-56                  Draft - Do Not Cite or Quote

-------
 1                  Another area of focus was the comparison of various lifestages and time periods of Pb
 2                  exposure with respect to increasing risk of neurodevelopmental deficits. Toxicological
 3                  studies clearly demonstrated that gestational Pb exposure with or without additional early
 4                  postnatal exposure resulted in neurodevelopmental impairments. Nonetheless, not all
 5                  neurodevelopmental effects in animals had a single defined window of risk; for example,
 6                  postnatal-only and lifetime Pb exposures also were shown to impair learning and
 7                  behavior. Epidemiologic studies observed decrements in cognitive function in children
 8                  ages 3 to 17 years in association with prenatal, peak childhood, cumulative childhood,
 9                  and concurrent blood Pb levels. Although examined in few studies, tooth or bone Pb
10                  levels were associated with cognitive function decrements and behavioral problems in
11                  children and adolescents (Wasserman et al.. 2003; Bellinger etal.. 1994b: Fergusson et
12                  al., 1993; Needleman et al., 1979). also pointing to an effect of cumulative childhood  Pb
13                  exposure. Among studies of children (ages 3-10 years) that examined blood Pb levels
14                  measured at multiple lifestages and time periods, several found that concurrent blood  Pb
15                  was associated with a similar magnitude or larger decrement in FSIQ than blood Pb
16                  levels measured earlier in childhood or averaged over multiple years (Lanphear et al..
17                  2005; Wasserman et al.. 1994; Dietrich et al.. 1993). A common limitation of prospective
18                  studies of children was the high correlation among blood Pb levels at different ages,
19                  making it difficult to identify an individual critical lifestage or duration of Pb exposure
20                  associated with risk of neurodevelopmental decrements (Lanphear et al.. 2005). Some
21                  evidence indicated the persistence of neurodevelopmental effects of Pb exposure, by
22                  associations of biomarkers of earlier childhood Pb exposure (e.g., deciduous tooth, blood
23                  at age 2 or 6 years) with cognitive function decrements and behavioral problems in
24                  adolescents  and young  adults (Ris et al.. 2004; Wasserman et al.. 2003; Bellinger et al..
25                  1994a: 1994b: Fergusson et al.. 1993: Baghurst et al.. 1992: Needleman et al.. 1979).
26                  Persistence of effects also was demonstrated by findings in some studies of rats and
27                  monkeys that gestational and/or early postnatal Pb exposures were associated with
28                  impairments in cognitive function and  behavior in animals evaluated as adults
29                  (Kuhlmann et al.. 1997: Altmann et al.. 1993: Rice. 1992b. 1990).

30                  In epidemiologic studies of adults, a range of nervous system effects (e.g., impaired
31                  memory, attention, reaction time, visuomotor tasks and reasoning, alterations in visual or
32                  brainstem evoked potentials, postural sway)  were mostly clearly indicated in Pb-exposed
33                  workers with blood Pb  levels in the range of 14 to 40 ug/dL (Iwataet al.. 2005; Bleecker
34                  et al.. 1997: Baker etal.. 1979: Cantarow and Trumper. 1944). In the smaller body of
35                  studies examining nonoccupationally-exposed adults, poorer cognitive performance was
36                  associated with bone Pb levels (Weisskopf et al.. 2004: Wright et al.. 2003) but not
37                  concurrent blood Pb levels (Krieg etal.. 2005; Nordberg et al.. 2000; Pavtonetal.. 1998;
38                  Muldoon et al.. 1996). These findings suggested the influence of past or cumulative Pb
39                  exposures on cognitive function decrements  in nonoccupationally-exposed adults. With

      November 2012                             5-57                  Draft - Do Not Cite or Quote

-------
 1                   regard to neurodegenerative diseases, whereas a few toxicological studies found
 2                   Pb-induced amyloid plaques, a pathology commonly found in the brains of adults with
 3                   Alzheimer's disease (Bashaet al., 2005; Zawia and Basha. 2005), epidemiologic studies
 4                   did not indicate that Pb exposure was associated with Alzheimer's Disease in adults.
 5                   Blood and bone Pb levels were inconsistently associated with amyotrophic lateral
 6                   sclerosis (ALS) in adults in the general population; however, in some case-control
 7                   studies, history of occupational Pb exposure was more prevalent among ALS cases than
 8                   controls (Kamel et al. 2002; Chancellor etal.. 1993). Associations were reported for
 9                   essential tremor and symptoms of anxiety and depression in adults, but each was
10                   examined in only a few studies.

11                   As discussed throughout this section, recent epidemiologic and toxicological studies
12                   continued to demonstrate associations of Pb exposure and biomarkers of Pb exposure
13                   with nervous system effects. The strongest evidence continued to be derived from
14                   associations observed for Pb exposure and blood Pb levels in young  animals and children,
15                   respectively, with cognitive function decrements. Several recent studies in children
16                   expanded the evidence for associations between concurrent blood Pb levels and attention-
17                   related behavioral problems. Recent epidemiologic studies in adults  focused primarily on
18                   cognitive function decrements but provided additional evidence for Pb-associated
19                   psychopathological effects, ALS, Parkinson's disease, and essential tremor. Recent
20                   toxicological studies supported evidence for the effects of prenatal and postnatal Pb
21                   exposure on learning, memory, and impulsivity in animals and examined interactions
22                   between Pb exposure  and stress. New or expanded areas of toxicological research related
23                   to Pb exposure included, neurofibrillary tangle formation and neurodegenerative effects
24                   after early life Pb exposures and effects potentially related to psychopathological effects.
25                   Recent toxicological studies added to the large extant evidence base  for Pb-induced
26                   effects on endpoints describing modes of action, including neurotransmitters, synapses,
27                   glia, neurite outgrowth, the blood brain barrier, and oxidative stress. The data detailed in
28                   the subsequent sections continue to enhance the understanding of the spectrum of nervous
29                   system effects associated with Pb exposure.
      November 2012                             5-58                  Draft - Do Not Cite or Quote

-------
      5.3.2  Cognitive Function

 1                  Epidemiologic studies have assessed cognitive function extensively by FSIQ and its
 2                  verbal and performance subscale components in children ages 3 to 17 years. FSIQ has
 3                  strong psychometric properties (i.e., reliability,  consistency, validity), is among the most
 4                  rigorously standardized cognitive function measures, is relatively stable in school-age,
 5                  and has been predictive of life success. In children ages 6 months to 3 years, mental
 6                  development has been assessed with the Bayley Scales of Infant Development. A large
 7                  body of evidence also comprises associations of blood and tooth Pb levels with memory
 8                  and learning, executive function, language, and visuospatial processing. Several of these
 9                  domains of cognitive function are evaluated in the subtests of FSIQ, and some are more
10                  comparable to endpoints examined in tests in animals. Fewer studies have examined
11                  academic performance  and achievement; however, these outcomes may provide
12                  information on the impact of Pb exposure on life success. In the subsequent sections, the
13                  epidemiologic evidence for each of these categories of outcomes is reviewed separately
14                  in order of strength of evidence as assessed by the following parameters. Emphasis was
15                  placed on prospective studies with repeated measurements of blood Pb levels  and
16                  cognitive function and on studies that examined blood Pb levels more similar to those of
17                  contemporary U.S. children (i.e., <5 (ig/dL) and children whose blood Pb levels were less
18                  influenced by higher past Pb exposures. Studies of chelation in children generally were
19                  not included because the high pre-chelation blood Pb levels may limit generalizability of
20                  results, and chelation itself has been linked to neurodevelopmental effects.

21                  Many factors have been shown to influence the  cognitive function of children, including
22                  parental SES, parental education, parental IQ, quality and stability of parental caregiving
23                  environment  (often measured as Home Observation for the Measurement of Environment
24                  inventory [HOME]), nutritional status, and birth weight (Nation and Gleaves. 2001;
25                  Wasserman and Factor-Litvak. 2001). These and other influences on neurodevelopment
26                  often are correlated with blood Pb levels. Thus, due to their association with both blood
27                  Pb level and causal association with outcome, these other risk factors potentially may
28                  bias or confound the associations observed between blood Pb level and indices of
29                  cognitive function. In the evaluation of the effects of Pb independent from the effects of
30                  the other risk factors, greater weight was given to studies that more extensively accounted
31                  for potential confounding in the study design or in statistical analyses. A detailed
32                  evaluation of control for potential confounding  in associations between indicators of Pb
33                  exposure and neurodevelopmental effects is located in Section 5.3.14.
      November 2012                             5-59                  Draft - Do Not Cite or Quote

-------
      5.3.2.1  Full Scale IQ in Children

                     Evidence from Prospective Studies

 1                   Prospective cohort studies that were initiated in the 1980s addressed some limitations of
 2                   cross-sectional studies, including better characterizing the temporal sequence between
 3                   blood Pb levels and cognitive function, examining the persistence of cognitive function
 4                   decrements to older ages, and comparing associations among blood Pb levels measured at
 5                   various life stages or representing various time periods. Recruitment of participants before
 6                   or at birth without consideration of Pb exposure or maternal IQ, high follow-up
 7                   participation (>70%), and nonselective loss-to-follow-up in most studies increase
 8                   confidence that the observed associations are not due to selection bias. Moreover,
 9                   cooperation among investigators to adopt similar study protocols (e.g., similar tests of IQ
10                   and consideration of similar potential confounding factors) strengthened inferences
11                   regarding the consistency of associations with blood Pb level by facilitating pooled
12                   analyses and by reducing sources of heterogeneity in evaluating results across
13                   populations that varied in geographic location, proximity to Pb sources, blood Pb level
14                   range, race/ethnicity, and SES.

15                   Individual cohort studies of varying sample sizes (n = 148-375) conducted in several
16                   different populations (e.g., Boston, MA; Cincinnati, OH; Rochester, NY; Cleveland, OH;
17                   Mexico City, Mexico; Port Pirie and Sydney, Australia; and Kosovo, Yugoslavia) were
18                   consistent in demonstrating associations of higher blood Pb measured prenatally
19                   (maternal or umbilical cord), earlier in childhood, or averaged over childhood with lower
20                   FSIQ measured later in childhood, i.e., 4 to 17 years (Schnaas et al. 2006; Ris et al..
21                   2004; Canfield et al.. 2003a: Schnaas et al.. 2000; Factor-Litvak et al.. 1999; Tong et al..
22                   1996: Wasserman et al.. 1994: Dietrich et al..  1993b: Baghurst et al.. 1992: Bellinger et
23                   al.. 1992: Bellinger et al.. 1991: McMichael et al..  1988) (Figure 5-2 and Table 5-3). Null
24                   or weak associations were limited to a few  cohorts, namely, the Cleveland and Sydney
25                   cohorts (Greene et al.. 1992; CooneyetaL 1991;  1989a. b; Ernhart et al.. 1988). In the
26                   prospective studies, lower FSIQ also was associated with higher concurrent blood Pb
27                   levels (Figure 5-2 and Table 5-3) and tooth Pb levels. These latter results were based on
28                   cross-sectional analyses; however, the pattern of associations observed for blood Pb
29                   levels measured at various lifestages or time periods does not indicate that reverse
30                   causation explains the FSIQ decrements observed in association with concurrent blood
31                   Pb or tooth Pb levels.

32                   In addition to better characterizing the temporal sequence between Pb exposure and
33                   decrements in FSIQ, a common strength of most prospective studies was the adjustment
34                   for several of the potential confounding factors noted above, including maternal  IQ and
      November 2012                             5-60                  Draft - Do Not Cite or Quote

-------
 1                  education, child sex and birth weight, SES, and HOME score (Table 5-3). Although not
 2                  considered as frequently, some studies also indicated lack of confounding by parental
 3                  smoking, birth order, and nutritional factors. Multiple testing of associations with blood
 4                  Pb levels and/or FSIQ was common in prospective studies that found and did not find
 5                  associations between blood Pb level and FSIQ. However, higher probability of
 6                  associations due to chance alone does not appear to unduly influence the evidence
 7                  because in studies that found associations, there  was a consistent pattern of blood
 8                  Pb-associated cognitive function decrements across the various ages of blood Pb level
 9                  and/or cognitive assessments evaluated (Table 5-3). Studies finding null or weak
10                  associations also tended to show a consistent pattern across the various analyses
11                  conducted.
      November 2012                             5-61                  Draft - Do Not Cite or Quote

-------
 Study             Blood PbTiming   Blood PbMean(SD)  Blood PbInterval  FSIQage(yr)
                                      (ug/dL)      examined"(ug/dL)
 Prospective Studies
 Lanphearetal. (2005)  Concurrent, peak< 7.5 3.2
 Canfield etal. (2003)   Concurrent, peak< 10 3.3
 Juskoetal. (2008)"    Peak

 Bellingeretal. (1992)   Age2 yr, peak< 10

 Dietrich etal. (1993)    Concurrent

 Schnaasetal. (2006)   Prenatal (maternal)

 Cooney etal.(1991)b   Age3-5yravg

 Wasserman etal. (1997) Oto7yravg

 Tongetal. (1996)     Oto 11-13yravg
                11.4(7.3)

                3.8

                11.8 (6.3) (Age 5)
               1.3-6

               0.5-8.4

               2.1-10

               1.9.3

               5.5-10
                7.8 geometric mean 3.2-10

                NR             NR

                16.2 geometric mean6.0-10

                14.0(1.2) geometric 12.7-18.1
 Kordasetal. (2011)


 Greene etal. (1992)

 Minetal. (2009)
Prenatal (cord)
Concurrent

Age2yr

Concurrent
4 yr
4 yr
6.6(3.3)
8.7(4.4)
3.2-10.8
4.2-10
15.6 (1.4) geometric  10.1-24.0

7.0(4.1)          3.0-10
 Cross-sectional Studies
 Kim etal. (2009)       Concurrent, low Mn  1.73(0.80)
                   Concurrent, high Mn
                               0.9-2.8
 Fultonetal. (1987)    Concurrent         11.5(range:3.3-34) 5.6-10

 Royetal. (2011)      Concurrent         11.4(5.4)        5.8-10
FSIQage(yr)

6 O

7 D
11-13 -D-

4.8 	 O


6-9 —*—












                                                                    Change in FSIQ (95% Cl) per 1 ug/dL increase in various intervals
                                                                                      of blood Pb level
aSee Table 5-3 for explanation of the blood Pb level interval examined. Where possible, effect estimates were calculated for the
lowest range examined or the 10th percentile of blood Pb level to a blood Pb level of 10 ug/dL.
bSufficient data were not provided to calculate 95% Cl.
Note: Results are presented for most of the cohorts examined in the literature and generally are presented in order of strength of
study design and representativeness of study  population. Evidence usually is presented for the oldest age examined in cohorts.
Multiple results from a cohort are grouped together. To facilitate comparisons among effect estimates across studies with different
distributions of blood Pb levels and model structures (e.g., linear, log-linear), effect estimates are standardized to a 1 ug/dL increase
for the lowest range of blood Pb levels examined or the interval from the 10th percentile of blood Pb level  to 10 ug/dL.  For
populations with 10th percentile near or above 10 ug/dL, the effect estimate was calculated for the 10th to 90th percentile of blood
Pb level. The percentiles are estimated using various methods and are only approximate values. Effect estimates are assumed to
be linear within the blood Pb level interval evaluated. The various tests used to measure FSIQ are scored on a similar scale
(approximately 40-160 FSIQ points). Black diamonds, blue circles, orange triangles, and gray squares represent effect estimates for
concurrent, earlier childhood, prenatal, and lifetime average blood Pb levels, respectively. The lines represent 95% confidence
intervals (Cl).


Figure  5-2         Associations of blood  Pb  levels with full-scale IQ  (FSIQ)  among

                        children.
November 2012
                                     5-62
                                                 Draft - Do Not Cite or Quote

-------
Table 5-3     Additional characteristics and quantitative results for studies
                 represented in  Figure 5-2
Study
Study Population and Methodological Details
 Blood Pb Data
 (Ma/dL)
FSIQ
Testing3
                                                                                               Effect
                                                                                               Estimate
                                                                                               (95% CIV3
Prospective Studies:
Lanphear et
al. (2005)
Canfield et al.
(2003a)
Jusko et al.
(2008)
103 children pooled from Boston, Cincinnati,
Cleveland, Mexico City, Port Pirie, Rochester, and
Yugoslavia cohorts.
Uniform analysis of cohorts from diverse locations
and SES. Blood Pb levels and FSIQ measured at
different ages. Several sensitivity analyses to
examine heterogeneity of results by cohort, model
specification, and confounding. Linear regression
model adjusted for HOME score, birth weight,
maternal IQ, maternal education. Also considered
potential confounding by child sex, birth order,
maternal age, marital status, prenatal smoking status
and alcohol use.
101 children born 1994-1995 followed from age 6 mo
to 5 yr, Rochester, NY
Recruitment from study of dust control. 73%
nonwhite. High follow-up participation, no selective
attrition. Linear regression model adjusted for
maternal race, IQ, education, and prenatal smoking
status, household income, HOME score, child sex,
Fe status, birth weight.
1 74 children born 1 994-1 995 followed from age 6 mo
to 6 yr, Rochester, NY
Same cohort as above. High follow-up participation.
Participants had higher maternal IQ. Nonparametric
regression model adjusted for maternal race, IQ,
education, and prenatal smoking status, HOME
score, family income, child sex, birth weight, and
Fe status.
Concurrent,
children with peak
<7.5
Mean: 3.2
Interval analyzed:
1.3-6.0 =
5th-95th
percentiles
Concurrent,
children with peak
<10
Mean: 3.3
Interval analyzed:
<1-8.4 = range
(1 = detection
limit)
Peak
Mean (SD):
11.4(7.3)
Interval analyzed:
2.1-10
WISC-III, -2.94
WISC-R, (-5.16, -0.71)
WPPSI,
WISC-S
Ages
4.8-1 Oyr
Stanford- -1 .8
Binet (-3.0, -0.60)
Age 5 yr
WPPSI-R -1.2°
Age 6 yr
Bellinger and
Needleman
(2003)
48 children followed from birth (1979-1981) to age
10yr, Boston, MA area
Recruitment at birth hospital. Moderate follow-up
participation. Participants had higher SES and HOME
scores. 95% white. Linear regression model adjusted
for HOME score (age 10 and 5 yr), maternal race, IQ,
and marital status, SES, child sex, birth order, and
stress, and number of residence changes. Also
considered potential confounding by family stress,
maternal age, psychiatric factors, child serum ferritin
levels.
Earlier childhood
(age 2 yr),
children with peak
<10
Mean: 3.8
Interval analyzed:
1-9.3 = range
WISC-R
Age 10 yr
-1.56
(-2.9, -0.20)
Mazumdar et    43 adults followed from birth (1979-1981) to age
al. (2011)       28-30 yr, Boston, MA area
               Same cohort as above. Small proportion of original
               cohort but no selective attrition. 93% white. Linear
               regression model adjusted for maternal IQ. Also
               found associations adjusted for maternal marital
               status at birth, education at birth, prenatal smoking
               status, or alcohol use, HOME score (mean across
               ages), subject sex, birth weight, birth order,
               gestational age, race,  concussion history, or current
               smoking status. Also considered potential
               confounding by subject alcohol  use.
                                                Earlier childhood
                                                avg (age 6 mo-
                                                10yr):NR
                                                Mean (SD), [age]:
                                                8.0 (5.3) [6 mo],
                                                10.0 (6.7) [1 yr],
                                                7.7 (4.0) [2 yr],
                                                6.7 (3.6) [4 yr],
                                                3.0 (2.7) [1 Oyr].
                  WAIS
                  Age 28-30
                  yr
             -1.1
             (-2.29, 0.06)d
November 2012
                                  5-63
            Draft - Do Not Cite or Quote

-------
Study
Dietrich et al.
(1993b)
Schnaas et al.
(2006)
Cooney et al.
(1991)
Wasserman et
al. (1997)
long et al.
(1996)
Kordaset al.
(2011)
Study Population and Methodological Details
253 children followed from birth (1979-1985) to age
6. Syr, Cincinnati, OH
Recruitment at prenatal clinic. High follow-up
participation. Participants had slightly higher age 1 yr
blood Pb levels. Primarily African-American. Linear
regression model adjusted for HOME score, maternal
IQ and prenatal cigarette smoking, child birth weight,
birth length, and sex. Also considered potential
confounding by perinatal complications, prenatal
maternal substance abuse, nutritional status.
150 children followed from birth (1987-1992) to age
10 yr, Mexico City, Mexico.
Recruitment at prenatal clinic. Low follow-up
participation. Participants had higher SES, FSIQ,
higher blood Pb level before age 5 yr, lower at older
ages. Log linear mixed effects model adjusted for
SES, maternal IQ, HOME score, child sex, birth
weight, and postnatal blood Pb, indicator of first
FSIQ, random slope for subject. Most covariates
assessed in pregnancy or within age 6 mo.
175 children followed from birth (1983) to age 7 yr,
Sydney, Australia
Recruitment at birth hospital. Moderate follow-up
participation but no selective attrition. 100% white.
Linear regression adjusted for maternal education
and IQ, paternal education and occupation, HOME
score, child gestational age.
258 children followed prenatally (1984-1985) to age
7 yr, Kosovo, Yugoslavia
50% subjects live near Pb sources. Low follow-up
participation. Participants had lower HOME score,
maternal IQ, higher early childhood blood Pb levels
and fewer subjects lived in town with Pb sources.
Generalized estimating equations with log blood Pb
adjusted for maternal age, education, and IQ, child
age, sex, sibship size, and birth weight, language
spoken in home, HOME score.
375 children followed from birth (1979-1982) to age
1 1-13 yr, Port Pirie, Australia
Residence near Pb smelter. Moderate follow-up
participation. Participants had higher parental
occupational prestige. Regression model adjusted for
maternal IQ and age, parental occupational prestige,
smoking, marital status, and education, HOME score,
family functioning score, family size, child sex, age,
school grade, birth weight, birth order, feeding
method, breastfeeding duration, life events,
prolonged absences from school. Also considered
potential confounding by maternal psychopathology,
child Fe status, medication use in previous 2 weeks,
length of residence in area.
186 children followed from birth (1994-1995) to age
4 yr, Mexico City, Mexico
Recruitment at prenatal clinic. Low follow-up
participation but no selective attrition. Linear
regression model adjusted for maternal
age, education, IQ, smoking status, and marital
status, crowding in home, type of floor in home, child
sex, birth weight, gestational age. Did not consider
potential confounding by parental caregiving quality.
Blood Pb Data
(ug/dL)
Concurrent NR
Age 5 yr
Mean (SD):
11.8(6.3)
Interval analyzed:
5.5 (10th
percentile)-10
Prenatal (maternal
28-36 weeks)
Geometric mean
(5th-95th): 7.8
(2.5-24.5)
Interval analyzed:
3.2 (10th
percentile)-10
Age 3-5 yr avg:
NR
Age 5 yr
Mean (Max):
8.3 (27)
Lifetime avg
(to age 7 yr)
Geometric mean:
16.2
Interval analyzed:
6.0 (10th
percentile)-10
Lifetime avg
(to age 11-1 Syr)
Geometric mean
(GSD): 14.0(1.2)
Interval analyzed:
12.7-18.1 =
10th-90th
percentiles
Prenatal (cord)
mean (SD):
6.6 (3.3),
Interval analyzed:
3.2-10.8 =
10th-90th
percentiles
Concurrent mean
(SD):
8.7 (4.4) ,
Interval analyzed:
4.2 (1 Oth
percentile)-10
FSIQ
Testing3
WISC-R
Age 6.5 yr
McCarthy
GCI
Ages
6-1 Oyr
WISC-R
Age 7 yr
WISC-III
Age 7 yr
WISC-R
Age
11-13yr
Prenatal
Concurrent
McCarthy
GCI
Age 4 yr
Effect
Estimate
(95% CIV3
-0.33
(-0.60, -0.06)
-1.05
(-1 .67, -0.43)
-0.07°
-0.48
(-0.68, -0.27)
-0.12
(-0.24, -0.003)
-0.20
(-0.79, 0.39)
-0.60
(-0.99, -0.21)
November 2012
5-64
Draft - Do Not Cite or Quote

-------
Study
Greene et al.
(1992)
Min et al.
(2009)
Study Population and Methodological Details
270 children followed from with age 4 yr 10 mo,
Cleveland, OH
Recruitment at birth hospital. High prevalence of
prenatal alcohol and drug exposure. High follow-up
participation. Participants tended to be Black,
exposed to marijuana. Log linear regression adjusted
for maternal IQ, weight, street drug use,
cigarettes/day, alcohol use, and age, parental
education, authoritarian scale, race, parity, gestation
duration, date of first prenatal visit, HOME score,
quality of physical environment peeling paint, home
cleanliness, pica behavior
267 children followed from birth (1994-1996) to age
11 yr, Cleveland, OH
Recruitment at birth hospital. 86% African American
with high prevalence of prenatal drug and alcohol
exposure. Moderate follow-up participation to age 4
yr, high retention to age 1 1 yr. Participation tended to
be African American and had married mothers. Linear
regression model adjusted for HOME score, head
circumference at birth (all ages), current caregiver
vocabulary score, maternal marital status, parity,
child sex (age 4 yr), maternal vocabulary score at
birth (age 9 and 1 1 yr), average prenatal cocaine use
(age 9 yr), prenatal 1 st trimester marijuana use (age
1 1 yr). Also considered potential confounding by
maternal education, Fe deficiency, maternal
psychological distress, and race.
Blood Pb Data
(ug/dL)
Earlier childhood
(age 2 yr)
Geometric mean
(GSD):
15.6(1.4)
Interval analyzed:
10.1-24.0 =
10th-90th
percentiles
Age 4 yr
Mean (range):
7.0 (1 .3-23.8)
Interval analyzed:
3.0 (10th
percentile) -10
FSIQ
Testing3
WPPSI
Age 4.8 yr
WPPSI,
age 4yr
(concurrent)
WISC-R,
age 9 yr
WISC-R,
age 11 yr
Effect
Estimate
(95% CIV3
-0.11
(-0.51,0.29)
-0.50
(-0.89, -0.11)
-0.41
(-0.78, -0.04)
-0.54
(-0.91, -0.17)
Cross-sectional Studies:
Fulton et al.
(1987)











Surkan et al.
(2007)







Kim et al.
(2009b)





Roy et al.
(2011)



501 children, ages 6-9 yr, Edinburgh, Scotland
Recruitment at schools. High participation rate,
representative of area population. Log linear
regression model adjusted for parental SES,
education, marital status, health, mental health,
cigarettes smoked, vocabulary and matrices test
scores, involvement, interest, communication, and
participation with child, family size and structure, child
age, sex, handedness, height, gestation length, birth
weight, medical history, absence from school, recent
school change, grade, and time of day of test, people
per room in home, car/phone ownership, consumer
goods ownership.
389 children, ages 6-10 years, Boston, MA,
Farmington, ME
Recruitment from trial of amalgam fillings. High
participation rate. Higher participation of white children
in Maine. Analysis of covariance adjusted for caregiver
IQ, child age, SES, race, and birth weight, Also
considered potential confounding by site, sex, birth
order, caregiver education and marital status,
parenting stress, and maternal utilization of prenatal
and annual health care but not parental caregiving
quality.
279 children (born 1996-1999) ages 8-1 1 yr, Seoul,
Seongnam, Ulsan, and Yeoncheon, Korea
Recruitment at schools. Moderate participation rate.
Log linear regression model adjusted for maternal
age, education and prenatal smoking status, paternal
education, yearly income, smoking exposure status
after birth, child age, sex, and birth weight. Did not
consider potential confounding by parental caregiving
quality or IQ.
71 7 children ages 3-7 yr, Chennai, India
Recruitment at schools. High participation rate. Log
linear model adjusted for mid-arm circumference, age,
sex, family income, parental education and IQ, family
size. Did not consider potential confounding by
parental caregiving quality.
Concurrent
Geometric mean
(range):
1 1 .5 (3.3- 34)
Interval analyzed:
5.6 (mean of 1 st
decile)-10






Concurrent
Group! : 1-2
Group 2: 3-4
GroupS: 5-10






Concurrent
Mean (SD):
1 .73 (0.80)
Interval analyzed:
n Q OR
u.y -z.o -
1 0th-90th
percentiles

Concurrent
Mean (SD):
1 1 .4 (5.4)
Interval analyzed:
5.8 (10th
percentile)-! 0
BASC
Age 6-9 yr











WISC-III
Age 6-1 0 yr







KEDI-WISC
Ages 8-1 1 yr
Blood Mn:
<1.4ug/dL
Blood Mn:
>1.4ug/dL

Binet-Kamat
Ages 3-7 yr



-0.22
(-0.37, -0.06)












Reference
-0.1 2 (-3.3, 3.1 )e
-6.0 (-10.7, -1.4)e








-2 4
(-6.0,1.1)
-3.2
(-6.1, -0.23)

-1.16
(-1 .94, -0.37)



November 2012
5-65
Draft - Do Not Cite or Quote

-------


Study
Chiodo et al.
(2007)









Chiodo et al.
(2004)










Study Population and Methodological Details
506 children (born, 1982-1984) age 7 yr, Detroit, Ml
area
Recruitment at prenatal clinic. 100% African -
American. High prevalence of prenatal drug exposure.
High follow-up participation. Linear regression model
adjusted for maternal concurrent psychopathology, IQ,
prenatal cigarettes/day, and prenatal use of marijuana,
SES, HOME score, caretaker education, number
children in home, child sex. Also considered potential
confounding by child age, caretaker marital status,
maternal age, custody, cocaine use, prenatal alcohol
use, concurrent alcohol/week, concurrent
cigarettes/day, and concurrent marijuana use.
237 children, age 7.5 yr, Detroit, Ml area
Recruitment at prenatal clinic. 100% African-
American. High prevalence of prenatal alcohol
exposure. High participation rate. Log linear
regression model adjusted for SES, maternal
education and vocabulary score, # children <18 yr,
HOME score, parity, family environment scale, child
sex. Also considered potential confounding by
prenatal alcohol, marijuana, smoking, or cocaine use,
crowding, child age and life stress, caregiver life
stress, conflict tactics.

Blood Pb Data FSIQ
(ug/dL) Testing3
Concurrent WISC-III
Mean(SD): Age 7 yr
5.0 (3.0)
Interval analyzed:
2.1-8.7 =
1 0th-90th
percentiles






Concurrent WISC-III
Mean(SD): Age 7.5 yr
5.4(3.3)
Interval analyzed:
2.3-9.5 =
1 0th-90th
percentiles




Effect
Estimate
ic\co/ /"*i\k
(95% Cl)
-0.19
(-0.30, -0.08)e'f
Standardized
regression
coefficient







-0.22
(-0.38, -0.05)e'f
Standardized
regression
coefficient






      aWISC = Wechsler Intelligence Scale for Children, WPPSI = Wechsler Preschool and Primary Scale of Intelligence,
      WAIS = Wechsler Adult Intelligence Scale, GCI = General Cognitive Index, BASC = British Ability Scales Combined,
      KEDI = Korean Educational Development Institute
      bEffect estimates are standardized to a 1 ug/dL increase in blood Pb level within the lowest range examined in study or 10th
      percentile to 10 ug/dL For populations with 10th percentiles near or above 10 ug/dL, effect estimates were calculated for the
      10th-90th percentile interval of blood Pb level. Effect estimates are assumed to be linear within the evaluated interval of blood Pb
      level. The percentiles are estimated using various methods and are only approximate values.
      ""Sufficient data were not provided to calculate 95% Cl.
      dResults not included in Figure 5-2 because FSIQ assessed in adults.
      eResults not included in Figure 5-2 because blood Pb level analyzed as categorical variable or because standardized regression
      coefficient reported.
      f95% CIs were constructed using a standard error that was estimated for the reported p-value of 0.01.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
Across the cohort studies, blood Pb-associated FSIQ decrements were found in
populations with mean blood Pb levels 5-10 (ig/dL. In analyses restricted to children in
the lower range of the blood Pb distribution (e.g., peak <10 (ig/dL), associations were
observed in groups of children with mean blood Pb levels 3-4 (ig/dL (Bellinger. 2008;
Canfield. 2008; Hornung. 2008). The analysis of the Rochester cohort is particularly
informative for lower blood Pb levels of children  (mean at age 5  years: 5.8 (ig/dL)
compared to other cohorts and the greater consideration for potential confounding by
factors such as sex, race, family income, maternal education, race, prenatal maternal
smoking, birth weight, maternal IQ, HOME score, and transferrin saturation which
indicates iron status and for providing unadjusted and covariate-adjusted results (Canfield
et al.. 2003a). At age 5 years, higher age 6-24 month average, peak, concurrent, and
lifetime average blood Pb levels (area under the curve calculation using repeat
measurements between age  6 months and 5 years) were associated with lower FSIQ, and
while effect estimates in the covariate-adjusted model were 40-45% smaller than
estimated in the unadjusted models, they remained statistically significant. A larger effect
was estimated for the 101 (59%) children whose peak blood Pb levels never exceeded
      November 2012
                              5-66
Draft - Do Not Cite or Quote

-------
 1                   10 (ig/dL, i.e., -1.8 points (95% CI: -3.0, -0.60) per 1 (ig/dL in concurrent blood Pb level.
 2                   Similarly, Bellinger and Needleman (2003) estimated a larger effect in the subset of the
 3                   Boston cohort (n = 48, 32%) with peak blood Pb levels <10 (ig/dL, i.e., -1.6 points (95%
 4                   CI: -2.9, -0.2) per 1 (ig/dL increase in age 2-year blood Pb level. The mean blood Pb
 5                   levels in these subsets of children were 3.3 (Rochester) and 3.8 (ig/dL (Boston) closer to
 6                   that of current U.S. children compared with other prospective studies.

 7                   Analyses of the Cincinnati and Port Pirie, Australia cohort also indicated associations
 8                   between blood Pb level and FSIQ decrements with as extensive consideration for
 9                   potential confounding (Table 5-3) albeit in populations with higher blood Pb levels
10                   (i.e., mean at age 5 years: 11.8 (ig/dL, lifetime average geometric mean: 14.0 (ig/dL)
11                   (Tong et al.. 1996; Dietrich et al.. 1993b). In contrast with other studies, in the Cleveland
12                   cohort, associations of blood Pb level (ages 2 and 3 years) and tooth Pb level with FSIQ
13                   (ages 3 and 4.8 years), became attenuated or were too imprecise to be informative with
14                   adjustment for a large number of potential confounding factors, including maternal
15                   substance abuse, home cleanliness, and pica behavior, which were not considered widely
16                   in other studies (Greene etal.. 1992; Ernhart et al.. 1988). HOME score accounted for a
17                   large proportion of the variance in FSIQ and was the major factor accounting for the
18                   attenuation of the effect estimates for Pb biomarkers. The association between tooth Pb
19                   level and FSIQ at age 4.8 years was attenuated with  additional adjustment for HOME
20                   score but was estimated with similar precision (-3.0  points [95% CI: -6.4, 0.32] per
21                   1 (ig/g increase in tooth Pb level) (Greene and Ernhart. 1993). The few weak or null
22                   associations do not mitigate the otherwise strong evidence provided by other studies.  The
23                   Cleveland (Greene et al.,  1992) and Sydney (Cooney et al., 1991) studies were not
24                   outliers with respect to population mean blood Pb levels or the  specific confounding
25                   factors considered (Table 5-3). and the Cleveland cohort had high prevalence of maternal
26                   prenatal substance abuse which may limit the representativeness of results.  Further, the
27                   blood Pb-FSIQ association in children was substantiated in a pooled analysis of seven
28                   prospective studies by Lanphear et al. (2005). which included the Cleveland cohort, as
29                   well as multiple meta-analyses that combined results across various prospective and
30                   cross-sectional studies, including those from the Cleveland and Sydney cohorts (Pocock
31                   et al..  1994; Schwartz. 1994; Needleman and Gatsonis. 1990). The meta-analysis by
32                   Schwartz (1994) demonstrated the robustness of evidence to potential publication bias.
33                   The addition of eight hypothetical  studies with a zero effect and with the  average weight
34                   of the eight published studies resulted in a 50% lower but still negative and precise
35                   (p <0.001) blood Pb-FSIQ effect estimate.

36                   The pooled analysis of seven prospective studies included individual-level data from
37                   1,333  children ages 4.8-10 years of age with a median (5th-95th percentile) concurrent
38                   blood Pb level of 9.7  (ig/dL (2.5-33.2 (ig/dL) (Lanphear et al.. 2005). In multivariate
      November 2012                              5-67                   Draft - Do Not Cite or Quote

-------
 1                   models that adjusted for study site, maternal IQ, HOME score, birth weight, and maternal
 2                   education, higher concurrent, peak, lifetime average, and early childhood blood Pb levels
 3                   were associated with lower FSIQ measured at age 4.8-10 years. Various models were
 4                   investigated to characterize the shape of the blood Pb-FSIQ concentration-response
 5                   relationship. Consistent with the supralinear concentration-response relationship found in
 6                   several individual cohort studies, Lanphear et al. (2005) found that a nonlinear (i.e., log-
 7                   linear) model fit the data better than a linear model. The nonlinear relationship was
 8                   indicated further by observations of a greater decrease in FSIQ for a 1 (ig/dL increase in
 9                   concurrent blood Pb for the 244 (18%) children who had peak blood Pb levels <10 (ig/dL
10                   (-0.80 points [95%  CI: -1.74, -0.14]) and the 103 (8%) children with peak blood Pb levels
11                   <7.5 (ig/dL (-2.9 points [95% CI: -5.2, -0.71]). Among children with peak blood Pb
12                   <10 (ig/dL and <7.5 (ig/dL, the median concurrent blood Pb levels were 4.2 (ig/dL and
13                   3.2 (ig/dL, respectively (Hornung. 2008).

14                   An additional strength of the pooled analysis by Lanphear et al. (2005) was the
15                   examination of several potential confounding factors related to SES and the caregiving
16                   environment. Variables such as HOME score, birth weight, maternal IQ, and maternal
17                   education were selected for inclusion in the final model with blood Pb level based on
18                   their statistically significant association with FSIQ. Child sex, maternal prenatal tobacco
19                   or alcohol use, maternal age at delivery, marital status, and birth order were not
20                   statistically significantly associated with FSIQ and did not  alter the effect estimate for
21                   concurrent blood Pb level adjusted for the four aforementioned covariates. While a
22                   smaller decrement in FSIQ was estimated for concurrent blood Pb level in the adjusted
23                   model than in the unadjusted model (-4.7 points [95% CI: -5.7, -3.6] versus -2.7 points
24                   [95% CI: -3.7, -1.7] per log increase in concurrent blood Pb level), the adjusted blood Pb
25                   level effect estimate did not lose precision.

26                   The few prospective studies published since the 2006 Pb AQCD continued to
27                   demonstrate associations between higher blood Pb level and lower FSIQ, in some cases,
28                   with additional follow-up of previous cohorts. Similar to studies reviewed in the
29                   2006 Pb AQCD, most recent studies demonstrated associations between blood Pb level
30                   and lower FSIQ in populations with mean blood Pb level between 5 to  10 (ig/dL. Jusko  et
31                   al. (2008) affirmed  the findings in the Rochester cohort previously reported by Canfield
32                   et al. (2003a), who  examined the cohort at age 5 years. Jusko et al. (2008) examined 174
33                   Rochester cohort subjects at age 6 years and similar to Canfield et al. (2003a), found that
34                   an increase in peak blood Pb level was associated with a larger decrease in FSIQ  among
35                   children with peak blood Pb levels <10 (ig/dL than among  children with peak blood Pb
36                   levels 10-20 (ig/dL (-1.2 points versus -0.32 points per 1  (ig/dL increase in blood). The
37                   age  6 year analysis  had similarly extensive consideration for potential confounding as did
      November 2012                             5-68                  Draft - Do Not Cite or Quote

-------
 1                   Canfield et al. (2003a) (Table 5-3) and also indicated associations with higher concurrent,
 2                   infancy average, and lifetime average blood Pb levels (effect estimates not reported).

 3                   Additional evidence was provided for children in Mexico City, albeit in a separate cohort
 4                   of children born later with lower blood Pb levels at corresponding ages. Among 150
 5                   children born 1987-1992, Schnaas et al. (2006) previously reported larger Pb-associated
 6                   decrements in FSIQ for prenatal maternal (28-36 weeks) blood Pb levels than for child
 7                   concurrent blood Pb levels between ages 1 and 10 years. In contrast, Kordas et al. (2011)
 8                   found that an increase in concurrent blood Pb level was associated with a larger
 9                   decrement in FSIQ at age 4 years than was an increase in cord blood Pb level with
10                   adjustment for several potential confounding factors (HOME score not examined) (Table
11                   5-3). The 186 children in the latter study were born 1994-1995 and at age 4 years had a
12                   mean blood Pb level of 8.7 (ig/dL. In Schnaas et al. (2006). the geometric mean blood Pb
13                   level at age 4 years was 10.3 (ig/dL. It is not clear whether different temporal patterns of
14                   Pb exposure or age of FSIQ assessment may have contributed to the contrasting
15                   associations for prenatal and concurrent blood Pb levels in these two Mexico City
16                   cohorts.

17                   Mazumdar et al. (2011) followed the Boston cohort (Bellinger etal.. 1992) to age
18                   28-30 years, and indicated that the effect of childhood Pb exposures may persist to
19                   adulthood. Only 43 of the original 249 subjects enrolled at birth were examined at age
20                   28-30 years, but they did not differ from the original cohort in demographic
21                   characteristics or blood Pb history.  Higher blood Pb levels measured at age 6 months,
22                   4 years, 10 years, and averaged over childhood (to age 10 years) (means: 3 (ig/dL at age
23                   10 years to 8 (ig/dL at age 6 months) were associated with lower FSIQ in adults with
24                   adjustment for sex, birth weight, birth order, gestational age, maternal marital status,
25                   maternal education, maternal IQ, race, maternal smoking and  alcohol use in pregnancy,
26                   average of childhood HOME score, concussion, and subject current smoking status. The
27                   effect estimates were similar in magnitude for all childhood blood Pb measures, except
28                   for age 6 month blood Pb level, which was associated with a smaller FSIQ decrement.

29                   Min et al. (2009) found higher earlier childhood blood Pb levels (age 4 year) to be
30                   associated with decrements in FSIQ in another cohort of children in Cleveland, OH
31                   between ages 4 and 11 years, indicating persistence of effects (Figure 5-2 and Table 5-3).
32                   However, similar to the other Cleveland cohort (Greene etal.. 1992). the recent cohort
33                   had high prevalence of prenatal alcohol and drug exposure. These exposures were weakly
34                   associated with FSIQ or did not influence the blood Pb-FSIQ association, indicating lack
35                   of strong confounding bias. However, because the population lacks representativeness,
36                   these findings are less of a consideration in drawing conclusions regarding the effects of
37                   Pb exposure on FSIQ of children.
      November 2012                             5-69                  Draft - Do Not Cite or Quote

-------
 1                   An important consideration in the evaluation of epidemiologic evidence is the precision
 2                   of effect estimates, both within and among studies. There was variability in precision
 3                   among studies, which appeared to be influenced by sample size rather than the age of
 4                   subjects or the extent of adjustment for potential confounding factors. Analyses of the
 5                   Port Pirie (n = 375, ages 11-13 years) and Yugoslavia (n = 258, age 7 years) cohorts
 6                   ("Wasserman et al.. 1997; Tong et al.. 1996) and the pooled analysis of 1,333 children
 7                   (Lanphear et al., 2005) estimated more precise effects compared to the Boston (n =  148,
 8                   age 10 years) and Rochester cohorts (n = 172, age 5 years) (Canfield  et al.. 2003a:
 9                   Bellinger et al.. 1992). Analyses  of the Yugoslavia and Port Pirie cohorts did not
10                   necessarily have more or less extensive adjustment for potential confounding.

11                   Among prospective studies, a wide range of blood Pb-associated FSIQ decrements was
12                   estimated (Figure 5-2 and Table  5-3). This wide range is not unexpected, given
13                   differences among studies in blood Pb level ranges, model specification (linear versus log
14                   linear), lifestage or time period of blood Pb level examined, and distribution of potential
15                   confounding factors. The pooled analysis examined study populations of diverse SES,
16                   maternal education, and cultural  backgrounds with the same model and indicated
17                   precision of effect (Lanphear et al.. 2005). A series of sensitivity analyses, in which one
18                   cohort was excluded at a time, revealed that no single study was responsible for the
19                   results. Per log increase in blood Pb  level, effect estimates excluding  one study at a time
20                   fell within a narrow range, -2.36  to -2.94 (blood Pb level ranges for sensitivity analyses
21                   not reported). Precision of effect also was indicated by the similar effect estimates found
22                   with similar model specifications, population blood Pb levels, and sample sizes in the
23                   Boston and Rochester cohorts (Table 5-3). but very different SES and racial distributions
24                   of the cohorts, and different ages of blood Pb level and FSIQ examined (Canfield et al..
25                   2003a; Bellinger etal.. 1992). These estimates were larger than those found in the
26                   Cincinnati, Port Pirie, and Yugoslavia cohorts but were based on similar extent of
27                   adjustment for potential confounding factors. Several of the smaller blood Pb-associated
28                   FSIQ decrements were based on  log-linear models that estimated effects at higher blood
29                   Pb levels (10th percentiles >5.5 versus 2 (ig/dL). The widely different null associations
30                   found in the Cleveland cohort have weaker implications because of the
31                   nonrepresentativeness of the cohort due to their high prevalence of prenatal alcohol and
32                   drug exposure (Greene etal.. 1992; Ernhart et al.. 1988).
      November 2012                             5-70                  Draft - Do Not Cite or Quote

-------
                     Evidence from Cross-sectional Studies

 1                   The smaller body of cross-sectional studies reviewed in the 2006 Pb AQCD (U.S. EPA.
 2                   2006b) found associations of higher concurrent blood (Fulton et al.. 1987) or tooth
 3                   (Needleman et al., 1979) Pb levels with FSIQ decrements in children ages 6-9 years, and
 4                   associations also were found in the few recent studies in children ages 3-11 years (Figure
 5                   5-2 and Table 5-3). Several cross-sectional studies had larger sample sizes (n = 279-717)
 6                   than the prospective studies and produced effect estimates with similar precision.
 7                   Previous meta-analyses produced similar combined blood Pb-FSIQ effect estimates for
 8                   prospective and cross-sectional studies (Pocock etal. 1994; Schwartz. 1994). However,
 9                   in this ISA, the cross-sectional findings were given less weight in conclusions regarding
10                   Pb-associated effects on cognitive function. The temporal sequence between Pb exposure
11                   and decreases in FSIQ is difficult to establish. Some studies had population-based
12                   recruitment, high participation rates, and did not indicate undue selection bias, but the
13                   evidence overall had less consideration for potential confounding by parental caregiving
14                   quality and/or parental IQ (Roy etal.. 2011: Kim et al.. 2009b: Zailina et al.. 2008:
15                   Surkan et al., 2007: Needleman et al., 1979). The meta-analysis by Pocock et al. (1994)
16                   noted the lack of adequate control for potential confounding factors in previous cross-
17                   sectional studies. Other cross-sectional studies lacked representative study populations
18                   because of high prevalence of prenatal alcohol (Chiodo et al.. 2004) or drug exposure
19                   (Chiodo et al.. 2007).

20                   Among the cross-sectional studies, Fulton  et al. (1987) had more extensive consideration
21                   for potential confounding. Among 501 children, ages 6-9 years, in Edinburgh, Scotland, a
22                   1 (ig/dL increase in concurrent blood Pb level in the interval between 5.6 and  10 (ig/dL
23                   was associated with a 0.22-point decrease (-0.37, -0.06) in FSIQ, after adjustment for
24                   several factors related to  SES, parental health and mental health, child health,  and
25                   parental caregiving quality (Table 5-3). The effect estimate from this study was among
26                   the smallest produced by cross-sectional studies. The study population was representative
27                   of the source population but had much higher blood Pb levels (geometric mean:
28                   11.4 (ig/dL) than those of most of the current U.S. population of children.

29                   Recent cross-sectional studies examined potential confounding by parental IQ and
30                   education and SES but not parental caregiving quality. Studies that examined populations
31                   with mean concurrent blood Pb levels <4 (ig/dL did not conclusively indicate
32                   associations with FSIQ decrements at lower blood Pb levels. Among 389 children from
33                   urban Boston, Massachusetts and rural Farmington, Maine with mean concurrent blood
34                   Pb level 2.2 (ig/dL, lower FSIQ was limited to children with blood Pb levels 5-10 (ig/dL,
35                   with children with blood Pb levels 1-2 (ig/dL serving as the referent group (Table 5-3)
36                   (Surkan et al., 2007). There was consideration for potential confounding by several
37                   factors, including age, race/ethnicity, birth weight, SES, primary caregiver IQ, SES,

      November 2012                             5-71                  Draft - Do Not Cite or Quote

-------
 1                   education and marital status, parenting stress, and maternal utilization of prenatal or
 2                   annual health care. Other recent cross-sectional studies found associations at lower
 3                   concurrent blood Pb levels; however, some of the children likely had higher earlier
 4                   childhood Pb exposures, which may have contributed to the associations observed with
 5                   relatively low concurrent blood Pb levels. In a group of 279 children ages 8-11 years
 6                   from four Korean cities with a mean concurrent blood Pb level 1.73 (ig/dL, Kim et al.
 7                   (2009b) found an association between higher concurrent blood Pb level and lower FSIQ
 8                   with adjustment for parental education, yearly income, prenatal and postnatal smoking
 9                   exposure, birth weight, age, and sex. The adjusted effect estimate was attenuated but
10                   similarly precise as the unadjusted estimate. The concurrent blood Pb-FSIQ and verbal
11                   IQ relationship was modified by concurrent blood manganese (Mn) levels. Blood Pb and
12                   Mn levels were not correlated (r = 0.03, p = 0.64). Higher concurrent blood Pb level was
13                   associated with lower FSIQ in both children in high Mn (above the median of 1.4 (ig/dL)
14                   and low Mn group (below the median of 1.4 (ig/dL) but a larger FSIQ decrement in the
15                   130 children in the high Mn group (-3.2 points [95% CI: -6.1, -0.23] per 1 (ig/dL increase
16                   in the 10th to 90th percentile interval 0.9-2.8 (ig/dL) compared with children in the low
17                   Mn group (-2.4 points [95% CI: -6.0, 1.1]). The biological plausibility for the Pb-Mn
18                   interaction is provided by observations that Mn has similar modes of action and cellular
19                   targets as does Pb, i.e., altering Ca2+ metabolism, inducing oxidative damage in neuronal
20                   cells, diminishing dopamine transmission. Among 169 children in Malaysia, ages 6-8
21                   years, concurrent blood Pb (mean ~4 (ig/dL) level but not parental education or family
22                   income was associated with FSIQ decrements, producing uncertainty as to whether these
23                   potential confounding factors were measured adequately (Zailinaet al..  2008).

24                   Other recent cross-sectional studies found associations in populations of children with
25                   relatively higher concurrent blood Pb levels (means >8 (ig/dL). Evidence demonstrates
26                   that Pb affects dopaminergic neurons and dopamine release (Section 5.3.11.8). Further,
27                   dopaminergic activity is a key mediator of cognitive  function. These findings suggest that
28                   variants in dopamine-related genes may modify Pb-associated effects on cognition.
29                   Epidemiologic evidence for effect modification is not consistent; however, subgroup
30                   analyses are subject to higher probability of findings by chance. The larger of these
31                   studies (n = 717 children ages 3-7 years in Chennai, India) with higher concurrent blood
32                   Pb levels (mean: 11.4 (ig/dL) found that  a 1 (ig/dL higher blood Pb level was associated
33                   with a larger decrease in FSIQ among the 72 children with the Taq A1/A1 dopamine
34                   receptor (DRD2) variant (-2.5 points [95% CI: -5.0, -0.04] within the blood Pb level
35                   interval 5.8-10 (ig/dL) than among the 651 children with the Taq A2/A2 variant (-1.1
36                   points [-2.3, -0.12]) (Roy et al.. 2011). Kordas et al. (2011) did not find effect
37                   modification in a smaller study of 186 children in Mexico City with a mean concurrent
38                   blood Pb level of 8.7 (ig/dL. Another difference between studies that may have
39                   contributed to the difference in effect modification by the DRD2 variant was an

      November 2012                             5-72                  Draft - Do Not Cite or Quote

-------
 1                   association observed between Taq A1/A1 genotype and higher mean FSIQ score in the
 2                   group in Mexico but no association in the group in India.

 3                   In summary, a majority of prospective and cross-sectional studies demonstrated
 4                   associations between higher blood Pb level and lower FSIQ in children ages 3-17 years
 5                   (e.g., Figure 5-2 and Table 5-3). While studies performed numerous tests, bias due to
 6                   increased probability of findings by chance was unlikely because most studies found a
 7                   consistent pattern of association across the ages of blood Pb level and FSIQ analyzed.
 8                   Across studies, FSIQ was measured with similar instruments scored on similar scales
 9                   with similar measurement error. The  key supporting evidence is provided by the
10                   prospective studies, which better indicated the temporal sequence between blood Pb
11                   levels measured earlier in childhood or averaged over multiple years or tooth Pb levels
12                   and FSIQ measured later in childhood. Prospective studies also had more extensive
13                   consideration for potential confounding by maternal IQ and education, SES, birth weight,
14                   smoking exposure, parental caregiving quality, and in a few cases, other birth outcomes
15                   and nutritional factors. Further, the representativeness of findings was supported by
16                   associations found in diverse populations (e.g., Boston, MA; Cincinnati, OH; Rochester,
17                   NY; Cleveland,  OH; Mexico City, Mexico; Port Pirie and Sydney, Australia; and
18                   Kosovo, Yugoslavia) and in studies examining populations recruited from prenatal
19                   clinics, hospital  maternity departments, or schools with high follow-up participation and
20                   lack of biased follow-up participation by blood Pb level and FSIQ. The few weak or null
21                   associations found in Cleveland and Sydney cohorts were adjusted for similar potential
22                   confounding factors and thus do not mitigate the otherwise strong evidence. The blood
23                   Pb-FSIQ association in children was  substantiated in a pooled analysis of seven
24                   prospective studies by  Lanphear et al. (2005) as well as multiple meta-analyses that
25                   combined results across various prospective and cross-sectional studies (Pocock et al..
26                   1994: Schwartz. 1994: Needleman and Gatsonis.  1990). with Schwartz (1994)
27                   demonstrating the  robustness of evidence to potential publication bias.

28                   Across the prospective studies, blood Pb-associated FSIQ decrements were found with
29                   concurrent, prenatal (maternal or cord), earlier childhood, multiple year average, or
30                   lifetime average blood Pb levels. Associations also were found with tooth Pb levels.
31                   There is no clear indication of a stronger association of FSIQ with blood Pb level
32                   measured at a particular lifestage or time period. Concurrent blood Pb level in children
33                   reflects recent and past Pb exposures. Thus, several observations point to an effect of
34                   cumulative childhood Pb exposure. Blood Pb-associated FSIQ decrements were found in
35                   populations with mean blood Pb levels 5-10 (ig/dL. A common finding was a supralinear
36                   concentration-response relationship, i.e., a larger decrement in cognitive function per unit
37                   increase in blood Pb level in children in the lower range of the population blood Pb level
38                   distribution. In analyses restricted to  children in the lower range of blood Pb levels
      November 2012                             5-73                  Draft - Do Not Cite or Quote

-------
 1                  (e.g., peak <10 (ig/dL), associations were found in groups of children with mean blood
 2                  Pb levels 3-4 (ig/dL (Bellinger. 2008: Canfield. 2008: Hornung. 2008). Precision of effect
 3                  estimates was demonstrated in the pooled analysis by the narrow range of estimates,
 4                  -2.36 to -2.94 points per log increase in blood Pb level, obtained by excluding one study
 5                  at a time (Lanphear et al., 2005). Across individual studies, there was a wide range of
 6                  effect estimates reported for blood Pb-associated FSIQ decrements. However, there was
 7                  variability in model specification and blood Pb level ranges examined among studies
 8                  (Figure 5-2 and Table 5-3). Similarly larger effect estimates were found in the Boston
 9                  and Rochester cohorts, which differed in racial and SES  distributions. Although these
10                  studies had smaller sample sizes, they had at least as extensive consideration for potential
11                  confounding as other studies (Canfield et al.. 2003a: Bellinger et al.. 1992). Each study
12                  estimated larger effects for children whose peak blood Pb levels never exceeded
13                  10 (ig/dL, -1.8 points (95% CI: -3.0, -0.60) per 1 (ig/dL increase in concurrent blood Pb
14                  level in the Rochester cohort (Canfield et al.. 2003a) and -1.6 points (95% CI: -2.9, -0.2)
15                  per 1 (ig/dL increase in age 2-year blood Pb level in the Boston cohort (Bellinger and
16                  Needleman. 2003). These subsets of children had mean blood Pb levels of 3.3
17                  (Rochester)  and 3.8 (ig/dL (Boston), lower than those examined in other prospective
18                  studies.
      5.3.2.2  Bayley Scales of Infant Development

19                  The Mental Development Index (MDI) of the Bayley Scales of Infant Development is a
20                  widely used test of infant mental development. Black et al. [(2004) and Pollit (2005)]
21                  asserted that the MDI is a reliable indicator of current development and cognitive
22                  functioning of the infant, integrating cognitive skills such as sensory/perceptual acuities,
23                  discriminations, and response; acquisition of object constancy; memory learning and
24                  problem solving; vocalization and beginning of verbal communication; and basis of
25                  abstract thinking. However, the MDI test is not an intelligence test, and MDI scores,
26                  particularly before ages 2-3 years, are not necessarily correlated with later measurements
27                  of FSIQ in children with normal development. In the review of the MDI evidence,
28                  emphasis was placed on examinations at ages 2-3 years, which have test items more
29                  similar to those in school-age IQ tests. Most of the prospective studies reviewed in the
30                  2006 Pb AQCD (U.S. EPA. 2006b) found associations of prenatal, earlier infancy, and
31                  concurrent blood Pb level with MDI score in children between ages 2 and 3 years, and
32                  recent studies examined and found associations with cord blood Pb level  (Table 5-4).
      November 2012                             5-74                  Draft - Do Not Cite or Quote

-------
 1

 2

 3

 4

 5

 6

 7

 8

 9

10

11

12

13

14

15

16
 The prospective studies found blood Pb-associated decrements in MDI in some large

 (n =  146-592) populations with mean blood Pb levels 5-10 (ig/dL. Recruitment of

 participants before or at birth without consideration of Pb exposure or maternal IQ, high

 to moderate follow-up participation, and nonselective loss-to-follow-up in most studies

 increase confidence that the observed associations are not due to selection bias.

 Comparisons of blood Pb levels measured at various lifestages did not clearly indicate a

 stronger effect on MDI of prenatal or postnatal childhood blood Pb levels. While the

 prospective studies adjusted for birth outcomes and maternal IQ and education, most did

 not adjust for other SES indicators or parental caregiving quality. Concurrent and cord

 blood Pb levels were associated with MDI, with additional adjustment for SES and

 HOME score in the Boston cohort and HOME score in the Yugoslavia cohort

 (Wasserman et al., 1992; Bellinger etal.. 1987). In the Cleveland cohort, associations of

 cord, age 6 month, and concurrent blood Pb levels with MDI at age 2 years became null

 after adjusting for covariates including HOME score (Ernhart et al., 1988; Ernhart et al.,

 1987). However, 50% of the cohort was born to alcoholic mothers and may be less

 representative of the general U.S. population of children.
      Table 5-4     Associations of blood Pb level with Bayley MDI in children ages
                      12 months to 3 years.
      Study3
Study Population and
Methodological Details
(Presented in order of strength of study      Blood Pb
design and consideration for potential       Timing and
confounding)3	Levels (ug/dL)
MDI Assessment
                    Effect Estimate
                    (95% Cl)b
      Bellinger et al.    249 children followed from birth
      (1987)          (1979-1981) to age 2 yr, Boston, MA.
                     Prospective. Recruitment from birth
                     hospital. High follow-up participation.
                     Participants had higher cord blood Pb
                     level, SES, HOME score, maternal
                     education and IQ, lower maternal age,
                     were nonwhite. Regression model
                     adjusted for maternal age, race, maternal
                     IQ, maternal education, years of smoking,
                     alcohol drinks/week in 3rd trimester, SES,
                     HOME score, sex, birth weight,
                     gestational age, birth order.
                                    Prenatal (cord)
                                    Low: <3
                                    Medium: 6-7
                                    High:> 10
Adjusted mean at age
2yr
  High vs. low cord
  blood

  High vs. medium
  cord blood
-4.8 (-7.3, -2.3)


-3.8 (-6.3, -1.3)
      Jedrychowski et   444 children born 2001 -2004 followed
      al. (2009b)       prenatally to age 3 yr, Krakow, Poland.
                     Prospective cohort examining multiple
                     exposures and outcomes. Recruitment
                     from prenatal clinic. High follow-up
                     participation. Log linear regression model
                     adjusted for maternal education, birth
                     order, prenatal smoking, sex, and within-
                     subject MDI correlation. Did not consider
                     potential confounding by HOME score.
                                    Prenatal (cord)
                                    Geometric mean
                                    (range): 1.29
                                    (0.44-5)
                                    Interval
                                    analyzed:!.2-1.3 =
                                    10th-90th
                                    percentiles
Age 2 yr

Age 3 yr
-2.6 (-5.0,-0.21)

-2.3 (-4.3, -0.30)
      November 2012
                              5-75
 Draft - Do Not Cite or Quote

-------


Study3
Tellez-Rojo et
al. (2006)
Tellez-Rojo
(2008)





Claus Henn et
al. (2012)



Hu et al. (2006)














Pilsner et al.
(2010)






Surkan et al.
(2008)





Study Population and
Methodological Details
(Presented in order of strength of study
design and consideration for potential
confounding)3
1 93 children born 1 997-1 999 followed
prenatally to age 2 yr, Mexico City, Mexico
Cross-sectional. Recruitment from
prenatal clinic or birth hospital.
Participants had older, more educated
mothers, lower cord blood Pb level, and
slightly higher MDI. Log linear regression
model adjusted for sex, maternal age,
birth weight, maternal IQ, cohort.
Considered potential confounding by other
unspecified factors.
455 children born 1997-2000 followed
prenatally to age 3 yr, Mexico City, Mexico
Prospective, same cohort as above. No
selective participation of subjects. Linear
mixed effects regression adjusted for sex,
hemoglobin, gestational age, maternal IQ,
maternal education, blood Pb-blood Mn
interaction. Did not consider potential
confounding by HOME score.
1 46 children born 1 997-1 999 followed
prenatally to age 2 yr, Mexico City, Mexico
Prospective. Recruitment from prenatal
clinic. Moderate follow-up participation.
Eligible similar to non-eligible. Log linear
regression model adjusted for sex,
maternal age, current weight, height-for-
age Z score, maternal IQ, concurrent
blood Pb (in models examining blood Pb
at other lifestages). Considered potential
confounding by other unspecified factors.








255 children born 1994-1995 followed
prenatally to age 2 yr, Mexico City,
Mexico.
Prospective. Recruitment from birth
hospital. Low but not selective
participation. Linear regression model
adjusted for maternal age, maternal IQ,
marital status, parity, gestational age,
inadequate folate intake, MTHFR
genotype. Did not consider potential
confounding by HOME score.
309 children born 1991-2004 followed
from birth to age 3 yr, Mexico City,
Mexico.
Cross-sectional. Recruitment from birth
hospital. High participation rate. Linear
mixed effects model adjusted for sex,
maternal age, IQ, education, and self-
esteem, parity, grams/day alcohol,
smoking status, cohort. Did not consider
potential confounding by HOME score.

Blood Pb
Timing and
Levels (ug/dL)
Concurrent
Geometric mean:
2.9
Interval analyzed:
0.8-4.9 = range





Age 1 yr
Mean (SD): 5.1
(2.6)
Interval analyzed:
2.5-8.4 =10th-90t
h percentiles


Prenatal maternal
1st trimester
Mean (SD): 7.1
(5 1)
W- ' /
Prenatal maternal
3rd trimester
Mean (SD): 6.9
I A ^>\
(4.Z)
Prenatal cord
blood
Mean (SD): 6.2
(3.9)
Concurrent
Mean (SD): 4.8
(3.7)

Prenatal (cord)
Mean (SD):
6.7 (3.6)
Interval analyzed:
3.5-10.5 =
1 0th-90th
percentiles



Concurrent
Mean (SD): 6.4
(4.3)
Interval analyzed:
2.0 (10th
percentile)-10





MDI Assessment
Age 2 yr








Ages 1 to 3 yr
Blood Mn:
<2.0 ug/dL
Blood Mn:
2.0-2.8 ug/dL
Blood Mn >2.8 ug/dL

Age 2 yr
Prenatal 1st trimester
Interval analyzed: 2.5
(10thpercentile)-10)

Prenatal 3rd trimester
Interval analyzed: 2.8
(10thpercentile)- 10

Prenatal Cord blood
Interval analyzed: 2.5
(10th percentile)- 10

Concurrent
Interval analyzed:
1.6-9.1 = 10th-90th
percentiles
Age 2 yr







Ages 1 to 3 yr
All subjects
High maternal self-
esteem
Low maternal self-
esteem



Effect Estimate
/oco/ ^i\k
(95% Cl)
-1 .71 (-3.0, -0.42)









-3.0 (-5.22, -0.78)

-0.07 (-0.39, 0.25)
-2.2 (0, 4.44)


-0.91 (-1.8, -0.04)


-0.49 (-1.3, 0.31)


-0.07 (-0.93, 0.79)



-0.23 (-0.92, 0.45)



-0.73 (-1 .2, -0.23)








-0.1 8 (-0.45, 0.09)
0.36 (-0.50, 1.2)
-0.31 (-0.60, -0.02)



November 2012
5-76
Draft - Do Not Cite or Quote

-------



Study3
Vimpani et al.
(1985)







Wasserman et
al. (1992)








Solon et al.
(2008)







Ernhart et al.
(1988: 1987)





Study Population and
Methodological Details
(Presented in order of strength of study
design and consideration for potential
confounding)3
592 children followed prenatally to age 2
yr, Port Pirie, Australia.
Prospective. Residence near Pb smelter.
High baseline participation rate. Linear
regression model adjusted for maternal
age, education, IQ, workplace, and
prenatal marital status, paternal education
and workplace, parental relationship, child
birth rank, mouthing activity, oxygen use
at birth, apgar score, neonatal jaundice,
size for gestational age. Did not consider
potential confounding by HOME score.
392 children followed prenatally (from
1985) to age 2 yr, Kosovska Mitrovica and
Pristina, Yugoslavia.
Prospective. 53% live near Pb sources.
High follow-up participation, no selective
attrition. Log linear regression model
adjusted for sex, birth order, birth weight,
ethnic group, HOME, maternal education,
maternal age, maternal IQ.


502 children born 1997-2004, Visayas,
Philippines.
Cross-sectional. Census based
recruitment. No selective participation of
subjects. Two-stage linear regression
model to account for determinants of
blood Pb (sex, roof materials, water
source, breastfed for > 4 months) and
cognitive function (HOME score, maternal
education, maternal smoking, born
premature, region of residence).
359 children, followed prenatally to age 2
yr, Cleveland, OH
Prospective. Recruitment at birth
hospital. High follow-up participation, more
white, higher IQ, nonalcoholic mothers not
followed. 50% born to alcoholic mothers.
Linear regression adjusted for age, race,
sex, birth order, parental education,
maternal IQ, Authoritarian Family
Ideology, HOME.


Blood Pb
Timing and
Levels (ug/dL)
20% subjects had
age 2 yr blood Pb
levels >30







Concurrent
Means:
35.5 (K. Mitrovica)
8.4 (Pristina)






Concurrent
Mean (SD):
7.1 (7.7)
Interval analyzed:
1.6 (10th
percentile) -10




Means (SD):
Prenatal cord:
6.0(2.1)
Age 6 mo:
10.1 (3.3)
Concurrent:
16.7(6.5)





MDI Assessment
Age 2 yr
Maternal avg prenatal
blood Pb

Cord blood Pb
Age 6 mo blood Pb
Age 2 yr blood Pb
Lifetime (to age 2 yr)
ava
** • a


Age 2 yr
Cord blood Pb
6 mo blood Pb

12 mo blood Pb

18 mo blood Pb
24 mo blood Pb
Ages 6 mo to 3 yr









Age 2 yr
Prenatal cord
Age 6 mo blood Pb
Concurrent blood Pb




Effect Estimate
/oco/ ^i\k
(95% Cl)

-0.64

0.03
-0.40, p <0.05
-0.06
-0.31, p<0.05


Per three-fold
increase in
blood Pb
-1.7, p = 0.12
-1.1, p = 0.34

-1.7, p = 0.17

-1.8, p = 0.16
-2.5, p = 0.03
-1 .07 for
population mean
serum folate of
225 ug/dL, 95%
Cl: not available





Variance
estimates
0.0003, t = -0.21
0.00, p = 0.95
0.00, p = 0.95


     MDI = Mental Development Index, MTHFR = methylenetetrahydrofolate reductase
     aStudies are presented in order of strength of study design and consideration for potential confounding. All Mexico City studies
     were kept together.
     bExcept where noted, effect estimates are standardized to a 1 ug/dL increase in blood Pb level for the lowest blood Pb range
     examined in the study or for blood Pb level up to 10 ug/dL.
1
2
o
J
4
5
6
7
A large study of 444 children in Krakow, Poland, found cord blood Pb-associated
decrements in MDI at ages 2 and 3 years with lower cord blood Pb levels (median
1.23 (ig/dL, 5th-95th:  1.24-1.34 (ig/dL) than examined in other studies (Jedrychowski et
al.. 2009b) (Table 5-4). However, cord blood Pb levels reflect the pregnancy blood Pb
levels of mothers. Evidence indicates increased mobilization of Pb from bone to blood in
pregnant women (Sections 4.2.2.4 and  4.3.5.2). Thus, there is uncertainty regarding the
Pb exposure scenarios that contribute to associations between cord blood  Pb level and
MDI in children. Jedrychowski et al. (2009b) estimated a larger decrease  in MDI per unit
     November 2012
                             5-77
Draft - Do Not Cite or Quote

-------
 1                  increase in cord blood Pb level among the 233 males than among the 223 females. Other
 2                  observations have indicated increased susceptibility of the developing male central
 3                  nervous system (CNS) to environmental insults (Moffitt et al., 2001). Although median
 4                  cord blood Pb levels were similar in males (1.35 (ig/dL) and females (1.41 (ig/dL), the
 5                  mean age 3-year MDI score was slightly lower among males than among females (101
 6                  and 105, respectively).

 7                  Multiple studies in various Mexico City cohorts reported associations of prenatal
 8                  (maternal or cord) or child postnatal blood Pb levels with decrements in MDI in children
 9                  between ages  1 and 3 years (Claus Henn et al.. 2012; Pilsner et al. 2010; Surkan et al..
10                  2008: Hu et al.. 2006: Tellez-Roio et al.. 2006). Hu et al. (2006) compared associations
11                  among prenatal maternal blood Pb levels measured at different trimesters among 146
12                  children at age 2 years. Increases in first trimester maternal  blood Pb levels (whole blood
13                  or plasma) were associated with larger decreases in MDI scores than increased in
14                  maternal third trimester, cord, or child concurrent blood Pb  levels (Table 5-4). These
15                  results were adjusted for sex, 2-year blood Pb level, height-for-age Z score, weight,
16                  maternal age, and maternal IQ. Model covariates did not include  SES, maternal
17                  education, or HOME score; however, a larger list of unspecified potential confounding
18                  factors was considered.

19                  Consistent with several findings for FSIQ, Tellez-Rojo et al. (2006) found larger effect
20                  estimates in children with lower blood Pb levels. In linear models, a 1 (ig/dL increase in
21                  concurrent blood  Pb level was associated with a -1.71 point (95% CI: -3.0, -0.42)
22                  decrease in MDI among 193 children with concurrent blood Pb levels <5 (ig/dL and a
23                  -1.0 point (95% CI: -1.8, -0.26) decrease among 294 children with concurrent blood Pb
24                  levels <10 (ig/dL. In a follow-up of the same cohort to age 3 years, Claus Henn et al.
25                  (2012) found inconsistent interactions between blood Mn and Pb levels. Investigators
26                  selected mid-range (2.0-2.8 (ig/dL) blood  Mn levels as the reference group based on
27                  previous observations that MDI scores were least affected by increases in blood Mn level
28                  in this group. Larger blood Pb-associated MDI decrements were found in the  91 children
29                  each with blood Mn levels <2.0 (ig/dL and >2.8 (ig/dL with age 1 year blood  Pb level but
30                  not age 2  year blood Pb level. Adjustment for sex, gestational age, hemoglobin, maternal
31                  IQ, maternal education, and visit produced more negative effect estimates. Kim et al.
32                  (2009b) also found effect modification by blood Mn levels for the association between
33                  blood Pb level and FSIQ, but in older children ages 8-11 years  (Section 5.3.2.1).

34                  Other recent studies in Mexico City examined effect modification by maternal self-
35                  esteem, genetic variants, and nutritional status. Surkan et al. (2008) stratified data by the
36                  level of maternal  self-esteem as reported by mothers. Higher age  2-year blood Pb level
37                  was associated with lower MDI score among children with mothers in the lowest three
      November 2012                             5-78                  Draft - Do Not Cite or Quote

-------
 1                   quartiles of self-esteem but not among children with mothers in the highest quartile of
 2                   self-esteem (Table 5-4). Model covariates included cohort, sex, parity, and maternal IQ,
 3                   age, education, smoking, alcohol consumption, and self-esteem. These findings indicated
 4                   that maternal psychosocial functioning may influence the effects of Pb on the mental
 5                   development of young children.

 6                   In another study in Mexico City, higher cord blood Pb level was associated with a lower
 7                   MDI score in children at age 2 years (-0.73 points [95% CI: -12, -0.23] in MDI per
 8                   1 (ig/dL increase in cord blood Pb level in a linear model) (Pilsner et al., 2010).
 9                   Investigators reported a lack of effect modification by genetic variants in the
10                   methylenetetrahydrofolate reductase (MTHFR) enzyme, which is involved in folate
11                   metabolism. The MTHFR 677 TT genotype produces an enzyme with lower metabolic
12                   activity, is associated with lower serum folate levels (Kordas et al.. 2009). and in this
13                   Mexico City cohort, was associated with lower MDI score at age 2 years. Results from
14                   stratified analyses were not reported, thus differences in the magnitude of association
15                   between genotypes could not be compared.

16                   Consistent with the prospective evidence, a recent cross-sectional analysis indicated an
17                   association between higher concurrent blood Pb level and lower MDI score children in
18                   the Philippines, ages 6 months to 3 years (Solon et al., 2008). Although HOME score,
19                   years of schooling of child, premature birth, region of residence, and maternal education
20                   and smoking were examined as potential confounding factors, adjustment was made in
21                   two stages: first, adjustment for blood Pb determinants and second, adjustment for MDI
22                   determinants. This method may not adequately control for the variance shared by blood
23                   Pb level and MDI. In this cohort, children with lower folate levels had larger
24                   Pb-associated decreases in cognitive function. Among children with folate levels
25                   < 230 (ig/mL, blood Pb level had an association with lower MDI scores in the range of
26                   0.80 to 2.44 points. Among children with higher folate levels, blood Pb level was not
27                   estimated to have a negative impact. The results from this study indicated a moderating
28                   effect of folate  on blood Pb levels as folate levels were not associated with MDI. Higher
29                   folate level has been associated with lower blood Pb level due to the role of folate in
30                   increasing Pb excretion by inhibiting the binding of Pb to blood elements.

31                   In summary, evidence consistently indicates associations of higher blood Pb levels with
32                   lower MDI scores in children ages 2-3 years (Table 5-4). Key evidence was provided by
33                   prospective studies, in particular those that adjusted for maternal IQ and education, SES,
34                   birth outcomes, and HOME score rWasserman et al.. 1992; Bellinger et al.. 1987).
35                   Several large studies contributed to the evidence (n = 146-592), and several studies had
36                   high to moderate follow-up participation, and nonselective  loss-to-follow-up, which
37                   reduces the likelihood of selection bias. Higher blood Pb level was associated with lower
      November 2012                             5-79                  Draft - Do Not Cite or Quote

-------
 1                  MDI scores in a few different cohorts in Mexico City and other populations, and while
 2                  most adjusted for maternal education and IQ, they did not examine potential confounding
 3                  by parental caregiving quality. The lack of association observed in the Cleveland cohort
 4                  does not mitigate the otherwise compelling evidence, given the high prevalence of
 5                  prenatal alcohol exposure in this cohort and the evidence in the Boston and Yugoslavia
 6                  cohorts that adjusted for several potential confounding factors. MDI was associated with
 7                  prenatal (cord or maternal) and concurrent blood Pb levels. Comparison among blood Pb
 8                  levels at different lifestages did not consistently indicate a stronger effect on MDI of
 9                  prenatal or postnatal Pb levels. Pb-associated decrements in MDI were found in
10                  populations with mean blood Pb levels 5-10  (ig/dL. However, a cross-sectional analysis
11                  in children in Mexico City found a larger decrement in age 2 year MDI per unit increase
12                  in concurrent blood Pb level among children with blood Pb levels <5 (ig/dL versus
13                  5-10 (ig/dL or <10 (ig/dL (Tellez-Rojo et al.. 2006). An association was found in children
14                  in Poland with lower cord blood Pb levels, median 1.23 (ig/dL (Jedrychowski et al.,
15                  2009b). However, since cord blood Pb levels reflect blood Pb levels of mothers, which in
16                  turn are influenced by Pb mobilized from bone to blood, the specific Pb exposure
17                  scenario contributing to the observed associations  is uncertain. Overall, while evidence
18                  indicates associations of higher prenatal and postnatal child blood Pb levels with lower
19                  MDI scores in young children (ages 2-3 years), the impact on later cognitive function is
20                  not certain since MDI scores do not necessarily predict later IQ in children with normal
21                  development.
      5.3.2.3  Learning and Memory in Children

                    Epidemiologic Studies of Learning and Memory in Children

22                  The small body of studies in the 2006 Pb AQCD did not clearly indicate associations
23                  between higher blood Pb level and poorer memory or learning (i.e., acquisition of new
24                  information) in children ages 5-17 years (Table 5-5). Studies used various tests to assess
25                  learning and memory, which may account for some of the heterogeneity observed in
26                  effect estimates. Evidence from prospective analyses in the Rochester, Boston, and
27                  Cincinnati cohorts was mixed, with an association found in the Rochester cohort at age 5
28                  years (Canfield et al.. 2004). associations in the positive and negative direction in the
29                  Boston cohort across the various tests and ages of blood Pb examined at ages 5 and  10
30                  years (Stiles and Bellinger.  1993; Bellinger et al.. 1991). and blood Pb-associated
31                  improved memory found in the Cincinnati cohort at age 15-17 years (Ris et al.. 2004).
32                  Previous cross-sectional studies found associations between higher concurrent blood Pb
33                  levels and poorer learning and memory, including the large (n = 4,852) study of children
      November 2012                             5-80                   Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5

 6
 7
 8
 9
10
11
12
13
ages 6-16 years participating in NHANES (Lanphear et al.. 2000). Several recent studies,
all cross-sectional, also found associations between higher concurrent blood Pb level and
poorer memory in children ages 6-16 years. Some were variants of previous studies
(Krieg et al.. 2010; Froehlich et al.. 2007): others had limited implications because of
little consideration for potential confounding (Counter et al., 2008; Min et al., 2007).
The prospective studies had smaller sample sizes (n = 148-195) than cross-sectional
studies (n = 246-4,852) but greater examination of potential confounding. Further,
recruitment of participants before or at birth, moderate to high follow-up participation,
and in most cases follow-up not biased to higher blood Pb levels and lower cognitive
function reduce the likelihood of selection bias (Table 5-5). Another strength was the
examination of earlier childhood or lifetime average blood Pb levels, which better
indicated the temporal sequence between Pb exposure and decrements in learning and
memory.
     Table 5-5    Associations between blood Pb levels and performance on tests of
                    learning and memory in children.




Study
Canfield et
al. (2004)










Froehlich et
al. (2007)





Study Population and Methodological
Details
(Prospective studies first, then cross-sectional
studies. Within each category, presented in
order of strength of study design)
174 children born 1994-1995 followed from age
6 mo to 5 yr, Rochester, NY
Prospective. Recruitment from study of dust
control. 73% nonwhite. High follow-up participation,
no selective attrition. Linear regression model
adjusted for neonatal intensive care unit (NICU)
admission, sex, age, spatial span length. Also
considered potential confounding by prenatal
smoking, household income, maternal IQ and
education, ethnicity, HOME, breastfeeding
duration, 1st prenatal visit, spatial working memory
problem, birth weight, marital status, household
crowding.
174 children born 1994-1995 followed from age
6 mo to age 5 yr, Rochester, NY
Cross-sectional. Same cohort as above. High
follow-up participation, no selective attrition. Linear
regression model adjusted for income (spatial
working memory), HOME, maternal IQ, race
(spatial span). Also considered potential
confounding by transferrin saturation, prenatal
smoking exposure, maternal education, age, NICU,
sex.


Blood Pb
Levels
(ug/dL)
Lifetime (to age
5 yr) avg
Mean (SD): 7.2
(3.6)
Interval
analyzed: 3.5
(10th
percentile)-10




Concurrent
Mean (SD): 6.1
(4.9)
Interval
analyzed: 1.9
/•] r\*u
^ i uin
percentile)-10




.
Memory/ SI?fV!n^'ma*e
Learning Test (95/oCI)
Spatial span total -0.1 1 (-0.20, -0.02)b
errors CANTAB
Age 5 yr









Spatial working -0.51 (-1.2, 0.1 6)b
memory,
total errors
Spatial span -0.02 (-0.04, 0)
length
CANTAB
Age 5 yr


     November 2012
                           5-81
Draft - Do Not Cite or Quote

-------
Study
Bellinger et
al. (1991)
Stiles and
Bellinger
(1993)
Riset al.
(2004)
Lanphear et
al. (2000)
Krieg et al.
(2010)
Study Population and Methodological
Details
(Prospective studies first, then cross-sectional
studies. Within each category, presented in
order of strength of study design)
1 70 children followed from birth (1 979-1 981 ) to age
5 yr, Boston, MA area
Prospective. Recruitment at birth hospital.
Moderate follow-up participation. More participants
were white, had higher age 2 yr HOME score,
higher postnatal blood Pb levels. Log linear
regression model adjusted for SES, maternal IQ
and marital status, preschool attendance, HOME,
out of home care, residence changes, medication
use in previous 12 mo, # adults in home, child sex,
race, birth weight, birth order.
1 48 children followed from birth (1 979-1 981 ) to age
10yr, Boston, MA area
Prospective. Same cohort as above. Moderate
follow-up participation, participants had higher SES
and HOME score. Linear regression model
adjusted for HOME, family stress, race, marital
status (earlier childhood blood Pb), HOME, family
stress, maternal age and race, birth weight, number
of daycare situations to age 57 mo (concurrent
blood Pb).
195 children followed prenatally (1979-1985) to age
15-1 7 yr, Cincinnati, OH
Prospective. Recruitment at prenatal clinic. High
follow-up participation, no selective attrition. Mostly
African-American. Linear regression model
adjusted for SES, maternal IQ, HOME, adolescent
marijuana use, and obstetrical complications. Also
considered potential confounding by birth
outcomes, maternal age, prenatal smoking,
alcohol, marijuana, and narcotics use, number of
previous abortions, stillbirths, gravidity, parity,
caregiver education, public assistance, child age,
sex, health, Fe status
4,852 children ages 6-1 6 yr (born 1972-1988), U.S.
NHANES III (1988-1994)
Cross-sectional. Large U.S. representative study
of multiple risk factors and outcomes. Linear
regression model adjusted for sex, race/ethnicity,
poverty index ratio, reference adult education,
serum ferritin and cotinine levels. Did not consider
potential confounding by parental cognitive
function, HOME score.
766-780 children ages 1 2-1 6 yr (born 1 975-1 982),
U.S. NHANES III (1991 -1994)
Cross-sectional. Large U.S. representative study
of multiple risk factors and outcomes. Log linear
regression model adjusted for sex, caregiver
education, family income, race/ethnicity, test
language. Did not consider potential confounding
by parental cognitive function, HOME score.
Blood Pb
Levels
(ug/dL)
Earlier
childhood (age
2yr)
Mean (SD):
7.0 (6.6)
Interval
analyzed: 1.8
(10th
percentile)-10
Concurrent
blood Pb levels
NR
Earlier
childhood
Exact levels NR
but concurrent
mean reported
to be <8
Memory/
Learning Test
Memory
Age 2 yr blood
Pb
Concurrent blood
Pb
McCarthy Scale
of Children's
Abilities,
Age 5 yr
Effect Estimate
(95% Clf
-0.1 4 (-0.52, 0.25)
0.10 (-0.41, 0.61)
Perseveration score, CVLT, Age 10 yr
Age 1 yr blood
Pb
Age 2 yr blood
Pb
0.02 (0, 0.04)b
-0.03 (-0.05, -0.01 )b
# trials to 1st category, WCST, Age 10 yr

Earlier
childhood (age
6.5 yr)
Mean (SD): NR
Concurrent
Geometric
mean: 1.9
(5th-95th:
1.70-2.10)
Interval
analyzed:
1 .74-2.06 =
1 0th-90th
percentiles
Concurrent
Mean
(5th-95th): 1 .95
(1 .63-2.27)
Interval
analyzed:
1.69-2.19 =
1 0th-90th
percentiles
Age 1 yr blood
Pb
Memory
composite of
SubtestsofCVLT
Ages 15-1 7 yr
Digit Span
WISC-R
Ages 6-1 6 yr
Digit span
WISC-R
Ages 12-1 6 yr
-0.44 (-0.93, 0.05)b
0.01 (-0.02, 0.05)
-0.05 (-0.89, -0.01)
-0.42 (-0.65, -0.18)
November 2012
5-82
Draft - Do Not Cite or Quote

-------
Study
Surkan et
al. (2007)
Kordas et
al. (2006)
Study Population and Methodological
Details
(Prospective studies first, then cross-sectional
studies. Within each category, presented in
order of strength of study design)
389 children ages 6-10 yr, Boston, MA,
Farmington, ME
Cross-sectional. Recruitment from trial of
amalgam fillings. High participation rate. Higher
participation by white children in Maine. Analysis of
covariance adjusted for caregiver IQ, child age,
SES, race, and birth weight, Also considered
potential confounding by caregiver education and
marital status, parenting stress, and maternal
utilization of prenatal or annual health care but not
HOME score.
293 children, age 7 yr, Torreon, Mexico.
Cross-sectional. Recruitment at prenatal clinic.
High participation rate. Residence near metal
Blood Pb
Levels
(ug/dL)
Concurrent
Mean (SD): 2.2
(1.6)
Concurrent
Interval
analyzed:
Memory/
Learning Test
General memory
index, WRAML
Ages 6-1 0 yr
Stern berg
memory
Age 7 yr
Effect Estimate
(95% Clf
-6.7 (-12.1, -1.2)°
Blood Pb group
5-10 ug/dL vs.
1-2 ug/dL
-0.1 6 (-0.37, 0.05)
      Chiodo et
      al. (2004)
                                          Concurrent
                                          Mean (SD):
                                          5.4 (3.3)
                                          Interval
                                          analyzed:
                                          2.3-9.5 =
                                          10th-90th
                                          percentiles
 Verbal learning,
 WRAML

 Corsi Backward
 Spatial Span
 Age 7.5 yr
-0.20, p>0.05d
                                                                                                  -0.22, p>0.05d
             foundry. Linear regression model adjusted for child    2.1-10.0
             sex, age, school, birth order, hemoglobin, forgetting
             homework, household possessions and crowding,
             house ownership, maternal education, family
             structure, urinary As, tester. Did not consider
             potential confounding by HOME score or parental
             cognitive function.

             246 children, age 7.5 yr, Detroit, Ml area
             Cross-sectional. Recruitment at prenatal clinic. All
             African-American. High prevalence of prenatal
             alcohol exposure. High participation rate. Log linear
             regression model adjusted for SES (all outcomes).
             SES, caregiver vocabulary, disruption in caregiver
             (verbal learning). HOME score, child age, child  sex,
             caregiver education, parity (spatial span). Also
             considered potential confounding by maternal
             prenatal marijuana, smoking,  or cocaine use,
             crowding, child life stress, caregiver age, life stress,
             and psychology, conflict tactics, prenatal alcohol
             exposure, family functioning, # children <18 years.

Note: Results are presented first for prospective studies then for cross-sectional studies. Results from the same cohort are kept
together. Within each category, results are presented in order of strength of study design.
CANTAB = Cambridge Neuropsychological Testing Automated Battery, CVLT = California Verbal Learning Test,
WCST = Wisconsin Card Sorting Test, WISC-R = Wechsler Intelligence Scale for Children Revised, WRAML = Wide Range
Assessment of Memory and Learning
"Effect estimates are standardized to a 1  ug/dL increase in blood Pb level in the lowest range of blood Pb levels examined in the
study or the interval from the 10th percentile to 10 ug/dL or the 90th percentile, whichever is lower.
bThe direction of the effect estimate was changed such that a negative estimate represents poorer performance and a positive
estimate represents better performance.
°Effect estimate compares test performance of children in higher blood Pb groups to children in lowest blood Pb group.
Sufficient data were not provided to calculate 95% Cl.
 1

 2

 3

 4

 5

 6

 7

 8

 9

10
There were contrasting associations between blood Pb levels and memory in the

Rochester and Boston cohorts at age 5 years (Canfield et al.. 2004; Bellinger et al.. 1991).

Although different time periods of blood Pb level were examined, the population means

were similar, ~7 (ig/dL. In the Rochester cohort, a 1 (ig/dL increase in higher lifetime

average blood Pb level was associated with 0.11 (95% CI: 0.02, 0.20) more total errors

on the spatial span memory task (i.e., errors in replicating a sequence pattern) with

adjustment for neonatal intensive care unit (NICU) admission, sex, age, and spatial span

length and consideration for several other factors (Table 5-5) (Canfield et al.. 2004).

Recent evidence extended findings to associations between poorer performance on a

spatial working memory tasks and higher concurrent blood Pb levels (Froehlich et al..
      November 2012
                                5-83
Draft - Do Not Cite or Quote

-------
 1                   2007). which also reflect represent past and recent Pb exposure. In the recent analysis,
 2                   associations were found with spatial span length (number of squares recalled correctly)
 3                   with adjustment for HOME score, maternal IQ, and race; and with spatial working
 4                   memory errors with adjustment for family income. In each analysis of the Rochester
 5                   cohort, multiple associations were examined; however, there were consistent patterns
 6                   blood Pb-associated decrements in cognitive function observed across the various indices
 7                   of memory, learning, and executive function examined. Coherence for associations with
 8                   performance on spatial span and spatial working memory tasks was found with evidence
 9                   in rodents for Pb-induced impaired performance on visual-spatial memory tasks in the
10                   Morris water maze and working memory tasks in the radial arm maze, respectively
11                   (discussed below). In contrast, in the Boston cohort at age 5 years, concurrent blood Pb
12                   level was not associated with poorer memory, as assessed using the McCarthy Scale of
13                   Children's Abilities (Bellinger et al.. 1991). These results were adjusted for more
14                   potential confounding factors than results from other studies (Table 5-5). including SES,
15                   maternal IQ, and HOME score. Higher age 2 year blood Pb level was associated with
16                   poorer memory at age 5 years, but the association lacked precision (Table 5-5).

17                   In the Boston cohort at age 10 years, associations were inconsistent across the  multiple
18                   tests of memory and learning and time periods of blood Pb levels (ages 1, 2, 5, 10 years)
19                   examined (Stiles and Bellinger. 1993). For example, higher age 1 year blood Pb level was
20                   associated with better memory (i.e., fewer errors in recalling word list) as assessed with
21                   the California Verbal Learning Test. In the Cincinnati cohort at age 15-17 years, higher
22                   earlier childhood (age 6.5-year) blood Pb level also was associated with better memory
23                   (composite score of various subtests of the California Verbal Learning Test with
24                   adjustment for similar potential confounding  factors plus adolescent marijuana use (Ris et
25                   al.. 2004). In two independent Boston-area cohorts examined at different ages, poorer
26                   learning (number of trials to sort cards properly or number of categories achieved) as
27                   assessed with the Wisconsin Card Sorting Test (WCST) was associated with higher age 1
28                   year blood Pb level in children ages 10 years (Stiles and Bellinger. 1993) and with higher
29                   childhood tooth Pb levels in young adults ages 19-20 years (-0.6 categories [95% CI:
30                   -1.0,  -0.21] per natural log unit increase in tooth Pb level [collected from age 5-8 years],
31                   with adjustment for parental IQ, maternal age and education, SES, sex, birth order,
32                   current smoking status, drug use, and alcohol use) (Bellinger et al.. 1994a). In the
33                   younger cohort age 10 years, decrements in learning as assessed by performance on the
34                   WCST were not consistently found across the various ages of blood Pb level examined
35                   (Stiles and Bellinger. 1993).

36                   The cross-sectional studies examined potential confounding by parental education and
37                   SES, but a notable omission was consideration for HOME score. Chiodo et al. (2004)
38                   found a concurrent blood Pb-associated decrement in spatial memory with adjustment for
      November 2012                             5-84                  Draft - Do Not Cite or Quote

-------
 1                  HOME score. However, the results may lack generalizability because of the high
 2                  prevalence of prenatal alcohol exposure. HOME score was not associated with memory
 3                  in every study. For example, in the Rochester cohort, only household income remained
 4                  significantly associated with total errors in the spatial working memory task and was
 5                  included in the final model (Froehlich et al., 2007). Therefore, the confounding factors
 6                  may vary among populations. Most cross-sectional studies  found Pb-associated
 7                  decrements in memory in populations with mean concurrent blood Pb levels 5-8 (ig/dL.
 8                  Despite the lack of information on HOME score, the cross-sectional analyses of children
 9                  participating in NHANES III had several strengths, including large sample sizes
10                  (n = 760-4,852), high participation rates, lower likelihood of selection bias due to the
11                  examination of multiple risk factors and outcomes, nationally-representative results, and
12                  examination of the shape of the concentration-response relationship (Krieg etal.. 2010;
13                  Lanphear et al.. 2000). An increase in concurrent blood Pb  level was associated with a
14                  larger decrement in digit span score in adolescents ages 12-16 years (Krieg et al., 2010)
15                  than children 6-16 years (Lanphear et al.. 2000); however, the influence of higher past Pb
16                  exposures is likely greater in older children. In the analysis of children ages 6-16 years,
17                  Lanphear et al. (2000) estimated the largest decrement in memory score per unit increase
18                  in blood Pb level in children with concurrent blood Pb levels <2.5 (ig/dL (-0.25 [95% CI:
19                  -0.58, 0.08] points per 1 (ig/dL increase in concurrent blood Pb level versus -0.05 [95%
20                  CI: -0.09, -0.01] among  all subjects). A nonlinear concentration response relationship
21                  also was found among children ages 7 years living near a metal foundry in Torreon,
22                  Mexico, with a larger Pb-associated decrement in memory  found among children with
23                  concurrent blood Pb levels <10 (ig/dL (Kordas et al..  2006).


                    lexicological Studies of Learning and  Memory

24                  As described in the preceding sections, blood Pb levels are  consistently associated with
25                  decrements in FSIQ in children but show more variable associations with performance on
26                  neuropsychological tests of learning and memory. A relationship between Pb exposure
27                  and cognitive function decrements is supported  further by evidence for Pb-induced
28                  impairments in memory and  learning in animals. The 2006 Pb AQCD (U.S. EPA. 2006b)
29                  reported Pb-induced impaired memory and learning primarily in animals with Pb
30                  exposures that resulted in blood Pb levels 30-50 (ig/dL; however, some studies observed
31                  impairments in rodents (pre- and/or post-natal Pb exposure) and monkeys (lifetime
32                  postnatal Pb exposure) with blood Pb levels 14-25 (ig/dL (Altmann et al.. 1993; Rice and
33                  Karpinski.  1988; Gilbert and Rice, 1987). Several recent studies added to the evidence for
34                  impaired learning and memory in animals with lower blood Pb levels in the range
35                  relevant to humans, 8-17 (ig/dL (Cory-Slechta et al., 2010;  Niu et al., 2009; Virgolini  et
36                  al.. 2008a; Stangle et al.. 2007). Effects in animals with lower blood Pb levels generally
      November 2012                             5-85                   Draft - Do Not Cite or Quote

-------
1                   were found with gestational/lactational Pb exposures. Results from toxicological studies
2                   on learning as well as other nervous system endpoints that provided concentration-
3                   response information (i.e., those with multiple Pb exposure concentrations) are shown in
4                   Figure 5-3 and associated Table 5-6. These results demonstrate the coherence among
5                   inter-related CNS changes induced by Pb exposure in animals, including deficits in CNS
6                   development and plasticity and alterations in neurotransmitters, which mediate cognitive
7                   function.
     November 2012                              5-86                   Draft - Do Not Cite or Quote

-------
                                Behavioral; Neonate; Rat; Fernale(l)
                                  Beh.ivior.il; Neonale; Ral; Bolh (2)
                                  BDhaviorat: Nconate; Rat; Male (i)
                                  Beli.ivior.il; Neonatc; Rat; Bolh(<1)
                                   Cognition; NeolhUD; Rat; Both (5)
                                     Cognition; Adull; Rat; Male(G)
                                  Cognition;Adull: Monkey; Bolh (7)
                                Cognition; Ncon.lte; Mouse; Both(8)
                                   l_ortniticin;N<>onate: Rat; Both (9)
                                   Cognition;Neonate; Rat; Both (1)
                            ' .,,(,. ,. ,1, i ..,,.• (J. ",,.,!•  Rat; Female (10)
                              • ,,ih, ,,'.1 ,',.,,„•  !!.•..,,,,i. Ral; Male(ll)
                                i ..rii. ..-.I.-(..!!,• .-...lull Ral; M ,i, i i •,
                                Corticosterone;Adult; Ral; Male(13)
                                Morphology; Neonale; Rat;Both I M)
                                   Morphology: Adult; Rat; Male (6)
                             Morphology; AduH: Montey; Female (15)
                                  Morphology; Adult; Ral;Male (1C)
                              Morphology. Neonale; Mouse; Bolh (8)
                                 Morphologv; Neonate; Rat; Both (9)
                           Motor lunction; Neon.ile; Mouse; M.ile(J7)
                                M..r,.r Inn. n,,i, '  I'lli II.,-  M i, I i  i
                              Motor (unction; Neonale; Rat; Both (5)
                                Mm or dinci inn; Adull: Rat; Male(13)
                              Molor (unction; Ncnn.ite; Ral; Both (1)
                              Mot or (unction; Neon ate; Rat; Male(3)
                            Ncurophysiology; Neonale; Rat; Both (1*1)
                         Neuro!ransmiller; Neonate; Mouse; Both(18)
                         Neurotransmilter, Neonate; Mouse: Both(17)
                            Neurotransmltter Adult; Rat; Female (13)
                              Ncurotr.insmillor; Adull; Ral; Male(13)
                          NeurotransmKten Neonale; Rat; Female (11)
                          Ond.Ttivc St ross; Adull; Monkey; Female (IS)
                           »...i'l.itivi;Sli<£.s; Neonale; Mouse; Bolh (8)
                         Physical Develop me nl; Adull; Ral; Female (14)
                        I'll"'.!, .il (>i,vi'li>|iMii'iil: H.,nii.ili-. R.it. (•i-ni.ili'tl)
                         F'livsii'-il ll.".'fli-.|im.Ti1; M.'on.il,-: fi.fl; Mal.-(i)
                            Physical [levebpmcnl; Adull; Ral; M.ile(C)
                       ProlKefMion/dlfl/survival; Neonate; Rat; Both (9)
                 Stress-Induced cort icosterone; Neonale; (tat: Fern ale (10)
                   Slrcss-induccd cofticosti;ronc;Adull; Ral; Female(13)
                 Slriiii-induced cort icosterone; Neonale; Ral; Fern ale (11)
                     Stress-induced motorfunction; Adull; Rat; Both(13)
              Stress-induced neurotransmiltcr: Neonate; Rat; Female (10)
                 Stress-induced ncumtransmltter. Adult; Ral; Female (13)
                  Slress-inducedneurotranimlller; Adult; Rat; Male(13)
                Siren-induced neu retransmit tec Neonale; Rat; Male (11)
                   • Highest Concentration
                   *Lowfl5tCooc. with Response
                   A Highest Cone with No Response
                   o LosvestConcentration
                                                                                10
                                                                                                   100
                                                                                                                    1,000
                   Blood Lead (ng/dl)
Note: This figure illustrates nervous system effects associated with Pb exposure in studies that examined multiple exposure
concentrations. Dosimetric representation reported by blood Pb level. (ID corresponds to studies described in Table 5-6)
Figure 5-3          Summary of Pb exposure-nervous system  concentration-
                          response information from toxicological  studies.
November 2012
5-87
Draft - Do Not Cite or Quote

-------
Table 5-6     Summary of findings from  neurotoxicological  concentration-
                  response array presented in Figure 5-3.
Study
 ID in
Figure
  5-3      Reference
                      Blood Pb
                        Level
                       (ug/dL)
                                                                       Outcome
1
10
         Beaudin et
         al. (2007)
         Grant et al.
         (1980)
                      13 & 31     Behavior, neonate: Lactational Pb exposure, offspring deficient in Reward Omission testing.

                                  Physical development; Postnatal Pb exposure (birth to 4 weeks of age): Pb-induced
                                  development of over-reactivity to reward omission and errors is reversible with chelation
                                  treatment.
                      57
                                  Behavior, neonate: chronic Pb exposure (drinking water) to dams and pups, Changed behavior,
                                  males.
         Kishi et al.    59 & 186    Behavior, neonate: Pb exposure (oral gavage of pups) during lactational period, Changed
         (1983)                    emotional behavior, males and females.

                                  Motor function, neonate: Pb exposure (oral gavage of pups) during lactational period, motor
                                  function (rotarod performance) impaired, both sexes.

                                  Physical development; Pb exposure during lactation (oral gavage): Delayed development of
                                  righting reflex in male rats.

         Overmann    33, 174 &   Behavior-Pb exposure (oral  gavage of pups) during lactation: aversive conditioning affected by
         (1977)        226        Pb exposure, male and females.

                                  Cognition-Pb exposure (oral gavage of pups) during lactation: Response inhibition impaired,
                                  both sexes.

                                  Motor function- Pb exposure (oral gavage of pups) during lactation: Increased motor activity
                                  and impaired motor coordination (rotarod), male and females.

         Stangle et    13 & 31     Cognition;  Developmental Pb exposure (PND1-PND30): Impaired learning with visual
         al. (2007)                 discrimination task, heightened response to errors, both sexes.

                                  Motor function; Developmental Pb exposure (PND1-PND30): Alcove latency and response
                                  latency significantly affected by  Pb exposure, both sexes.
         Gong &
         Evans
         (1997)
                      38 & 85     Cognition-Adult male 21 day Pb exposure: Hyperactivity with Habituation to new cage
                                  environment.

                                  Morphology; 21 day Pb exposure to adult males: Marker of neuronal injury-elevated
                                  hippocampal glial fibrillary acidic protein (GFAP).

                                  Physical development; Adult male rats (21 day Pb exposure): Neurotoxicity measured with
                                  brain glial fibrillary acidic protein (GFAP).
         Rice (1990)   32 & 36
                                  Cognition-Chronic Pb exposure from birth: Spatial discrimination reversal task impairment, both
                                  sexes.
         Lietal.
         (2009C)
                      80 & 100    Cognition-Gestational & lactational Pb exposure: Morris water maze performance impaired.

                                  Morphology; Gestational & lactational Pb exposure: Increased levels of inflammatory cytokines
                                  & exocytosis related proteins in brains of pups at weaning, both sexes.

                                  Oxidative stress-gestational and lactational Pb exposure: Elevated hippocampal TNF levels in
                                  offspring, males and females.

                      80 & 102    Cognition- Gestational & lactational Pb exposure: Morris water maze performance impaired.

                                  Morphology: Increased levels of Alzheimer disease-associated proteins in mice with gestational
                                  and lactational Pb exposure, both  sexes.

                                  Proliferation/diff/survival, gestational & lactational Pb exposure: Increased hippocampal
                                  expression of P-tau and amyloid beta in male and female pups.

                      10 & 13     Corticosterone: Lifetime Pb +/- stress: Correlation between 9-month old's corticosterone level
                                  and frontal cortex dopamine levels in behaviorally tested female offspring.

                                  Stress: Corticosterone-Lifetime Pb plus stress: Affects Fl performance, dopamine and serotonin
                                  levels in  female offspring.

                                  Stress: Corticosterone-neurotransmitter-Lifetime Pb exposure in female rats plus stress:
                                  Dopamine homeostasis affected.

11        Virgolini et    25          Corticosterone: Maternal Pb plus stress: Elevated corticosterone in male offspring with prenatal
         al. (2008b)                stress + offspring stress was further  enhanced with Pb exposure.

                                  Stress: Corticosterone-Maternal Pb plus stress:  Affects Fl performance.
         Lietal.
         (201 Ob)
         Cory-
         Slechta et
         al. (2010)
November 2012
                                                   5-8
Draft - Do Not Cite or Quote

-------
Study
 ID in
Figure
  5-3      Reference
16
18
Blood Pb
  Level
 (ug/dL)
                                                                        Outcome
                                  Neurotransmitter; Gestational and lactational Pb exposure: Induced dopamine and serotonin
                                  changes in rat offspring.

                                  Stress induced neurotransmitter effects, Maternal Pb plus stress: serotonin and 5-HIAA
                                  (serotonin metabolite), and dopamine turnover were significantly affected in males.

12       Virgolini et     15&27     Corticosterone: Chronic Pb exposure from weaning: Pb exposure alone decreased basal
         al. (2005)                 plasma corticosterone levels at 5 months of age, males.

                                  Motor function: Chronic Pb exposure from weaning: Locomotor activity significantly decreased
                                  Fl response rates & increased post-reinforcement pause period in a concentration-dependent
                                  manner, males.

                                  Stress: Corticosterone-Chronic Pb plus stress: Affects neurotransmitters & Fl performance

13       Virgolini et     31          Corticosterone: Maternal Pb exposure (gestation and lactation) +/- stress: Differential basal
         al. (2008a)                corticosterone levels between behavioral and non-behavioral tested rats, females.

                       11 &/or     Stress: Corticosterone-Maternal Pb  plus stress: Affects Fl performance, dopamine, serotonin,
                       31          and NE levels.

                                  Motor function: Maternal Pb +/- stress: Increased locomotor activity (run  rate) with Pb and
                                  stress exposure.

                       31          Neurotransmitter; Gestational and lactational Pb exposure: Induced NE aberrations in adult rat
                                  offspring (both sexes).

                                  Stress induced motor function: Maternal Pb +/- stress: Increased locomotor activity (run rate)
                                  with Pb and stress exposure.

                                  Stress induced neurotransmitter; Gestational and lactational Pb exposure + stress: Induced
                                  HVA (monoamine neurotransmitter metabolite) and NE aberrations in female adult rat offspring.

14       Hu et al.       4&12      Morphology; Gestational Pb exposure: Neurite outgrowth marker PSA-NCAM decreased in rat
         (2008b)                  pups, both sexes.

                                  Neurophysiology; gestation Pb exposure: decreased hippocampal sialyltransferase activity,
                                  both sexes.

                                  Physical development; t-Gestational Pb exposure:  Early  brain synapse development impaired
                                  (hippocampal PSA-NCAM).

15       Wuetal.       19&26     Morphology: Elevated expression of Alzheimer's disease-related genes and Tc factors in aged
         (2008a)                  brains of female monkeys (exposed to Pb as infants).

                                  Oxidative stress: Elevated oxidative DNA damage  in aged brains of female monkeys (exposed
                                  to Pb as infants).
         Tavakoli-
         Nezhad et
         al. (2001)
18, 29, &    Morphology; 3 to 6 weeks of Postnatal (starting at PND22) Pb exposure in males: Decreased
54          number of spontaneously active midbrain dopamine neurons.
17       Leasure et     10 & 42     Motor function; Mouse maternal (dam) Pb exposure: Induced decreased rotarod performance
         al. (2008)                 in offspring (1  year-old male offspring).

                                  Neurotransmitter; Mouse maternal (dam) Pb exposure: Affects 1 -year old male offspring
                                  dopamine homeostasis,  both sexes.
         Fortune &
         Lurie (2009)
                      8&43
            Neurotransmitter; Mouse maternal (dam) Pb exposure: Affects offspring superior olivary
            complex (auditory) neurotransmitters, both sexes.
                  Learning and Memory - Morris Water Maze
November 2012
                             5-89
Draft - Do Not Cite or Quote

-------
 1                  In the Rochester cohort of children, blood Pb level was associated with poorer
 2                  performance on tests of spatial memory (Table 5-5). In animals, spatial memory has been
 3                  tested using the Morris water maze apparatus. The Morris water maze can be used to test
 4                  spatial memory and learning ability by assessing the time or distance taken for a rodent to
 5                  swim to a submerged platform using visual cues and in subsequent trials with the
 6                  platform removed, the time spent in the previous location of the platform. The
 7                  2006 Pb AQCD (U.S. EPA. 2006b) reported mixed effects of Pb on memory; some
 8                  studies found Pb-induced impaired memory in animals whereas others found improved
 9                  memory with Pb exposure (U.S. EPA. 2006b). Evidence was more consistent for
10                  Pb-induced impairments  in long-term memory, i.e., stored information, which was found
11                  in animals with gestational, gestational/lactational, or gestation/lifetime Pb exposure
12                  producing blood Pb levels 23-32 (ig/dL (Yang etal.. 2003; Jettetal.. 1997; Kuhlmann et
13                  al..  1997). Using the Morris water maze, Jett et al. (1997) found an effect of gestational
14                  plus lactational Pb exposure of female rats (Pb acetate in maternal feed 10 days prior to
15                  mating to PND21) on long-term memory but not working memory, which is memory for
16                  information that changes frequently that is not stored permanently. Kuhlmann et al.
17                  (1997) compared various lifestages of Pb exposure and found impaired learning and long-
18                  term memory using the Morris water maze in Long Evans rats exposed to Pb during
19                  gestation and lactation (via maternal diet) or over a lifetime from gestation through
20                  adulthood each producing peak blood Pb levels of 59 (ig/dL, which are higher than those
21                  relevant to humans. Pb exposure during only the post-weaning period, producing a more
22                  relevant blood Pb level of 26 (ig/dL did not affect memory.

23                  In contrast with Kuhlmann et al.  (1997). recent evidence points to impairments in
24                  memory and learning as assessed with the Morris water maze with postweaning Pb
25                  exposure, albeit with higher blood Pb levels than relevant to humans. Impaired learning,
26                  i.e., slower decrease in time to escape from the Morris water maze across training trials,
27                  was found in weanling Sprague-Dawley male rats exposed to 400 or 800 mg/L  Pb acetate
28                  for 8 weeks in drinking water (Fan etal.. 2010; Fan et al.. 2009a). Further, various dietary
29                  supplements (oral gavage of methionine, taurine, Zn, ascorbic acid or glycine) or
30                  methionine-choline given before or concomitantly with Pb exposure in these weanling
31                  males mitigated the effect of Pb on escape latency to resemble that of control pups at the
32                  end of training (Fan et al.. 2010:  Fan et al.. 2009a). In Fan et al. (2009a). Zn and
33                  methionine given before  Pb exposure attenuated Pb-induced impairments in spatial
34                  memory and learning, whereas glycine, taurine, vitamin C, vitamin Bl, tyrosine had no
35                  effect (blood Pb levels ranged from 7 to  70 (ig/dL depending on nutrient status  and
36                  recovery time). In Fan et al. (2010). Pb-exposed rat pups (blood Pb level 50.2 (ig/dL)
37                  showed deficits in retaining information about the platform location after test day 2 and
38                  3. Concurrent methionine-choline treatment mitigated these impairments in spatial
39                  memory. The action of methionine, a source of sulfur, may be attributable to its function

      November 2012                            5-90                  Draft - Do Not Cite or Quote

-------
 1                  as a chelator and/or free-radical scavenger. Choline is important for cell membranes and
 2                  neurotransmitter synthesis (Zeisel and Blusztajn. 1994). and unrelated to Pb exposure,
 3                  choline supplementation of rats PND16-PND30 was shown to attenuate normal age-
 4                  related declines in spatial memory (Meek et al.. 2007).

 5                  Consistent with Kuhlmann et al. (1997). other recent  studies found impairments in
 6                  learning and memory in animals as assessed using the Morris water maze with
 7                  gestational/lactational Pb exposures, albeit at higher concentrations than those relevant to
 8                  humans. In Li et al. (2009c). rodents were exposed GD1-PND21 to Pb acetate via the
 9                  drinking water of dams (1,000-10,000 ppm) and had corresponding blood Pb levels of
10                  40-100 (ig/dL at PND21. Beginning at weaning, learning and spatial memory were
11                  assessed in pups with a reversal procedure in  the Morris water maze. Pb-exposed pups
12                  with blood Pb levels of 80 and 100 (ig/dL had statistically significant increases in escape
13                  latency, indicating impaired spatial memory and learning (Li et al.. 2009c). The pups in
14                  Li et al. (2009c) were not separated by sex. Cao et al. (2009) found that long-term
15                  postnatal Pb exposure from birth (2,000 ppm  Pb acetate) impaired spatial memory in
16                  male Wistar rats as adults (PND81-90), and these effects were exacerbated by long-term
17                  administration of melatonin (3 mg/kg by gastric gavage, 60 days from weaning).
18                  Mechanistic support for effects on learning and memory was provided in this study by
19                  observations in the hippocampal dentate gyrus that Pb exposure also impaired long-term
20                  potentiation  (LTP), a major cellular mechanism underlying learning and memory.


                    Working Memory - Delayed Spatial Alternation

21                  Working memory also can be measured by testing delayed spatial alternation (DSA). In
22                  DSA, an animal receives rewards by alternating responses between two locations or
23                  levers. This test requires working memory because the correct response changes between
24                  trials, and the animal must determine which response is correct based on memory of its
25                  previous response. Impaired working memory is indicated by increased response errors,
26                  decreased percent of correct responses, and increased response perseverative errors
27                  (i.e., repeatedly pressing the same lever without moving to the other lever when the
28                  reward is moved. Studies detailed in earlier Pb AQCDs showed that Pb-exposed animals
29                  had deficits in working memory as assessed with DSA (Alberand Strupp. 1996; Rice and
30                  Gilbert. 1990b; Rice and Karpinski. 1988; Levin etal. 1987).  Studies in nonhuman
31                  primates showed that there were multiple lifestages and durations of Pb exposure that
32                  induced poorer performance on DSA tasks, including lifetime  from birth or later
33                  postnatal (i.e., post-weaning to time of testing) (Rice  and Gilbert. 1990b: Rice and
34                  Karpinski. 1988; Levin etal.. 1987). Pb-induced impairments in DSA task performance
35                  have been observed less consistently in rats with juvenile only or juvenile to adult
      November 2012                            5-91                  Draft - Do Not Cite or Quote

-------
 1                  exposure. In fact, postweaning Pb exposure was shown to increase accuracy in
 2                  performance on DSA tasks in rodents (Cory-Slechta et al.. 1991).


                    Learning and Memory - Y Maze

 3                  The three-branch radial Y-maze test evaluates learning as the number of days required to
 4                  learn the maze (90% correctly). The Y-maze has a light at the end of one of the branches.
 5                  The branch with the illuminated light is the safe area whereas the other two branches are
 6                  electrified and cause a mild electric shock when entered. The spatial memory test
 7                  assessed by the Y-maze test shares homology with the spatial working memory test
 8                  conducted by Froehlich et al. (2007) in the Rochester cohort, which require children to
 9                  search boxes for a reward and avoid returning to boxes where the reward was previously
10                  found. A recent study using the Y-maze showed impaired learning in Wistar albino rat
11                  offspring exposed to Pb from lactation to adulthood up to 12 weeks of age (300 mg/dL
12                  Pb acetate in dam drinking water and then in offspring drinking water postweaning) (Niu
13                  et al.. 2009). Pb induced statistically significant impairments in learning at 8, 10 and
14                  12 weeks of age but not age 6 weeks. These effects on learning were  found with blood Pb
15                  levels relevant to humans, 17 (ig/dL, as measured at age 6 weeks. Mechanistic support for
16                  Pb-induced learning impairments was provided by concomitant observations of
17                  Pb-induced attenuation in levels of hippocampal glutamate (Section 5.3.11.4). which
18                  mediates signaling pathways involved in LTP.


                    Learning - Schedule-Controlled Behavior Testing

19                  The 2006 Pb AQCD described the effects of Pb exposure on learning in animals as
20                  measured with schedule-controlled behaviors using Fixed Interval (FI) or Fixed Ratio
21                  (FR) operant conditioning reinforcement schedules and indicated differential effects by
22                  Pb exposure concentration, with low-level  (e.g., 11 (ig/dL) and high-level Pb (peak levels
23                  of 115 (ig/dL) exposures increasing and decreasing FI response rate, respectively. This
24                  nonlinear response has been examined further in recent work by the Cory-Slechta
25                  laboratory, much of which also examined the interaction between stress and Pb exposure.
26                  Impaired performance in FI testing with Pb exposure also supports the effects of Pb on
27                  attention-related behavioral problems (Section 5.3.3.1).

28                  Recent evidence indicated that certain learning impairments induced by certain levels of
29                  Pb exposure were modifiable. Female rats were exposed to 300 ppm Pb acetate via dam
30                  drinking water from birth through lactation PND1-PND21 and 30 ppm via their own
31                  drinking water to PND30. Rats subsequently administered succimer PND31-PND52
32                  (twice daily by oral gavage, resulting in blood Pb level 2.8 (ig/dL) performed better on
33                  visual discrimination tasks than did the rats exposed to Pb alone (blood Pb level at

      November 2012                            5-92                  Draft - Do Not Cite or Quote

-------
 1                  PND52: 12.6 (ig/dL) (Stangle et al.. 2007). Brain Pb levels also were lower in the
 2                  Pb+succimer group (196 ng/g) than Pb-only group (1,040 ng/g). Succimer alone in the
 3                  absence of Pb exposure resulted in some cognitive impairment in this study, and rats
 4                  given succimer after a higher Pb concentration (300 ppm, blood Pb level 8.5 (ig/dL) did
 5                  not have better learning ability than rats exposed to 300 ppm Pb alone (blood Pb level:
 6                  31 (ig/dL). Therefore,  succimer administration may not completely alleviate the effects of
 7                  Pb exposure on learning impairments.


                    Learning Ability with Stress

 8                  The combined paradigm of Pb exposure and stress experienced by a laboratory animal
 9                  has been examined by the Cory-Slechta laboratory, which has focused on the common
10                  pathway of altered HPA axis and brain neurotransmitter levels. Depending on the timing
11                  of stress and Pb exposure concentration, greater impairments in learning were found in
12                  animals with dietary Pb exposure that resulted in blood Pb levels relevant to humans
13                  when combined with stress. Evidence additionally indicates that associations of Pb
14                  exposure and stress with learning deficits (multiple schedule of repeated learning and
15                  performance in females) may be related to aberrations in corticosterone and dopamine.
16                  As indicated in Figure 5-3 and Table 5-6. Pb exposure with stress has been shown to
17                  increase corticosterone levels and exacerbate Pb-induced dopamine release and learning
18                  impairments. For example, learning deficits in female rat offspring at age 2 months were
19                  enhanced  following lifetime Pb exposure  combined with prenatal stress, i.e., maternal
20                  restraint (Cory-Slechta et al.. 2010). This  exposure paradigm involved exposure of dams
21                  to Pb acetate from 2 months prior to mating through  lactation and exposure of their pups
22                  from a mixed sex litter via drinking water Pb (50 ppm) through the remainder of their
23                  lifetime (2 months). The peak blood Pb levels of pups (age 5-6 days) ranged from 10 to
24                  13 (ig/dL. Learning impairments were found in repeated learning assessments but not
25                  performance assessments. Pb/stress was found to increase the total number of responses
26                  required to learn a sequence. Pb/stress exposure also affected dopamine from the frontal
27                  cortex and dopamine turnover in the nucleus accumbens, which are processes underlying
28                  cognition. Also, Pb-exposed offspring with and without maternal stress  exposure had
29                  statistically significant decreases in hippocampal nerve growth factor versus controls.

30                  Another study of lifetime Pb exposure (50 or 150 ppm  Pb acetate in drinking water to
31                  dams from 2 weeks before pregnancy though lactation  and to offspring thereafter)
32                  indicated a potentiation of effects on learning with Pb and stress co-exposure, with stress
33                  given prenatally via dams or postnatally to offspring. Lifetime  50 ppm Pb exposure plus
34                  prenatal or postnatal stress resulting in blood Pb levels 11-16 (ig/dL decreased the post-
35                  reinforcement pause (PRP) period in female offspring when examined starting at age
36                  2 months (Rossi-George et al.. 2011) (Table 5-7. rightmost column). Animals with

      November 2012                            5-93                   Draft - Do Not Cite or Quote

-------
 1                   150 ppm Pb exposure and blood Pb levels 25-33 (ig/dL had decreased PRP only with
 2                   prenatal stress. Within the FI schedule, the PRP represents timing capacity or proper
 3                   temporal discrimination and refers to the period during which the animal waits or pauses
 4                   before depressing the lever for a reward. In this case, decreased pause or PRP interval in
 5                   Pb plus stress-exposed animals indicates that they started responding earlier than did
 6                   controls. These results also point to an effect of Pb on increasing impulsivity
 7                   (Section 5.3.3.1). Separately, the overall FI response rate, which also indicates
 8                   impulsivity (i.e., rate of not withholding responses), was significantly increased with
 9                   50 ppm Pb exposure alone and with co-exposure to maternal or offspring stress. At
10                   150 ppm, Pb increased FI response rate only with co-exposure to stress (maternal or
11                   offspring). Biochemical analysis revealed alterations in frontal cortex norepinephrine,
12                   reductions in dopamine homeostasis in the nucleus accumbens, and enhancement of the
13                   striatal monoamine system as possible mechanistic contributions to Pb-induced
14                   impairments in learning.
      November 2012                             5-94                  Draft - Do Not Cite or Quote

-------
Table 5-7     Summary of effects of maternal and lifetime Pb exposure on
                 Fl performance observed by Cory-Slechta and  colleagues.
Maternal Pbb
Pb (ppm)
Overall rate3
PRPa
Lifetime Pb"
Overall rate3
PRP"
0 ppm:
0-PS
0-OS
No Significant Effect
No Significant Effect
No Significant Effect
*i -23%
No Significant Effect
No Significant Effect
No Significant Effect
No Significant Effect
50 ppm:
50-NS
50-PS
50-OS
No Significant Effect
No Significant Effect
*t 64.9%
No Significant Effect
No Significant Effect
No Significant Effect
*t 95%
*t 79.2%
*t 74.7%
No Significant Effect
*i -42%
* |-39.3%
150 ppm:
1 50-NS
1 50-PS
1 50-OS
*t 42.4%
No Significant Effect
*t 59.2%
*| -30.3%
*i -25.7%
No Significant Effect
No Significant Effect
*t 90.7%
*t 78.5%
No Significant Effect
*i -44.7%
No Significant Effect
"Based on calculation of group mean values across session block post-stress challenge for both maternal and lifetime Pb exposure
studies. All calculations represent percent of 0-NS control values; f represents increase; J, represents decrease.
bData from Virgolini et al. (2005). 'Denotes significant effect versus 0 ppm control (p <0.05).
°Data from current study, Rossi-George et al. (2011)
'Denotes significant effect vs. 0 ppm control (p <0.05).

Source: Reprinted with permission of Elsevier Science, Table 1 of Rossi-George et al. (2011).

Notes:
PRP = Post-reinforcement pause; PS = Prenatal (maternal stress); OS= Offspring stress.
Overall results demonstrated that lifetime Pb exposure (right column) with or without prenatal stress induced learning deficits in
female rats as demonstrated by an increase in overall rate and decreased PRP. Mechanistically, these authors proposed that
associations of Pb and stress with learning deficits may be  related to aberrations in corticosterone and dopamine. Prenatal Pb
exposure alone (left column) induced similar responses during testing at age 2 months.
November 2012
5-95
Draft - Do Not Cite or Quote

-------
 1                   A separate investigation from the same laboratory similarly indicated a potentiation of
 2                   effects with Pb and stress co-exposures but with developmental Pb exposure from
 3                   two months prior to mating through lactation (50 or 150 ppm Pb acetate in drinking
 4                   water) (Virgolini et al.. 2008a). Dams were subject to restraint stress at GD16-GD17.
 5                   Prenatal stress or Pb exposure alone did not affect FI performance in offspring.
 6                   Compared with controls, marked increases in response rates on FI performance were
 7                   found in the 50 ppm Pb plus prenatal stress female offspring at age 2-3 months, whose
 8                   mean blood Pb level was 19 (ig/dL at weaning. Using the same Pb exposure protocol,
 9                   Virgolini et al.  (2008b) expanded evidence for Pb-stress interactions through the
10                   examination of the effects of additional adult intermittent stress (cold, novelty or
11                   restraint) on FI performance, corticosterone, and dopamine. Compared with females with
12                   stress but no Pb, female offspring with adult intermittent restraint and cold stress and
13                   higher dose Pb (150 ppm) had statistically significant increases in FI response rate and
14                   decreased PRP, i.e., increased impulsivity (Figure 5-4, A panel). Male offspring showed
15                   decreased FI response rates due to decreased run rate with restraint stress at the lower Pb
16                   dose (50 ppm) (Figure 5-4. B panel). At the higher dose of Pb, males  showed increased
17                   FI response rates and increased run rates with  cold stress.

18                   Pb exposure over various developmental windows in rodents has been shown to affect the
19                   HPA axis, as measured by the levels of corticosterone, the major glucocorticoid involved
20                   in stress responses in rodents. Thus, modulation of corticosterone may provide a
21                   mechanistic explanation for learning deficits (FI testing in females) found with Pb and
22                   stress co-exposure in rodents. As examined by the Cory-Slechta laboratory, exposure to
23                   Pb induced differential changes in corticosterone levels in each sex, depending on the age
24                   of the animal and the timing of exposure, developmental (gestational  and lactational),
25                   post-weaning, or lifetime (Rossi-George et al.. 2011; Cory-Slechta et al.. 2010; Virgolini
26                   et al.. 2008aV
      November 2012                             5-96                  Draft - Do Not Cite or Quote

-------
 A.
        B.
                                       Novelty
                                                                               Novelty

         C  60  164
                      0  SO  150
                                   o  50  1SQ
         0  50  160
      Pastreinforcement Pause Time
                      0  SO  150
                                   0  60  1SO
                          1ED      *    120-i



                            III     III
                            0  50 150       0  SO 150
              iJtete
                          150      +    120-j


              iJin     Hi
                0  50 150      0  50 150       0  50 150
                                                  Postreinforcemenl Pause Time
      SOO -I           MO -|           12S -i                150 -,           180-1          150 -,   _!_



        I     L     III           ll     III     ill
                                                     0  60  ISO       0  40  150      0 SO 150
         °   M  ,!"       °  "  "°       °  »  1M            Pb(ppm)]        [Pbfepn,)]        [Pblppm,]
         [Pb(ppm)]         [Pb(ppm)]        [Pb (ppm)]
Note: * denotes significantly different from 0 ppm Pb. ~denot.es p = 0.07. + denotes significantly different from the 50 ppm Pb group.
Each column presents results for a particular stressor (restraint, cold, novelty). In females (panel A), gestational/lactational 50 and
150 ppm Pb exposure increased overall rate and decreased post-reinforcement-pause time over that of cold or restraint stress given
in adulthood. In males (panel B), gestational/lactational Pb exposure did not alter the effects of the stressors.


Source: Reprinted with permission of Elsevier Science, Virgolini et al. (2008b).


Figure 5-4      Changes in Fixed Interval performance in (A) female and (B) male
                offspring with gestational/lactational Pb exposure plus various

                stressors given in adulthood.
November 2012
5-97
Draft - Do Not Cite or Quote

-------
 1                   Because animals that are used for FI testing are regularly handled by laboratory personnel
 2                   and often participate in other tests of cognition, their baseline level of stress may be
 3                   skewed from that of a laboratory animal that constantly remains in a cage without daily
 4                   handling. Because effects on the HPA axis are of interest to Pb researchers, the baseline
 5                   corticosterone levels of animals that have participated in behavior testing (FI) and those
 6                   who have not (NFI) have been compared after gestational/lactational Pb exposure.
 7                   Virgolini et al. (2008b) found that baseline corticosterone levels were significantly
 8                   different between FI and NFI animals. Also, the effect of combined
 9                   gestational/lactational Pb exposure plus maternal stress on corticosterone was compared
10                   in FI and NFI animals. At the baseline age of 4-5 months, Pb exposure with or without
11                   stress did not induce differences in corticosterone levels in FI females but did in males
12                   (Virgolini et al., 2008b). In the FI males, 50 ppm Pb exposure decreased corticosterone
13                   versus control (no Pb exposure), and 150 ppm Pb exposure elevated corticosterone versus
14                   control. In male NFI animals, a U shaped concentration-response was found, with 50 ppm
15                   Pb exposure reducing corticosterone over than in the controls or with 150 ppm Pb
16                   exposure. In the NFI males, stress did not affect corticosterone levels or interact with the
17                   effect of Pb exposure. NFI females exposed to 150 ppm Pb had significantly elevated
18                   corticosterone versus control (no Pb exposure). These data demonstrate that behaviorally
19                   tested animals have altered HPA axis and altered responses to Pb  exposure versus
20                   animals that are housed under conditions without daily handling by caregivers.

21                   Lifetime Pb exposure beginning in gestation (150 ppm drinking water of dams from
22                   2 months prior to mating through lactation, then continuing in offspring water) induced
23                   increases in basal (age 2 months, before behavioral testing) corticosterone only in female
24                   offspring but not final (age 10 months,  after testing) corticosterone in either female or
25                   male offspring (Rossi-George et al., 2011) (Figure 5-5). The Pb-related increase in
26                   corticosterone was found in animals with blood Pb levels of 19 and 31 (ig/dL measured at
27                   PND21.  Pb-stress  interactions were observed at age 2 months but not 10 months (final,
28                   after behavioral testing). At age 2 months, Pb plus stress attenuated the Pb-induced
29                   elevations in corticosterone to baseline levels (Figure 5-5). By 10 months of age, these
30                   offspring had lower corticosterone concentrations versus control animals. In males,
31                   corticosterone levels were not affected  significantly by Pb and/or  stress at 2 (basal) or
32                   10 (final) months of age (Figure 5-5) (Rossi-George et al.. 2011).
      November 2012                             5-98                  Draft - Do Not Cite or Quote

-------
                                 Female
              900n
                         Basal
                                            Final
                                       PbxS
              600-
           O)
              300-
                                                           900
                                                                                        Final
                                                           600
                                                           300
                    0     50    150        0    50   150
                           Pb Exposure (ppm)
                     so    iso       o    so   iso
                      Pb Exposure (ppm)
      Note: Corticosterone levels are noted on the y-axis. 'denotes significantly different from NS control (Black bars); # denotes
      significantly different from corresponding Pb-NS value; + differs from 50-NS. (0, 50, 150 ppm) and/or stress (PS [dam stress, white
      bar with black dots] or OS [PS followed by offspring stress, black bar with white spots]). Basal measurements were taken at
      2 months of age, prior to the initiation of behavioral testing and final measurements were taken at 10 months, after behavioral
      testing.
      Source: Rossi-George et al. (2011)

      Figure 5-5       Mean basal and final corticosterone levels of female and male
                         offspring exposed to lifetime Pb.

 1                   Pb-stress effects on corticosterone have not been consistent. In another study, gestational
 2                   plus lactational Pb (50, 100 ppm) did not affect baseline corticosterone levels in females
 3                   and there was no interaction with stress (Virgolini et al., 2008a). In males, stress
 4                   increased baseline corticosterone in the 150 ppm Pb group. In animals given intermittent
 5                   stress as adults and not behaviorally tested, Pb decreased corticosterone levels in females
 6                   but not males, which may explain the  observations of Pb plus stress increases in
 7                   decreases in FI response rates in  females (Virgolini et al., 2008b). Post-weaning exposure
 8                   of male rodents to Pb (PND21-age 5 months) produced a U-shaped concentration-
 9                   response for corticosterone prior to FI testing, with a significant decrement in basal
10                   corticosterone levels in the 50 ppm exposure group versus control and the 150 ppm Pb
11                   group (Virgolini et al., 2005). Lifetime Pb exposure and prenatal stress both reduced
12                   corticosterone in animals behaviorally tested at age 4 months but not in animals at age
13                   11 months not behaviorally tested (Cory-Slechta et al., 2010).

14                   Another study examined the effects of Pb acetate on the HPA axis but examined the
15                   interaction with an outside stress administered using control vehicle injections (Rossi-
16                   George et al., 2009). The corticosterone response to this vehicle injection stress was
      November 2012
5-99
Draft - Do Not Cite or Quote

-------
 1                  prolonged in a nonlinear concentration-dependent manner in both sexes with the most
 2                  profound effects observed at the lower 50 ppm Pb dose. Maternal stress also prolonged
 3                  the corticosterone stress response to vehicle injection and enhanced the Pb effect in
 4                  males. To test the negative feedback of the HPA axis, exogenous dexamethasone (DEX)
 5                  was administered to suppress endogenous corticosterone. The DEX test revealed HPA
 6                  axis hypofunction. Specifically, Pb and Pb plus maternal stress initially reduced the
 7                  ability of DEX to suppress corticosterone. With time, the effect of DEX in males induced
 8                  prolonged corticosterone suppression or failure to return to baseline as was observed in
 9                  control animals. Rossi-George et al. (2011) additionally found that Pb and/or maternal
10                  stress significantly impacted the negative feedback by increasing nuclear glucocorticoid
11                  receptor levels. In summary, prenatal Pb exposure induced HPA negative feedback
12                  hypofunction in both male and female offspring with an inverse U concentration-
13                  response relationship. This negative feedback loop was impacted more at the lower Pb
14                  dose (50 ppm) versus the higher dose (150 ppm) (Rossi-George et al., 2011).

15                  To summarize the results for Pb-stress interactions in animals (Table 5-6). lifetime Pb
16                  exposure when combined with stress was found to exacerbate learning impairments
17                  compared with Pb exposure or stress alone, although not across all tests or Pb doses. The
18                  interaction between Pb and stress may be mediated via effects on corticosterone and
19                  dopamine. Lifetime Pb exposure was found to increase basal (at age 2 months)
20                  corticosterone levels in females, and co-exposure to stress attenuated the response (Rossi-
21                  George etal.. 2011; 2009). Males given lifetime Pb exposure had no statistically
22                  significant corticosterone response to Pb exposure; whereas males with
23                  gestational/lactational Pb exposure had statistically significant decreases in corticosterone
24                  at 5 months of age in the 50 ppm exposure group only (but not in 150 ppm Pb exposure
25                  group). On the other hand, females had concentration-dependent corticosterone responses
26                  to Pb exposure in both exposure models (lifetime Pb exposure and gestational/lactational
27                  Pb exposure). Maternal stress alone also led to HPA axis negative feedback hypofunction
28                  in offspring. Pb exposure plus maternal stress enhanced negative feedback in males and
29                  attenuated this effect in females. Pb exposure with or without maternal stress prolonged
30                  the effect of DEX-dependent corticosterone suppression in males. These data together
31                  show that HPA axis alterations could provide a link for interactions found between Pb
32                  and stress in impairing learning.
      November 2012                             5-100                  Draft - Do Not Cite or Quote

-------
      5.3.2.4  Executive Function in Children

 1                   Epidemiologic evidence presented in the 2006 Pb AQCD (U.S. EPA. 2006b) indicated
 2                   associations between higher blood or tooth Pb levels and poorer performance on tests of
 3                   executive function in children and young adults. Associations were found with indices of
 4                   executive function such as strategic planning, organized search, flexibility of thought and
 5                   action to a change in situation, and control of impulses (described in greater detail in
 6                   Section 5.3.3.1). Prospective analyses in two Boston area and the Rochester cohorts
 7                   provided key evidence with  examination of blood Pb levels preceding executive function
 8                   testing and adjustment for several potential confounding factors (Canfield et al.. 2004;
 9                   2003b: Bellinger et al.. 1994a: Stiles and Bellinger. 1993). Further, recruitment of
10                   participants before or at birth, moderate to high follow-up participation, and in most cases
11                   follow-up not biased to higher blood Pb levels and lower cognitive function increase
12                   confidence that the  observed associations are not due to selection bias (Table 5-8).
13                   Among the few recent cross-sectional studies, most found concurrent blood Pb-associated
14                   decrements in executive function, including an analysis of the Rochester cohort
15                   (Froehlich et al.. 2007). Evidence from other recent studies had weaker implications due
16                   to the limited consideration of potential confounding (Nelson and Espy. 2009; Vega-
17                   Dienstmaier et al.. 2006). Several of the prospective and cross-sectional studies
18                   performed multiple tests of cognitive function, including executive function. However,
19                   except for Stiles and Bellinger (1993). a consistent pattern of association was found
20                   across the various tests performed. Thus, the evidence does not appear to be biased by
21                   associations found by chance alone.

22                   Studies in children found Pb-associated decreases in executive function using various
23                   tests including the Intra-Extra Dimensional Set Shift, WCST, and Stoop test (Table 5-8).
24                   As discussed below, studies  in animals also demonstrated Pb-induced decrements in
25                   executive function,  including rule learning and reversal, which also were associated with
26                   blood Pb levels in children. This coherence between findings in animals and humans for
27                   analogous domains further supports a relationship between  Pb exposure and decrements
28                   in cognitive function. Additional biological plausibility for Pb-associated decrements in
29                   executive function is provided by toxicological evidence for Pb-induced changes in the
30                   availability of dopamine (Section 5.3.11). a neurotransmitter that affects executive
31                   functions mediated  by the prefrontal cortex. Recent work shows that executive function
32                   in animals is affected by N-Methyl-D-aspartic acid or N-Methyl-D-aspartate (NMDA)
33                   receptors and dopamine-like receptors (Herold. 2010). which are two well-characterized
34                   targets of Pb.
      November 2012                             5-101                  Draft - Do Not Cite or Quote

-------
 1                   In the Boston cohort, children tested at a relatively older age (10 years), when testing is
 2                   more reliable, a 1 (ig/dL increase in age 5 year blood Pb level was associated with 0.05
 3                   (95% CI: 0.01, 0.09) more perseverative errors on the WCST (errors in sorting cards
 4                   according to a change in rule) (Stiles and Bellinger. 1993). In this cohort, results were
 5                   inconsistent across the various cognitive tests. However, associations were more
 6                   consistent for executive functions assessed by the WCST. In another cohort ages
 7                   19-20 years from towns around Boston, higher tooth Pb levels (from ages 5-8 years) were
 8                   associated with more errors on the WCST in sorting by the set rules and poorer
 9                   performance on the Stroop Color and Color-word tests, which test the ability of subjects
10                   to shift focus to another dimension of stimulus that defines correct responding (Bellinger
11                   etal.. 1994a).
      November 2012                             5-102                  Draft - Do Not Cite or Quote

-------
Table 5-8   Associations between blood or tooth Pb levels and performance of
            tests of executive function in children and young adults.



Study
Stiles
and
Bellinger
I A QQO\
(1 b)b)O)




Bellinger
etal.
(1994a)






Canfield
etal.
(2004)










Froehlich
etal.
(2007)






Canfield
etal.
(2003b)







Study Population and Methodological
Details
(Presented first for prospective studies then for
cross-sectional studies. Within each category,
results are presented in order of strength of study
design)
148 children followed from birth (1979-1981) to age
1 0 yr, Boston, MA area
Prospective. Recruitment at birth hospital.
Moderate follow-up participation, participants had
higher SES and HOME score. Linear regression
model adjusted for HOME score, family stress,
race, marital status (5 yr blood Pb), HOME score,
family stress, maternal age and race, birth weight,
# daycare situations to age 57 mo (concurrent).
79 young adults, born 1970, followed from 1st grade
to age 19-20 yr, Boston, MA area
Prospective. Moderate follow-up participation.
Participation from higher SES, females, higher initial
IQ but no affect on association with tooth Pb level.
Regression model adjusted for parental IQ, sex,
SES, current drug, alcohol and illicit drug use,
maternal education and age, birth order. Also
considered potential confounding by other
unspecified factors.
174 children born 1994-1995 followed from age
6 mo to 5 yr, Rochester, NY
Prospective. Recruitment from study of dust
control. 73% nonwhite. High follow-up participation,
no selective attrition. Linear regression model
adjusted for NICU admission, HOME, prenatal
maternal smoking, household income, child sex,
average crowding in home (I ED); maternal IQ,
HOME, prenatal smoking, household income, child
sex (Stockings of Cambridge). Also considered
potential confounding by breastfeeding duration,
maternal ethnicity, first prenatal visit, spatial working
memory problem, age at testing, birth weight,
marital status, maternal education and spatial span
length
174 children born 1994-1995 followed from age
6 mo to age 5 yr, Rochester, NY
Cross-sectional. Same cohort as above. High
follow-up participation, no selective attrition. Linear
regression model adjusted for NICU, sex (rule
learning). Also considered potential confounding by
income, HOME score, maternal IQ and education,
prenatal smoking exposure, race, and age,
transferrin saturation.
150 children born 1994-1995 followed from age
6 mo to age 4.5 yr, Rochester, NY
Cross-sectional. Same cohort as above. High
follow-up participation, no comparison of
nonparticipants. Linear mixed effects model
adjusted for: child sex, maternal IQ, education, and
prenatal smoking, household income, marital status,
HOME score. Also considered potential
confounding by age, birth order, attention rating,
race, gestational age, color/shape knowledge, child
IQ

Blood Pb
Levels
(ug/dL)
Earlier
childhood
(age 5 yr)
Concurrent
mean <8
Exact levels
NR mean
reported to
be<8
Deciduous
tooth (age
5-8 yr)
Mean (SD):
13.7
(11.1)ug/g
10th-90th:
4.3-26.4


Lifetime (to
age 5 yr) avg
Mean (SD):
7.2 (3.6)
10th-90th:
3.5-11.8








Concurrent
Mean (SD):
6.1 (4.9)
10th-90th:
1.9-11 .7




Concurrent
Mean: 6.5
10th-90th:
data not
available







Executive
Function Test
Perseverative errors,
Age 5 yr blood Pb
Concurrent blood Pb




Mean time to
complete color-word
test, Stroop test
Perseverative
responses, WCST
Ages 1 9-20 yr



Stages Completed -
Intra-Extra
Dimensional (IED)
Ortt OKift
bet bhitt
Stockings of
Cambridge problems
solved in minimum
moves

CANTAB
Age 5 yr




Stages Completed -
Intra-Extra
Dimensional Set
C Kift
onlTt
CANTAB
Aae 5 vr
nyc \j yi


Inhibit Efficiency
(# correct-
incorrect)/phase
duration
Shape School Task
Repeated measures
at ages 4 and 4.5 yr





Effect Estimate
/nco/ /^i\3
(95% Cl)
WCST, Age 10yr
-0.05 (-0.09, -0.01 )b
-0.05 (-0.11,0.01)"




-0.68 (0.28, 1 .08)b


-0.37(0.10,0.64)"





-0.11 (-0.21, -0.01)



-0.08 (-0.1 7, 0.001)








-0.06 (-0.1 2,0)°








-0.01 9 (-0.03, -0.007)









November 2012
5-103
Draft - Do Not Cite or Quote

-------
Study
Surkan
etal.
(2007)
Study Population and Methodological
Details
(Presented first for prospective studies then for
cross-sectional studies. Within each category,
results are presented in order of strength of study
design)
389 children ages 6-10 yr, Boston, MA ,Farmington,
ME
Cross-sectional. Recruitment from trial of
Blood Pb
Levels Executive
(ug/dL) Function Test
Concurrent
Mean (SD):
2.2(1.6)
Effect Estimate
(95% Cl)a
Blood Pb level 5-10
vs. 1-2 ug/dLd
                amalgam fillings. High participation rate. Higher
                participation of white children in Maine. Analysis of
                covariance adjusted forcaregiver IQ, child age,
                SES, race, and birth weight, Also considered
                potential confounding by caregiver education and
                marital status, parenting stress, and maternal
                utilization of prenatal or annual health care but not
                parental caregiving quality.
                                                    rciacvciauuii ciiuia,    a 1 a / 1/I c -3 7\
                                                    WCST              -9.19 (-14.6,-3.7)
                                                    Stroop color-word
                                                    interference
                0.75 (-1.6, 3.1)
Chiodo
etal.
(2004)
Choet
al.
(2010)
246 children, age 7.5 yr, Detroit, Ml area
Cross-sectional. Recruitment at prenatal clinic. All
African American High prevalence of prenatal
alcohol exposure. High participation rate. Log linear
regression model adjusted for SES, family
functioning, # children <1 8 yr, caregiver vocabulary,
prenatal alcohol use, caregiver education, child sex.
Also considered potential confounding by HOME,
maternal prenatal marijuana, smoking, or cocaine
use, crowding, child life stress, caregiver age, life
stress, and psychology, conflict tactics, disruption in
caregiver, parity, child age.
667 children ages 8-1 1 yr, born 1 997-2000, 5
Korean cities
Cross-sectional. School-based recruitment,
moderate participation rate. Log linear regression
model adjusted for age, sex, parental education,
maternal IQ, child IQ, birth weight, urinary cotinine.
Did not consider potential confounding by parental
caregiving quality.
Concurrent
Mean (SD):
5.4 (3.3)
Interval
analyzed:
2.3-9.5= 10t
h-90th
percentiles
Concurrent
Mean (SD):
1.9(0.67)
Interval
analyzed:
1.2-2.8=101
h-90th
percentiles
Perseverative errors,
WCST
Age 7.5 yr
Color-word score
Stroop test
Ages 8-1 1 yr
-0.49, p >0.05e
0 (-0.09, 0.08)
      Note: Results are presented first for prospective studies then for cross-sectional studies. Within each category, results are
      presented in order to strength of study design.
      WCST = Wisconsin Card Sorting Test, CANTAB = Cambridge Neuropsychological Testing Automated Battery.
      "Effect estimates are standardized to a 1 ug/dL increase in blood Pb level or 1 ug/g in tooth Pb level in the 10th-90th percentile
      interval.
      bThe direction of the effect estimate was changed such that a negative estimate represents poorer performance.
      °95% Cl: was constructed using a standard error that was estimated from the reported p-value.
      dEffect estimates compare test performance of children in  higher blood Pb groups to children in lowest blood Pb group.
      Sufficient data were not provided to calculate 95% Cl.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
Results from the Rochester cohort at ages 4 and 5 years indicated associations of
concurrent and lifetime average blood Pb level with lower inhibition efficiency in the
Shape School task (i.e., giving correct responses and withholding incorrect responses)
(Canfield et al.. 2003b). poorer problem solving on a spatial planning task (Canfield et
al.. 2004). and poorer rule learning and reversal (Froehlich et al.. 2007). Associations
with Shape School tasks were attenuated and lost precision with adjustment for attention
ratings, color/shape knowledge, and child IQ. These results suggest that the effect of Pb
exposure on executive function may be mediated through effects on knowledge.
Froehlich et al. (2007) found a larger concurrent blood Pb-associated decrement in a rule
learning and reversal task in the Intra-Extra Dimensional Set Shift in children with the
DRD4 exon III 7-repeat microsatellite (assessed using a blood Pb-DRD4-7 interaction
term, p = 0.042). While this evidence for effect modification is based on a smaller subset
      November 2012
                               5-104
Draft - Do Not Cite or Quote

-------
 1                   of subjects (n = 34/174), they add support for Pb-associated decreases in executive
 2                   function because dopamine is a key neurotransmitter that regulates executive function,
 3                   the DRD4-7 variant is associated with reduced dopamine-induced signaling in
 4                   downstream pathways (e.g., cyclic AMP), and the DRD4-7 variant was associated with
 5                   poorer executive function in this cohort. The association of concurrent blood Pb level
 6                   with impaired rule learning and reversal also was greater in boys, who had lower mean
 7                   scores than girls.

 8                   In addition to assessment of earlier or cumulative Pb biomarkers, a strength of the
 9                   prospective studies was the consideration for numerous potential confounding factors.
10                   The potential confounding factors varied among studies based on their association with
11                   executive function and/or influence on the Pb-executive function relationship. Some
12                   prospective studies demonstrated Pb-associated decrements in executive function with
13                   adjustment for SES, maternal IQ, and HOME score (Canfield et al., 2004; Stiles and
14                   Bellinger. 1993). Others considered and excluded potential confounding by HOME score,
15                   parental smoking, maternal education, or birth outcomes (Table 5-8).

16                   The prospective studies indicated blood Pb-associated decrements in executive function
17                   in populations with a mean lifetime average blood Pb level of 7.2 (ig/dL and a mean
18                   concurrent blood Pb level of 6.5 (ig/dL. Associations in populations  with lower mean
19                   blood Pb levels (2-5 (ig/dL), as assessed in cross-sectional studies with concurrent blood
20                   Pb level, were not as clearly demonstrated. While these studies adjusted for SES and
21                   parental cognitive function, most did not examine potential confounding by parental
22                   caregiving quality, i.e., HOME score. Among children in New England ages 6-10 years
23                   with mean concurrent blood Pb level 2.2 (SD: 1.6) (ig/dL, Pb-associated decrements in
24                   executive function assessed by WCST, Trail-making, and Verbal cancellation tests were
25                   observed primarily in the group with blood Pb levels 5-10 (ig/dL. In this study, higher
26                   blood Pb level was not associated with poorer color-word score in the Stroop test. Cho
27                   et al.  (2010) did not find a Pb-associated lower color-word score among children in five
28                   Korean cities ages 8-11 years with mean concurrent blood Pb level 1.9 (SD: 0.7) (ig/dL.
29                   Other studies found associations in children with mean blood Pb levels 2-5 (ig/dL but had
30                   limited implications because of lack of representativeness of a population with high
31                   prevalence of prenatal alcohol exposure (Chiodo et al.. 2004) or lack of consideration for
32                   potential confounding (Nelson and Espy. 2009).

33                   The associations observed in children between blood or tooth Pb levels and poorer
34                   executive function as assessed by the rule learning and reversal components of the Intra-
35                   Extra Dimensional Set Shift, Stroop Test, and WCST are supported by observations in
36                   animals of Pb-induced  impairment in analogous measures of cognitive flexibility, tested
37                   with discrimination reversal learning and concurrent random interval (RI-RI) scheduling.
      November 2012                             5-105                  Draft - Do Not Cite or Quote

-------
 1                  In animals, some of these observations were made with Pb exposures relevant to humans.
 2                  These tests of cognitive flexibility measure the ability of humans and animals to adjust
 3                  their responses in reaction to changes in reinforcement. Poorer performance in both
 4                  children and animals is indicated by increased response errors, decreased percent correct
 5                  responses, and perseverative responding (e.g., persistence in making a previously-
 6                  rewarded response after a new shift in reinforcement). As reviewed in the
 7                  2006 Pb AQCD (U.S. EPA. 2006b). several lines of evidence indicated Pb-induced
 8                  impairments in executive function in animals. Lifetime dietary Pb exposures beginning at
 9                  birth or after weaning that produced peak blood Pb levels of 19-36 (ig/dL were found to
10                  induce poorer performance on discrimination reversal learning tasks in monkeys ages
11                  5-10 years (Rice and Gilbert. 1990b: Gilbert and Rice. 1987). Recent work has shown
12                  that discrimination reversal learning involves NMDA receptors and dopamine-like
13                  receptors (Herold. 2010). which are two well-characterized targets of Pb. Gestational Pb
14                  exposure (blood Pb of dams >40 (ig/dL) was found to impair cognitive flexibility in
15                  squirrel monkeys, ages 5-6 years, as indicated by a slower shift or lack of a shift to the
16                  lever reinforced more frequently under RI-RI scheduling (Newland et al., 1994). Rats
17                  also showed Pb-induced impairments on discrimination reversal tasks, but the authors
18                  attributed the changes to learning-related problems instead of impaired executive function
19                  (Garavan et al.. 2000: Hilson and Strupp. 1997V
      5.3.2.5 Academic Performance and Achievement in Children

20                  As described in preceding sections, a large body of evidence demonstrates Pb-associated
21                  decrements in FSIQ, with more variable findings for performance on tests of learning and
22                  memory. Lower FSIQ and learning are linked with poorer academic performance and
23                  achievement, which may have important implications for success later in life. Further,
24                  academic performance may better assess the knowledge of an individual in the actual
25                  subject areas studied, whereas aptitude tests are used to predict future performance. In
26                  addition to FSIQ, the 2006 Pb AQCD described associations between blood Pb levels in
27                  children ages 5-18 years and poorer performance on tests of math and reading skills,
28                  vocabulary, and spelling, objective measures  such as high school completion and class
29                  rank, and teacher ratings of academic functioning. Associations continued to be reported
30                  in recent studies, including prospective  studies examining performance on academic
31                  achievement tests and an additional analysis of adolescents participating in NHANES
32                  (Table 5-9).  Findings from other recent studies had weaker implications because of the
33                  lack of representativeness of populations with high prevalence of prenatal alcohol or drug
34                  exposure (Min et al.. 2009; Chiodo et al.. 2007). Multiple testing was common in studies;
35                  however, the consistent pattern of blood Pb-associated decrements in academic
      November 2012                            5-106                  Draft - Do Not Cite or Quote

-------
 1                  performance across the various tests conducted increases confidence that the evidence is
 2                  not unduly biased by a higher probability of associations found by chance alone.

 3                  Key evidence supporting associations between blood Pb level and performance on tests
 4                  of quantitative, reading, vocabulary, and spelling skills was provided by previous
 5                  prospective studies in the Boston and Cincinnati cohorts (Ris et al.. 2004; Bellinger etal..
 6                  1991). Associations with earlier childhood blood Pb levels better characterized the
 7                  temporal sequence between Pb exposure and poorer academic performance. Evidence
 8                  from prospective studies did not strongly indicate selection bias with recruitment of
 9                  participants before or at birth, moderate to high follow-up participation, and in most cases
10                  follow-up not biased to higher blood Pb levels and lower cognitive function (Table 5-9).
11                  An additional strength of the prospective  studies was the consideration of several
12                  potential confounding factors (Table 5-9). including birth outcomes, exposure to smoking
13                  and drugs, and nutritional status and the adjustment for SES, parental education and IQ,
14                  and HOME score. Evidence for associations between blood Pb levels and reading, math,
15                  and vocabulary skills provides coherence for the associations observed between blood Pb
16                  levels and FSIQ, which includes components of quantitative reasoning and language
17                  ability.
      November 2012                             5-107                  Draft - Do Not Cite or Quote

-------
Table 5-9     Associations between  blood or tooth Pb levels and measures of
                 academic performance and achievement in children and young
                 adults.
Study
Study Population and
Methodological Details
(Presented first for prospective studies then
for cross-sectional studies. Within each
category, results are presented in order to
strength of study design)
Blood or
Tooth Pb
Levels
(ug/dL)
Indicator of
Academic
Performance/
Achievement
Effect Estimate
(95% Cl)a
Studies of neuropsychological testing of academic performance
Bellinger et al.
(1991)
170 children followed from birth (1979-1981)
to age 5 yr, Boston, MA area
Prospective. Recruitment at birth hospital.
Moderate follow-up participation. More
participants were white, had higher age 2 yr
HOME score, and higher postnatal blood Pb
levels. Log linear regression model adjusted
for SES, maternal IQ and marital status,
preschool attendance, HOME,  out of home
care, residence changes, medication use in
previous 12 mo, number of adults in home,
child sex, race, birth weight, birth order.
Earlier
childhood
(age 2 yr)
Mean (SD):
7.0 (6.6)
Interval
analyzed:
1.8 (1 Oth
percentile)-
10
                                                                       Verbal
-0.09 (-0.51, 0.34)
                                                                       Quantitative
                                                                       McCarthy Scale of
                                                                       Children's Abilities
                                                                       Age 5 yr
-0.30 (-0.65, 0.05)
Dietrich et al.      258 children followed prenatally (1979-1985)
(1991)            to age 4 yr, Cincinnati, OH
                 Prospective. Recruitment at prenatal clinic.
                 High follow-up participation, no selective
                 attrition. Mostly African American. Linear
                 regression adjusted for SES, birth weight,
                 maternal IQ, prenatal marijuana use, HOME,
                 child race, preschool attendance. Also
                 considered potential confounding by birth
                 outcomes, maternal age, prenatal smoking,
                 alcohol use and narcotics use, # previous
                 abortions, stillbirths, gravidity, parity,
                 caregiver education, public assistance, child
                 age, sex, health, Fe status
                                         Earlier
                                         childhood
                                         (Age 2 yr):
                                         NR
             Achievement score,
             KABC
0.06, p >0.05b
                                         Concurrent:
                                         NR
                                 0.01,p>0.05b
                                         Lifetime
                                         avg: NR
                                 0.07, p >0.05b
Ris et al. (2004)    195 children followed prenatally (1979-1985)
                 to age 15-17 yr, Cincinnati, OH
                 Prospective. Same cohort as above. High
                 follow-up participation, no selective attrition.
                 Linear regression adjusted for SES, maternal
                 IQ, HOME, adolescent marijuana use, and
                 obstetrical complications. Also considered
                 potential confounding by birth outcomes,
                 maternal age, prenatal smoking, alcohol,
                 marijuana, and narcotics use, # previous
                 abortions, stillbirths, gravidity, parity,
                 caregiver education, public assistance, child
                 age, sex, health, Fe status
                                         Earlier
                                         childhood
                                         (age 6.5 yr)
                                         Mean  (SD):
                                         NR
             Reading, spelling,
             math,
             vocabulary
             composite
             WRAT-III and
             WISC-III
             Ages 15-17 yr
-0.081 (-0.17, 0.003)
Chandramouli     488 children followed from age 30 mo (born
et al. (2009)       1991 -1992) to 7-8 yr, Avon, U. K.
                 Prospective. All births in area eligible.
                 Similar characteristics as U.K. census, high
                 participation at baseline and follow-up.
                 Participants had better educated mothers,
                 who smoked less, better home environment.
                 Regression model adjusted for maternal
                 education and smoking, home ownership,
                 home facilities score, family adversity index,
                 paternal SES, parenting attitudes at 6 mo,
                 child sex. Also considered potential
                 confounding by child IQ.
                                         Age 30 mo
                                         Mean (SD):
                                         NR
                                         Group 1:
                                         0-2
                                         Group2: 2-5
                                         Group 3:
                                         5-10
                                         Group 4:
             Standardized
             Achievement Test
             Ages 7-8 yr
Per doubling blood Pb
-0.3 (-0.5,-0.1)°
November 2012
                                 5-108
                  Draft - Do Not Cite or Quote

-------
Study
Miranda et al.
(2009)
Lanphear et al.
(2000)
Krieg et al.
(2010)
Surkan et al.
(2007)
Kordas et al.
(2006)
Study Population and
Methodological Details
(Presented first for prospective studies then
for cross-sectional studies. Within each
category, results are presented in order to
strength of study design)
57,568 children, 4th grade, all counties, NC
Prospective: based on data from
surveillance databases. Quantile regression
adjusted for sex, age of blood Pb
measurement, race, enrollment in free lunch
program, parental education, charter school.
Did not consider potential confounding by
parental caregiving quality or cognitive
function.
4,852 children ages 6-16 yr (born
1972-1988), U.S. NHANES III (1988-1994)
Cross-sectional. Large U.S. representative
study of multiple risk factors and outcomes.
High, non-selective participation. Linear
regression model adjusted for sex,
race/ethnicity, poverty index ratio, reference
adult education, serum ferritin and cotinine
levels, Did not consider potential
confounding by parental cognitive function or
parental caregiving quality.
766-780 children ages 12-16 yr (born
1975-1982), U.S. NHANES III (1991-1994)
Cross-sectional. Large U.S. representative
study of multiple risk factors and outcomes.
Log linear regression model adjusted for sex,
caregiver education, family income, race-
ethnicity, test language. Did not consider
potential confounding by parental cognitive
function or caregiving quality.
389 children ages 6-10 yr, Boston, MA,
Farmington, ME
Cross-sectional. Recruitment from trial of
amalgam fillings. High participation rate.
Higher participation of white children in
Maine. Analysis of covariance adjusted for
caregiver IQ, child age, SES, race, and birth
weight, Also considered potential
confounding by caregiver education and
marital status, parenting stress, and
maternal utilization of prenatal or annual
health care but not parental caregiving
quality.
294 children, age 7 yr, Torreon, Mexico.
Cross-sectional. Recruitment at prenatal
clinic. High participation rate. Residence
near metal foundry. Linear regression model
adjusted for child sex, age, school, birth
order, hemoglobin, forgetting homework,
household possessions and crowding, house
ownership, maternal education, family
structure, urinary As, tester. Did not consider
potential confounding by parental cognitive
function or caregiving quality.
Blood or
Tooth Pb
Levels
(ug/dL)
Earlier
childhood
(ages 9-36
mo)
Median
(25th-75th):
4.8 (3-6)
Concurrent
Geometric
mean: 1.9
(5th-95th:
1.70, 2.10)
Interval
analyzed:
1 .74-2.06
= 1 0th-90th
percentile
Concurrent
Mean
(5th-95th):
1.95
(1 .63-2.27)
Interval
analyzed:
1.69-2.19 =
10th-90th
percentile
Concurrent
Mean (SD):
2.2(1.6)
Concurrent
Geometric
mean
(range):
10.2
(2-43.8)
Interval
analyzed:
2.1-10.0
Indicator of
Academic
Performance/
Achievement
Reading
4th grade end-of-
grade test score
Math score
Reading score
WRAT-R
Ages 6-1 6 yr
Math score
Reading score
WRAT-R
Ages 1 2-1 6 yr
Reading score
Math score
WIAT
Ages 6-1 0 yr
Math achievement
test
PPVT
Age 7 yr
Effect Estimate
(95% Cl)a
Score vs. blood Pb
1 ug/dL
2 ug/dL:
-0.30 (-0.58, -0.01 )d
3 ug/dL:
-0.46 (-0.73, -0.1 9)d
4 ug/dL:
-0.52 (-0.79, -0.24)d
5 ug/dL:
-0.80 (-1.08, -0.51 )d
-0.70 (-1.0, -0.37)
-0.99 (-1 .4, -0.62)
-2.5 (-4.5, -0.50)
-2.9 (-4.3, -1.5)
Blood Pb5-10 ug/dL
vs. 1-2:
-5.20 (-9.45, -0.95)d
-4.02 (-7.6, -0.43)
-0.42 (-0.92, 0.08)
-0.71 (-1 .4, 0.02)
November 2012
5-109
Draft - Do Not Cite or Quote

-------
Study
Study Population and
Methodological Details
(Presented first for prospective studies then    Blood or     Indicator of
for cross-sectional studies. Within each        Tooth Pb    Academic
category, results are presented in order to      Levels       Performance/
strength of study design)	(ug/dL)       Achievement
                                                                                               Effect Estimate
                                                                                               (95% Cl)a
Chiodo et al.      506 children, age 7 yr (born 1982-1984),
(2007)           Detroit, Ml area.
                 Cross-sectional. All African American. High
                 prevalence of prenatal drug exposure. High
                 follow-up participation. Linear regression
                 model adjusted for caregiver education,
                 SES, HOME, maternal IQ, child sex, prenatal
                 marijuana use (all outcomes), Caregiver
                 concurrent psychological symptoms (Math),
                 Child age, maternal  custody (Reading). Also
                 considered potential confounding by prenatal
                 cigarettes/day, alcohol use, cocaine use, #
                 children in home, caretaker marital status,
                 concurrent alcohol/week, current maternal
                 cigarettes/day, and current marijuana use.
                                           Concurrent
                                           Mean (SD):
                                           5.0 (3.0)
                                           Interval
                                           analyzed:
                                           2.1-8.7=10
                                           th-90th
                                           percentiles
                                                                          Math
                                  -0.17 (-0.27, -0.07)°
              Reading
              Metropolitan
              Aptitude Test
              Age 7 yr
                     -0.06, p>0.05b
Chiodo et al.
(2004)
246 children, age 7.5 yr, Detroit, Ml area       Concurrent
Cross-sectional. Recruitment at prenatal      Mean (SD):
clinic. All African American. High prevalence    5.4 (3.3)
of prenatal alcohol exposure. High             Interval
participation rate. Log linear regression        analyzed'
model adjusted  for SES (all outcomes).        23-95 =10
HOME, caregiver vocabulary, prenatal         th-90th
alcohol use (arithmetic). Caregiver             percentiles
vocabulary, disruption in caregiver (verbal      H
learning). Also considered potential
confounding by  maternal prenatal marijuana,
smoking, or cocaine use, crowding, child life
stress, caregiver age, life stress, and
psychology, conflict tactics, family
functioning, # children <18 years, caregiver
education, child sex  and age, parity.
              Verbal learning
              (WRAML)
                     -0.20, p>0.05b
              Arithmetic (WISC-
              III)
              Age 7.5 yr
                                                                                               -0.1 7, p>0.05b
Min et al. (2009)   267 children, age 11 yr (born 1994-1996),
                 Cleveland, OH
                 Prospective, Recruitment at birth hospital.
                 86% African American with high prevalence
                 of prenatal drug and alcohol exposure.
                 Moderate follow-up participation to age 4 yr,
                 high retention to  age 11 yr. Higher
                 participation from African American and
                 married mothers. Linear regression model
                 adjusted for HOME score, maternal birth
                 vocabulary score, head circumference at
                 birth (both outcomes), prenatal cocaine use
                 (math), child sex, prenatal cocaine and
                 alcohol use, current caregiver alcohol use
                 (reading  age 11 yr). Also considered
                 potential confounding by maternal education,
                 Fe deficiency, maternal psychological
                 distress,  race.
                                           Earlier
                                           childhood
                                           (Age 4 yr)
                                           Mean
                                           (range): 7.0
                                           (1.3-23.8)
                                           Interval
                                           analyzed:
                                           3.0 (1 Oth
                                           percentile)-
                                           10
                                                                          Math
                                  -0.45 (-0.84, -0.06)
              Reading
              WJTA
              Age 11 yr
                     -0.58 (-1.0, -0.13)
Studies of School Performance
Fergusson et al.
(1997)
881 children followed from birth to age 16-18
yr, Christchurch, New Zealand
Prospective. Moderate follow-up
participation, attrition did not affect results.
Regression model adjusted for maternal age,
punitiveness, standard of living,
breastfeeding duration, parental conflict,
class level, residence on busy roads. Also
considered potential confounding by sex,
ethnicity, maternal education, family size,
HOME, SES, # schools attended, ethnicity,
paternal education, parental smoking, child
birth outcomes, weatherboard housing.
Tooth Pb
(age 6-8 yr)
Mean (SD):
6.2
(3.7) ug/g
Percent leaving
school with no
qualifications
Age 16-18 yr
0-2 ug/g: 15.6
                                                                                               3-5 ug/g: 16.7
                                                                                               6-8 ug/g: 18.1
                                                                                               9-11 ug/g: 19.7
                                                                                               12+ug/g: 24.1
November 2012
                                  5-110
                   Draft - Do Not Cite or Quote

-------
Study
Needleman et
al. (1990)
Study Population and
Methodological Details
(Presented first for prospective studies then
for cross-sectional studies. Within each
category, results are presented in order to
strength of study design)
132 young adults followed from 1st/2nd
grade to age 18 yr, Chelsea, Sommerville,
MA
Prospective. Recruitment at schools. Low
follow-up participation. Participants had
lower tooth Pb, higher parental education,
SES, maternal IQ. Participation status did
not alter tooth Pb-childhood IQ association.
Logistic regression adjusted for maternal age
at birth, education, and IQ, family size, sex,
age at testing, birth order, alcohol use,
mother and child left hospital together. Did
not examine potential confounding by
parental caregiving quality.
Blood or
Tooth Pb
Levels
(ug/dL)
Tooth Pb
(1st/2nd
grade)
distribution
<10 ppm,
50%
10-1 9.9 pp
m: 22.7%
>20 ppm:
27.3%
Indicator of
Academic
Performance/
Achievement
Failure to graduate
high school
Highest grade
achieved
Effect Estimate
(95% Cl)a
OR >10 ppm vs.
<10 ppm
7.4 (1 .4, 40.8)d
-0.03 (-0.05, 0) per
natural log increase in
tooth Pb
Study of teacher ratings of academic performance
Leviton et al.
(1993)
1923 children followed from birth
(1979-1980) to age 8 yr, Boston, MA
Prospective. Recruitment from birth
hospital. High participation at baseline and
follow-up. Regression model adjusted for
single parent family, gestational age,
maternal education, ethnicity, #
children, daycare in first 3 years. Also
considered potential confounding by other
unspecified factors.
Prenatal
(cord)
Mean: 6.8
Tooth Pb
(Age 6 yr)
Mean: 3.3
Reading, BTQ, Age
Syr
Prenatal (cord)
Tooth Pb
RR (yes/no) per loge
increase
Girls: 1.7(0.9, 3.3)
Boys: 1 .3 (0.8, 2.2)
Girls: 2.2 (1.1, 4.2)
Boys: 1 .2 (0.7, 2.2)
      Note: Results are organized by method of outcome assessment then by prospective or cross-sectional design. Within each
      category, results are presented in order to strength of study design.
      KABC = Kaufman Assessment Battery for Children, WRAT = Wide Range Achievement Test, WISC = Wechsler Intelligence Scale
      for Children, BTQ = Boston Teacher's Questionnaire, WIAT = Wechsler Individual Achievement Test, PPVT = Peabody Picture
      Vocabulary Test, WRAML = Wide Range Assessment of Memory and Learning, WJTA = Woodcock Johnson-Ill Tests of
      Achievement.
      aEffect estimates are standardized to a 1  ug/dL increase in blood Pb level in the interval from the 10th percentile of blood Pb level
      to the 90th percentile or 10 ug/dL, whichever is lower.
      bSufficient data were not provided to calculate 95% Cl.
      °95% Cl: was constructed using a standard error that was estimated from the reported p-value.
      dEffect estimates compare test performance of children in higher blood Pb groups to children in lowest blood Pb group.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
The ages at which associations between blood Pb level and performance on academic
achievement tests were found varied between prospective studies. In the Boston cohort
with lower blood Pb levels (mean: 7.0 (ig/dL), a 1  (ig/dL increase in age 2 year blood Pb
level was associated with a -0.30-point (95% Cl: -0.65, 0.05) change in quantitative skills
score age at 5 years in the blood Pb interval 1.8-10 (ig/dL (Bellinger et al..  1991) with
adjustment for SES, maternal IQ and marital status, preschool attendance, HOME, out of
home care, residence changes, medication use in previous  12 months, number of adults in
home, child sex, race, birth weight, and birth order. Evidence did not strongly indicate an
association with verbal skills. In the Cincinnati cohort with higher blood Pb levels, age
6.5 year blood Pb level was associated with decrements in academic performance at ages
15-17 years and 5 years (Ris et al.. 2004; Dietrich  et al.. 1992) but not 4 years (Dietrich et
al.. 1991). These differences within the Cincinnati cohort could be attributed to changes
in blood Pb levels overtime or age-related differences in reliability  of tests of learning.
      November 2012
                             5-111
Draft - Do Not Cite or Quote

-------
 1                   Recent prospective studies found associations for earlier childhood blood Pb levels but
 2                   did not have blood Pb measurements available at other time periods for comparison. The
 3                   records-based analysis (Miranda et al., 2009; 2007a), multi-factorial nature, or high
 4                   participation rate of recent studies (Chandramouli et al.. 2009) do not indicate a strong
 5                   influence of selection bias. Miranda et al. (2009; 2007a) linked higher blood Pb levels
 6                   measured at ages 0-5 years, as ascertained from a surveillance database, with lower end-
 7                   of-grade (EOG) test scores in 8,600 fourth grade children in seven of the largest counties
 8                   in North Carolina and then in 57,678 children in the entire state. A strength of the
 9                   analyses was the availability of individual-level data on a large number of children
10                   representative of the North Carolina fourth grade population.  The large numbers of
11                   children with blood Pb levels 2-5 (ig/dL provided greater power to estimate the effects of
12                   Pb in the lower range of blood Pb levels. In each analysis, children with an earlier blood
13                   Pb level of 2 (ig/dL had lower EOG scores (p < 0.05) compared with children with a
14                   blood Pb level of 1 (ig/dL. Further, across deciles of blood Pb level, the decrease in EOG
15                   score generally was monotonic (Figure 5-6). Because these children were born in the
16                   early- to mid-1990s and blood Pb levels were measured earlier in childhood, it is less
17                   likely that associations were influenced by higher past Pb exposures.

18                   Due to the records-based study design, investigators had a smaller set of potential
19                   confounding factors available than those considered in the prospective studies described
20                   above. Results were adjusted for sex, race, school-type, school district, age of blood Pb
21                   measurement, parental education, participation in a free or reduced lunch program as a
22                   measure of SES, and in the analysis of seven North Carolina counties, daily use of a
23                   computer as a measure of a stimulating home environment (Miranda et al., 2007a). While
24                   there may be no complete single measure of SES and parental caregiving quality, the
25                   covariates examined in these analyses are not as well characterized,  and the results may
26                   be subject to residual confounding.
      November 2012                             5-112                 Draft - Do Not Cite or Quote

-------
                     £-0,
                     u
                     
-------
 1
 2
 3
       risk (lower parental education, enrollment in a school lunch program) had a greater
       magnitude of negative association with EOG score, blood Pb level was independently
       associated with EOG score decrements that were as large as 1 to 2 points.
        «   o
           -2

        I
        8
        W
        n
        UJ
        c
        •  -8
        I
        8-10
        o
          -12
 Cummulative Deficit: Decrease in EOG scores by multiple risk factors
5%  10% 15%  20% 25% 30% 35% 40% 45% 50%  55% 60% 65%  70% 75% 80% 85%  90% 95%
                                                              t
     Baseline (BLL=lMg/dL, no school
     lunch program, parents completed college)
     Effect of reduction in parental education
   > from completed college
     to only completed high school
                                                                Income effect as indicated by
                                                                enrollment in school lunch program
                                                              } Effect of increased BLL from 1 to 5 pg/dL
                                         Quantile
      Note: Baseline score calculated for a hypothetical referent individual with a blood Pb level of 1 ug/dL, parents completed college,
      and not enrolled in the school lunch program (i.e., model covariates = zero). Vertical bars indicate the decrease in EOG score
      associated with blood Pb level and covariates in various percentiles of EOG score (lowest to highest, left to right). An increase in
      earlier childhood blood Pb level is associated with a larger decrease in EOG score (larger black bars on left) among children in the
      5th and 10th percentiles of EOG score than children in the 90th and 95th percentiles (smaller black bars on right).
      Source: Reprinted with permission of Elsevier Science, Miranda et al. (2009)

      Figure 5-7        Greater reduction in End-of-Grade (EOG) scores with increasing
                          blood Pb level in lower percentiles of the test  score distribution.
 4
 5
 6
 1
 8
 9
10
11
12
       Similar to Miranda et al. (2009). Chandramouli et al. (2009) found associations between
       earlier childhood blood Pb levels (age 30 months) and later academic performance
       (Standard Assessment Test [SAT] at age 7 years). In this study of 488 children in the
       U.K., who had similar sociodemographic characteristics as those found in the U.K.
       census, a doubling of age 30 month blood Pb level was associated with a 0.3-point (95%
       CI: 0.1, 0.5) decline in SAT grade. Results were adjusted for maternal education and
       smoking, home ownership, parental SES and several factors related to caregiving quality
       including home facilities score, family adversity index, and parenting attitudes. In
       analyses of blood Pb level categories,  lower SAT scores were most clearly indicated in
      November 2012
                                    5-114
Draft - Do Not Cite or Quote

-------
 1                   children with age 30 month blood Pb levels >5 (ig/dL. Children with blood Pb levels
 2                   2-5 (ig/dL generally did not have lower SAT scores than children with blood Pb levels
 3                   0-2 ng/dL.

 4                   Consistent with prospective studies, cross-sectional studies found associations between
 5                   higher concurrent blood Pb level and lower scores on tests of math and reading, including
 6                   large studies of children participating in NHANES. While cross-sectional studies
 7                   considered potential confounding by SES and caregiver education, few considered
 8                   parental cognitive function, and none considered parental caregiving quality. Lanphear et
 9                   al (2000)  and Krieg et al. (2010) found concurrent blood Pb-associated decrements in
10                   math and  reading score among 4,852 children ages 6-16 years and 766-780 children ages
11                   12-16, respectively, participating in NHANES. The examination of multiple exposures
12                   and outcomes in NHANES increases confidence that associations are not unduly
13                   influenced by selection bias. While the mean blood Pb levels were low in these study
14                   populations, ~2 (ig/dL, the influence of higher past Pb exposures on findings cannot be
15                   excluded. Consistent with studies of FSIQ, Lanphear et al. (2000) found a supralinear
16                   concentration-response relationship. A 1 (ig/dL increase in concurrent blood Pb level was
17                   associated with a change in reading score of-0.70-points (95% CI: -1.0,  -0.37)  among all
18                   subjects and -1.1-points (95% CI: -1.54, -0.58) among the 4, 043 children with  blood Pb
19                   levels <5  (ig/dL. A supralinear concentration-response relationship also was found in
20                   children ages 7 years in Mexico living near a metal foundry as indicated by larger blood
21                   Pb-associated decrements in math and vocabulary scores among children with concurrent
22                   blood Pb levels <10 (ig/dL (Kordas et al.. 2006). In contrast with these findings, among
23                   children ages 6-10 years in New England decrements in reading and math scores were
24                   found in association with higher blood Pb levels, i.e., blood Pb levels 5-10 (ig/dL
25                   compared with blood Pb levels 0-2 (ig/dL (Surkan et al.. 2007).

26                   Prospective studies in a Boston, MA area cohort and New Zealand cohort found
27                   associations of tooth Pb levels measured at an earlier age (ages 6-8 years) with  school
28                   performance ascertained at ages 16-18 years from school records (Fergusson et al.. 1997,
29                   1993; Needleman et al.. 1990). suggesting the effect of early exposure to Pb may  be
30                   persistent. In the New Zealand cohort at ages 12-13 and 18 years, recruitment rate and
31                   follow-up participation were high, and model correction for nonrandom sample attrition
32                   produced  robust results, indicating lack of undue selection bias (Fergusson et al..  1997.
33                   1993). Further, associations observed at ages 12-13 years between higher tooth Pb level
34                   and lower teacher ratings of math, reading, and writing abilities (Fergusson et al.. 1993).
35                   which are subject to greater measurement error, were supported by associations observed
36                   at age 18 years with more objective measures such as lower probability of completion of
37                   high school and lower scores on school exams (Fergusson et al.. 1997). In this cohort,
38                   Pb-associated decrements in school performance were found with consideration for
      November 2012                             5-115                  Draft - Do Not Cite or Quote

-------
 1                  potential confounding by several factors including SES, parental education, HOME
 2                  score, sex, ethnicity, number of school changes, perinatal history, breastfeeding, maternal
 3                  age, and residence in weatherboard housing and near busy roads.

 4                  In one Boston-area cohort, age 6-8 years tooth Pb level >20 (ig/g was associated with
 5                  dropping out of high school at age 18 years with an odds ratio of 7.4 (95% CI:  1.4, 40.7)
 6                  (Needleman et al.. 1990). The relatively small sample size (n = 132) and adjustment for
 7                  several potential confounding factors, including maternal education, IQ, and age, SES,
 8                  and subject alcohol use may have contributed to the imprecision of the effect estimate.
 9                  Parental caregiving quality was not considered. Participation was biased to children with
10                  lower tooth Pb levels and higher SES. This selection bias likely did not produce a
11                  spurious association; however, the results may be less generalizable to the original study
12                  population. In another Boston-area cohort, higher tooth Pb level at age 6 years  was
13                  associated with higher teacher ratings of age 8 year spelling and reading difficulties in
14                  girls but not boys (Leviton et al.. 1993). Despite the large sample  size (n = 1923) and
15                  high follow-up participation, the study did adjust for SES or parental caregiving quality.
16                  However, other unspecified potential confounding factors were considered.
      5.3.2.6  Integrated Summary of Cognitive Function in Children

17                  Results from recent epidemiologic and animal studies add to the strong evidence base
18                  reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b) demonstrating that Pb exposure is
19                  associated with decrements in cognitive function in children, based on associations
20                  observed with FSIQ, and also executive function, and academic performance.
21                  Associations with performance on tests of learning and memory were less consistently
22                  found (Table 5-5). A large epidemiologic evidence base demonstrates associations of
23                  higher blood Pb level with lower FSIQ in school-aged children (Figure 5-2 and Table
24                  5-3). with smaller bodies of evidence indicating associations with lower scores on tests of
25                  executive function, and academic performance in children ages 4 to 18 years (Table 5-8
26                  and Table 5-9). There was no clear indication that blood Pb level was more strongly
27                  associated with performance in a particular domain of cognitive function. The
28                  Pb-associated decrements in cognitive function observed in children were strongly
29                  supported by observations in animals of decrements in learning, memory, and executive
30                  function with relevant dietary Pb exposures. In particular, coherence was found between
31                  observations of Pb-associated decrements in performance on spatial span tasks in
32                  children and Morris water maze in animals both of which test visual spatial memory on
33                  spatial working memory tasks in children and the radial arm maze in animals
34                  (Section 5.3.2.3). Coherence also was found with Pb-associated changes in performance
35                  on tests of rule learning and reversal in humans and animals (Section 5.3.2.4) both of


      November 2012                            5-116                  Draft - Do Not Cite or Quote

-------
 1                   which assess cognitive flexibility. Additional biological plausibility for Pb-associated
 2                   cognitive function decrements was provided by toxicological evidence for the effects of
 3                   Pb on neurophysiological and neurochemical processes that mediate cognition.

 4                   Compelling epidemiologic evidence for Pb-associated cognitive function decrements was
 5                   described in the 2006 Pb AQCD (U.S. EPA. 2006b) for FSIQ (see Section 5.3.2.1 of this
 6                   ISA also). Across studies, FSIQ was measured with similar instruments (i.e., WISC-R,
 7                   WISC-III, WPPSI, Stanford-Binet) scored on a similar scale with similar measurement
 8                   error. Associations were found in most of the prospective studies, conducted in the U.S.,
 9                   Mexico, Europe, and Australia, in representative populations with high follow-up
10                   participation without indication of selective participation among children with higher
11                   blood Pb levels and lower cognitive function (Figure 5-2 and Table 5-3) that could
12                   produce spurious associations. The prospective studies found associations of blood Pb
13                   levels measured concurrently with FSIQ (ages  4-17 years) and earlier in life (i.e., prenatal
14                   cord or maternal, age 2 year) or averaged over  multiple years, better establishing the
15                   temporal sequence between Pb exposure and cognitive function decrements than cross-
16                   sectional studies, multiple testing was common; however, the consistent pattern of
17                   association observed across the ages of blood Pb level and/or cognitive test examined in
18                   most previous and recent studies increases confidence that the evidence is not unduly
19                   biased by a higher probability of associations found by chance alone. Another strength of
20                   the prospective evidence was the consideration of several potential confounding factors.
21                   As indicated in Table 5-3. results from  most cohorts were adjusted for maternal IQ and
22                   education, child sex and birth weight, SES, and HOME score. Although not considered as
23                   frequently, some studies also indicated lack of confounding by parental smoking, birth
24                   order, and nutritional status. The robustness of the blood Pb-FSIQ association in children
25                   was substantiated in a pooled analysis of seven prospective studies by Lanphear et al.
26                   (2005) as well  as multiple meta-analyses that combined results across various prospective
27                   and cross-sectional studies (Pocock et al., 1994; Schwartz,  1994; Needleman and
28                   Gatsonis. 1990). with Schwartz (1994)  demonstrating the robustness of evidence to
29                   potential publication bias.

30                   Comparisons of effect estimates across studies  are difficult because of the variability in
31                   population blood Pb distributions, lifestage of blood Pb level examined, type of model
32                   examined (linear versus nonlinear), and tests conducted. The pooled analysis of seven
33                   prospective cohorts demonstrated precision of effect estimates by finding a relatively
34                   narrow range of effect estimates, -2.36  to -2.94 points per natural log increase in blood Pb
35                   level, excluding one study at a time (Lanphear  et al.. 2005). In a linear model, a greater
36                   decrease in FSIQ estimated for a 1 (ig/dL increase in concurrent blood Pb for the 244
37                   children who had peak blood Pb levels  <10 (ig/dL (-0.80 points [95% CI: -1.74, -0.14])
38                   and the 103 children with peak blood Pb levels <7.5 (ig/dL (-2.9 points [95% CI: -5.2,
      November 2012                             5-117                  Draft - Do Not Cite or Quote

-------
 1                   -0.71]). Among children with peak blood Pb levels <10 (ig/dL and <7.5 (ig/dL, the
 2                   median concurrent blood Pb levels were 4.2 (ig/dL and 3.2 (ig/dL, respectively (Hornung.
 3                   2008). Among individual studies, a wide range of effect estimates was reported.
 4                   However, studies varied in model specification and the blood Pb level range examined.
 5                   Similarly large effects were estimated in the Boston and Rochester cohorts, which
 6                   differed widely in racial and SES distributions (Canfield et al.. 2003a: Bellinger et al..
 7                   1992). While the sample sizes were smaller, these studies had least as extensive
 8                   consideration for potential confounding as other studies. Further, each study estimated
 9                   larger effects for children whose peak blood Pb levels never exceeded 10 (ig/dL, -1.8
10                   points (95% CI: -3.0, -0.60) per 1 (ig/dL increase in concurrent blood Pb level in the
11                   Rochester cohort (n = 101, 59%) (Canfield et al.. 2003a) and -1.6 points (95% CI: -2.9,
12                   -0.2) per 1  (ig/dL in age 2-year blood Pb level in the Boston cohort (n = 48, 32%)
13                   (Bellinger and Needleman. 2003). These subsets of children had mean blood Pb levels of
14                   3.3 (Rochester) and 3.8 (ig/dL (Boston). Some recent cross-sectional studies estimated
15                   smaller effects but with examination of populations with higher concurrent blood Pb
16                   levels (means: 7, 8.7 (ig/dL) using a linear model (Kordas et al.. 2011; Min et al.. 2009).
17                   Other recent studies estimated similar effects as previous studies although the log-linear
18                   models make comparisons difficult. Among children ages 3-7 years in India, a  1 (ig/dL
19                   increase in concurrent blood Pb level was associated with a 1.2-point decrease (95% CI:
20                   -1.9, -0.37) in FSIQ from the 10th percentile of blood Pb level 5.8 to 10 (ig/dL  (Roy et
21                   al.. 2011). Kim et al. (2009b) found that a 1 (ig/dL increase in concurrent blood Pb level
22                   was associated with a 3.2-point decrease (95% CI: -6.1, -0.23) in FSIQ among children
23                   ages 8-11 years in Korea with blood Mn levels >1.4 (ig/dL in the  10th-90th percentile
24                   interval of blood Pb level (0.9-2.8 (ig/dL). In this study, the potential influence of higher
25                   past Pb exposures cannot be excluded. Further, while these recent studies adjusted for
26                   parental education and SES, parental caregiving quality was not examined. The relatively
27                   low blood Pb levels in the Rochester and Boston cohorts, consideration of peak blood Pb
28                   levels, and the adjustment for several potential confounding factors indicate that their
29                   results may be more  representative.

30                   Previous prospective  studies, several of which contributed to the FSIQ evidence,
31                   provided key evidence for associations of blood or tooth Pb level  with decrements in
32                   executive function and academic performance for the reasons described for FSIQ.
33                   Endpoints associated with blood or tooth Pb level included rule learning and reversal,
34                   reading and math skills assessed using neuropsychological tests and school performance
35                   assessed from school records. Higher concurrent blood Pb level was associated with
36                   lower scores on tests of math and reading in the large study of children participating in
37                   NHANES (Lanphear et al.. 2000). Recent studies conducted in the U.S., Mexico, Europe,
38                   and Asia, most of which were cross-sectional, also found associations between higher
39                   blood Pb level and lower cognitive function. The few recent prospective studies indicated

      November 2012                             5-118                  Draft - Do Not Cite or Quote

-------
 1                   associations between higher earlier childhood blood Pb level, ages 9-36 month and age
 2                   30 months, respectively, with poorer academic performance in children in North Carolina
 3                   at age 9 years (Miranda et al.. 2009) and in children ages 7 years in the U.K.
 4                   (Chandramouli et al.. 2009).

 5                   In most studies that provided unadjusted and adjusted effect estimates, blood Pb level
 6                   was associated with a smaller but statistically significant decrement in FSIQ after
 7                   adjusting for potential confounding factors (Palaniappan et al.. 2011; Kim et al.. 2009b:
 8                   Lanphear et al., 2005; Canfield et al., 2003a). The consideration for potential
 9                   confounding varied among studies. Most studies adjusted for SES-related variables such
10                   as the Hollingshead Index, household income, and/or parental education. Several, in
11                   particular the prospective studies, adjusted for parental cognitive function and parental
12                   caregiving quality commonly evaluated as HOME score. Overall, recent studies
13                   considered potential confounding by SES and parental IQ or education but not parental
14                   caregiving quality. Analyses of associations between potential confounding factors and
15                   blood Pb level and cognitive function indicated that the confounding factors may vary
16                   across populations and endpoints. In the Cleveland cohort, adjustment for HOME score
17                   attenuated the blood or tooth Pb level-cognitive function relationships (Greene and
18                   Ernhart. 1993: Greene et al.. 1992: Ernhart et al.. 1988). In the Rochester cohort, HOME
19                   score met the criteria for adjustment in models for FSIQ (Canfield et al., 2003a) but not
20                   all measures of memory and executive function (Froehlich et al.. 2007: Canfield et al..
21                   2004: 2003b). Adjustment for SES is difficult as it is highly correlated with Pb exposure
22                   and there is no single measure that represents SES. Residual confounding also is likely by
23                   factors not considered. The combination of evidence from prospective studies that
24                   considered several well-characterized potential confounding factors plus evidence  that Pb
25                   exposure induces impairments in cognitive function in animals, in particular, spatial
26                   memory and executive function, which are associated with blood or tooth Pb levels in
27                   children increases confidence that the associations between blood and tooth Pb levels and
28                   cognitive function observed in children represent a relationship with Pb exposure.

29                   With regard to important lifestages and durations of Pb exposure, toxicological evidence
30                   clearly demonstrates impaired learning and memory in animals exposed to Pb
31                   gestationally with or without early postnatal exposure. Impairments in learning and
32                   memory observed with lower blood Pb levels (8-17 (ig/dL) were found with Pb exposures
33                   that began during the gestational or lactation period. The effect of early life Pb exposures
34                   is supported by evidence that processes such as neurogenesis and synaptic pruning are
35                   highly active during the first few years of life (Rice and Barone. 2000: Landrigan et al..
36                   1999). However, evidence in a group of monkeys also indicates impaired learning  with
37                   Pb exposure beginning later during the juvenile period, indicating that Pb exposure in
38                   infancy is not necessary to induce impairments in cognitive function (Rice.  1992b. 1990:
      November 2012                             5-119                 Draft - Do Not Cite or Quote

-------
 1                   Rice and Gilbert. 1990b). Epidemiologic studies also point to cognitive function
 2                   decrements associated with blood Pb levels measured at various lifestages and time
 3                   periods. Among studies of young children <3 years,  several found stronger associations
 4                   of MDI with prenatal (maternal or cord) blood Pb than with postnatal child blood Pb (Hu
 5                   et al., 2006; Bellinger et al.. 1987; Dietrich et al.. 1987a; Vimpani et al.. 1985). However,
 6                   in older children, ages 4-17 years, in whom cognitive function is more stable and reliably
 7                   measured, decrements in cognitive function were associated with more strongly with
 8                   indicators of postnatal Pb exposure, i.e., concurrent,  early childhood, and cumulative
 9                   average blood Pb levels as well with tooth Pb levels. Evidence did not clearly identify an
10                   individual critical postnatal time period or duration of Pb exposure in terms of risk of
11                   developing cognitive function decrements. Because of the contribution of bone Pb levels
12                   to concurrent blood Pb levels in children, associations with concurrent blood Pb levels
13                   may reflect an effect of past and recent Pb exposures.

14                   Previous prospective  studies found blood Pb-associated decrements in cognitive function
15                   in populations with mean blood Pb levels 5-10 (ig/dL (Table 5-3). In analyses restricted
16                   to children in the lower range of the blood Pb distribution (e.g., <10 (ig/dL), associations
17                   with FSIQ were found in groups of children with mean age 2 year or concurrent blood Pb
18                   levels 3-4 (ig/dL with consideration of peak blood Pb levels (Bellinger. 2008; Canfield.
19                   2008; Hornung. 2008). Several  recent studies found associations of FSIQ with lower
20                   blood Pb levels (primarily concurrent), population means 2-5 (ig/dL, (Kim et al.. 2009b;
21                   Jusko et al.. 2008; Zailinaet al., 2008; Chiodo et al., 2007) for FSIQ but not consistently
22                   for other indices of cognitive function (Cho et al.. 2010; Miranda et al.. 2010;
23                   Chandramouli et al., 2009; Surkan et al., 2007). Many of these recent studies had
24                   uncertainties related to the influence of higher past Pb exposures, high prevalence of
25                   prenatal drug exposure, or potential confounding. Several recent toxicological studies
26                   added to the evidence for impaired learning and memory in animals with lower blood Pb
27                   levels, 8-17 (ig/dL (Cory-Slechta et al.. 2010; Niu et al.. 2009; Virgolini et al.. 2008a;
28                   Stangle et al.. 2007). Recent evidence from the Cory-Slechta laboratory found learning
29                   impairments with lower lifetime Pb exposures when combined with  stress, which
30                   potentially may be mediated via effects on corticosterone and dopamine (Rossi-George et
31                   al.. 2011; Cory-Slechta et  al.. 2010; Rossi-George et al.. 2009; Virgolini et al.. 2008a).

32                   The biological plausibility for epidemiologic and toxicological evidence linking Pb
33                   exposure to decrements in cognitive function is provided by the well-characterized
34                   toxicological evidence for Pb exposure interfering with development of the brain and
35                   activity of neurochemical processes that mediate cognitive function (Section 5.3.11). Pb
36                   has been shown to increase the permeability of the blood-brain barrier and deposit in the
37                   CNS. Pb has been shown to impair neurogenesis, synaptic architecture, and neurite
38                   outgrowth. Cognitive function is mediated by the cortical and subcortical structures of the
      November 2012                             5-120                 Draft - Do Not Cite or Quote

-------
 1                  brain that integrate function in the hippocampus, prefrontal cortex, and nucleus
 2                  accumbens using dopamine and glutamate as primary neurotransmitters. Experimental
 3                  studies have shown that Pb induces changes in dopamine and glutamate release in these
 4                  regions and decreases long-term potentiation, which is a major cellular mechanism
 5                  underlying synaptic plasticity and learning and memory.
      5.3.2.7  Epidemiologic Studies of Cognitive Function in Adults

                    Adults without Occupational Pb Exposures

 6                  As described in the preceding section, Pb exposure that begins in gestation and lasts
 7                  through the early postnatal period or for a lifetime or begins after infancy has been shown
 8                  to induce learning impairments in adult animals. Less well characterized are learning
 9                  impairments in adult animals due to adult-only Pb exposures. As reported in the
10                  2006 Pb AQCD, epidemiologic studies have examined cognitive performance in adults
11                  without occupational Pb exposure primarily in association with concurrently measured
12                  blood and bone Pb levels and have found associations with bone Pb levels but not blood
13                  Pb levels (U.S. EPA. 2006b). Recent studies produced similar findings and provided new
14                  evidence from prospective analyses (Table 5-10).

15                  Evidence was provided by large cohorts examining multiple exposures and outcomes,
16                  reducing the likelihood of selective participation of subjects with higher Pb exposures
17                  and cognitive deficits. Most studies performed multiple tests of cognitive function.
18                  However, associations with bone Pb level were not isolated to a few tests. Several
19                  publications are available; however, many are variant analyses in the same population
20                  (e-g-, Normative  Aging Study [NAS], NHANES) and are not considered as all
21                  independent assessments of the Pb-cognitive function relationship. Further, although
22                  evidence is available from longitudinal cohorts, most analyses are cross-sectional
23                  examining the association between one measurement of cognitive function and a
24                  concurrent measure of blood or bone Pb level. Because temporality cannot be
25                  determined, causal inference regarding the effects of Pb exposure is limited. In analyses
26                  of bone Pb level, this limitation is mitigated somewhat because bone Pb level reflects
27                  several years of exposure. Additionally, with blood and bone Pb level, it is difficult to
28                  characterize the specific timing, duration, frequency, and level of Pb exposure that
29                  contributed to associations observed with cognitive function. This uncertainty may apply
30                  particularly to assessments of blood Pb levels, which in nonoccupationally-exposed
31                  adults, reflect both current exposures and cumulative Pb stores in bone that are mobilized
32                  during bone remodeling (Sections 4.3 and 4.7.3). Although studies adjusted for age, a
33                  common limitation is the potential for residual confounding by age because of the strong


      November 2012                           5-121                 Draft - Do Not Cite or Quote

-------
 1                  correlation between increasing bone Pb levels and increasing age. However, the
 2                  coherence with evidence for cognitive function decrements associated with long-term Pb
 3                  exposure in animals provides support for associations observed in adults. Because of the
 4                  difficulty in establishing the temporal sequence between Pb exposure and cognitive
 5                  function in cross-sectional studies, in the review of evidence, emphasis was placed on
 6                  prospective analyses. Emphasis also was placed on studies that considered several
 7                  potential confounding factors such as age, education, SES, smoking, and alcohol use.


                    Evidence from Prospective Studies

 8                  Key evidence for the effects of Pb exposure on cognitive function of adults has been
 9                  provided by recent prospective analyses of the large  Baltimore Memory Study (BMS)
10                  and NAS. Strengths of these studies include comparisons of associations between bone
11                  and blood Pb levels, the repeated assessment of Pb biomarker levels and cognitive
12                  performance, the high follow-up participation of subjects, and lack of selective attrition
13                  by Pb biomarker levels and demographic characteristics. In particular, the repeated
14                  assessments permitted the examination of associations of bone Pb levels with changes in
15                  cognitive function over time, which better established the temporal sequence between Pb
16                  exposure and subsequent changes in cognitive function. The BMS and NAS differed in
17                  many respects, including sex and race of subjects, the test instruments used, and potential
18                  confounding factors considered. The BMS included  men and women, 50-70 years of age,
19                  residing in Baltimore, MD. A total of 1,140 out of 2,351 (48.5%) subjects participated
20                  from neighborhoods that represented a diversity of race and SES. This study was unique
21                  in that it included a large proportion of African-Americans (n=395). In comparison, the
22                  NAS involved only men (original n = 2,280) residing in the Greater Boston area. Subjects
23                  primarily were white and at enrollment were ages 21 to 80 years and had no current or
24                  past chronic medical conditions. Both studies adjusted for age and education. The BMS
25                  additionally adjusted for household wealth, and the NAS additionally adjusted for
26                  smoking and alcohol intake. Results from both of these cohorts with different
27                  demographics and methodology indicated Pb-associated cognitive function decrements.

28                  In the BMS, longitudinal analyses involved repeat cognitive testing at 14-month
29                  intervals. Most subjects completed follow-up; 91% at the second round of testing and
30                  83% at the third round (Bandeen-Roche et al., 2009). An interquartile range higher
31                  baseline tibia Pb level (12.7 ug/g) was associated with a 0.019 unit (95% CI:  -0.031,
32                  -0.007) per year decrease in eye-hand coordination z-score, with adjustment for age, sex,
33                  race, SES, and interviewer, with a larger decrease estimated for African Americans than
34                  for whites (Table 5-10). Results were not homogeneous across the various tests
35                  performed. Tibia Pb levels were more weakly associated with time-related decreases in
36                  language, processing speed, and executive function;  however, most effect estimates were

      November 2012                            5-122                  Draft - Do Not Cite or Quote

-------
                 negative in direction. Further, for language and executive function, tibia Pb level was
                 associated with greater decreases in scores among whites than African Americans.
Table 5-10   Associations of blood and bone Pb levels with cognitive function in
                 adults.
Study  Population and
Methodological Details
    (Presented first for
    prospective analyses then
    for cross sectional analyses.
    Within categories, studies
    are presented in order of
    strength of methodology)
                   Subgroup
                   (where
 Cognitive Test   examined)
                Blood Pb Effect
                Estimate
                (95% Cl)
Bone Pb Effect Estimate
(95% Cl)
Prospective Studies:
Bandeen-Roche et al. (2009)
943, ages 50-70 yr at baseline,
BMS, Baltimore, MD
Large sample of men and
women of various
races/ethnicities with repeated
measures of cognitive function
and tibia Pb. High follow-up
participation over 28 mo.
Marginal linear regression
models adjusted for age, sex,
household wealth, education,
race/ethnicity, interviewer. Did
not consider potential
confounding by history of
smoking or alcohol use.
Tibia Pb Mean (SD):
19 (12.7) ug/g
                                                                 NOT EXAMINED
Longitudinal associations
  Eye/hand         African-
  coordination -      Americans
  Purdue pegboard
                                     Change in Z-scores per
                                     IQR increase:

                                     -0.032 (-0.052, -0.012)/yr


                                     -0.009 (-0.024, 0.006)/yr
Cross-sectional associations
  Verbal
  memory/learning
  - Rey auditory
  verbal learning
  test
  Language -
  Boston naming
  test
African-
Americans
White
African-
Americans
White
0.006 (-0.09, 0.10)


-0.076 (-0.15, 0.001)


0.065 (-0.010,0.14)


-0.024 (-0.12, 0.07)
Weisskopfetal. (2007b)
405-749 males, mean age 68.7
yr at baseline, NAS, Boston,
MA area.
Large sample, only men,
primarily white. Repeated
measures of cognitive function
and tibia Pb. High follow-up
participation over 3.5 yr.
Linear repeated measures
analysis adjusted for age, age2,
education, smoking status,
current alcohol intake, yr
between bone Pb
measurement and first
cognitive test, yr between
cognitive tests. Also evaluated
language, computer
experience,  physical activity.
Mean (IQR)
Tibia Pb:20(15) ug/g
Patella Pb: 25 (20) ug/g
                                                                 NOT EXAMINED
Longitudinal associations

  Visuospatial skills - pattern
  comparison (+ = poorer
  performance), NES2

  Executive function - verbal fluency,
  WISC-R

  Short-term memory - word list,
  CERAD
                                     Change in score over 3.5 yr
                                     per IQR increase:

                                     Tibia: 0.079 (0.04, 0.12)
                                     Patella: 0.073 (0.04, 0.12)

                                     Tibia:-0.04 (-0.16, 0.08)
                                     Patella: -0.086 (-0.20, 0.03)

                                     Tibia:-0.028 (-0.12, 0.06)
                                     Patella:-0.081 (-0.17, 0.005)
Cross-sectional associations
  Visuospatial skills - pattern
  comparison latency
  (+ = poorer performance), NES2

  Executive function - verbal fluency,
  WISC-R

  Short-term memory - word list,
  CERAD
                                     Tibia:-0.03 (-0.17, 0.11)
                                     Patella:-0.02 (-0.14, 0.11)


                                     Tibia:-0.27 (-0.70, 0.16)
                                     Patella:-0.22 (-0.62, 0.17)

                                     Tibia: 0.12 (-0.20, 0.32)
                                     Patella: 0.012 (-0.18, 0.41)
November 2012
                    5-123
                          Draft - Do Not Cite or Quote

-------
Study3 Population and
Methodological Details
    (Presented first for
    prospective analyses then
    for cross sectional analyses.
    Within categories, studies
    are presented in order of
    strength of methodology)
 Cognitive Test
Subgroup
(where
examined)
Blood Pb Effect
Estimate
(95% Cl)
Bone Pb Effect Estimate
(95% Cl)
Wang et al. (2007a)
358 males, median age: 67 yr,
MAS, Boston, MA area
Same cohort as above. Subset
representative of full cohort.
Linear regression adjusted for
age, years of education,
smoking status, pack-years
smoking, nondrinker,
grams/day alcohol
consumption, English as first
language, computer
experience, diabetes.
Tibia Pb: Median (IQR) 19
(15)ug/g
                                                                  NOT EXAMINED
Mini Mental State
Exam Score
HFE wildtype
One HFE
variant
Two HFE
variants
                     Change in Score per
                     IQR increase:

                     -0.02 (-0.10, 0.07)/yr

                     -0.14 (-0.33, 0.04)/yr

                     -0.63 (-1.04,-0.21 )/yr
Cross-sectional Studies:
Shih et al. (2006)
985 adults, mean age: 59 yr,
BMS, Baltimore, MD
Large sample. Subjects with
tibia Pb measured were more
educated and white. Compared
blood/bone associations.
Linear regression adjusted for:
Model A: age, sex, technician,

presence of APOE-E4 allele
Model B: Model I, years of
education, race/ethnicity, wealth
Did not consider potential
confounding by history of
smoking or alcohol use.
Mean (SD)
Concurrent blood Pb: 3.5
(2.2) ug/dL
TibiaPb:18.7(11.2)ug/g



Language -
Boston naming
test

Eye-hand
coordination -
Purdue Pegboard,
trail making

Executive
functioning -
Purdue Pegboard,
Stroop and trail
making test
Visuoconstruction
- Rey complex
figure copy




Model A
Model B


Model A
Model B



Model A
Model B


Model A
Model B

Score per 1 |jg/dL
increase:

-0.006 (-0.03, 0.017)
-0.002 (-0.02, 0.016)


-0.011 (-0.03, 0.01)
-0.008 (-0.02, 0.002)



-0.01 4 (-0.03, 0.005)
-0.010 (-0.03, 0.007)


-0.01 9 (-0.05, 0.008)
-0.01 4 (-0.04, 0.01)


Score per 1 |jg/g increase:

-0.008 (-0.01 , -0.004)
0.0006 (-0.003, 0.004)


-0.008 (-0.01 , -0.004)
-0.008 (-0.02, 0.002)



-0.008 (-0.01 , -0.004)
-0.003 (-0.006, 0.0008)


-0.01 2 (-0.02, -0.007)
-0.004 (-0.01 , 0.0003)

Glass etal. (2009)
1,001  adults, mean age 59 yr,
BMS,  Baltimore, MD
Large sample. High participation
rate: 91 %.
Multilevel hierarchical regression
model adjusted for age, sex,
race/ethnicity, education, testing
technician, time of day.
Investigator assessed NPH.
Did not consider potential
confounding by history of
smoking or alcohol use.
Tibia PbMean (SD): 18.8
                                                                  NOT EXAMINED
 Language -
 Boston naming
 test
 Eye-hand
 coordination -
 Purdue Pegboard,
 trail making test
 Executive
 functioning -
 Purdue Pegboard,
 Stroop, trail
 making test
 Visuoconstruction
 - Rey complex
 figure copy
2nd tertile NPH
SrdtertileNPH


2nd tertile NPH
3rd tertile NPH
2nd tertile NPH
3rd tertile NPH
                                                  2nd tertile NPH
                                                  3rd tertile NPH
                     Score per 1 |jg/g increase:

                     0.001 (-0.008, 0.009)b
                     -0.009 (-0.017,-0.0001 )b


                     -0.004 (-0.012, 0.005)b
                     -0.006 (-0.015, 0.002)b
                     -0.002 (-0.010,0.006)"
                     -0.010 (-0.018, -0.002)b
                                     -0.003 (-0.014, 0.008)b
                                     -0.006 (-0.017, 0.005)b
November 2012
                    5-124
                           Draft - Do Not Cite or Quote

-------
Study3 Population and
Methodological Details
    (Presented first for
    prospective analyses then
    for cross sectional analyses.
    Within categories, studies
    are presented in order of
    strength of methodology)
                   Subgroup
                   (where
Cognitive Test    examined)
Blood Pb Effect
Estimate
(95% Cl)
Bone Pb Effect Estimate
(95% Cl)
Weuve et al. (2006)
720 males, ages > 45 yr, MAS,
Boston, MA area
Large sample. High follow-up
participation. Compared
blood/bone associations.
Linear mixed effects regression
adjusted for smoking status,
grams/day alcohol consumption,
calorie adjusted calcium intake,
regular energy expenditure on
leisure time physical activity,
diabetes.
Additional adjustment for dietary
factors.
Median (IQR)
Concurrent blood Pb: 5.2
(3) ug/dL
Tibia: 19(15) ug/g
Patella: 27 (21) ug/g
Mini Mental State
Exam Score
                   ALAD wildtype
                   ALAD-2 carrier
                   ALAD wildtype
                   ALAD-2 carrier
Score per IQR
increase:

-0.05 (-0.16, 0.06)
-0.29 (-0.56, -0.02)
Score per IQR increase:


Tibia:-0.05 (-0.21, 0.12)
 -0.16 (-0.58, 0.27)

Patella: -0.07 (-0.23, 0.09)
 -0.26 (-0.64, 0.12)
Rajanetal. (2008)
486-959 males, ages > 45 yr,
MAS, Boston, MA area
Large sample. Compared
blood/bone associations. Linear
regression adjusted for blood Pb
main effect, ALAD genotype,
age at cognitive test, education,
grams/day alcohol consumption,
pack-years smoking, English as
first language. Also considered
smoking status, income,
physical activity, diabetes,
coronary heart disease.
Concurrent blood Pb Mean
(SD): 5.3(2.9) ug/dL(ALAD
wildtype), 4.8 (2.7) ug/dL
(ALAD2 carriers)
Tibia Mean (SD): 21.9
(13.8) ug/g (ALAD wildtype),
21.2(11.6) ug/g (ALAD2
carriers)
Patella Mean (SD): 29.3
(19.1) ug/g (ALAD wildtype),
27.9 (17.3) ug/g (ALAD2
carriers)
Visuospatial - constructional Praxis,
CERAD

Executive function - verbal fluency,
CERAD

Verbal memory - word list memory,
CERAD

Perceptual speed - mean latency
continuous performance, NES
Score*ALAD2 per
IQR increase:

-0.05 (-0.23, 0.13)°


-0.03 (-0.22, 0.16)°


0.003 (-0.18, 0.19)°


-0.18 (-0.42, 0.06)°
Score*ALAD2 per IQR
increase:

Tibia: -0.25 (-0.49, -0.02)°
Patella: 0.02 (-0.19, 0.23)°

Tibia:-0.11 (-0.34,0.13)°
Patella:-0.03 (-0.24, 0.19)°

Tibia: 0.08 (-0.15, 0.31)°
Patella: 0.14 (-0.07, 0.34)°

Tibia: -0.25 (-0.59, 0.08)°
Patella:-0.16 (-0.44, 0.12)°
November 2012
                   5-125
           Draft - Do Not Cite or Quote

-------
Study3 Population and
Methodological Details
    (Presented first for
    prospective analyses then
    for cross sectional analyses.
    Within categories, studies
    are presented in order of
    strength of methodology)
                   Subgroup
                   (where
Cognitive Test    examined)
                 Blood Pb Effect
                 Estimate
                 (95% Cl)
                     Bone Pb Effect Estimate
                     (95% Cl)
Weuve et al. (2009)
587 females, ages 47-74 yr,
Nurses' Health Study, Boston,
MA area
Large sample of only females.
Compared blood/bone
associations. Generalized
estimating equations adjusted
forage and age2 at Pb
assessment, age at cognitive
assessment, education,
husband's education, alcoholic
drinks/week, smoking status,
physical activity, use of aspirin,
ibuprofen, Vitamin  E
supplements, menopausal
status, and postmenopausal
hormone use. Additional
adjustment for nutrition factors,
medication use, mental health.
Assessed outcomes over
telephone but a mean 5 years
after Pb measured.
Mean (SD)
Concurrent blood Pb 2.9
(1.9)ug/dL
Tibia Pb 10.5(9.7) ug/g
Patella Pb 12.6(11.6) ug/g
Orientation, registration, immediate
verbal memory with TICS. Immediate
and delayed paragraph recall,
category fluency, digit span
backwards (working memory,
attention) with EBMT

Composite Z-score
Composite except letter fluency
                                   Z-score per SD
                                   increase:
                                      Z-score per SD increase:
                 -0.015
                 (-0.069, 0.039)


                 0.016
                 (-0.071, 0.039)
                     Tibia: -0.040 (-0.09, 0.004)
                     Patella:-0.012 (-0.06, 0.03)


                     Tibia:-0.05 (-0.10,-0.003)
                     Patella:-0.033 (-0.08, 0.014)
Krieg and Butler (2009)
2,823 adults, ages 20-59 yr,
Large U.S. representative
NHANES III (1991-1994).
Log-linear regression model
adjusted forage, sex, education,
family income, race-ethnicity,
computer or video-game
familiarity, alcohol use within the
last 3 h, test language, sampling
unit and stratum. Did not
consider potential confounding
by smoking.
Concurrent Blood mean (SD):
2.88(6.91) ug/dL
Symbol Digit
Substitution
(mean total
latency, sec)
Serial digit
learning total
score, NES
Ages 20-39 yr
Ages 40-59 yr


Ages 20-39 yr
Ages 40-59 yr
                                   Z-score per 1
                                   increase:
-0.097 (-0.42, 0.23)d
-0.290 (-0.60, 0.02)d


-0.1 17 (-0.46, 0.23)d
0.401 (-0.19, 1.0)d
                                                        NOT EXAMINED
November 2012
                   5-126
                           Draft - Do Not Cite or Quote

-------
Study3 Population and
Methodological Details
(Presented first for
prospective analyses then
for cross sectional analyses.
Within categories, studies
are presented in order of
strength of methodology)
Krieg et al. (2009)
2,090 adults, ages 20-59 yr,
1 976 adults, ages > 60 yr. Large
U.S. representative NHANES III
(1991-1994).
Log linear regression model
adjusted for sex, age, education,
family income, race-ethnicity,
computer or video game
familiarity, alcohol use in the last
3 h, test language (20-59 yr) and
sex, age, education, family
income, race-ethnicity, test
language (> 60 yr), sampling
unit and stratum. Did not
consider potential confounding
by smoking.
Concurrent Blood Pb Mean (SD)
Age 20-59 yr: 2.85 (7.31) ug/dL
Age > 60 yr: 4.02 (3.56) ug/dL
Krieg et al. (201 0)
2,093 adults, ages 20-59 yr,
1 ,799 adults, ages > 60 yr.
Large U.S. representative
NHANES III (1991-1994).
Log linear regression model
adjusted for sex, age, education,
family income, race-ethnicity,
computer or video game
familiarity, alcohol use in the last
3h, test language, sampling unit,
stratum (20-59 yr) and sex, age,
education, family income, race-
ethnicity, test language,
sampling unit, and stratum (s 60
yr). Did not consider potential
confounding by smoking.
Concurrent blood Pb Mean (SD)
Age 20-59 yr: 2.85 (7.32) ug/dL
Age > 60 yr: 4.02 (3.39) ug/dL









Cognitive Test


Symbol Digit
Substitution
(mean total
latency)

Serial digit
learning total
score
Word recall,
number correct


Story recall,
number correct,
Neurobehavioral
Evaluation
System



Symbol Digit
Substitution
(mean total
latency, sec)

Serial digit
learning total
score


Word recall,
number correct



Story recall,
number correct
Neurobehavioral
Pvsli istinn
QVCtlUCt UUI 1
System


Subgroup
(where
examined)


Ages 20-59 yr
ALAD wildtype
ALAD-2 carrier

Ages 20-59 yr
ALAD wildtype
ALAD-2 carrier
Ages > 60 yr
ALAD wildtype
ALAD-2 carrier

Ages > 60 yr
ALAD wildtype
ALAD-2 carrier


Age group and
VDR haplotype
Ages 20-59 yr
CC haplotype
CT haplotype
TC haplotype
TT haplotype
Ages 20-59 yr
CC haplotype
CT haplotype
TC haplotype
TT haplotype
Ages > 60 yr
CC haplotype
CT haplotype
TC haplotype
TT haplotype
Ages > 60 yr
CC haplotype
CT haplotype

TC haplotype
TT haplotype


Blood Pb Effect
Estimate Bone Pb Effect Estimate
(95% Cl) (95% Cl)
Z-score per 1 [tgldL
increase: NOT EXAMINED
-0.1 32 (-0.358, 0.095)d
-0.526 (-1.11 8, 0.066)d


-0.022 (-0.526, 0.482)d
0.025 (-0.406, 0.456)d

-0.075 (-0.285, 0.135)
0.025 (-0.406, 0.456)

0.085 (-0.0997, 0.271)
-0.466 (-1 .072, 0.139)


Score per log
increase: NOT EXAMINED
-20 (-44, 4.0)d
0.73 (-1 .4, 2.9)d
-2.6 (-5.3, 0.07)d
-3.6 (-7.2, 0.05)d

8.0(0.61, 15.4)
1 .0 (-0.89, 2.9)
-1.4 (-3.1, 0.29)
-0.01 4 (-2.8, 2.5)

-0.65 (-1.5, 0.25)
-0.08 (-0.34, 0.19)
-0.03 (-0.40, 0.33)
-0.08 (-0.76, 0.60)

0.29 (-3.3, 3.9)
0.01 (-0.38, 0.40)

0.07 (-0.64, 0.78)
-0.22 (-0.86, 0.43)
November 2012
5-127
Draft - Do Not Cite or Quote

-------
Study3 Population and
Methodological Details
    (Presented first for
    prospective analyses then
    for cross sectional analyses.
    Within categories, studies
    are presented in order of
    strength of methodology)
Cognitive Test
Subgroup
(where
examined)
Blood Pb Effect
Estimate
(95% Cl)
Bone Pb Effect Estimate
(95% Cl)
Van Wijngaarden et al. (2009)
47 adults, mean age 61.5 yr,
Rochester, NY
Small sample size, without
consideration for potential
confounding  by smoking or
alcohol use. Linear regression
adjusted forage, sex,
educational level, history of
hypertension. Excluded subjects
with BMI >32 kg/m2.
Mean (SD)
Tibia: 2.0 (5.2) ug/g
Calcaneus: 6.1 (8.5) ug/g
Delayed matching
to sample, %
correct
CANTAB
                                   NOT EXAMINED
Paired Associate
Learning, total
trials adjusted
(increase = poorer
performance)
CANTAB
                                      Calcaneus
                                      Lowest fertile: 87.56e
                                      Medium tertile: 86.67
                                      Highest Tertile: 80.67, p=0.03
                                      Tibia
                                      Lowest tertile: 85.42e
                                      Medium tertile: 87.08
                                      Highest tertile: 82.44, p=0.25


                                      Calcaneus
                                      Lowest tertile: 2.54e
                                      Medium tertile: 2.61
                                      Highest tertile: 2.72, p = 0.21


                                      Tibia
                                      Lowest tertile: 2.62e
                                      Medium tertile: 2.59
                                      Highest tertile: 2.66, p = 0.79
Gao et al. i
188 adults, mean age 69.2 yr,
Rural Sichuan and Shandong
Provinces, China.
Small sample size. Subset was
younger, had more education,
and higher BMI than full cohort.
Separate ANCOVA adjusted for
age, sex, education, BMI, or
APOE £4. History of smoking
and alcohol consumption not
associated with cognitive score.
Concurrent plasma Pb Mean
(SD): 0.39 (0.63) ug/dL
Composite
cognitive Z-score
Word list learning,
word recall
(CERAD), CSID,
IU story recall,
Animal naming
fluency test, IU
token test of
language and
working memory
                 Z-score per 1 |jg/dL
                 plasma Pb
                 increase:


                 42.8(21.4, 64.2)
                                                                                       NOT EXAMINED
Note: Effect estimates in bold indicate the stronger association between blood Pb and bone Pb level. IQR = Interquartile range,
BMS = Baltimore Memory Study, NAS = Normative Aging Study, NES2 = Neurobehavioral Evaluation System 2, WISC-R =
Wechsler Adult Intelligence Scale-Revised, CERAD = Consortium to Establish Registry for Alzheimer's disease, HFE = Human
Hemochromatosis protein, NPH = Neighborhood Psychosocial Hazard, TICS = Telephone Interview for Cognitive Status,
EBMT = East Boston Memory Test, CANTAB = Cambridge Neuropsychological Test Automated Battery, CSID = Community
Screening Instrument for Dementia, IU = Indiana University.
"Studies are presented first for prospective analyses then for cross sectional analyses. Within categories, studies are presented in
order of strength of methodology.
bEffect estimates indicate interactions between Pb and category of NPH, with the lowest tertile of NPH serving as the reference
group.
°Effect estimates indicate interactions between Pb and ALAD genotype.
dThe directions of effect estimates were changed to indicate a negative slope as a decrease in cognitive performance.
eResults refer to mean cognitive function scores among tertiles of bone Pb. Tertile concentrations not reported.
November 2012
                   5-128
                           Draft - Do Not Cite or Quote

-------
 1                   Similar to the BMS, among NAS men, higher baseline tibia Pb levels were associated
 2                   with decreases in cognitive performance overtime in longitudinal analyses with repeated
 3                   measures of cognitive function plus a bone Pb-time interaction term in order to estimate
 4                   the association between baseline bone Pb level and decline in cognitive test score over
 5                   time (Weisskopf et al. 2007b). This NAS analysis expanded the evidence base by also
 6                   finding associations with patella Pb levels. Two measurements of cognitive function,
 7                   collected approximately 3.5 years apart were available for 60-70% of participants. Both
 8                   tibia and patella Pb levels were associated with decrements in executive function, short-
 9                   term memory, and visuospatial skills (as indicated by increased response latency on a
10                   pattern comparison test). The strongest effect was estimated for the latter. Weisskopf et
11                   al. (2007b) also found a nonlinear association with patella Pb, with latency times
12                   becoming worse over time (i.e., larger values indicating slower response time) up to
13                   approximately 60 ug/g patella Pb then leveling off at higher levels (Figure 5-8). A
14                   20 ug/g difference in patella Pb level was associated with an increase in latency of 0.073
15                   ms (95% CI: 0.04, 0.12) among all men and a 0.15 ms increase among men with patella
16                   Pb level <60 ug/g. Both patella and tibia Pb were associated with fewer errors on the
17                   pattern comparison test. The authors proposed that this may be related to slowing reaction
18                   time to improve accuracy. When the nine men with the highest bone Pb levels were
19                   removed, the association with fewer errors was no longer statistically significant.
20                   However, the authors did not indicate whether the point estimate changed.
      November 2012                             5-129                  Draft - Do Not Cite or Quote

-------
                       .™   CSj
                        E   o
                        |
                       I   P
                        o   o
                       cc   °
                       1
                        c   2
                       I
                        P   CO
                        8   ? J
                                                 20        40         60
                                                    Patella lead (M9/g)
                                 80
      Note: Models are adjusted for age, age squared, education, smoking, alcohol intake, years between bone Pb measurement and first
      cognitive test, and years between the cognitive tests. The 9 subjects with the highest patella Pb levels (>89 ug/g bone mineral) were
      removed. The estimated concentration-response is indicated by the solid line and the 95% confidence interval by the dashed lines.
      The patella Pb level-associated increase in response latency is larger among men with patella Pb levels <60 ug/g. Patella Pb levels
      of all individual subjects are indicated by short vertical lines on the abscissa, (reference = 0 at mean of patella  Pb level).
      Source: Reprinted with permission of Williams & Wilkins, Weisskopf et al. (2007b).

      Figure 5-8        Nonlinear association between patella Pb level and the relative
                         change over 3.5 years in response latency on the pattern
                         comparison test in men from the Normative Aging Study.

 1                   Longitudinal analysis of the NAS cohort also indicated that hemochromatosis (HFE)
 2                   gene variants modified the blood Pb-cognition association (Wang et al.. 2007a). In
 3                   models adjusted for age, years of education, smoking status, pack-years  smoking,
 4                   nondrinker, grams/day alcohol consumption, English as first language, computer
 5                   experience, and diabetes, an interquartile range higher tibia Pb level (15 ug/g) was
 6                   associated with a 0.22 point steeper annual decline (95% CI: -0.39, -0.05) in Mini-Mental
 7                   State Examination score (MMSE, which assesses cognitive impairment  in a number of
 8                   domains) among the 130 (36%) men with either the H63D or C282Y variant. The
 9                   association was found to be nonlinear, with larger Pb-associated declines observed at
10                   higher tibia Pb levels (Figure 5-9. solid line).  The change in MMSE score associated with
11                   15 ug/g higher tibia Pb levels was comparable to that found between NAS men who were
12                   4 years apart in age. Tibia Pb level was not associated with a decline in MMSE score in
13                   men with the HFE wildtype genotype (Figure 5-9. dashed line). Bone Pb levels did not
      November 2012
5-130
Draft - Do Not Cite or Quote

-------
1
2
3
4
differ widely by HFE variant. HFE variants, H63D and C282Y, are associated with
hemochromatosis, a disease characterized by higher iron body burden. Iron metabolism
has been hypothesized to affect neurodegenerative diseases, which may explain the
observed effect modification. However, firm conclusions are not warranted.
                         o
                         o
                        uj  _n?
                        GO   U'^
                        •-  -0.4
                         o
                         O)
                         c
                         CD
                        "§  -0.6
                        <  -0.8
               	WFfwildtype
               	 HFE variant allele
                                0        10        20        30       40       50
                                          Tibia lead biomarker (ug/g)
     Note: The lines indicate curvilinear trends estimated from the penalized spline method. Among hemochromatosis (HFE) wild-types,
     the association between tibia Pb and annual cognitive decline was nearly null (dashed line). Among variant allele carriers, the
     association tended to deviate from linearity (solid line, p = 0.08), with a greater tibia Pb-associated decline in MMSE observed
     among men with higher tibia Pb levels. The model was adjusted for age, years of education, smoking status, pack-years smoking,
     nondrinker, grams/day alcohol consumption, English as first language, computer experience, and diabetes.
     Source: Wang et al. (2007a).

     Figure 5-9        Nonlinear association of tibia Pb level with annual rate of
                        cognitive decline, by hemochromatosis genotype in men from the
                        Normative Aging Study.
5
6
7
Evidence from Cross-sectional Studies
Associations between bone Pb levels and decrements in cognitive function in adults also
are supported by evidence from several cross-sectional studies conducted in the BMS and
NAS cohorts and other populations. The cross-sectional studies have contributed
evidence for stronger associations of cognitive function decrements with bone Pb levels
than blood Pb level and for associations with adjustment for additional potential
     November 2012
                            5-131
Draft - Do Not Cite or Quote

-------
 1                   confounding factors such as diet and medication use. While cross-sectional studies
 2                   examined factors that may potentially increase risk of Pb-associated cognitive function
 3                   decrements in adults, they each examined different factors and did not produce
 4                   conclusive evidence. These subgroup analyses also are subject to higher probability of
 5                   finding an association by chance.

 6                   In addition to comparisons of blood and bone Pb levels, cross-sectional analyses of the
 7                   BMS included detailed analysis of potential confounding, although smoking and alcohol
 8                   use were not examined. Among 991 adults, both higher concurrent blood and bone Pb
 9                   level were associated with poorer performance in tests of language, processing speed,
10                   eye-hand coordination, executive function, verbal memory and learning, visual memory,
11                   and visuoconstruction; however, associations with tibia Pb level tended to be larger in
12                   magnitude (Table 5-10) (Shih et al. 2006).  Mean (SD) blood and tibia Pb levels were
13                   3.46 (2.23)  ug/dL and  18.7 (11.2) ug/g, respectively. Tibia Pb levels were associated with
14                   worse performance on tests in all domains with adjustment for age, sex, testing
15                   technician, and presence of the  apolipoprotein (APO)E-e4 allele (potential risk factor for
16                   Alzheimer's Disease). The magnitudes of associations were attenuated with additional
17                   adjustment for education, race,  and household wealth; however, in these more fully-
18                   adjusted models, higher tibia Pb levels remained associated with poorer performance in
19                   all domains except language and processing speed. The strongest association was found
20                   for visuoconstruction, which assesses visuospatial skills and motor skills. A 1 ug/g bone
21                   higher tibia Pb level was associated with a 0.0044 SD (95% CI:  -0.0091, 0.0003) lower
22                   visuoconstruction score. Analysis of tibia Pb as a quadratic term did not indicate a
23                   nonlinear relationship with visuoconstruction.

24                   In contrast with longitudinal results in BMS, race-stratified analyses of persistent effects
25                   in cross-sectional analyses indicated that tibia Pb levels were associated with greater
26                   decreases in performance on tests of eye-hand coordination, executive function, and
27                   verbal memory and learning among whites than among African Americans (Bandeen-
28                   Roche et al., 2009). Among all  subjects, tibia Pb-associated decrements in cognitive
29                   performance were modified by  neighborhood level psychosocial stress. Specifically,
30                   higher tibia Pb levels were associated with larger decrements, particularly in language,
31                   eye-hand coordination, and executive function, among subjects living in neighborhoods
32                   with a greater number psychosocial hazards (e.g., number of violent crimes, emergency
33                   calls, off-site liquor licenses as  assessed by investigators) (Glass et al.. 2009) (Table
34                   5-10). Results were adjusted for age, sex, race/ethnicity, education, testing technician,
35                   and time of testing. Subjects living near more psychosocial hazards had slightly higher
36                   tibia Pb levels. In support of these results, several studies have found Pb-stress
37                   interactions in  impaired learning and memory of adult animals with Pb exposures
      November 2012                             5-132                  Draft - Do Not Cite or Quote

-------
 1                  beginning in gestation and lasting through post-weaning or to the time of testing
 2                  (Section 5.3.2.3).

 3                  The 2006 Pb AQCD (U.S. EPA. 2006b) described cross-sectional associations of both
 4                  blood and tibia Pb levels with poorer cognitive performance among 141 NAS men
 5                  (Pavton et al.. 1998). Several recent, larger cross-sectional NAS analyses corroborated
 6                  previous findings for bone Pb but generally indicated weak associations with concurrent
 7                  blood Pb levels and only in groups with specific genetic variants. In contrast with the
 8                  longitudinal analyses, Weisskopf et al. (2007b) found that repeat measures of bone Pb
 9                  levels were inconsistently associated with cognitive function (improved and poorer
10                  performance) in cross-sectional analyses. Among 720 NAS men 45 years of age and
11                  older, higher concurrent blood and bone Pb levels were associated with lower MMSE
12                  scores among  149 ALAD-2 carriers (Weuve et al.. 2006). with a larger decrease found
13                  for an increase in blood Pb level. A 3 (ig/dL higher concurrent blood Pb level (the
14                  interquartile range) was associated with a 0.26 point lower mean MMSE score (95% CI:
15                  -0.54, -0.01) among ALAD-2 carriers and a 0.04 point lower score (95% CI: -0.16, -0.07)
16                  among noncarriers. A subsequent NAS analysis (n = 486-959) did not find a consistent
17                  direction of modification of the association of blood or bone Pb levels with tests of
18                  cognitive function in various domains by ALAD genotype (Rajan et al.. 2008). An
19                  interaction between higher tibia Pb level and ALAD-2 genotype was found only for
20                  visuospatial skills (constructional praxis test), and between patella Pb level and ALAD-2
21                  genotype for perceptual speed (pattern comparison test). The potential direction of effect
22                  modification by the ALAD-2 genotype is not clear as the greater affinity of the ALAD-2
23                  enzyme subunit for Pb may increase risk of Pb-associated health effects by increasing
24                  blood Pb levels, or it may diminish Pb-associated health effects by sequestering Pb in the
25                  bloodstream and decreasing its bioavailability.

26                  Cross-sectional studies examined a larger number of potential confounding factors than
27                  did longitudinal analyses. The NAS found blood and bone Pb-associated decrements in
28                  cognitive function with adjustment for dietary  factors, physical activity, medication use,
29                  and comorbid conditions (Raian et al.. 2008; Weuve et al.. 2006) (Table 5-10). As in the
30                  BMS and NAS, tibia and patella Pb levels were more consistently associated with
31                  cognitive performance than was blood Pb levels in 587 healthy women in the Boston,
32                  MA area participating in the Nurses' Health Study (Table 5-10) (Weuve et al.. 2009).
33                  Additional potential confounding factors examined in this group included use of aspirin,
34                  ibuprofen, or vitamin E, mental health, and antidepressant use. Blood, patella, and tibia
35                  Pb levels were measured between ages 47 and 74 years and an average of 5 years before
36                  cognitive testing. Contrary to expectation, higher patella and tibia Pb levels were
37                  associated with higher scores on the "f" naming test (naming words that begin with f). In
38                  separate models, the "f' naming test was omitted from a composite index of all cognitive
      November 2012                             5-133                  Draft - Do Not Cite or Quote

-------
 1                  tests performed by phone, and a one SD higher tibia Pb level was associated with
 2                  0.051-point lower (95% CI: -0.10, -0.003) composite cognitive function z-score (Table
 3                  5-10). A similar magnitude of decrease was estimated for an increase in age of 3 years in
 4                  these women. The magnitude of association was smaller for an SD increase in patella Pb
 5                  level (-0.033 [95% CI: -0.080, 0.014]), and a weak association was found for an SD unit
 6                  increase in blood Pb level (-0.016 [95% CI: -0.071, 0.039]).

 7                  Several analyses of the large, U.S.-representative NHANES III (1991-1994) population
 8                  of men and women investigated effect modification by age and genetic variants. Only
 9                  blood Pb levels were available and were measured in samples collected concurrently with
10                  cognitive testing. These analyses adjusted for several of the same potential  confounding
11                  factors as other studies, with the exception of smoking. Krieg and Butler (2009) found
12                  blood Pb level to be associated weakly with poorer performance on tests of learning and
13                  visuospatial skills among adults ages 20-39 years and inconsistently  in adults ages 40-59
14                  years. Krieg et al. (2009) further found inconsistent associations with word and story
15                  recall in adults ages > 60 years. Because of the different types and numbers of tests
16                  administered, it is difficult to compare findings between adults less than and greater than
17                  age 60 years. In the subset of the population with genetic analysis, blood Pb-cognitive
18                  function associations were not found to be modified by ALAD genetic variants in a
19                  consistent direction (Krieg et al.. 2009). Among adults ages 20-59 years and > 60 years,
20                  higher concurrent blood Pb level was associated with a larger decrement in performance
21                  on some tests in ALAD-2 carriers and other tests in ALAD -1 subjects (Table 5-10).
22                  Krieg et al. (2010) found differences in the association between concurrent blood Pb level
23                  and scores  on a symbol-digit substitution test by the VDR variants, rs731236 and VDR
24                  rs2239185, and by the VDR haplotype, which have unclear functional relevance. Similar
25                  to observations in adolescent NHANES participants (Section 5.3.2.5). results were not
26                  uniform across the various tests. However, for several tests, blood Pb level  was
27                  associated  with greater decrements in cognitive performance among  adults  with the CC
28                  genotypes  of VDR variants.

29                  Other cross-sectional  studies with fewer subjects generally produced results consistent
30                  with those  from the larger studies described above. A study of 188 rural Chinese men and
31                  women found a weak association between higher plasma Pb levels and a lower composite
32                  cognitive score based on a battery of in-person administered tests (Gao et al.. 2008).
33                  Results were adjusted for age, sex, education, BMI, or APOE-e4 in individual ANCOVA
34                  analyses. Smoking and alcohol use were not associated with cognitive performance in
35                  this group. Pb in plasma is not bound to erythrocytes, as is about 99% of blood Pb,  and is
36                  the fraction delivered directly to soft tissue (Chuang et al.. 2001; Hernandez-Avila et al..
37                  1998). The results of Gao et al. (2008) may provide information on the cognitive effects
38                  of a more bioavailable fraction of Pb dose; however, because there is little investigation
      November 2012                             5-134                  Draft - Do Not Cite or Quote

-------
 1                  of plasma Pb, firm conclusions are not warranted. Among 47 men and women in
 2                  Rochester, NY (age 55-67 years), subjects in the highest two tertiles of calcaneal bone
 3                  (heel bone with higher turnover rate than tibia) Pb level performed worse on delayed
 4                  matching-to-sample and paired associated learning tasks than subjects in the lowest tertile
 5                  (exact Pb levels in tertiles not reported) (van Wijngaarden et al. 2009). In analyses of
 6                  tibia Pb levels, subjects in the highest tertile of tibia Pb level did not consistently perform
 7                  worse  on the various cognitive tests (Table 5-10).


                    Adults with Occupational Pb Exposures

 8                  The 2006 Pb AQCD (U.S. EPA. 2006b) concluded that in adults, blood Pb levels were
 9                  associated with cognitive function more consistently among those with occupational Pb
10                  exposures. These findings were  supported by results from a few recent studies of
11                  occupationally-exposed adults. Several of these associations were found with adjustment
12                  for fewer but a similar set of potential confounding factors as in nonoccupational studies;
13                  however, other occupational exposures were not considered. A prospective analysis was
14                  conducted in former male Pb battery workers whose occupational exposure had ceased
15                  0.02 to 16 years (median: 6) before follow-up testing in 2001-2004  (Khalil et al.. 2009a).
16                  Subjects included 83 of 288 workers (in 2004 mean age: 54 years, median tibia Pb level:
17                  57 ug/g) and 51 of 181 controls  (mean age: 55 years, median tibia Pb  level: 12 ug/g)
18                  from the 1982 Lead Occupational Study in Pennsylvania. While the follow-up
19                  participation was low, participation was not biased to poor performers on cognitive tests
20                  at baseline. In former Pb-exposed workers, a 10 ug/g higher peak tibia Pb levels was
21                  associated with a -0.352 change in total cognitive function score (e.g., learning, memory,
22                  executive function, general intelligence, spatial function, psychomotor speed) between
23                  1982 and 2004. In controls, higher tibia Pb levels were associated with improved
24                  performance on several tests. Results were  adjusted for age, education, income, blood
25                  pressure, years of employment, years since last worked, smoking, alcoholic drinks/week,
26                  and baseline score. Cross-sectional associations indicated stronger associations of
27                  concurrent tibia Pb level than concurrent blood Pb level (median: 12 ug/dL) with poorer
28                  cognitive performance in former Pb-exposed workers. In controls, higher concurrent
29                  blood Pb levels were associated with larger decrements in cognitive performance. As in
30                  nonoccupationally-exposed adults, the stronger findings for tibia  Pb levels in former
31                  Pb-exposed workers indicate stronger effects of long-term cumulative Pb exposures than
32                  recent exposures on cognitive function. The associations for concurrent blood Pb levels
33                  in controls also may reflect effects of past exposures.
      November 2012                             5-135                  Draft - Do Not Cite or Quote

-------
 1                  Blood and tibia Pb levels also were associated with cognitive performance in a follow-up
 2                  of 652 Pb-exposed workers (mean age: 43.4 years, mean blood Pb level: 30.9 ug/dL) in
 3                  Korea, whose patella Pb levels were measured (Dorsey et al., 2006). Higher patella Pb
 4                  levels were associated with poorer manual dexterity, executive function, and verbal
 5                  memory with adjustment for age, sex, education, and job duration. The associations for
 6                  patella Pb level were not as strong as those previously found for either blood or tibia Pb
 7                  levels in these workers (Schwartz et al., 2005; Schwartz et al., 2001).

 8                  Other occupational studies aimed to characterize factors that either mediate or modify the
 9                  association between Pb biomarkers and cognitive function. Both a working lifetime time-
10                  weighted integrated blood Pb level (an index of cumulative exposure) (p = 0.09) and tibia
11                  Pb level (p = 0.08) were associated with longer times to complete the grooved pegboard
12                  test among current Pb smelter workers (Bleecker et al.. 2007b). In the same workers
13                  (n = 112, mean age: 38 years), higher time-weighted integrated blood Pb level was
14                  associated with decrements in executive function, learning, and memory among those
15                  with lower cognitive reserve (i.e., > 12th grade reading level by Wide Range
16                  Achievement Test-R) (Bleecker et al.. 2007a). Subjects with lower and higher cognitive
17                  reserve were matched by blood Pb level (mean: 26 ug/dL), and results were adjusted for
18                  age, depression scale, and current alcohol use.

19                  Apolipoprotein E is a transport protein for cholesterol  and lipoproteins and has been
20                  found to regulate synapse formation (connections between neurons). A genetic variant,
21                  called the ApoE-e4 allele is a haplotype between 2 exonic SNPs and has been associated
22                  with a two-fold increased risk of developing Alzheimer's disease, although the majority
23                  of such individuals still do not develop the disease. Thus, it is biologically plausible that
24                  ApoE-e4 carriers may be biologically susceptible to cognitive dysfunction. A study of
25                  529 U.S. male, former tetra-ethyl Pb workers found that higher peak tibia Pb levels were
26                  associated with lower scores on tests of executive function, vocabulary, and memory
27                  (Stewart et al.. 2002). and for several tests, larger decrements among the 118 men with at
28                  least one ApoE-e4 allele. Results were adjusted for age, race, education, depression,
29                  testing technician, and visit number. The group with at least one ApoE-e4 allele had
30                  slightly higher peak tibia Pb levels (mean: 26.2 versus 23.1 ug/g) and a larger percentage
31                  of non-white subjects but were similar in age, education, and time since employment.


                    Summary of Cognitive Function in Adults

32                  In summary, consistent with evidence described in the 2006 Pb AQCD, recent studies
33                  found that higher bone Pb levels were associated decrements in cognitive function in
34                  adults without occupational Pb exposure (Table 5-10). Much of this evidence was
35                  provided by analyses of the BMS and NAS,  with additional findings reported in the
      November 2012                            5-136                  Draft - Do Not Cite or Quote

-------
 1                  Nurses' Health Study and smaller populations. Nonetheless, the multiple risk factors and
 2                  health outcomes examined in most of these cohorts reduces the likelihood of selection
 3                  bias by Pb exposure or cognitive function. While the NAS and Nurses' Health Study
 4                  included primarily white men and white women, respectively, the BMS examined a more
 5                  diverse population of men and women of several different race/ethnicities. There was
 6                  variability in associations across the various domains of cognitive function tested within
 7                  studies; however, higher bone Pb levels were associated with decrements in most of the
 8                  tests performed. In several populations, higher bone Pb levels were associated with
 9                  decrements in executive function, visuospatial skills, learning, and memory.

10                  Key evidence for bone Pb-associated cognitive decrements was provided by recent
11                  prospective analyses that demonstrated that higher tibia (means 18.8, 20 (ig/g) and patella
12                  (mean 25 (ig/g) bone Pb levels measured at baseline were associated with subsequent
13                  declines in cognitive function over 2- to 4-year periods (Bandeen-Roche et al., 2009;
14                  Weisskopf et al.. 2007b). These findings indicate that long-term Pb exposure may
15                  contribute to ongoing declines in cognitive function in adults. These associations were
16                  found with adjustment for potential confounding by age,  education, smoking, and alcohol
17                  use in the NAS and age, sex,  race, household wealth, and education in the BMS.

18                  Supporting evidence was provided by most cross-sectional analyses that adjusted for
19                  several of the potential confounding factors described above plus dietary factors, physical
20                  activity, medication use,  and  comorbid conditions (Rajan et al., 2008; Weuve et al..
21                  2006). Cross-sectional studies generally demonstrated larger decrements in cognitive
22                  function in adults in association with tibia or patella Pb levels than with concurrent blood
23                  Pb levels. In comparisons of associations  with patella and tibia Pb levels in the NAS and
24                  Nurses' Health Study, tibia Pb levels were not consistently associated with larger
25                  decreases in cognitive performance (Weuve et al.. 2009; Weisskopf et al.. 2007b). In
26                  NHANES analyses, concurrent blood Pb levels were associated with lower cognitive
27                  function in particular age and genetic variant subgroups but not consistently across the
28                  various cognitive tests evaluated (Krieg et al.. 2010; Krieg and Butler. 2009; Krieg et al.,
29                  2009). NHANES did not have bone Pb measures for comparison.

30                  Because bone Pb is a major contributor to blood Pb levels, blood Pb level also can reflect
31                  longer term exposures, including higher past exposures, especially in adults without
32                  occupational exposures. Thus, in the NHANES results, it is difficult to characterize the
33                  relative contributions of recent and past Pb exposures to the  associations observed
34                  between concurrent blood Pb level and cognitive function. In other cohorts, the
35                  discrepant findings for blood and bone Pb levels indicate that cumulative Pb exposure
36                  that likely included higher past exposures, may be a better predictor of cognitive function
37                  in adults than is blood Pb level. Additional support  for the effects of cumulative or past
      November 2012                             5-137                 Draft - Do Not Cite or Quote

-------
 1                   Pb exposure is provided by analyses of a few child cohort as adults, which indicated that
 2                   childhood tooth (from ages 5-8 years) and blood Pb levels (e.g., age 10 years) were
 3                   associated with decrements in cognitive function in adults ages 19-30 years) (Mazumdar
 4                   et al.. 2011; Bellinger et al.. 1994a). An uncertainty related to the evidence for cognitive
 5                   function decrements associated with bone Pb levels is the potential residual confounding
 6                   by age. Although studies adjusted for age, the high correlation between increasing age
 7                   and bone Pb levels (Section 4.3.5) makes it difficult to distinguish the independent effect
 8                   of Pb exposure. However, the coherence with evidence for cognitive function decrements
 9                   associated with long-term Pb exposure in animals provides support for associations
10                   observed in human adults.

11                   Cross-sectional analyses provided information on potential effect modification of bone
12                   Pb- and blood Pb-associated decrements in  cognitive  function in adults by race,
13                   psychosocial stress, and genetic variants. Inconsistencies were found for effect
14                   modification by race in the BMS, ALAD-2  genotype  in the NAS and NHANES, and
15                   VDR genotype in NHANES. Larger tibia Pb-associated decrements in cognitive function
16                   was found in NAS men with HFE variants and in BMS subjects living near more
17                   psychosocial hazards. Evidence does not clearly indicate whether the observed effect
18                   modification reflects chance, a change in the toxicokinetics of Pb that alters Pb dose at
19                   the biological site of action, or a direct biological interaction that increases the toxicity of
20                   Pb in the target tissue. However, such effect modification serves to strengthen inferences
21                   about associations between Pb biomarkers and cognitive function since it is unlikely that
22                   potential confounding factors vary by levels of the modifying factor, particularly genetic
23                   variants. However, because there is  little available evidence and inconsistent evidence for
24                   some factors, firm conclusions regarding effect modification are not warranted.

25                   In contrast with nonoccupationally-exposed adults, in adults with former and current
26                   occupational Pb exposures, cognitive function decrements were associated with both
27                   blood (means: 12 [former workers]-31 ug/dL) and bone Pb levels. Thus, among
28                   Pb-exposed workers,  both current and cumulative Pb  exposures may affect cognitive
29                   function. Several of these studies considered confounding by a  similar set of potential
30                   confounding factors as studies of adults  without occupational Pb exposures but did not
31                   consider other occupational exposures. In the prospective study of former Pb workers,
32                   peak tibia Pb levels were associated more strongly with cognitive performance than were
33                   blood Pb levels (Khalil  et al.. 2009a). Thus, in the absence of higher current Pb
34                   exposures, cumulative Pb exposures may have a greater effect on cognitive function in
35                   adults.
      November 2012                             5-138                  Draft - Do Not Cite or Quote

-------
      5.3.3  Attention-related Behavioral Problems in Children

 1                   The effects of Pb exposure on attention-related behavioral problems such as inattention,
 2                   impulsivity, hyperactivity, and ADHD have not been examined as extensively as effects
 3                   on cognition. Behavioral effects are more complex to study than are cognitive effects,
 4                   particularly FSIQ. There are fewer objective tests of attention-related behavioral
 5                   problems with as strong psychometric properties or as rigorous validation as IQ tests. In
 6                   several studies, attention-related behavioral problems were assessed using teacher and/or
 7                   parent ratings which are subject to greater measurement error. However, domain-specific
 8                   neuropsychological assessments are advantageous as they may provide greater insight
 9                   into whether there is a particular domain more susceptible to the effects of Pb exposure.
10                   As with cognitive function, in the evaluation of epidemiologic evidence for attention-
11                   related behavioral problems, greater emphasis was placed on  evidence from
12                   neuropsychological tests than from parent or teacher ratings and prospective studies with
13                   repeated assessments of blood Pb levels and behavior, studies of older children in whom
14                   outcomes are more reliably measured, and studies of children whose blood Pb levels are
15                   less influenced by higher past Pb exposures. Similar to cognitive function, associations
16                   between blood Pb levels  and attention-related behavioral problems potentially may be
17                   confounded by factors such as parental  SES, education, and IQ, quality and stability of
18                   the caregiving environment, and nutritional status. Accordingly, greater weight was given
19                   to studies with greater consideration for potential confounding in the study design or in
20                   statistical analyses. Consideration also was given to studies assessing effects relevant to
21                   blood Pb levels in contemporary U.S. children (i.e., <5 (ig/dL).

22                   Some studies that found associations between concurrent blood Pb levels and attention-
23                   related behavioral problems were  not major considerations in drawing conclusions and
24                   are not discussed in detail in the following sections. These studies examined populations
25                   that have high prevalence of prenatal alcohol or drug exposure (Chiodo et al.. 2007;
26                   2004). or had earlier childhood  chelation (Chen et al.. 2007).  and thus may not be
27                   representative of current children  in the U.S. population. Others had limited consideration
28                   for potential confounding (Liu et al.. 20 lib), or examined infants in whom inattention
29                   ratings may not predict behavior in later childhood (Plusquellec et al.. 2007).
      November 2012                             5-139                 Draft - Do Not Cite or Quote

-------
      5.3.3.1  Inattention and Impulsivity

                     Epidemiologic studies of Inattention and Impulsivity in Children

 1                   Attention is the ability to maintain a consistent focus on an activity or relevant stimuli
 2                   and can be assessed by examining sustained attention, impulsivity, or distractibility.
 3                   Several epidemiologic studies reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b)
 4                   reported associations between blood, tooth, or bone Pb levels with inattention in children
 5                   ages 8-17 years, including prospective studies described in previous sections for
 6                   cognitive function (Ris et al., 2004; Fergusson et al., 1993; Leviton et al., 1993). As
 7                   described in this section, recent studies also found associations of blood Pb levels with
 8                   inattention in children ages 8-17 years. Many previous studies of inattention included
 9                   children with higher blood Pb levels than those of most current U.S. children. Recent
10                   studies, most of which were cross-sectional, provided evidence of blood Pb-associated
11                   inattention in populations of children with mean concurrent blood Pb levels 2 to 5 (ig/dL
12                   (Cho et al., 2010; Nicolescu et al.. 2010; Plusquellec et al., 2010); however, limitations
13                   include the cross-sectional design of studies and potential influence of higher past Pb
14                   exposures. In the collective body of literature, most evidence was for inattention rated by
15                   teachers, parents, or blinded examiners; however, associations were consistently found
16                   for more objective measures such as the continuous performance test (CPT) (Figure 5-10
17                   and Table 5-11). Thus, evidence does not indicate undue influence by biased reporting of
18                   inattention by parents of children with high Pb exposures. Epidemiologic findings for
19                   inattention and impulsivity are supported by the coherence of findings in Pb-exposed
20                   animals of poorer response inhibition  in Schedule Controlled Behavior Tests and poorer
21                   performance on signal detection tests with distracting stimuli (discussed below). In
22                   particular, both evidence in children and animals indicates Pb-associated poorer
23                   performance on test of response inhibition, i.e., continued responses to stop signals.

24                   Most studies that assessed inattention with the objective CPT found associations with
25                   blood Pb level (Figure 5-10 and Table 5-11). including prospective studies in Cincinnati
26                   and in Chelsea/Sommerville, MA, which indicated associations of higher prenatal or
27                   earlier childhood blood Pb levels or tooth Pb levels with increases in commission and
28                   omission errors or reaction time in adolescents and young adults (Ris et al.. 2004;
29                   Bellinger et al.. 1994a). In the CPT, subjects are assessed for their ability to maintain
30                   focus during a repetitive task and respond to targets or inhibit responses. These findings
31                   from prospective studies characterized the temporal sequence between Pb exposure and
32                   inattention better than cross-sectional  studies and made reverse causation a less likely
33                   explanation for observed associations. These studies recruited cohorts from schools or
34                   prenatal clinics and had moderate to high follow-up participation that was not conditional
35                   on blood or tooth Pb levels, which reduces the likelihood of selection bias. Among

      November 2012                             5-140                  Draft - Do Not Cite or Quote

-------
 1                   primarily white, higher SES young adults ages 19-20 years, Bellinger et al. (1994a) found
 2                   that compared with the group with age 5-8 year tooth Pb levels 2.9-5.9 ppm, the group
 3                   with tooth Pb levels >19.9 ppm had fewer correct response on the CPT and had longer
 4                   reaction times for correct responses but did not commit more commission errors
 5                   (responding to a nontarget). In the mostly African-American, lower SES Cincinnati
 6                   cohort, Ris et al. (2004) found increased inattention (composite of CPT outcomes) in
 7                   association with prenatal maternal, age 3-60 month average, and age 78 month blood Pb
 8                   levels in adolescents ages  15-17 years, particularly among males. Although blood Pb
 9                   levels at older ages were not examined and results do not exclude an effect of more recent
10                   Pb exposures, the combined evidence from these prospective studies points to an effect
11                   on inattention of cumulative earlier childhood Pb exposures. Both studies considered
12                   several potential confounding factors, including SES, parental IQ, maternal education,
13                   and self drug use. HOME score was considered only in the Cincinnati cohort. Ris et al.
14                   (2004) also considered potential confounding by prenatal drug and alcohol exposure,
15                   birth outcomes, and iron status.

16                   Recent studies that assessed inattention with neuropsychological tests, primarily in non-
17                   U.S. populations, found Pb-associated increases in inattention, although all were cross-
18                   sectional design and did not consider potential confounding by parental caregiving
19                   quality. Further, sufficient data were not provided to assess whether participation was
20                   biased to those with higher Pb exposure and inattention. A study in children ages 8-11
21                   years in Korea demonstrated poorer performance on some indices of the CPT with
22                   relatively low blood Pb levels (mean 1.9 (ig/dL) (Cho et al.. 2010): however, the
23                   contribution of higher earlier Pb exposures cannot be excluded.  Specifically, higher
24                   concurrent blood Pb levels were associated with more commission errors, but weakly
25                   with other parameters of the CPT (Figure 5-10 and Table 5-11). Results were adjusted for
26                   age, sex, paternal education, maternal IQ, child IQ, city of residence, birth weight, and
27                   urinary cotinine, the latter of which was more strongly associated with CPT performance
28                   than blood Pb level and the primary cause of attenuation of blood Pb effect estimates.
29                   Cho et al. (2010) found that mean blood Pb levels were similar in children with and
30                   without (1.80 and 1.93 (ig/dL, respectively, p = 0.32) parental report of history of
31                   neuropsychiatric disease (e.g., ADHD, learning disability, depression, obsessive-
32                   compulsive disorder); however, history may not accurately represent current parental
33                   caregiving quality.
      November 2012                             5-141                  Draft - Do Not Cite or Quote

-------
Study
Risetal. (2004)
Choetal. (2010)
Nicolescuetal. (2010)
Niggetal.(2008)
Chiodoetal.(2007)a
Chiodoetal.(2004)a
Bumsetal. (1999)
Wasserman etal. (2001)
Canfieldetal. (2003b)
Nicolescuetal. (2010)
Silvaetal. (1988)
Plusquellecetal.(2010)
Kordasetal.(2007)
Royetal. (2009a)
Chiodoetal.(2007)a
Chiodo etal. (2004)=
Nicolescuetal. (2010)
Silvaetal. (1988)
Plusquellecetal.(2010)
Royetal. (2009a)
Niggetal. (2008)=
Chiodoetal.(2007)a
Chen etal. (2007)
Choetal. (2010)
Chen etal. (2007)
Mean (SD) blood Blood Pb
Pb(ug/dL) interval analyzed
NR NR
1.9(0.67)
3.7 (2.6)
1.04(0.53)
5.0 (3.0)
5.4 (3.3)
5.4 (3.3)
GM: 14.3(13.5-15.1)
GM: 13.9(13.2-14.6)
6.5(1.5)
6.5
3.7 (2.6)
11.1 (4.9)
5.4 (5.0)
11.5(6.1)
11.4(5.3)
5.0 (3.0)
5.4 (3.3)
3.7 (2.6)
11.1 (4.9)
5.4 (5.0)
11.4(5.3)
1.04(0.53)
5.0 (3.0)
12.0(5.2)
1.9(0.67)
12.0(5.2)
1.2-2.8
2.0-8.5
0.5-1.7
2.1-8.7
2.3-9.5
2.3-9.5
13.7-14.9
13.3-14.4
4.7-8.4
Data N/A
2.0-8.5
5.9-10
1.4-10.8
5.4-10
5.8-10
2.1-8.7
2.3-9.5
2.0-8.5
5.9-10
1.4-10.8
5.8-18.3
0.5-1.7
2.1-8.7
6.5-10
1.2-2.8
6.5-10
Outcome
Inattention composite, Continuous Performance Test
Comissions errors, Continuous Performance Test
Response time — <
Go/no go, Test of Attentional Performance
Stoptask
Comission errors, Continuous Performance Test ^
Omission errors
Number of errors, Continuous Performance Test
Comissionserrors
Reaction Time
Attention problems, boys, Child BehaviorChecklist —
Attention problems, girls
Attention problems, Child BehaviorChecklist H
Inattention, control phase, Shape School
Inattention , com pleted phases
Inattention, parent, Conners —
Inattention, teacher
Inattention, parent, Rutter —
Inattention, teacher
Off task duration, InfantBehavioral Rating Scale
Impulsivity
Off task passive
Inattention, Conners
Inattention, AchenbachTeacher Report Form
Inattention, Child Behavior Checklist
Hyperactivity, parent, Conners
Hyperactivity, teacher
Hyperactivity, parent, Rutter
Hyperactivity, teacher
Global activity scale, InfantBehavioral Rating Scale
Hyperactivity, Conners
Hyperactivity/impulsivity, Child BehaviorChecklist
Hyperactivity, teacher, PROBS-14
Hyperactivity,' indirect —
Total ADHD rating, teacher, Korean ADHD Rating Scale
Total ADHD rating, parent
ADHD index, indirect
Inattention/
» * Tmpulsivity
-*-
— » 	
*fc
w&
A It
* . '
a —
i
•
•—
-•—
— • 	
•
•
-•-
-m-
	 • 	
	 • 	
« Hyperactivity
:
•
•
0


t* Total ADHD
^ Rating

                                                                -0.5
                                                                       -0.3
                                                                               -0.1
                                                                                       0.1
                                                                                              0.3
                                                                                                      0.5
                                                               Standard deviation change perl |jg/dL increase in
                                                               blood Pb level in various intervals of blood Pb level


aStandard errors were estimated from p-values or sufficient data were not provided to calculate 95% CIs.
Note: Regression coefficients were scaled to their standard deviation to facilitate comparisons among tests with different scales.
Small effect estimates should not necessarily indicate lack of effect or weak effect. Results are categorized by outcome category:
inattention/impulsivity (with objective tests presented first), hyperactivity ratings, total ADHD ratings. Within categories, results
generally are presented in order of strength of study design. Effect estimates are  standardized to a 1 ug/dL increase in blood Pb
level in the interval from the 10th percentile to the 90th percentile or 10 ug/dL, whichever was lower. For studies with 10th
percentiles of blood Pb level > 10 ug/dL, effect estimates are standardized to a 1  ug/dL increase in blood Pb level in the interval
from the 10th to 90th percentile of blood Pb level. The percentiles are estimated using various methods and are approximate values.
Effect estimates are assumed to be linear within the 10th to 90th  percentile interval of blood Pb level. Gray, orange, blue, and black
symbols represent associations with lifetime average, prenatal (maternal),  earlier childhood, and concurrent blood Pb levels,
respectively.


Figure 5-10      Associations of blood Pb levels with attention-related behavioral
                      problems in children.
November 2012
5-142
Draft - Do Not Cite or Quote

-------
Table 5-11   Additional characteristics and quantitative results for studies
             presented in Figure 5-10.


Study
Bellinger et al.
(1994a)







Risetal.
(2004)












Cho et al.
(2010)





Nicolescu et al.
(2010)











Study Population and
Methodological Details
79 young adults, born 1970, followed
from first grade to age 1 9-20 yr, Boston,
MA area
Prospective. Moderate follow-up
participation. Participation from higher
SES, females, higher initial IQ but no
affect on association with tooth Pb level.
Regression model adjusted for parental
IQ, sex, SES, current drug, alcohol and
illicit drug use, maternal education and
age, birth order. Also considered
potential confounding by other
unspecified factors.
195 children followed prenatally
(1 979-1 985) to age 1 5-1 7 yr, Cincinnati,
OH
Prospective. Recruitment at prenatal
clinic. High follow-up participation, no
selective attrition. Mostly African
American. Linear regression model
adjusted for SES, maternal IQ, HOME,
adolescent marijuana use, obstetrical
complications. Also considered potential
confounding by birth outcomes, maternal
age, prenatal smoking, alcohol,
marijuana, and narcotics use, gravidity,
# previous abortions, stillbirths, parity,
caregiver education, public assistance,
child age, sex, health, Fe status.
667 children ages 8-1 1 yr, born
1 997-2000, 5 Korean cities
Cross-sectional. School-based
recruitment, moderate participation rate.
Log linear regression model adjusted for
age, sex, parental education, maternal
IQ, child IQ, birth weight, urinary
cotinine. Did not consider potential
confounding by parental caregiving
quality.

83 children ages 8-1 2 yr (born
1995-1999), Bucharest and Pantelimon,
Romania
Cross-sectional. Pantelimon near
former metal processing plant. Low
correlations blood Pb with blood Al, Hg.
No information on participation rate. Log
linear regression model adjusted for city,
sex, age, computer experience,
handedness, eye problems, # siblings,
parental education, prenatal smoking
and alcohol use, parental ever having
psychological/psychiatric problem.
Blood /
tooth Pb
Levels
(ug/dL
or ug/g)
Deciduous
tooth (age
5-8 yr)
Q1:
2.9-5.9 ppm
Q2:
6.0-8.7 ppm
Q3:
8.8-19.8 ppm
Q4: 19.9-51.8
ppm

Multiple time
periods
Mean (SD):
Not Reported











Concurrent
|\yia~,n /onV
IvIcaM ^OU^.
1.9(0.67)
Interval
analyzed:
1.2-2.8 =
1 Oth-90th
percentiles

Concurrent
Median
(IQR): 3.7
(2.6)
Interval
analyzed:
2.0-8.5 =
1 0th-90th
percentiles





Outcome
Mean (SE) Correct
Responses per quartile



Mean (SE) Reaction time
errors per quartile
Continuous Performance
Test (CPT)
Ages 19-20yr

Inattention composite,
CPT
Prenatal (maternal)
3-60 mo avg

78 mo
Ages 15-17 yr







Commission errors, CPT

Response time, CPT
Total ADHD rating,
teacher
Total ADHD rating,
parent
Korean ADHD Rating
Scale IV
Ages 8-1 1 yr
Go/no go, KITAP

Inattention parent rating
Inattention teacher rating

Hyperactivity, parent
rating

Hyperactivity, teacher
rating
German Conners Rating
Ages 8-1 2 yr

Effect Estimate
/OCO/ ^l\^
(95% Cl)
Q1: 98.0(1.0)b
Q2: 97.6(1.1)

Q3: 96.9(1.1)
Q4: 94.6 (1.1)
Q1: 361.2(16.5)
Q2: 374.2(17.3)
Q3: 370.7(17.9)
Q4: 385.0(17.1)



0.16(0.04,0.27)
0.11 (0.04,0.19)

0.12(0.02,0.22)








0.03 (-0.01 , 0.07)°

-0.01 (-0.05, 0.03)°
0.042(0.017,0.067)°
0.010 (-0.01 3, 0.033)°


8.9% (-1.3, 19.3)

1 .3% (-3.3, 5.9)d
4.5% (-1.3, 10.3)d

6.3% (-0.60, 13.2)d


5.2% (-2.4, 12.8)d



November 2012
5-143
Draft - Do Not Cite or Quote

-------



Study
Nigg et al.
(2008)








Needleman et
al. (1979)









Chiodo et al.
(2007)


















Chiodo et al.
(2004)

















Study Population and
Methodological Details
1 50 children ages 8-1 7 yr, Birth yr and
location NR
Case-control study of ADHD.
Recruitment with advertisements. Could
have biased participation by Pb
exposure. Regression-based path
analysis adjusted for sex and income.
Did not consider potential confounding
by parental education or caregiving
quality.


158 children in 1st/2nd grade (born
1968-1971), Chelsea, Sommerville, MA
Cross-sectional. Recruitment from
schools. Only 6.7% selected based on
low and high tooth Pb levels. Moderate
participation rate but no selective
participation based on tooth Pb or
teacher ratings. Analysis of covariance
adjusted for paternal SES, maternal age,
# pregnancies, maternal education and
parental IQ. Did not consider potential
confounding by parental caregiving
quality.
506 children, age 7 yr (born 1982-1984),
Detroit, Ml area.
Cross-sectional. Recruitment at
prenatal clinic. All African American.
High prevalence prenatal drug exposure.
High follow-up participation. Linear
regression model adjusted for child sex,
prenatal marijuana use (commission
errors), caregiver education, HOME,
maternal IQ, cocaine use, prenatal
alcohol use and cigarettes/day (omission
errors), child age and sex (hyperactivity),
child age, caretaker education, SES,
HOME, maternal age and IQ, prenatal
alcohol use, current marijuana use
(inattention). Also considered potential
confounding by # children in home,
caretaker marital status, concurrent
alcohol/week, current maternal
cigarettes/day, caregiver concurrent
psychological symptoms, maternal
custody.
246 children, age 7.5 yr, Detroit, Ml area
Cross-sectional. Recruitment at
prenatal clinic. All African American.
High prevalence prenatal alcohol
exposure. High participation rate. Log
linear regression model adjusted for
SES (all outcomes). Prenatal smoking
exposure, caregiver vocabulary (# errors
CPT), caregiver vocabulary, child age
(commission errors, CPT). HOME,
prenatal alcohol and smoking exposure,
disruption in caregiver (Inattention
rating). Also considered potential
confounding by caregiver education,
family functioning, # children <1 8 years,
maternal prenatal marijuana, smoking,
or cocaine use, parity, crowding, child
sex, child life stress, caregiver age, life
stress, and psychology, conflict tactics.
Blood /
tooth Pb
Levels
(ug/dL
or ug/g)
Concurrent
|\yia~,n /onV
IvIcaM ^OU^.
8-11 yr: 1.04
(0.53)
1 2-1 7 yr:
1 .03 (0.54)

Interval
analyzed:
0.5-1 .7 =
1 0th-90th
percentiles
Tooth Pb
(1st/2nd
grade)
High
>27.0 ppm
(n = 58)
Low
<5.1 ppm
(n = 1 00)



Concurrent
Mean (SD):
5.0 (3.0)
Interval
analyzed:
2.1-8.7 =
1 Oth-90th
percentiles













Concurrent
Mean (SD):
5.4 (3.3)
Interval
analyzed*
2.3-9.5 =
1 0th-90th
percentiles













Outcome
Stop task

Hyperactivity/impulsivity
Teacher, parent rating,
Child Behavioral
Checklist, ADHD Rating
Scale
Ages 8-1 7 yr.




Reaction time, 12 sec
delay


% negative response,
teacher rating

Impulsive

Hyperactive

Commission Errors (%),
CPT
Omission Errors (%),
CPT

Inattention, PROBS-14

Hyperactivity, Achenbach
Teacher Report Form
Anp 7 vr
y y











Number of errors, CPT
Commission errors, CPT
Reaction time, CPT

Inattention, Examiner
rating
Child Behavior Checklist
Age 7.5 yr











Effect Estimate
/OCO/ ^l\^
(95% Cl)
0.38(0.16,0.60)°

0.21 (0, 0.42)°







Mean (SD)
Low tooth Pb: 0.35
(0.08)
High tooth Pb: 0.37
CO 09)
\\j.\j*jf
High vs. low tooth Pb

25 vs. 9%, p = 0.01

16 vs. 6%, p = 0.08

-0.08, p >0.05e

0.18(0.07,0.29)°

0.13(0.03,0.23)°

0.13(0.03,0.23)°













0.35 (0, 0.69)°
0.05, p >0.05e
0.25, p >0.05e

0.15(0.04,0.26)°












November 2012
5-144
Draft - Do Not Cite or Quote

-------
Study
Fergusson et
al. (1993)
Chandramouli
et al. (2009)
Leviton et al.
(1993)
Burns et al.
(1999)
Study Population and
Methodological Details
878 children followed from birth to age
13yr, Christchurch, New Zealand
Prospective. Moderate follow-up
participation, attrition did not affect
results. Log linear regression model
adjusted for sex, ethnicity, maternal
education, family size, HOME, SES, #
schools attended. Also considered
potential confounding by ethnicity,
maternal age, paternal education,
breastfeeding duration, parental
smoking, child birth outcomes, residence
on busy roads, weatherboard housing.
488 children followed from age 30 mo
(born 1 991 -1992) to 7-8 yr, Avon, U.K.
Prospective. All births in area eligible.
Similar characteristics as U.K. census,
high participation at baseline and follow-
up. Participants had better educated
mothers, who smoked less, better home
environment. Regression model
adjusted for maternal education and
smoking, home ownership, home
facilities score, family adversity index,
paternal SES, parenting attitudes at 6
mo, child sex. Also considered potential
confounding by child IQ.
1 ,923 children followed from birth
(1979-1980) to age 8 yr, Boston, MA
Prospective. Recruitment from birth
hospital. High participation at baseline
and follow-up. Log linear regression
model adjusted for single parent family,
gestational age, maternal education,
ethnicity, # children, daycare in first
3 years. Also considered potential
confounding by other unspecified
factors.
322 children followed from birth
(1979-1 982) to age 11-1 Syr, Port Pirie,
Australia.
Prospective. Moderate follow-up
participation. Participants had higher
birth weight, older mothers, less
educated fathers. Log linear regression
model adjusted for maternal age,
prenatal smoking status, birth weight,
type of feeding, length of breastfeeding,
maternal education, maternal IQ,
paternal education, concurrent maternal
psychopathology, birth order, family
functioning, paternal occupation, parent
smoking, marital status, HOME, child IQ.
Blood /
tooth Pb
Levels
(ug/dL
or ug/g)
Tooth Pb
(age 6-8 yr)
Mean (SD):
6.2 (3.7) ug/g
Age 30 mo
Mean (SD):
Not Reported
Group 1 : 0-2
Group2: 2-5
Group 3:
5-10
Group 4: >10
Prenatal
(cord) Mean:
6.8
Tooth Pb
(Age 6 yr)
Mean: 3.3
Lifetime avg
(to age 11-13
yr) blood
Boys:
GM: 14.3
(5th-95th)
(13.5-15.1)
10th-90th:
13.7-14.9
Girls:
GM13.9
(5th-95th)
13.2-14.6)
10th-90th:
13.3-14.4
Outcome
Inattention/restlessness
Rutterand Conners'
ratings
Age 13yr

Selective inattention
Test of Everyday
Attention for Children
Ages 7-8 yr
Hyperactivity, teacher
Strengths and Difficulties
Questionnaire,
Ages 7-8 yr
Daydreaming
Prenatal (cord)
Tooth Pb
Boston Teacher
Questionnaire, Age 8 yr
Attention problems, boys
Attention problems, girls
Maternal rating by Child
Behavior Checklist
Age 11-1 Syr
Effect Estimate
(95% Cl)a
0.06(0, 0.12)b'°
OR vs. 0-2 ug/dLas
reference
2-5 ug/dL:
0.97 (0.62, 1 .52)'
5-10 ug/dL:
1.01 (0.64, 1.61)f
>10 ug/dL:
0.88(0.42,1.85)''
2-5 ug/dL:
0.84 (0.47, 1 .52)'
5-10 ug/dL:
1.25(0.67,2.33)'
>10ug/dL:
2.82(1.08,7.35)'
RR (yes/no) per loge
increase
Girls: 1.3 (0.8, 2.2)'
Boys: 1.0(0.6, 1.5)'
Girls: 1 .5 (0.9, 2.6)b'f
Boys: 1.1 (0.7, 1.7)"''
0.02 (-0.04, 0.09)
0.07(0.02,0.12)
November 2012
5-145
Draft - Do Not Cite or Quote

-------



Study
Wasserman et
al. (2001)






Canfield et al.
(2003b)













Silva et al.
(1988)







Plusquellec et
al. (201 0)











Study Population and
Methodological Details
191 children followed from birth to age
5 yr, Pristina, Yugoslavia.
Cross-sectional. High follow-up
participation. Participants had lower
maternal education, were Albanian, had
higher age 4 blood Pb. Generalized
estimating equations with log blood Pb
adjusted for child sex and age, ethnicity,
HOME, maternal education, birth weight,
maternal smoking. Did not consider
potential confounding by maternal IQ.
150 children born 1994-1995 followed
from age 6 mo to age 4.5 yr, Rochester,
NY
Cross-sectional. Recruitment from
study of dust control. 73% nonwhite.
High follow-up participation, no
comparison of nonparticipants. Linear
mixed effects model adjusted for age,
gestational age, maternal IQ and
education, HOME, race, color/shape
knowledge, child IQ (control phase),
birth order, marital status, race
(completed phases). Also considered
potential confounding by child sex, birth
weight, household income, prenatal
smoking exposure.
535 children age 1 1 yr (born
1972-1973), Dunedin, New Zealand
Cross-sectional. Moderate participation
rate. Participants were of higher SES
and non-Maori. Log linear regression
adjusted for SES, maternal verbal skills,
change in residence and school, solo
parenting, child/parent separation,
maternal age at first birth, family
relations, marriage guidance sought,
maternal mental health symptoms, child
sex, IQ. Did not consider potential
confounding by parental caregiving
quality.
90-98 children, ages 5-6 years (born
1993-1996), Inuit communities, Quebec,
Canada
Cross-sectional. Study of multiple
exposures. Low but no selective
participation by Pb, PCBs, Hg. Log linear
regression model adjusted for birth
weight, sex, parity, caregiver education
(impulsivity) and birth weight, SES, child
blood hemoglobin (off task duration).
Also examined potential confounding by
# children in home, residents per room,
caretaker psychological distress,
nonverbal reasoning, and linguistic
acculturation, HOME, prenatal alcohol
and illicit drug use and cigarettes/day,
serum Se, fatty acids.
Blood /
tooth Pb
Levels
(ug/dL
or ug/g)
Concurrent
Mean (SD):
6.5 (1 .48)
Interval
analyzed:
4.7-8.4 =
1 Oth-90th
percentiles


Concurrent
Mean: 6.5
10th-90th:
data not
available











Concurrent
Mean (SD):
11.1 (4.91)
Interval
analyzed: 5.9
(10th
percentile)-
10





Concurrent
Mean (SD):
5.4 (5.0)
Interval
analyzed:
1.4-10.8 =
1 0th-90th
percentiles











Outcome
Attention problems
Maternal rating, Child
Behavior Checklist
Repeated measures
ages 4-5 yr.




Inattention, control phase
Inattention, inhibit phase
Examiner rating during
Shape School Task
Repeated measures at
ages 4 and 4.5 yr









Inattention, parent
Inattention, teacher
Hyperactivity, parent
Hyperactivity, teacher
Rutter Behavior
Questionnaire





Off task duration
Impulsivity
Global activity rate
Examiner ratings
modified Infant
Behavioral Rating Scale
at ages 5-6 yr.










Effect Estimate
/OCO/ ^l\^
(95% Cl)
0 (-0.02, 0.02)







0.01 (-0.01,0.04)
0.008 (-0.02, 0.04)













0.06 (-0.03, 0.16)
0.15(0.06,0.25)
0.13(0.03,0.23)
0.12(0.01,0.22)






0.02 (0, 0.039)
0.019(0.001,0.036)
0.014(0.006,0.022)








November 2012
5-146
Draft - Do Not Cite or Quote

-------



Study
Kordas et al.
(2007)









Roy et al.
(2QQ9a)







Rabinowitz et
al. (1992)








Chen et al.
(2007)











Study Population and
Methodological Details
1 57 children ages 6-8 yr (born
1993-1995), Torreon, Mexico.
Cross-sectional. 26% of larger study
selected for classroom observation.
Residence near metal foundry. Linear
regression model adjusted forage, sex,
SES, home ownership, crowding in
home, maternal education, family
structure, forgetting homework. Also
considered potential confounding by
micronutrients but not parental
caregiving quality.
756 children ages 3-7 yr (born
1998-2003), Chennai, India
Cross-sectional. Recruitment at
schools. No information provided on
participation. Log linear regression
model adjusted for age, sex,
hemoglobin, average monthly income,
parental education, number of other
children, clustering in school and
classroom. Did not consider potential
confounding by parental caregiving
quality.
493 children, grades 1-3, Taiwan
Cross-sectional. Some reside near
smelter. High participation rate. Logistic
regression model adjusted for sex,
grade, # adults at home, child longest
hospital stay. Also considered potential
confounding by parental education, SES,
birth outcomes, handedness, language
at home, prenatal medicine, alcohol,
smoking but not parental caregiving
quality.
780 children in TLC trial followed ages
2-7 yr, Baltimore, MD; Cincinnati, OH;
Newark, NJ; Philadelphia, PA
Cross-sectional. Mostly African
American. 50% given chelation at ages
12-33 mo, blood Pb levels 20-44 ug/dL
No information on participation rate.
Regression-based path analysis
adjusted for city, race, sex, language,
parental education, parental
employment, single parent, age at blood
Pb measurement, caregiver IQ.
Considered potential confounding by
chelation but not parental caregiving
quality.
Blood /
tooth Pb
Levels
(ug/dL
or ug/g)
Concurrent
Mean (SD):
11.5(6.1)
Interval
analyzed: 5.4
(10th
percentile)-
10




Concurrent
Mean (SD):
11.4(5.3)
Interval
analyzed: 5.8
(10th
percentile)-10




Tooth Pb
(grades 1 -3)
Mean (SD):
4.6 (3.5)






Concurrent
Mean (SD)'
12.0(5.2) '
Interval
analyzed: 6.5
(10th
percentile)-
10









Outcome
Off task passive behavior
Examiner rating,
instrument developed by
investigator
Ages 6-8 yr






Inattention z-score
Hyperactivity z-score

Teacher ratings,
Conners' ADHD/DSM-IV
Scales Ages 3-7 yr.




Hyperactivity Syndrome
Boston Teacher
Questionnaire
Grades 1-3






Hyperactivity Index



ADHD index



Parent ratings, Conners
Scale-Revised, Age 7 yr




Effect Estimate
/OCO/ ^l\^
(95% Cl)
0.034 (0.005, 0.063)










0.031 (0.006, 0.056)
0.01 7 (-0.005, 0.039)







OR vs. <2.3 ug/g
reference
9 "V7 Lin/rr
^.o / HU'U- k
1.9(0.53,7.8)"
>7 ug/g:
2.8 (0.68, 2.8)b




Direct:
0.08 (-0.06, 0.22)
Indirsct'
0.04 (-0.06, 0.13)
Direct:
0.04 (-0.1 0,0.18)
Indirect:
0.07(0.03,0.11)
Direct = independent
of IQ.
Indirect = mediated
through IQ
November 2012
5-147
Draft - Do Not Cite or Quote

-------



Study
Froehlich et al.
(2009)












Study Population and
Methodological Details
2,588 children, ages 8-15 yr (born
1986-1996), U.S. NHANES 2001-2004
Cross-sectional. U.S. representative
results, study of multiple risk factors and
outcomes, high participation rate.
Logistic regression adjusted for current
household ETS exposure, sex, age,
race/ethnicity, income, preschool
attendance, maternal age, birth weight,
and interaction terms for Pb and prenatal
ETS interaction. Did not consider
potential confounding by parental
education orcaregiving quality.
Blood /
tooth Pb
Levels
(ug/dL
or ug/g)
Concurrent
Tertile 1: <0.8
Tertile 2:
0.9-1.3
Tertile 3: >1.3










Outcome
ADHD DSM-IV criteria
met
Parental rating, ADHD
DISC module
Age 8-1 5 yr









Effect Estimate
/OCO/ ^l\^
(95% Cl)
OR vs. <0.8 ug/dL
0.9-1 .3 ug/dL:
1.7(0.97,2.9)'

>1.3 ug/dL:
2.3(1.5,3.8)'







      "Effect estimates are standardized to a 1 ug/dL increase in blood Pb level in the interval from the 10th percentile of blood Pb level
      to 10 ug/dLorthe 90th percentile, whichever is lower and scaled to the standard deviation of the test score to facilitate
      comparisons among tests that are scored on different scales. For studies with 10th percentiles of blood Pb level > 10 ug/dL, effect
      estimates are standardized to a 1  ug/dL increase in blood Pb level in the interval from the 10th to 90th percentile of blood Pb level.
      ""Results for tooth Pb not presented in Figure 5-10.
      ""Standard error was estimated from the reported p-value.
      dResults represent the change in false alarm rate.
      Sufficient data were not provided to calculate 95% CIs.
      'Results not presented in Figure 5-10 because OR or RR reported in papers.
 1
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17

18
19
20
21
22
Response inhibition is a measure of impulsivity and in children has been assessed with
stop signal tasks, which measures the execution of action and the inhibition of that action
when given a stop signal. Recent cross-sectional studies found that children with higher
concurrent blood Pb levels had increased responses with stop signals. Among children
ages 8-12 years in Romania, a 1  (ig/dL increase in concurrent blood Pb level was
associated with a 8.9% (95% CI: -1.3, 19.3) increased false-alarm rate in responses to
stop signals with adjustment for city, sex, age, computer experience, handedness, eye
problems, number of siblings, parental education, prenatal alcohol and smoking exposure,
and parental report of parental psychopathology (Nicolescu et al.. 2010). It is uncertain
how history of parental psychopathology may be related to current caregiving quality.
Children in one town lived near  a metal processing plant; however, blood Pb  levels of Al
and Hg (other neurotoxic metals), were not associated with the stop signal task. In a case-
control study of children with ADHD, Nigg et al. (2008) found that higher concurrent
blood Pb level was associated poorer response inhibition on a stop task, which in turn
was associated with higher hyperactivity/impulsivity ratings. Path analysis showed that
the association between blood Pb level and hyperactivity/impulsivity ratings was
mediated by poorer performance on the stop task.

In addition to objective tests of attention, studies found Pb-associated increases in
inattention as rated by teachers, parents, and independent examiners (Figure 5-10 and
Table 5-11). Results from prospective studies indicated associations in children ages 7-13
years in Australia, New Zealand, and Boston, MA (Burns etal., 1999; Fergusson et al..
1993; Leviton et al.. 1993) but less  so in younger children ages 4-5 years in Rochester
      November 2012
                             5-148
Draft - Do Not Cite or Quote

-------
 1                   and Yugoslavia (Canfield et al.. 2003b; Wasserman et al.. 2001). These studies had
 2                   population-based recruitment and moderate to high follow-up, without indication of
 3                   biased participation by children with higher blood Pb levels and attention problems.
 4                   Among older children, inattention was associated with higher lifetime (to age 11-13
 5                   years) average (Burns etal.. 1999). and tooth (collected at ages 6-8 years) Pb levels
 6                   (Fergusson et al.. 1993; Leviton et al.. 1993). In separate cohorts, lifetime average blood
 7                   Pb level (Burns et al.. 1999) and tooth Pb level (Leviton et al.. 1993) were associated
 8                   with higher inattention ratings among girls than boys. With the exception of the Boston,
 9                   MA analysis, consideration of potential confounding did not differ widely among the
10                   prospective studies, with most adjusting for maternal education, SES, and parental
11                   caregiving quality (e.g. HOME) (Table 5-11). Prospective studies that found associations
12                   examined higher blood Pb levels. The mean lifetime (to age 11-13 years) average blood
13                   Pb level in the Port Pirie, Australia cohort was ~14 (ig/dL (Burns et al.. 1999).
14                   Chandramouli et al. (2009) did not find an association between higher age 30 month
15                   blood Pb levels and higher ratings of inattention in U.K. children ages 7-8 years.
16                   Associations with inattention were not found in the Rochester and Yugoslavia cohorts
17                   with lower concurrent blood Pb levels, means 6-7  (ig/dL (Canfield et al.. 2003b:
18                   Wasserman et al., 2001). The studies in the Rochester and Yugoslavia cohorts examined
19                   lower blood Pb levels but also had smaller sample sizes and examined younger children
20                   ages 4-5 years, in whom patterns of behavior are less well established and in whom
21                   inattention ratings may be less reliably measured. Nonetheless, these few weak findings
22                   do not mitigate the otherwise compelling evidence, including that for attention-related
23                   behavioral problems assessed with neuropsychological tests and that observed in animals.
24                   Canfield et al. (2003b) found associations between higher concurrent blood Pb level and
25                   higher ratings of inattention, but they were attenuated with adjustment for child color and
26                   shape knowledge and FSIQ, which suggested that poorer knowledge of the task
27                   parameters may increase distraction.  However, other studies found associations between
28                   blood Pb level and inattention with adjustment for child IQ (Cho etal.. 2010; Nigg et al..
29                   2008). indicating the relationship between inattention and cognitive function may vary
30                   across populations.

31                   Several cross-sectional studies found associations  between higher concurrent blood Pb
32                   level and inattention among children, with several studies examining older children, 8-17
33                   years (Nicolescu et al.. 2010; Nigg et al.. 2008;  Silvaetal.. 1988). These  studies had
34                   population-based recruitment but did not provide sufficient information to assess
35                   potential selection bias. While most of these studies examined potential confounding by
36                   parental education or cognition, and parental history of psychopathology, none
37                   considered parental caregiving quality. Whereas a previous study in New Zealand
38                   examined children with relatively high blood Pb levels (mean 11 (ig/dL) (Silva et al..
39                   1988). some recent studies provided evidence of association between blood Pb level and

      November 2012                             5-149                  Draft - Do Not Cite or Quote

-------
 1                   higher ratings of inattention in populations with relatively low concurrent blood Pb levels
 2                   (means: 1, 3.7 (ig/dL) (Nicolescu et al.. 2010; Nigg et al.. 2008). However, contributions
 3                   from higher past Pb exposures cannot be excluded. Past Pb exposures especially may
 4                   have an influence in the study of children ages 8-12 years in Romania, 55% (n = 46/83)
 5                   of whom lived near a former metal processing plant and had higher concurrent blood Pb
 6                   levels (mean 5.1 versus 3.2 (ig/dL) (Nicolescu et al.. 2010). Among all subjects, a
 7                   1 (ig/dL increase in concurrent blood Pb level was associated with a 4.5% higher (95%
 8                   CI: -1.3, 10.3%) teacher rating of inattention and a weaker, imprecise 1.3% (95% CI:
 9                   -3.3,5.9) higher parent rating. Blood Pb levels were not correlated with blood levels of
10                   Al or Hg, and neither of these other metals were associated with inattention ratings. The
11                   association did not change substantially in an analysis that excluded the 5 children with
12                   blood Pb levels  > 10 (ig/dL. Adjustment for city, sex, age, computer experience,
13                   handedness, eye problems, number of siblings, parental education, prenatal smoking
14                   exposure, prenatal alcohol exposure, and parental history of psychological or psychiatric
15                   problems resulted in less precise blood Pb level effect estimates, although investigators
16                   did not report the magnitude of change in the effect estimate.

17                   Consistent with Nicolescu et al. (2010). Nigg et al. (2008) found associations of
18                   concurrent blood Pb levels with parent and teacher ratings of a composite
19                   hyperactivity/impulsivity index in a group of children (location not reported)  with and
20                   without ADHD  (ages 8-17 years) with a mean blood Pb level of ~1 (ig/dL.  The case-
21                   control design of the study could have resulted in biased participation of ADHD children
22                   with higher blood Pb levels. Nigg et al. (2008) also found the Pb-associated increase in
23                   hyperactivity/impulsivity to be independent of the association with IQ using regression-
24                   based path analysis, a more rigorous method to characterize the impact of one variable on
25                   the association of another in the model after controlling for other previous variables. With
26                   adjustment for sex and income, concurrent blood Pb level was directly associated with
27                   hyperactivity/impulsivity, and the association was not completely mediated by the blood
28                   Pb-IQ association. Instead, the association between blood Pb level and IQ was found to
29                   be mediated by the association with hyperactivity/impulsivity. Other potential
30                   confounders including parental IQ and caregiving quality were not examined. Other
31                   recent cross-sectional studies of older children found associations between higher
32                   concurrent blood or hair Pb level and higher ratings of inattention (independent examiner
33                   or parent) in children in Mexico and China living near Pb sources (Bao et al., 2009;
34                   Kordas et al.. 2007). As in most other cross-sectional studies, these results were adjusted
35                   for SES and parental education.

36                   Cross-sectional  studies that included younger children (ages 3-5 years) also found
37                   associations between concurrent blood Pb level and higher inattention as  rated by
38                   teachers or study examiners (Plusquellec et al., 2010; Roy et al.. 2009a). In these younger
      November 2012                             5-150                  Draft - Do Not Cite or Quote

-------
 1                   children, attention-related behaviors may be less reliably measured and may not predict
 2                   later childhood behavior. A study conducted in Inuit children, ages 5-6 years, living in
 3                   Quebec, Canada reported associations between blood Pb levels and measures of
 4                   inattention with consideration of several potential confounding factors (Plusquellec et al.
 5                   2010). Concurrent blood Pb level but not cord blood Pb level was associated with
 6                   impulsivity and duration of off task behavior as rated by study examiners (Plusquellec et
 7                   al., 2010). Fraser et al. (2006) additionally indicated that at ages 5-6 years, the
 8                   relationship between concurrent blood Pb level and motor function (i.e., transversal sway,
 9                   reaction time) may be mediated by the association between blood Pb level and
10                   inattention/impulsivity. The various associations were adjusted for different factors but
11                   included SES, caregiver education, birth weight, and blood hemoglobin. HOME score
12                   and micronutrient levels were not associated with inattention or impulsivity and thus
13                   were not included in models.  In this population that has high consumption offish, blood
14                   levels of poly chlorinated biphenyls and Hg were not associated with inattention or
15                   impulsivity ratings. Other recent cross-sectional studies that included younger children
16                   (ages 3-7 years) found associations between higher concurrent blood Pb level and higher
17                   parent or teacher ratings of inattention in populations with higher blood Pb levels (mean
18                   11.4 (ig/dL or median 13.2 (ig/dL) (Liu et al.. 20 lib: Roy et al.. 2009a). While these
19                   studies adjusted for or considered potential confounding by  SES, child age, and parental
20                   education, they did not examine potential confounding by parental caregiving quality.


                     lexicological Studies of  Inattention and Impulsivity

21                   The associations described in the preceding section between blood Pb level and
22                   inattention and impulsivity in children are supported by findings in animals for
23                   Pb-induced impaired ability to inhibit inappropriate responses and increased
24                   perseveration. In animals, tests of response inhibition include  Signal Detection with
25                   Distraction, Differential Reinforcement of Low Rates of Responding (DRL), Fixed
26                   Interval (FI) testing, FI with Extinction, or Fixed Ratio (FR)/waiting-for-reward, with
27                   impulsivity indicated by premature responses, decreased pause time between two
28                   scheduled events, and increased perse veration. Some of these  tests also have been used to
29                   assess learning (Section 5.3.2.3). and the  interactions observed between Pb exposure and
30                   maternal or offspring stress also  may apply to effects on impulsivity. Multiple earlier
31                   studies and those included in  the 2006 Pb AQCD showed that early life Pb exposure
32                   impaired response inhibition  as assessed with these aforementioned tests, and recent
33                   studies provide supporting evidence. Discrimination reversal,  which also measures
34                   response inhibition by rewarding the withholding  of responses, has been shown to be
35                   affected by Pb exposure. Spatial and non-spatial discrimination reversal (i.e., reversal of a
36                   previously learned habit) was significantly affected after developmental Pb exposure and
      November 2012                             5-151                  Draft - Do Not Cite or Quote

-------
 1                  was exacerbated with distracting stimuli. The collective evidence indicates that
 2                  impulsivity in rodents and nonhuman primates is significantly affected by Pb exposure
 3                  that results in blood Pb levels in the range relevant to humans, i.e., 11-31 (ig/dL.

 4                  Toxicological studies provide more consistent evidence for the effects of Pb exposure on
 5                  impulsivity in animals than on sustained attention. The 2006 Pb AQCD (U.S. EPA.
 6                  2006b) reported inconsistent findings for the effects of Pb exposure on sustained attention
 7                  deficits in animals as assessed by a signal detection test with distracting stimuli, a test
 8                  recording omissions after exposure to an external distraction. In this test, animals earn
 9                  food rewards by discriminating correctly between a target and distracter light.
10                  Postweaning Pb exposure that produced blood Pb levels of <5, 16, or 28 ug/dL did not
11                  affect performance in the signal detection test with distracting stimuli in adult rats
12                  (Brockel and Cory-Slechta. 1999a). A similar lack of effect was reported in a recent study
13                  of female rats exposed to 20 or 300 ppm Pb acetate in drinking water during lactation
14                  (PND1-PND30) with resultant blood Pb levels on PND52 of 13 or 31 (ig/dL, respectively
15                  (Stangle et al., 2007). However, in this study, Pb exposure induced impulsivity as
16                  indicated by premature responses in a discrimination reversal task. Impulsivity was not
17                  improved with the chelator succimer, indicating persistence of effects. Impulsivity was
18                  found in monkeys exposed to Pb from birth to time of testing at age 3-4 years with blood
19                  Pb levels  15 and 25 (ig/dL; however, effects were reversible, as Pb-exposed monkeys did
20                  not improve performance as quickly but eventually acquired reinforcement rates equal to
21                  that in controls (Rice and Gilbert. 1985). Previous evidence indicated that Pb exposure of
22                  laboratory animals induces distractibility. Spatial and non-spatial discrimination reversal
23                  was significantly  affected after lifetime Pb exposure in monkeys ages 9-10 years (blood
24                  Pb levels  15, 25 (ig/dL) and was  exacerbated with distracting stimuli. Repeated learning
25                  testing revealed that these deficits likely were not due to sensory or motor impairment
26                  (Gilbert and Rice. 1987).

27                  The effects of Pb exposure on inattention and impulsivity in the 1986 Pb AQCD were
28                  indicated by aberrant performance on operant conditioning tasks in rodents and
29                  non-human primates (U.S. EPA.  1986b). The 2006 Pb AQCD reported consistent
30                  findings for Pb exposure (producing blood Pb levels: 58-94 (ig/dL) affecting FI response
31                  rates, by means of decreased interresponse times. Some studies indicated decreased
32                  interresponse times in animals with blood Pb levels 11-15 (ig/dL (U.S. EPA, 2006b). The
33                  effects of Pb exposure on impulsivity also have been demonstrated as shorter time
34                  Pb-exposed animals will wait for reward in FR/waiting for reward testing. In this test,
35                  animals can obtain food by pressing a lever for a set number of times. Free food is
36                  delivered with increasingly long  time intervals so long as animals inhibit additional lever
37                  presses. Animals  can reset the schedule to return to the FR component at any time.
38                  Brockel and Cory-Slechta (1998) exposed male Long-Evans rats to 0, 50, or 150 ppm
      November 2012                             5-152                  Draft - Do Not Cite or Quote

-------
 1                  Pb acetate in drinking water from weaning, which produced respective blood Pb levels of
 2                  <5, 11, and 29 ug/dL after 3 months of exposure. After 40 days of exposure, the 150 ppm
 3                  Pb-exposed rats responded more quickly in the FR component and reset the schedule
 4                  (thus shortening the waiting period) more frequently than did the 50 ppm Pb-exposed rats
 5                  and controls. In the waiting component,  wait time was significantly lower in both Pb
 6                  exposure groups compared to controls. The behavior of the 150 ppm Pb-exposed rats
 7                  suggested a low tolerance for waiting, but 150 ppm Pb exposure also yielded more
 8                  reinforcers per session and a higher response-to reinforcement-ratio than achieved by the
 9                  50 ppm Pb group and controls. Mechanistic understanding of the aforementioned
10                  Pb-induced impulsivity was provided by a study with similar postweaning dosing of 0,
11                  50, and 150 ppm Pb that yielded respective blood Pb levels of <5, 10, and 26 ug/dL after
12                  3 and 7 months of exposure. Administration of a  D2 receptor agonist reversed the
13                  Pb-induced parameters assessed by FR schedule testing, suggesting a role for dopamine-
14                  like receptors in Pb-induced impulsivity (Brockel and Cory-Slechta. 1999b).

15                  Pb-induced impulsivity appears to be related to emotionality, which was found in
16                  Pb-exposed rats trained to perform an olfactory discrimination task,  albeit at higher Pb
17                  exposures than those relevant to humans. In this study, rats were given early postnatal Pb
18                  exposure (300 ppm Pb acetate via dam drinking water PND1-PND17 then either 20 or
19                  300 ppm PND18-PND30 in their own drinking water) which produced blood Pb levels of
20                  40-60 and 100-140 (ig/dL). The offspring were tested as young adults on a food-
21                  motivated olfactory discrimination task in which  rewards for correct responses were
22                  occasionally and unpredictably omitted.  Pb-exposed animals were more sensitive both to
23                  their own errors and to reward omission than controls, suggesting a lowered capacity for
24                  regulating arousal and emotion. Administration of succimer, a chelating agent, after the
25                  Pb exposure period (PND31-PND52) normalized reactivity to reward omission and errors
26                  in the Pb-treated rats, but increased the reactivity in the control animals (Beaudin et al..
27                  2007). Similar observations were made by the same laboratory for heightened reactivity
28                  to errors in tests of visual discrimination and visual sustained attention in rats exposed to
29                  20 or 300 ppm Pb acetate in drinking water PND1-PND30 (Stangle et al.. 2007)

30                  In animals, Pb-induced increases in inattention have been indicated using tests that are
31                  not direct assessments of inattention but that examine behaviors linked to inattention. For
32                  example, a study reported that impaired  performance on auditory threshold tasks in
33                  Pb-exposed monkeys was likely due to inattention (Laughlin et al.. 2009). Rhesus
34                  monkeys were exposed to Pb acetate from gestation (drinking water of mothers, 3 months
35                  prior to mating) to birth or postnatally from birth to age 5.5 months at weaning and had
36                  resultant bone Pb levels at 11 years of 7  and  13 (ig/g for prenatal and postnatal groups,
37                  respectively, and blood Pb levels during Pb exposure of 35 and 46 (ig/dL, respectively.
38                  Animals were tested at age 13 years when blood Pb levels had returned to baseline levels.
      November 2012                            5-153                 Draft - Do Not Cite or Quote

-------
 1                   The inability of some of the monkeys to engage or focus attention on the task at hand
 2                   yielded fewer available measurements in Pb-exposed animals versus control rhesus
 3                   monkeys. These observations were made in monkeys with higher peak blood Pb levels
 4                   than those relevant to humans.

 5                   In summary, several studies in animals indicate that early developmental Pb exposure of
 6                   rodents and non-human primates, some producing blood Pb levels relevant to humans,
 7                   increases impulsivity as indicated by impaired response inhibition. Evidence for
 8                   Pb-induced decrements in sustained attention is less consistent. The observations for
 9                   Pb-induced increases in impulsivity in animals provide support for associations observed
10                   in children between blood and tooth Pb levels and impaired response inhibition and also
11                   higher ratings of inattention and impulsivity.
      5.3.3.2  Hyperactivity

12                   Studies reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b) indicated associations
13                   between higher concurrent blood Pb level or tooth Pb level and higher parent and teacher
14                   ratings of hyperactivity in children ages 6-11 years in the U.S. and Asia (Rabinowitz et
15                   al.. 1992: Silvaetal.. 1988: Gittleman and Eskenazi. 1983: Needleman et al.. 1979:
16                   David et al.. 1976). The case-control or cross-sectional design of studies limited
17                   understanding of the temporal sequence between Pb exposure and hyperactivity. Several
18                   recent studies, including a prospective study (Chandramouli et al., 2009). also found
19                   associations between blood Pb level and hyperactivity as rated by teachers and parents
20                   (Figure 5-10 and Table 5-11). Overall studies indicated associations with mean or group
21                   blood Pb levels >10 (ig/dL.

22                   The recent prospective study of children in the U.K. addressed some of the limitations of
23                   previous cross-sectional studies by demonstrating an association between higher earlier
24                   childhood (age 30 months) blood Pb level and higher teacher ratings of hyperactivity
25                   later in childhood at age 7-8 years with adjustment for several potential  confounding
26                   factors, including home facilities score and family adversity index (Chandramouli et al..
27                   2009). In addition to the prospective design, the study had a high participation rate at
28                   baseline and follow-up from a population with similar characteristics as reported in the
29                   U.K. census. Increases in hyperactivity were found primarily in the group of children
30                   with blood Pb levels >10 (ig/dL and were independent of associations with IQ.

31                   Among cross-sectional studies, adjustment for SES, maternal education and IQ was
32                   common; however, few adjusted for parental caregiving quality. Silva et al. (1988) and
33                   Nicolescu et al. (2010) respectively, adjusted for current and history of maternal
34                   psychopathology, whose relationships with parental caregiving quality are not well

      November 2012                            5-154                  Draft - Do Not Cite or Quote

-------
 1                   characterized. Both studies found associations of concurrent blood Pb level with
 2                   hyperactivity as rated by teachers and parents, and Silva et al. (1988) found the
 3                   association in children age 11 years in New Zealand to persist with adjustment for child
 4                   IQ. The group of children in New Zealand (Silva et al.. 1988) had a higher mean
 5                   concurrent blood Pb level than the group in Romania (Nicolescu et al.. 2010) (11.1 versus
 6                   3.7 (ig/dL). Other cross-sectional studies found associations with hyperactivity in
 7                   younger children, in whom behavior may be rated less reliably. Plusquellec et al. (2010)
 8                   found an association in children ages 5-6 years with relatively low concurrent blood Pb
 9                   levels, mean 5.4 (ig/dL, and found that HOME score and caretaker distress were not
10                   associated with hyperactivity. Roy et al. (2009a) found a Pb-associated increase in
11                   hyperactivity in children ages 3-7 years in India with a mean concurrent blood Pb level of
12                   11.4(ig/dL.

13                   Pb also has been associated with hyperactivity in animals, but the relevance to
14                   observations in children is not clear. In a recent study, Pb exposure from gestation to the
15                   early postnatal period (PND10) (low and high dose Pb: 10 and 42 (ig/dL blood Pb level at
16                   PND10,  respectively) increased activity of male mice  at age 1 year with co-treatment
17                   with amphetamines but not female mice (Leasure et al.. 2008). Without amphetamines,
18                   Pb induced less activity of mice, and the low Pb dose inhibited activity more than the
19                   high Pb dose did. In addition to the effects of Pb on impulsivity,  Stangle  et al. (2007)
20                   found Pb-induced decreases in arousal. In one theory of ADHD,  low arousal levels
21                   contribute to excessive self-stimulation or hyperactivity such that an optimal level of
22                   arousal can be attained (Swanson et al.. 2011).
      5.3.3.3  Ratings of Attention Deficit Hyperactivity Disorder-related Behaviors

23                   The 2006 Pb AQCD (U.S. EPA. 2006b) did not examine ADHD specifically. However,
24                   in addition to finding associations with inattention, impulsivity, and hyperactivity, some
25                   of the recent epidemiologic studies described in the preceding sections found associations
26                   between higher concurrent blood Pb level and higher parental and teacher ratings of
27                   ADHD-related behaviors (Cho et al.. 2010: Nicolescu etal.. 2010: Roy et al.. 2009a).
28                   which are a composite of the various behaviors that are evaluated in the diagnosis of
29                   ADHD. The strengths and limitations of these studies have been described in the
30                   preceding sections. Main limitations were the  cross-sectional design, lack of
31                   consideration for potential confounding by parental caregiving quality, and lack of
32                   validation of ADHD ratings with a clinical diagnosis. Thus, the evidence specifically for
33                   these total ADHD index ratings were emphasized less than evidence for inattention and
34                   impulsivity in drawing conclusions about the effects of Pb exposure on attention-related
35                   behavioral problems.


      November 2012                             5-155                  Draft - Do Not Cite or Quote

-------
 1                   The large, U.S. representative analysis of children participating in NHANES 2001-2004
 2                   found an association between concurrent blood Pb level in children ages 8-15 years and
 3                   parental assessment of child ADHD-related behaviors using the Diagnostic Interview
 4                   Schedule for Children which uses DSM-IV criteria to identify children at increased risk
 5                   of meeting diagnostic criteria for ADHD (Froehlich et al.. 2009). Compared with children
 6                   with concurrent blood Pb levels <0.8 (ig/dL, children with concurrent blood Pb levels
 7                   >1.3 (ig/dL had elevated odds of parentally-rated ADHD-related behaviors with an OR of
 8                   2.3 (95% CI: 1.5, 3.8). These results were adjusted for current household smoking
 9                   exposure,  sex, age, race/ethnicity, income, preschool attendance, maternal age, and birth
10                   weight. A similar OR was estimated when children with concurrent blood Pb levels
11                   >5.0 (ig/dL were excluded from the highest tertile. The strongest association was
12                   observed in children with both high blood Pb level and prenatal tobacco smoke exposure.
13                   Compared to children with blood Pb levels <0.8 (ig/dL with no exposure to prenatal
14                   tobacco smoke, children with blood Pb levels >1.3 (ig/dL and exposure to prenatal
15                   tobacco smoking had the highest odds  of parentally-rated ADHD-related behavior  (OR:
16                   8.1 [95% CI: 3.5, 18.7]). Although ADHD-related behavior was associated with low
17                   concurrent blood Pb levels (1.3-5 (ig/dL), the contribution of higher past Pb exposures of
18                   adolescents born in the late 1980s cannot be excluded. Roy et al. (2009a) also found an
19                   association with teacher ratings of ADHD-related behaviors using DSM criteria in
20                   children in Chennai, India; however, the study population included some very young
21                   children (i.e., age 3 years) and had relatively high blood Pb levels (mean: 11.4 (ig/dL).

22                   Other recent cross-sectional studies found Pb-associated higher ratings of ADHD-related
23                   behaviors using instruments that do not follow DSM criteria. Among children ages
24                   8-11 years in Korea, Cho et al. (2010) found a stronger relationship with a total ADHD-
25                   related behaviors index as rated by teachers than parents.  Mean ADHD ratings by teacher
26                   and parents were similar (both 9.1); however, parental ratings had greater variability (SD:
27                   11.5 for parents and 8.6 for teachers), which may have contributed to differences in
28                   association. Among children in Romania, concurrent blood Pb level was associated
29                   similarly with parent and teacher ratings of ADHD-related behaviors (Nicolescu et al..
30                   2010). As with individual attention-related behaviors described in preceding sections,
31                   blood Al and Hg levels were not associated with ratings of ADHD-related behaviors.
32                   Based on a log-linear model, a 1 (ig/dL increase in concurrent blood Pb level within the
33                   10th-90th percentile interval  (1.8-7.1 (ig/dL) was associated with a 4% increase (95% CI:
34                   0, 10) in rating of ADHD-related behavior. The association did not change substantially
35                   in an analysis that excluded the 5 children with blood Pb levels > 10 (ig/dL. Both studies
36                   considered potential confounding by parental history of psychopathology. In Cho et al.
37                   (2010). children of parents with history of psychiatric disease had lower blood Pb levels.
38                   In Nicolescu et al. (2010). parental history of psychological or psychiatric problems was
39                   weakly correlated with parental (r = 0.24, p < 0.05) and teacher (p = 0.12) rating of

      November 2012                             5-156                  Draft - Do Not Cite or Quote

-------
 1                  ADHD-related behavior, and ORs were fairly similar for ADHD score rated by teachers
 2                  and parents. Although parental history of psychopathology was examined in a few
 3                  studies, its relationship with current parental caregiving quality is not well characterized.
      5.3.3.4  Attention Deficit Hyperactivity Disorder in Children

 4                  The 2006 Pb AQCD (U.S. EPA, 2006b) did not review studies of prevalence or incidence
 5                  of ADHD diagnosis but noted lack of conclusive evidence for the effect of Pb exposure
 6                  on ADHD based on a few small studies comparing blood Pb levels between children with
 7                  and without hyperactivity as identified by parents, teachers, or schools (Gittleman and
 8                  Eskenazi. 1983; David et al., 1972). As described in the previous section, several recent
 9                  cross-sectional studies  found associations of concurrent blood Pb level with parent and
10                  teacher ratings of a total ADHD index, a composite index of inattention, impulsivity, and
11                  hyperactivity. Results from a small body of recent studies also indicate associations of
12                  higher concurrent blood Pb level with prevalence of diagnosed ADHD in children ages
13                  4-17 years (Nigg et al.. 2010: 2008: Wang et al. 2008d: Braun et al.. 2006). All of the
14                  studies were cross-sectional; thus, the temporal sequence  between Pb exposure and
15                  ADHD incidence cannot be established. While there is coherence with evidence from
16                  prospective  studies in other populations for associations of blood Pb levels with
17                  inattention, hyperactivity, and impulsivity, evidence specifically for ADHD prevalence
18                  was emphasized less than evidence for inattention and impulsivity in drawing
19                  conclusions about the effects of Pb exposure on attention-related behavioral problems.

20                  Associations between concurrent blood Pb level and ADHD prevalence were found in
21                  case-control studies conducted in different populations of children. While a potential
22                  limitation of these studies is selection bias arising from the nonrandom population
23                  sample, a common strength is their independent diagnosis of ADHD in a structured
24                  manner using parental and teacher ratings of behavior followed by independent
25                  assessment by multiple clinicians using DSM-IV criteria (Nigg et al.. 2010;  2008; Wang
26                  et al., 2008d). Nigg et al. (2010; 2008) found an association between concurrent blood Pb
27                  level and ADHD diagnosis in relatively small  (n = 150, 236) groups of children ages
28                  6-17 years from the same community, with controls selected from healthy children who
29                  responded to community advertisements. Wang et al. (2008d_) found an association in a
30                  larger (n = 1,260) group of children in China, with controls selected from children
31                  attending the same pediatric clinic for respiratory infections.

32                  Braun et al.  (2006) found an association in children ages 4-15 years participating in
33                  NHANES 1999-2002. ADHD was ascertained by parent-report of ADHD diagnosis or
34                  use of stimulant medication, which is subject to reporting bias; however, the examination
      November 2012                             5-157                  Draft - Do Not Cite or Quote

-------
 1                   of multiple risk factors and outcomes in NHANES reduces the likelihood of biased
 2                   participation and reporting of ADHD by parents of children with higher Pb exposure.
 3                   NHANES is not a random sample, but a strength over other studies that examined the
 4                   prevalence of ADHD diagnosis is the large (n = 4,704) sample size and the nationally-
 5                   representative results produced with adjustment for sampling weights in models.
 6                   Surveillance data indicate that  states with a higher percentage of children with blood Pb
 7                   levels > 10 (ig/dL have lower prevalence of diagnosed ADHD (CDC. 2012, 201 Ib).
 8                   These data reduce the potential for confounding of associations observed in the NHANES
 9                   population by regional  differences in blood Pb levels and ADHD prevalence.

10                   With respect to blood Pb levels associated with ADHD diagnosis,  analyses of the
11                   concentration-response relationship indicated monotonic increases in ORs across blood
12                   Pb level groups (Wang et al.. 2008d: Braun et al.. 2006). In the analysis of children in
13                   NHANES, compared to children with concurrent blood Pb level <0.8 (ig/dL, children
14                   with concurrent blood Pb level >2.0 (ig/dL (maximum not reported) had higher
15                   prevalence of ADHD with an OR of 4.1 (95% CI:  1.2, 14.0). A similar OR was estimated
16                   for children with blood Pb levels 2.0-5.0 (ig/dL (Braun et al.. 2006V In the study of
17                   children in China, the highest OR was found in children with concurrent blood Pb levels
18                   > 10 (ig/dL but also was elevated in the group with blood Pb levels 5-10 (ig/dL (OR: 4.92
19                   [95% CI: 3.47, 6.98] compared with children with blood Pb level <5 (ig/dL) (Wang et al..
20                   2008d). Other evidence indicated associations at lower blood Pb levels, i.e., population
21                   means ~1 (ig/dL or group with levels >0.8 (ig/dL (Nigg etal.. 2010; 2008; Braun et al..
22                   2006). However, the examination of adolescents adds uncertainty regarding the relative
23                   contributions of higher past Pb exposures and current exposures to the observed
24                   associations. Blood Pb  levels are higher in early childhood, and among children
25                   participating in NHANES who were born 1984-1998, some likely had higher early-life
26                   Pb exposures from the use of leaded gasoline in the U.S. (Braun et al.. 2006).

27                   Consideration for potential confounding varied among studies. In three-way analyses of
28                   covariance, Nigg et al.  (2008) adjusted for sex and household income, and Nigg et al.
29                   (2010) adjusted for maternal IQ and prenatal smoking exposure. However, in preliminary
30                   analyses, Nigg et al. (2010) considered blood hemoglobin, household income, age, sex,
31                   and race/ethnicity as potential confounding factors. The analysis of children participating
32                   in NHANES adjusted for age, race, prenatal smoking exposure, postnatal smoker in the
33                   home, preschool/child care attendance, health insurance coverage, and ferritin levels but
34                   initially considered poverty to income ratio, birth weight, and admission to the neonatal
35                   intensive care unit (Braun et al.. 2006). The results for children in  China were adjusted
36                   for similar covariates and also family (parent and sibling) history of ADHD diagnosis,
37                   ascertained from clinical records (Wang et al.. 2008d). Family history of ADHD was
38                   selected as a covariate based on its association with child ADHD;  no information was
      November 2012                             5-158                  Draft - Do Not Cite or Quote

-------
 1                   provided on its association with child blood Pb level. None of the studies of ADHD
 2                   prevalence considered potential confounding by current parental caregiving quality.

 3                   In recent commentaries to studies reporting associations between blood Pb level and
 4                   ADHD in children, Brondum (2011, 2007) asserted the need for studies to consider
 5                   confounding by parental history of ADHD. Given the highly heritable nature of ADHD,
 6                   parental ADHD is a strong risk factor for ADHD in children (Faraone and Doyle. 2001);
 7                   however, data have not characterized well associations of parental history of ADHD and
 8                   blood Pb level in the child. Therefore, it is uncertain whether the lack of adjustment for
 9                   parental history of ADHD produces spurious associations between blood Pb level and
10                   ADHD in children. Further, because parental history of ADHD likely explains a large
11                   portion of variance in child ADHD, not removing that variance with statistical adjustment
12                   may mask the smaller magnitude of risk due to other factors, including Pb, not produce
13                   spurious associations. Studies that examined parenting behaviors in parents with current
14                   ADHD have indicated that parents with ADHD show negative parenting control,
15                   i.e., over-reactive disciplining, lack of planning, and disorganization but have not
16                   consistently indicated that parents with ADHD have poorer emotional responsiveness,
17                   i.e., involvement with the child (as reviewed in Johnston et al.. 2012). Thus, the potential
18                   for parental ADHD to produce spurious associations between child blood Pb level and
19                   child ADHD is not well characterized.
      5.3.3.5  Integrated Summary of Attention-related Behavioral Problems

20                   Although not examined as extensively as cognitive function, epidemiologic studies have
21                   found associations of childhood blood and tooth Pb levels with attention-related
22                   behavioral problems in children and young adults, with more compelling evidence for
23                   increases in inattention and impulsivity than hyperactivity, ratings of ADHD-related
24                   behaviors, or ADHD diagnosis. The evidence for inattention and impulsivity is provided
25                   by both prospective and cross-sectional studies, whereas evidence for hyperactivity,
26                   ratings of ADHD-related behaviors, and ADHD diagnosis is provided primarily by cross-
27                   sectional studies. With analysis of earlier childhood blood or tooth Pb levels and later
28                   childhood behavioral problems, the prospective studies better characterized the temporal
29                   sequence between exposure and outcome. In cross-sectional associations with concurrent
30                   blood Pb levels, there is greater uncertainty regarding the potential for reverse causation.
31                   Associations between blood or tooth Pb levels and attention-related behavioral problems
32                   were found in diverse populations in North America, Europe, Asia, Australia, and New
33                   Zealand. Most studies had population-based recruitment from prenatal clinics, hospitals
34                   at birth, or schools and had moderate to high participation. A few prospective studies had
35                   increased loss-to-follow-up in certain groups, for example, lower SES, lower earlier


      November 2012                             5-159                   Draft - Do Not Cite or Quote

-------
 1                   FSIQ, lower HOME score. This potential selection bias can reduce the generalizability of
 2                   findings to the original study population, but there was not strong indication that
 3                   participation was biased to those with higher blood Pb levels and greater attention-related
 4                   behavioral problems. Multiple testing was common; however, in most studies, the
 5                   consistent pattern of association observed across the ages of blood Pb level and/or
 6                   behavior examined increases confidence that the evidence is not unduly biased by the
 7                   increased probability of finding associations by chance alone.

 8                   A large evidence base indicates associations of blood  and tooth Pb levels with inattention
 9                   and impulsivity as assessed using neuropsychological tests or ratings by parents or
10                   teachers (Section 5.3.3.1). Observations of associations across the various methods of
11                   assessment increase confidence that the collective evidence is not unduly influenced by
12                   biased reporting of inattention by parents of children with higher blood Pb levels. Most
13                   studies that examined inattention with the CPT found  associations with blood Pb level
14                   (Figure 5-10 and Table 5-11). including prospective studies, which indicated increases in
15                   commission and omission errors or reaction time in association with higher prenatal
16                   (maternal) and earlier childhood (age 3-60 month average, age 78 month) blood Pb levels
17                   in the Cincinnati cohort at ages 15-17 years  (Ris et al., 2004) and with higher tooth Pb
18                   (from ages 5-8 years) levels in Boston-area young adults at ages 19-20 years (Bellinger et
19                   al.,  1994a). Results from prospective studies also indicated higher parental and teacher
20                   ratings of inattention in association with higher lifetime average blood Pb levels in
21                   children ages 11-3 years in Port Pirie, Australia (Burns et al.,  1999) and with tooth Pb
22                   levels (from ages 6-8 years) in children ages 8-13 years in New Zealand, and Boston, MA
23                   (Fergusson et al.,  1993; Leviton et al., 1993). The mean blood Pb levels (prenatal cord,
24                   early childhood, lifetime average) in these populations were 7-14 (ig/dL. In children,
25                   inattention was associated with biomarkers of Pb exposure representing several different
26                   lifestages and time periods. Prospective studies did not examined a detailed blood Pb
27                   history, and results do not identify an individual critical lifestage, time period, or duration
28                   of Pb exposure associated with inattention in children. Associations in prospective studies
29                   with tooth Pb level, earlier childhood average and lifetime average blood Pb levels point
30                   to an effect on inattention of cumulative childhood Pb exposure. Indicators of more
31                   recent Pb exposures were not examined. Evidence did not strongly indicate associations
32                   between concurrent blood Pb  levels and inattention ratings in the Rochester and
33                   Yugoslavia cohorts (Canfield et al., 2003b; Wasserman et al., 2001).  This latter group of
34                   studies examined lower blood Pb levels, means 6.5 (ig/dL, but younger children ages 4-5
35                   years, in whom behaviors may be less reliably measured.

36                   An additional strength of the prospective studies was their more extensive consideration
37                   for potential confounding. Although the specific factors varied by study, prospective
38                   studies of inattention adjusted for factors such as SES, parental  IQ, maternal education,
      November 2012                             5-160                  Draft - Do Not Cite or Quote

-------
 1                   HOME score, self drug use, prenatal drug and alcohol exposure, and birth outcomes.
 2                   Adjustment for SES is difficult as it is highly correlated with Pb exposure and there is no
 3                   single measure that represents SES. Residual confounding also is likely by factors not
 4                   considered. The combination of evidence from prospective studies that considered
 5                   several well-characterized potential confounding factors plus coherence with evidence
 6                   that Pb exposure induces impulsivity in animals increase confidence that the associations
 7                   between blood and tooth Pb levels and inattention and impulsivity observed in children
 8                   represent a relationship with Pb exposure.

 9                   Several recent cross-sectional studies provided evidence of associations of higher
10                   concurrent blood Pb level with increases in impulsivity using response inhibition tests
11                   and higher inattention ratings in children ages 8-17 years. These associations were found
12                   in populations with mean concurrent blood Pb levels 1 and 4 (ig/dL (Nicolescu et al..
13                   2010; Nigg etal.. 2008). However, the contribution of higher Pb exposures earlier in
14                   childhood cannot be excluded. Further, while these recent studies considered potential
15                   confounding by parental education, they had less consistent consideration for other SES-
16                   related factors or parental caregiving quality than prospective studies. Some considered
17                   parental history of psychopathology (Cho  etal.. 2010; Nicolescu et al.. 2010); however,
18                   its relationship with parental caregiving quality is not well characterized. Recent cross-
19                   sectional studies that included younger children (ages 3-5 years) also found associations
20                   between concurrent blood Pb level and higher inattention as rated by teachers or study
21                   examiners (Plusquellec et al., 2010; Roy et al., 2009a); however, ratings in young
22                   children may be less reliably measured. With the exception of the Rochester cohort study
23                   (Canfield et al., 2003b). several studies found associations between blood Pb level and
24                   inattention with adjustment for child IQ (Cho et al.. 2010; Nigg et al.. 2008; Silva et al..
25                   1988). supporting an effect of Pb exposure on inattention independent of effects on
26                   cognitive function.

27                   The epidemiologic findings for impulsivity are supported by observations in rats and
28                   monkeys of Pb-induced impaired response inhibition in tests of discrimination reversal
29                   learning and FR/waiting for reward (Stangle et al.. 2007; Brockel and Cory-Slechta.
30                   1998; Rice and Gilbert. 1985). Coherence is found particularly with associations found
31                   between blood Pb levels and impaired response inhibition in children as assessed using
32                   the stop signal task. Impulsivity in animals was found with early postnatal  (lactation) and
33                   lifetime dietary Pb exposures relevant to humans, i.e., resulting in blood Pb levels
34                   11-31 (ig/dL. The effects of early postnatal Pb exposure are consistent with the
35                   continuing development of the nervous system and greater Pb absorption and retention
36                   during early life. The findings in children and animals for Pb-associated impulsivity are
37                   supported by observations that Pb affects dopaminergic neurons of the frontal cortex and
38                   striatum of the brain by altering dopamine release and receptor density. The circuitry in
      November 2012                             5-161                  Draft - Do Not Cite or Quote

-------
 1                  these regions is thought to mediate response inhibition. In animals, the effects of Pb on
 2                  sustained attention have been inconsistent; however, studies find Pb-induced increases in
 3                  distractibility. Attention-related behavioral problems also have been linked with changes
 4                  in the hippocampus, and evidence describing the effects of Pb on hippocampal functions
 5                  further supports the mode of action for Pb-associated increases in attention-related
 6                  behavioral problems.

 7                  Although examined in fewer studies, hyperactivity has been linked with higher blood and
 8                  tooth Pb levels in children (Section 5.3.3.2). Previous findings were limited to  cross-
 9                  sectional and case-control  studies. However, a recent prospective study found higher
10                  teacher ratings of hyperactivity in children in the U.K. ages 7-8 years with age 30 month
11                  blood Pb levels >10 (ig/dL, with adjustment for maternal education and  smoking, SES,
12                  home facilities score, family adversity index, plus other factors (Chandramouli et al..
13                  2009). Among cross-sectional studies, associations were found with adjustment for SES
14                  and maternal education; however, parental caregiving quality was examined infrequently.

15                  Recent studies also found associations with higher parental ratings of a composite index
16                  of ADHD-related behaviors, including a large NHANES analysis that used DSM-IV
17                  criteria (Froehlich et al.. 2009). In the few available studies, concurrent blood Pb levels
18                  were associated with prevalence of diagnosed ADHD in children (Section 5.3.3.4).  There
19                  is coherence with evidence from prospective studies for associations of blood and tooth
20                  Pb levels with inattention,  hyperactivity, and impulsivity, which comprise ADHD.
21                  However, the small number of studies, their cross-sectional or case-control design,  and
22                  lack of consideration for potential confounding by parental caregiving quality preclude
23                  conclusions regarding the relationship between Pb exposure and ADHD specifically.
      5.3.4  Conduct Problems


      5.3.4.1  Epidemiologic Studies of Conduct Problems in Children

24                  The 2006 Pb AQCD (U.S. EPA. 2006b) described several prospective studies that
25                  demonstrated associations of higher blood, tooth, and bone Pb levels with conduct
26                  problems in children as rated by parents or teachers [(U.S. EPA. 2006b). and see Table
27                  5-12 from this ISA]. A few previous studies found associations with criminal offenses in
28                  adolescents or young adults.  Supporting evidence from recent prospective studies
29                  included follow-up of previous cohorts to older ages (Table 5-12). Recent cross-sectional
30                  studies found associations between concurrent blood Pb level and ratings of misconduct,
31                  but several had limitations aside from establishing temporality, including prenatal drug
32                  and alcohol exposure, treatment with chelators earlier in childhood, and less extensive

      November 2012                            5-162                  Draft - Do Not Cite or Quote

-------
1                  consideration for potential confounding. Previous studies found associations with mean
2                  blood Pb levels >10 (ig/dL. Recent evidence indicated associations with lower blood Pb
3                  levels, means 1-8 (ig/dL. However, in these children and young adults, the influence of
4                  higher Pb exposures earlier in childhood cannot be excluded. In the evaluation of
5                  epidemiologic evidence for conduct problems, greater emphasis was placed on evidence
6                  from prospective studies and studies with greater consideration for potential confounding.
     November 2012                             5-163                  Draft - Do Not Cite or Quote

-------
Table 5-12  Associations between blood Pb level and misconduct in children and
            young adults.



Study3
Prospective
Dietrich et al.
(2001 )







Burns et al.
(1999)














Chandramouli
et al. (2009)









Needleman et
al. (1996)










Study Population and
Methodological Details
Studies of Ratings of Misconduct
195 children followed from birth (1979-1985)
to age 15-17 yr, Cincinnati, OH
Recruitment at prenatal clinic. High follow-up
participation, no selective attrition. Primarily
African American. Linear regression model
adjusted for HOME score, parental IQ,
current SES, birth weight. Also considered
potential confounding by maternal age, other
birth outcomes, prenatal smoking, alcohol
use, and marijuana use, Fe status, ear
infections, sex, age, caregiver education,
public assistance, attendance at preschool
program, # children and adults in home.
322 children followed from birth (1979-1982)
to age 11-1 Syr, Port Pirie, Australia.
Moderate follow-up participation. Participants
had higher birth weight, older mothers, less
educated fathers. Log linear regression model
adjusted for maternal age, prenatal smoking
status, birth weight, type of feeding, length of
breastfeeding, maternal education, IQ, and
concurrent psychopathology, paternal
education, birth order, family functioning,
paternal occupation, parent smoking, marital
status, HOME, child IQ.






488 children followed from birth (1991-1992)
to age 7-8 yr, Avon, U.K.
All births in area eligible. Similar
characteristics as U.K. census, high
participation at baseline and follow-up.
Participants had better educated mothers,
who smoked less, better home environment.
Regression model adjusted for maternal
education and smoking, home ownership,
home facilities score, paternal occupation,
family adversity index, parenting attitudes at 6
mo. Also considered potential confounding by
child IQ.
301 boys selected from prospective cohort
followed from first grade to age 11 yr,
Pittsburgh, PA
Nested case-control. Moderate participation
rate. Participants had higher SES, lower
maternal IQ, smaller family size, higher IQ,
were nonwhite. ANCOVA adjusted for
maternal age, IQ, occupation, and education,
presence of both parents in home, Number
(#) of children in home, race, history of
medical problems, age, score at age 7 yr. Did
not consider potential confounding by
parental caregiving quality.
Blood
(ug/dL),
Tooth or
Bone (ug/g)
Pb Levels

0-6 yr avg
blood: NR







Lifetime (age
11-13 yr) avg
blood
GM (5th-95th)
Boys 1 4.3
(13.5-15.1)
Girls: 13.9
(13.2-14.6)
Intervals
analyzed:
Boys
13.7-14.9,

Girls
13.3-14.4 =
10th-90th
percentiles
Age 30 mo
blood
Mean (SD): NR
Reference: 0-2

2nd group: 2-5
3rd group: 5-10
4th group: >10




Bone at 10.2yr:
NR
Bone Pb levels
in high and low
groups NR









Outcome

Self-Report of
Delinquent
Behavior Score
Parental Report
of Predelinquent
and Delinquent
Bshsvior SCOTS
at ages 15-1 7 yr




Aggressive
Score, boys
Aggressive
Score qirls

Destructive
score, boys

Destructive
score, girls


Maternal rating
by Child
Behavior
Checklist at
ages 11-1 Syr



Antisocial
sctivitiss


Parent or
teacher rating by
Antisocial
Behavior
Interview at age
Syr

Delinquency
score (square
root)
Aggression
score (square
root)

Parent rating by
Child Behavior
Checklist at age
11 yr



Effect Estimate
/oco/ ^i\k
(95% Cl)

0.10(0.01,0.193)


0.09 (-0.02, 0.20)





0.17(0.08, 0.26)
0.10(0, 0.21)

0.06 (0.02, 0.09)

0.01 (-0.01 , 0.04)









ORs for increase in
score vs. reference
2nd: 0.93 (0.47, 1 .83)°
3rd: 1.44(0.73, 2.84)°

4th: 2.90 (1.05, 8.03)°





Low Pb group: 1.18

High Pb: 1.45, p=0.04
Low Pb: 2.43
High Pb: 2.98, p=0.009






November 2012
5-164
Draft - Do Not Cite or Quote

-------



Study3
Wasserman
et al. (2001)





Bellinger et
al. (|994b)











Study Population and
Methodological Details
191 children followed prenatally (from 1985)
to age 4-5 yr, Pristina, Yugoslavia.
Recruitment from prenatal clinics. High follow-
up participation, participants had less
educated mothers, higher concurrent blood
Pb, were Albanian. Log linear regression
model adjusted for sex, ethnicity, age,
maternal education, HOME, birth weight,
maternal smoking history.
1 ,782 children followed from birth
(1 979-1 980) to age 8 yr, Boston, MA area
Recruitment at birth hospital. High follow-up
participation. More participants were white,
had lower cord blood Pb levels, better birth
outcomes. Log linear regression model
adjusted for prepregnant weight, race,
Cesarean section, maternal marital status,
prenatal care, paternal education, colic, child
current medication use, sibship size, sex,
birth weight. Also considered potential
confounding by public assistance, prenatal
smoking, maternal education but not parental
caregiving quality.
Blood
(ug/dL),
Tooth or
Bone (ug/g)
Pb Levels
Lifetime (to age
4-5 yr) avg
blood
Mean (SD):
9.6(1.5)
Interval
analyzed: 7.8
(A r\iu
(i utn
percentile)-10
Tooth (age 6 yr)
Mean (SD):
3.4 (2.4) ug/g
10th-90th:
•1 o c o
1 .Z-D.O










Outcome
Aggressive
Score
Delinquent
Score

Maternal rating
by Child
Behavior
Checklist at
ages 4-5 yr
Total
externalizing
score T score,
inattentive,
nervous/
overact ive,
aggressive)
Teacher rating
by Child
Behavior Profile
at age 8 yr





Effect Estimate
/oco/ ^i\k
(95% Cl)
0.0 (-0.01, 0.01 7)
0.016(0.001,0.03)




0.51 (0.19,0.83)










Cross-sectional studies of Ratings of Misconduct
Braun et al.
(2008)





Chiodo et al.
(2007)










Sciarillo et al.
(1992)








2,867 children ages 8-15 yr (born
1986-1996), U.S. NHANES 2001-2004
Large multi-location study of multiple risk
factors and outcomes. Subjects with available
data were older, white, higher SES, with lower
blood Pb levels, had higher birth weight, and
fewer had household smokers. Logistic
regression adjusted for child age, poverty
income ratio, maternal age, sex, race, and
prenatal smoke exposure, cotinine levels. Did
not consider potential confounding by
parental caregiving quality.
451-460 African-American children, age 7 yr
(born 1989-1 991), Detroit, Ml
Recruitment at prenatal clinic. High
prevalence prenatal drug and alcohol
exposure. High participation rate. Linear
regression model adjusted for sex (both
outcomes), caretaker education, HOME,
maternal prenatal alcohol use, current
marijuana use (delinquent behavior),
maternal age, # children in home (social
problems). Also considered potential
confounding by SES, child age, maternal
prenatal and current maternal drug and
alcohol use, and IQ, current caretaker
psychopathology.
201 children (born 1984-1987) ages 2-5 yr,
Baltimore, MD.
High participation rate. Linear regression
adjusted for maternal education, employment
status, marital status, current depressive
symptom score, preschool children in the
home, child age, sex, Fe deficiency.







Concurrent
blood
Q1:<0.8
Q2: 0.8-1.0
Q3: 1.1-1.4
Q4: 1.5-10



Concurrent
blood
Mean (SD):
5.0 (3.0)
Interval
analyzed:
2.1-8.7 =
1 0th-90th
percentiles






Concurrent
blood
Mean (SD)
Low: 9.2 (2.9)
High: 27.8
(10.4)
Interval
analyzed: 5.9
(10th percentile
of lowgroup)-10




Conduct
disorder;
Parental report
using Diagnostic
Interview
Schedule for
Children-
Caregiver
Module at ages
8-1 5 yr


Delinquent
behavior
Social problems
Teacher rating
byAchenbach
Teacher Report
Form at age 7 yr







Total Behavioral
Problem score
(aggressive,
destructive,
somatic
problems, sleep
problems,
depressed,
social
withdrawal, etc)
Maternal rating
by Child
Behavior
Checklist at
ages 2-5 yr
ORs (yes/no) vs.
Q1 as reference
Q2: 7.24 (1.06, 49.5)°
Q3: 12.4(2.37, 64.6)°
Q4: 8.64 (1.87, 40.0)°



0.09(0, 0.1 8)d
0.10(0, 0.20)d









0.18(0.04,0.32)









November 2012
5-165
Draft - Do Not Cite or Quote

-------



Study3
Chen et al.
(2007)










Nigg et al.
(2008)






Study Population and
Methodological Details
622 children participating in TLC trial, age 7
yr, Baltimore, MD; Cincinnati, OH; Newark,
NJ; Philadelphia, PA.
Multi-city, high participation rate. High age 1-3
yr blood Pb levels 20-44 ug/dL that resulted in
chelation. Regression-based path analysis
adjusted for city, race, sex, language,
parental education and employment, single
parent, age at blood Pb measurement,
caregiver IQ. Considered potential
confounding by chelation treatment but not
parental caregiving quality. Direct=
independent of IQ. lndirect= mediated
through IQ
150 children, ages 8-1 7 yr, location NR.
Case-control study of ADHD. Recruitment by
community advertisements. High participation
rate. Did not consider potential confounding
factors.


Blood
(ug/dL),
Tooth or
Bone (ug/g)
Pb Levels
Concurrent
blood Mean
(SD): 8.0 (4.0)
Interval
analyzed: 3.9
(10th
percentile)-10





Concurrent
blood
Mean (SE):
Age 8-1 1 yr'
1 .04 (0.09) '
Age 1 2-1 7 yr:




Outcome
Externalizing
behavior

Direct, parent
Direct, Teacher

Indirect, parent
Indirect, Teacher
Behavior
Assessment
System for
Children

Oppositional
defiant disorder
Index
Parent, teacher
rating
Conners Rating
Scale-Revised


Effect Estimate
/oco/ ^i\k
(95% Cl)
OR for score > 60:


1 .024 (0.996, 1 .053)
1 .036 (1 .003, 1 .069)

1.008(1.002,1.014)
1.004(0.998,1.010)




r = 0.18, p O.05





Studies of Criminal Offenses
Wright et al.
(2008)







Fergusson et
al. (2008)












250 adults followed from birth (1979-1985) to
age 19-24yr, Cincinnati, OH
Prospective. Recruitment at prenatal clinic.
High follow-up participation. No selective
attrition. Negative binomial regression model
adjusted for maternal IQ and education, sex,
SES. Also considered potential confounding
by maternal prenatal smoking and prior
arrests, marijuana use, narcotic use, HOME,
birth weight, # children in the home, public
assistance in childhood.
911 children followed from birth (1977) to age
21 yr, Christchurch, New Zealand
Prospective. High follow-up participation.
Participants had lower SES. Negative
binomial regression model adjusted for
maternal education and prenatal
cigarettes/day, SES, ethnicity, family conflict,
physical abuse in childhood, parental alcohol
problems. Also considered potential
confounding by traffic density in childhood,
maternal age, paternal education, average
family income, maternal use of punishment,
parental drug use, parental bonding, child
marijuana use.
Age 6 yr
blood Pb
Median
(5th-95th):
6.8(3.4-18.3)
Age 0-6 yr avg
Median
(5th-95th):
12.3(6.0-26.3)

Tooth
(age 6-8 yr)
Mean:
6.2 ug/g










Criminal arrests
From county
records at ages
1 9-24 yr
Age 6 yr blood
Age 0-6 avg
blood



Convictions for
property and
violent offenses
From police
records at age
21 yr








RRs (yes/no):



1.05(1.01, 1.09)
1.01 (0.97, 1.05)



0.49(0.16,0.82)













November 2012
5-166
Draft - Do Not Cite or Quote

-------
Study3
Needleman et
al. (2002)
Study Population and
Methodological Details
340 adolescents, ages 12-18 yr (born
1974-1986), Pittsburgh, PA area.
Cross-sectional. 194 cases (county) and 146
controls (Pittsburgh high schools) from
different sources. Low participation from
cases. Logistic regression adjusted for race,
parental education and occupation, both
parents in home, # children in home,
neighborhood crime rate. Did not consider
potential confounding by parental caregiving
quality.
Blood
(ug/dL),
Tooth or
Bone (ug/g)
Pb Levels
Concurrent
Bone
Mean (SD)
in ppm
Cases:
11.0(32.7)
Controls:
1.5(32.1)
Outcome
Delinquent
status
From Juvenile
Court records at
ages 12-1 Syr
Effect Estimate
(95% Cl)b
ORs (yes/no) for bone
Pb level >25 ppm
White:
3.7(1.3, 10.5)°
African-American:
2.2(0.5, 10.0)°
      aResults are organized according to outcome category, behavior ratings then documented criminal offenses. Within each category,
      studies are organized by strength of design and analysis.
      bUnless otherwise specified, effect estimates are standardized to a 1 ug/dL increase in blood Pb level or 1 ug/g increase in bone
      or tooth Pb level in the interval from the 10th percentile to 10 ug/dL blood Pb or bone Pb or the 90th percentile, whichever is
      lower.
      °Odds in higher quantile of blood or bone Pb level compared to that in lowest quantile of blood or bone Pb level (reference).
      d95% Cl was estimated from a reported p-value of 0.05.
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
Prospective studies provided key evidence for associations of earlier childhood and
lifetime average blood Pb levels and tooth Pb levels (from age 6-8 years) with conduct
problems such as delinquent behavior, aggression, antisocial activities, and destructive
behavior as rated by parents and/or teachers in children ages 8-17 years (Table 5-12).
These studies had moderate to high follow-up participation that was not conditional on
blood Pb levels, reducing the likelihood of selection bias. Across studies, behaviors were
assessed using various tests, but several studies used the Child Behavior Checklist.
Associations were found with both parent and teacher ratings, increasing  confidence that
biased reporting of conduct problems  by parents of children with higher Pb exposures did
not unduly influence the collective body of evidence. The evidence from  prospective
studies is substantiated by associations observed in diverse populations (i.e., U.K.,
Cincinnati, Port Pirie, Australia) that considered several potential confounding factors
including multiple SES-related factors, parental caregiving quality, smoking exposure,
and birth outcomes (Chandramouli et  al.. 2009; Dietrich et al.. 2001; Burns etal. 1999).
In the Yugoslavia cohort, lifetime average blood Pb level was associated  with parent
ratings of aggressive and delinquent behavior with consideration for similar potential
confounding factors but in children from ages 4-5 years, in whom behaviors may be less
reliably measured or predictive of behavior at older ages (Wasserman et al.. 2001).
      November 2012
                             5-167
Draft - Do Not Cite or Quote

-------
 1                   Collectively, the evidence from prospective studies indicated associations of teacher and
 2                   parental ratings of aggressive, destructive, antisocial, and delinquent behavior with
 3                   measures of cumulative Pb exposure, i.e., age 0-6 year average blood, lifetime average
 4                   blood, tooth, and bone Pb level. Associations with tooth (Bellinger et al.. 1994b) and
 5                   bone (Needleman et al., 1996) Pb level, collected prior to or at the same time as behavior
 6                   assessment, respectively, were found with adjustment for several potential confounding
 7                   factors as noted above, with the exception of parental caregiving quality. Collective
 8                   evidence from prospective studies most clearly indicated associations between blood Pb
 9                   level and ratings of conduct problems with population mean or group blood Pb levels
10                   > 10 ug/dL (Chandramouli et al.. 2009; Dietrich etal.. 2001; Wasserman et al.. 2001;
11                   Burns et al.. 1999). Among children ages 7-8 years in the U.K. born in the 1990s,
12                   Chandramouli et al. (2009) recently found that compared with children with age
13                   30 month blood Pb levels 0-2 ug/dL, children with blood Pb levels >10 ug/dL had
14                   increased odds of greater antisocial activities as rated by parents or teachers with an  OR
15                   of 2.9 (95% CI: 1.05, 8.0). The Boston cohort was found to have lower childhood blood
16                   Pb levels (Table 5-3); however, tooth Pb level was associated with a higher total
17                   externalizing behavior score, which also included inattention (Bellinger et al.. 1994b).

18                   As described above, cross-sectional studies also indicated blood Pb-associated higher
19                   ratings of conduct problems; however,  because of their various limitations discussed
20                   below, their results were less of a consideration in drawing conclusions about the effects
21                   of Pb exposure on conduct problems. In the recent analysis of 2,867 children ages 8-15
22                   years participating in NHANES 2001 -2004, Braun et al. (2008) analyzed blood Pb level
23                   as a categorical variable and found higher prevalence of conduct disorder as ascertained
24                   by parental questionnaire with concurrent blood Pb levels in the range of 0.8 to
25                   1.0 ug/dL. Compared with children with blood Pb levels <0.8  ug/dL, the OR in children
26                   with blood Pb levels 0.8-1.0 ug/dL was 7.24 (95% CI:  1.06, 49.5). The wide 95% CIs
27                   likely were due to the small numbers of cases of conduct disorder. For example, there
28                   were 22 children rated as having conduct disorder in the group with blood  Pb levels
29                   0.8-1.0 ug/dL. Nigg et al.  (2008) found a blood Pb-associated higher rating (parent or
30                   teacher) of oppositional defiant disorder in a population with similarly low concurrent
31                   blood Pb levels, means ~1 ug/dL. However, the implications are limited because of the
32                   case-control design (Section 5.3.3.4) and lack of consideration of potential confounding
33                   factors. Further, because these studies examined adolescents who likely had higher earlier
34                   childhood Pb exposures, there is uncertainty regarding the level, timing, frequency, and
35                   duration of Pb exposure that contributed to the observed associations.
      November 2012                             5-168                  Draft - Do Not Cite or Quote

-------
 1                   Consideration for potential confounding varied among the cross-sectional studies of
 2                   conduct problems. With the exception of Nigg et al. (2008). most considered SES.
 3                   However, only a few considered parental caregiving quality (i.e.,. HOME score) or
 4                   current maternal psychopathology. An analysis of children in Baltimore, MD adjusted
 5                   results for current maternal depressive symptoms but examined children less than age 5
 6                   years and analyzed misconduct as a component of total behavior problem score, which
 7                   included internalizing behaviors (Sciarillo et al., 1992). Chiodo et al. (2007) found that
 8                   higher concurrent blood Pb level was associated with higher teacher ratings of social
 9                   problems and delinquent behavior in children ages 7 years in Detroit, MI with adjustment
10                   for HOME score and initial consideration of current maternal psychopathology; however,
11                   the study population had high prevalence of prenatal drug and alcohol exposure. Neither
12                   of these exposures met the criteria for inclusion in the model, indicating lack of
13                   confounding by these factors. Nonetheless, the results may not be representative of the
14                   general U.S. population of children. Lack of representativeness also may pertain to the
15                   results of Chen et al. (2007). who examined children who were given chelators at ages
16                   1-3 years because of high earlier childhood blood Pb levels (20-44 ug/dL).

17                   While few in number, evidence from prospective studies also indicated associations of
18                   biomarkers  of earlier childhood Pb exposure with delinquent and criminal acts as
19                   objectively assessed from government records. These studies of delinquent and criminal
20                   acts examined Pb levels in blood or tooth samples collected in the 1980s when Pb
21                   exposures were much higher than those of the current U.S. population (Fergusson et al..
22                   2008; Wright et al.. 2008). However, the prospective study design and consideration for
23                   several potential confounding factors increase confidence that the observed associations
24                   represent a relationship with Pb exposure.

25                   In the Cincinnati cohort, prenatal cord and age 0-6 year average blood Pb levels were
26                   associated with self- and parent-reported delinquent and social acts at ages 16-17 years
27                   (Dietrich et al.. 2001). Wright et al. (2008) recently extended these findings to include
28                   associations of prenatal cord and age 6 year blood Pb levels with criminal and violent
29                   criminal arrests at ages 19-24 years. In models that adjusted for maternal IQ, sex, SES
30                   score, and maternal education, the relative risks (RRs) for total arrests per 1 ug/dL
31                   increment in blood Pb level were 1.07 (95% CI: 1.01, 1.13) for prenatal blood Pb level,
32                   1.01 (95% CI: 0.97,  1.05) for age 0-6 year average blood Pb level, and 1.05 (95% CI:
33                   1.01, 1.09) for blood Pb level at age 6 years. Interactions terms for blood Pb by sex were
34                   not statistically significant; however, the attributable risk was considerably higher for
35                   males (0.85 arrests/year [95% CI: 0.48, 1.47]) than for females (0.18 [95% CI: 0.09,
36                   0.33]). A strength of Wright et al. (2008) was the detailed examination of potential
37                   confounding by a large number of variables (Table 5-12). All of the examined covariates
38                   were weakly correlated with blood Pb levels (r = 0.24-0.35), thereby reducing the
      November 2012                             5-169                  Draft - Do Not Cite or Quote

-------
 1                  potential for confounding by the examined factors. Nonetheless, variables such as
 2                  maternal IQ, maternal education, and SES were included in the model because they were
 3                  associated with arrests in the full multivariate model or changed the blood Pb level
 4                  estimate by more than 10%. HOME score was similar between subjects with and without
 5                  criminal arrest records and did not meet the criteria for inclusion in final models.

 6                  The study of the New Zealand cohort also considered several potential confounding
 7                  factors such as family functioning and parental bonding (Table 5-12) (Fergusson et al..
 8                  2008). Per 1 ug/g Pb in teeth obtained between ages 6 and 8 years, there was a 0.49 (95%
 9                  CI: 0.16, 0.82) increase in the number of documented violent or property convictions at
10                  ages 14-21 years. Results were adjusted for SES, ethnicity, maternal education, family
11                  conflict, prenatal smoking exposure, physical abuse in childhood, and parental
12                  alcoholism. The effect estimate for tooth Pb level decreased in adjusted models and was
13                  found to account for <1% in the variance of criminal convictions; however, the
14                  association remained statistically significant.

15                  The epidemiologic studies described above employed different designs and assessed
16                  conduct problems using different behaviors and methods. The consistency of association
17                  of Pb biomarker levels with conduct problems was corroborated in a recent meta-analysis
18                  (Marcus et al.. 2010) that included 19 studies (several of which are described above) with
19                  a total of 8,561 children and adolescents (mean ages ranging from 3.5 years to
20                  18.4 years). Effect estimates were converted to Pearson correlation coefficients, and the
21                  combined effect estimate was r = 0.19 (95% CI: 0.14, 0.23). The key finding of this study
22                  was the robustness of associations to between-study sources of heterogeneity. In the
23                  meta-analysis, effect sizes did not differ significantly between prospective and cross-
24                  sectional studies, among studies that examined different conduct problems
25                  (i.e., opposition defiance, delinquency, externalizing problems), or among studies that
26                  assessed conduct disorders using self-report, teachers report, or criminal records.
27                  Adjustment for covariates such as SES, birth weight, parental IQ, and home environment
28                  did not attenuate the relationship between blood Pb level and conduct problems. In
29                  addition to strengthening the evidence for the independent associations of Pb biomarker
30                  levels with conduct problems, the results indicated that the lack of adjustment for any
31                  particular covariate, including HOME score, does not warrant limiting inferences from a
32                  particular study. The major source of heterogeneity in effect estimates was the biomarker
33                  of Pb examined. A larger magnitude of effect was estimated for hair Pb levels compared
34                  with bone or blood Pb levels, which had similar effect sizes. The authors suggested that
35                  hair Pb may be a better indicator of cumulative Pb exposure compared to bone Pb levels
36                  due to the high turnover of bone throughout childhood and into adolescence; however, an
37                  empirical basis for interpreting hair Pb measurements in terms of body burden or
38                  exposure has not been firmly established (Section 4.3.4.2).
      November 2012                             5-170                  Draft - Do Not Cite or Quote

-------
 1                  Several studies of misconduct aimed to characterize whether associations with
 2                  biomarkers of Pb exposure were independent of effects on IQ and educational attainment.
 3                  Most studies found that associations of Pb biomarkers with conduct problems remained
 4                  statistically significant in a model that additionally adjusted for child IQ or educational
 5                  attainment, indicating that Pb exposure may have a direct effect on misconduct
 6                  independent of its effect on cognitive function (Chandramouli et al.. 2009; Fergusson et
 7                  al., 2008; Burns et al.. 1999). However, simple statistical adjustment for cognitive
 8                  function indices may result in an underestimate of the effect on misconduct because a
 9                  decrement in cognitive function may lie on the causal pathway to behavioral problems.
10                  Chen et al. (2007) used path analysis to characterize the direct effects and indirect effects
11                  (mediated through child IQ) of blood Pb level on total externalizing problem ratings at
12                  age 7 years; however, results were inconsistent. A direct effect was estimated for
13                  externalizing problems rated by teachers, and an indirect effect was estimated for
14                  problems rated by parents (Table 5-12). These findings may have limited applicability to
15                  the general U.S. population given that some children in the study had been treated with
16                  chelators at ages 1-3 years because of high blood levels, and it is uncertain whether the
17                  observed associations were due to the residual effect of high earlier blood Pb levels
18                  (20-44 jig/dL).
      5.3.4.2  lexicological Studies of Aggression

19                  While recent studies were not identified, evidence available in the 2006 Pb AQCD (U.S.
20                  EPA. 2006b) pointed to the effects of Pb on changes in social behavior in rodents and
21                  nonhuman primates. Most observations comprised Pb-induced increases in social and
22                  sexual investigation, as indicated by sniffing, grooming, following, mounting, and
23                  lordosis behavior. In animals, the social behavior most comparable to epidemiologic
24                  findings in children is aggression; however, the effects of Pb on aggression in animals
25                  were inconsistent. In animals, aggression was assessed as threats, attacks, bites, chases,
26                  and offensive posture in encounters with other animals, and Pb exposure was found to not
27                  affect, decrease, and increase aggression. Pb exposure generally was not found to affect
28                  aggression in juvenile animals; however, increased aggression was found in adult animals
29                  with high concentration (>2,500 ppm) gestational  plus postnatal dietary Pb exposure.

30                  Delville (1999) found Pb-induced increases in aggression with the lowest concentration
31                  Pb exposure examined among all animal studies. Golden hamsters exposed to 100 ppm
32                  Pb acetate GD8-PND42 in drinking water had blood Pb levels of 10 to  15 ug/dL at
33                  PND42. As adults at PND45, Pb-exposed animals displayed more aggression as
34                  measured by attacking and biting an intruder put in the cage. In mice, higher Pb exposure
35                  produced mixed findings. BK:W mice exposed to  1,300 ppm Pb acetate in drinking water


      November 2012                             5-171                  Draft - Do Not Cite or Quote

-------
 1                  from gestation through age 18 weeks displayed increased social and sexual investigation
 2                  but not aggression in males (femur Pb level at 18 weeks: 5,364 uM Pb/g ash) or females
 3                  (femur Pb level at 18 weeks: 4,026 uM Pb/g ash) (Donald et al.. 1986). Additional
 4                  investigation from the same laboratory exposed BK:W mice to 2,500 ppm Pb acetate in
 5                  drinking water from gestation through age 17-18 weeks, and found shorter latencies to
 6                  aggression in Pb-exposed mice than  in controls (Donald et al.. 1987). In juvenile Long
 7                  Evans hooded rats, lactation-only (PND1-PND21) exposure to 670 ppm Pb chloride in
 8                  drinking water increased "rough and tumble" play behavior at PND36 which was not
 9                  characterized as aggression because  of the lack of injury, submissive posturing, or escape
10                  attempts in encounters with other animals (Holloway and Thor. 1987).
      5.3.4.3  Summary of Conduct Problems

11                  Although not examined as extensively as cognitive function, previous and recent
12                  prospective studies consistently demonstrate Pb-associated increases in delinquent
13                  behavior, aggression, antisocial activities, and destructive behavior as rated by parents
14                  and teachers in children and as assessed with government records of criminal offenses in
15                  adolescents and young adults (Table 5-12). Most studies examined multiple behaviors;
16                  however, the consistent pattern of association observed across the ages of blood Pb level
17                  and/or behaviors examined increases confidence that the evidence is not unduly biased by
18                  an increased probability of associations by chance alone. Recent cross-sectional studies
19                  found associations between concurrent blood Pb level and ratings of misconduct, but
20                  several had additional limitations aside from study design, including prenatal drug and
21                  alcohol exposure, treatment with chelators, or limited consideration for potential
22                  confounding (Nigg et al.. 2008: Chen et al.. 2007: Chiodo et al.. 2007). The most
23                  informative cross-sectional study was that finding a 7.24 (1.06, 49.47) higher odds of
24                  conduct disorder in adolescents ages 8-15 years participating in NHANES with
25                  concurrent blood Pb levels 0.8-1.0 (ig/dL compared with blood Pb levels <0.8 (ig/dL with
26                  adjustment for age, sex, race, poverty to income ratio, and smoking exposure (Braun et
27                  al.. 2008). However, the association was imprecise and could have been influenced by
28                  higher past Pb exposures  of the adolescents. Further, potential confounding by parental
29                  caregiving quality was not examined. Evidence of Pb-induced aggression in animals was
30                  mixed in adult animals with lifetime Pb exposure beginning in gestation and not indicated
31                  in juvenile animals.

32                  The evidence in children  from prospective studies is substantiated by analyses of school-
33                  aged children (ages 8-17 years) in populations from various locations and SES (i.e., U.K.,
34                  Cincinnati, Port Pirie, Australia) with high participation rates, lack of indication of
35                  substantial selection bias, and consideration of several potential confounding factors


      November 2012                            5-172                  Draft - Do Not Cite or Quote

-------
 1                   including multiple SES-related factors, parental caregiving quality, smoking exposure,
 2                   and birth outcomes (Chandramouli et al.. 2009; Dietrich et al.. 2001; Burns etal. 1999).
 3                   Pb biomarker levels were associated with both parent and teacher ratings of conduct
 4                   problems, reducing the likelihood of biased reporting of conduct problems by parents of
 5                   children with higher Pb biomarker levels. Recent prospective studies of criminal offenses
 6                   in young adults, ages 19-24 years, strengthened previous evidence with consideration for
 7                   potential confounding by factors such as SES, smoking, drug, and alcohol exposure, and
 8                   parental caregiving quality (Fergusson et al.. 2008; Wright et al.. 2008). In the Cincinnati
 9                   cohort, a 1 ug/dL increase in age 6 year blood Pb level was associated with an increased
10                   risk of total arrests with an RR of 1.05 (95% CI: 1.01, 1.09) with adjustment for maternal
11                   IQ, sex, SES, and maternal education (Wright et al.. 2008). In the New Zealand cohort, a
12                   1 ug/g Pb in teeth obtained between ages 6 and 8 years was associated with a 0.49 (95%
13                   CI: 0.16, 0.82) increase in the number of documented violent or property convictions at
14                   ages 14-21 years (Fergusson et al., 2008) with adjustment  for SES, ethnicity, maternal
15                   education, family conflict, prenatal smoking exposure, physical abuse in childhood, and
16                   parental alcoholism. Further support for Pb-associated increases in conduct problems was
17                   provided by a recent meta-analysis that found that evidence was robust to heterogeneity
18                   in study design, definition and assessment method of conduct problems, potential
19                   confounding variables examined, and population mean blood Pb levels (Marcus et al..
20                   2010).

21                   Associations of conduct problems (parent/teacher ratings and criminal offenses) with
22                   earlier childhood blood (e.g., age 30 month, age 6 year), early childhood average blood
23                   (e-g-, age 0-6 year), lifetime average blood (to age 11-13 years), tooth, and bone Pb levels
24                   pointed to the effects of early childhood or cumulative Pb  exposures. Associations were
25                   found with a mean lifetime (to age 11-13 years) average blood Pb level of 14 (ig/dL,
26                   mean age 6 year blood Pb level  of 6.8 (ig/dL, and age 30 month blood Pb levels
27                   >10 (ig/dL.  Recent cross-sectional studies found associations with concurrent blood Pb
28                   level and lower blood Pb levels, means 1-8 (ig/dL, but the  study limitations detailed
29                   above  limit inferences regarding the effects of Pb exposure on conduct problems in these
30                   populations. Most prospective studies did not analyze Pb biomarker levels at multiple
31                   lifestages and time periods, including later childhood and more recent adult exposures, or
32                   examine differences in association between Pb exposures at various lifestages and time
33                   periods. The evidence does not identify an  individual critical lifestage, time period, or
34                   duration of Pb exposure associated with conduct problems in children or exclude an
35                   effect of more recent Pb exposures.
      November 2012                             5-173                  Draft - Do Not Cite or Quote

-------
      5.3.5  Internalizing Behaviors in Children


      5.3.5.1  Epidemiologic Studies of Internalizing Behaviors in Children

 1                   A majority of investigations of the effects of Pb on behavior in children has focused on
 2                   externalizing behaviors such as inattention, hyperactivity, aggression, and delinquency.
 3                   However, several studies also have linked biomarkers of Pb exposure in children with
 4                   internalizing behaviors characterized by directing feelings and emotions inward,
 5                   i.e., withdrawn behavior, symptoms of depression, fearfulness, and anxiety. Whereas
 6                   some studies found stronger associations for externalizing behaviors than for
 7                   internalizing behaviors (Plusquellec et al.. 2010; Wasserman et al.. 2001; Bellinger et al..
 8                   1994a; Sciarillo et al.. 1992). others did not find a clear difference in the strength of
 9                   association  (Roy et al.. 2009b; Chiodo et al.. 2004; Bellinger et al.. 1994b). Internalizing
10                   behaviors were assessed frequently using the Child Behavior Checklist, and as with
11                   attention-related behavioral problems and conduct problems, were rated by parents and/or
12                   teachers. Associations with both parent and teacher ratings increase confidence that
13                   biased reporting of internalizing behaviors by parents of children with higher blood Pb
14                   levels did not unduly influence the collective body of evidence. Most studies had
15                   moderate to high follow-up participation. With the exception of the Yugoslavia cohort,
16                   participation was not biased to higher blood Pb levels. Additionally,  in most studies, a
17                   consistent pattern of association was observed across the ages of blood Pb level and/or
18                   multiple behaviors examined, which increases confidence that the evidence is not
19                   strongly biased by an increased probability of associations found by chance alone.

20                   Key evidence was provided by prospective studies in various populations, i.e., Boston,
21                   Port Pirie, Australia, and Yugoslavia. Collectively, these studies found higher ratings for
22                   internalizing behaviors in children (n = 322-1,511, ages 3-13 years) in association with
23                   concurrent blood Pb level, lifetime average blood, and tooth Pb levels (Wasserman et al..
24                   2001; Burns etal. 1999; Wasserman et al.. 1998; Bellinger et al.. 1994b). In the Port
25                   Pirie cohort, Burns  et al. (1999) found that higher lifetime average blood Pb levels (mean:
26                   -14 (ig/dL) were associated with parental ratings of externalizing behaviors more
27                   strongly in boys and with internalizing behaviors (i.e., withdrawn, anxious/depressed)
28                   more strongly in girls ages 11-13 years, which may indicate sex differences in the effect
29                   of Pb or differences in the types of behaviors that are observed and reported in girls
30                   versus boys. Based on a log-linear model, a 1 (ig/dL increase in lifetime average blood Pb
31                   level was associated with increased odds of an anxious/depressed rating above the
32                   median of 1.04 (95% CI: 1.0, 1.09) among 159 boys ages 11-13 years (in the 10th-90th
33                   percentile interval of blood Pb level 13.7-14.9 (ig/dL) and 1.08 (95% CI: 1.01, 1.15)
34                   among 163  girls ages 11-13 years (in the 10th-90th percentile interval of blood Pb level).
      November 2012                             5-174                  Draft - Do Not Cite or Quote

-------
 1                   These associations were found with the adjustment for factors related to SES and parental
 2                   caregiving including HOME score, family functioning score, and current maternal
 3                   psychopathology (General Health Questionnaire).

 4                   Differences between externalizing and internalizing behaviors also were found in the
 5                   Yugoslavia cohort according to age of assessment and blood Pb levels. This cohort was
 6                   examined between ages 3 and 5 years, ages at which behaviors may be less reliably
 7                   assessed. Among 379 children ages 3 years from the higher and lower Pb exposure
 8                   towns, higher cord and concurrent blood Pb levels were associated with higher maternal
 9                   ratings of anxious-depressed, withdrawn, and externalizing behaviors, with stronger
10                   associations found for concurrent blood Pb level (mean: 25.8 (ig/dL) (Wasserman et al..
11                   1998). Among 191 children ages 4-5 years from the lower Pb exposure town, higher
12                   lifetime average blood Pb level (mean: 9.6 (ig/dL), was associated with higher ratings of
13                   delinquent behavior and internalizing behaviors, with stronger associations found for
14                   delinquent behavior (Wasserman et al.. 2001). A log unit increase in higher lifetime
15                   average blood Pb level was associated with a 0.22 log (95% CI: -0.04, 0.47) higher rating
16                   of withdrawn behavior and 0.19 log (95% CI: -0.05, 0.43) higher rating of anxious-
17                   depressed behavior in children at ages 4-5 years. Results at each age were adjusted for
18                   age, sex, HOME score, and maternal education. Results at age 3 years and 4-5 years were
19                   additionally adjusted for residence type and maternal history of smoking, respectively.

20                   With regard to important lifestages or durations of Pb exposure, results from prospective
21                   studies did not clearly demonstrate differences in association among biomarkers
22                   measured at various lifestages or time periods. The importance of childhood cumulative
23                   exposure was indicated by associations with lifetime average blood Pb levels in the Port
24                   Pirie cohort (to age 11-13 years) and in the Yugoslavia cohort (to age 5 years) and with
25                   tooth Pb (from age 6 years) levels in the Boston cohort at age 8 years. In the Boston
26                   cohort, Pb levels measured in teeth (mean: 3.4 (ig/g) but not cord blood were associated
27                   with a higher teacher rating of internalizing behaviors at age 8 years (Bellinger et al..
28                   1994b). In another Boston-area cohort, tooth (collected at first or second grade) Pb levels
29                   were not associated with self-rated symptoms of depression (Profile of Mood States
30                   questionnaire) at ages 19-20 years (Bellinger etal., 1994a).  Prospective studies did not
31                   analyze a detailed history of Pb biomarker levels to evaluate persistence of effects of
32                   early exposure or to identify an individual critical lifestage or time period of Pb exposure
33                   associated with increases in internalizing behaviors. The available evidence does not
34                   preclude an effect of later childhood or more recent Pb exposure.

35                   In the Cincinnati cohort, using structural equations, Dietrich et al. (1987b) found that the
36                   associations of prenatal maternal and infant age 10 day blood Pb level (respective means:
37                   8.3 and 4.9  (ig/dL) with poorer mood in infants (n = 185) ages 6 months were indirect,
      November 2012                             5-175                  Draft - Do Not Cite or Quote

-------
 1                   meaning they were mediated through lower birth weight and/or shorter gestation. These
 2                   results suggested that Pb may exert its effects by impairing nervous system development.
 3                   The fetal period is an active period for neuronal differentiation, dendritic branching, and
 4                   synaptogenesis, which if impaired by Pb exposure, could have broad implications on a
 5                   wide range of subsequent neurodevelopmental effects. There are few such analyses, and
 6                   the findings are limited by the lower reliability of mood assessed in infancy.

 7                   Cross-sectional studies found associations between concurrent blood or hair Pb levels and
 8                   teacher and parent ratings of internalizing behaviors in children. Associations were found
 9                   in children ages 3-16 years (n = 303-756) in China and India with mean concurrent blood
10                   Pb levels 9-14 (ig/dL (Liuetal.. 2011b: Bao et al.. 2009: Roy et al.. 2009a). Results were
11                   adjusted for family income and parental education but not caregiving quality. In the few
12                   studies of populations with mean blood Pb levels ~5 (ig/dL, results were inconsistent.
13                   Chiodo et al. (2004) found an association with internalizing behaviors with adjustment
14                   for child and caregiver life stress and marital status in 246 children age 7 years in Detroit,
15                   MI who had high prevalence of prenatal alcohol or drug exposure. HOME score, SES,
16                   maternal education, prenatal alcohol exposure, drug exposure were not found to influence
17                   associations with blood Pb level; however, the results may lack generalizability to the
18                   general population of U.S. children. A study that examined 79-91 Inuit children (age
19                   5 years) in Quebec, Canada, did not find associations between concurrent blood Pb level
20                   and internalizing behaviors with consideration of potential confounding by HOME score,
21                   caregiver education and IQ, blood Hg levels, and prenatal smoking and alcohol exposure
22                   (Plusquellecetal.. 2010).

23                   Associations of Pb biomarkers with internalizing behaviors in children were observed
24                   with consideration for a wide range of potential confounding factors, most commonly,
25                   age, birth outcomes, parental education, and other SES-related factors. Parental
26                   caregiving quality was evaluated in few studies. Blood Pb-associated higher ratings of
27                   internalizing behaviors were found with adjustment for HOME score in the Yugoslavia
28                   cohort (Wasserman et al., 2001; 1998). and HOME, family functioning, and current
29                   maternal psychopathology (General Health Questionnaire) in the Port Pirie cohort (Burns
30                   et al.. 1999). Several studies, each of which adjusted for a different set of covariates,
31                   found similar or slightly  attenuated effect estimates in univariate and multivariate models
32                   (Wasserman etal.. 2001; Burns etal.. 1999: Bellinger et al.. 1994b). Collectively, these
33                   observations increase confidence that the observed associations with Pb biomarkers
34                   reflect a relationship with Pb exposure.
      November 2012                             5-176                  Draft - Do Not Cite or Quote

-------
      5.3.5.2  Toxicological Studies of Internalizing Behaviors

 1                  As in epidemiologic studies, toxicological studies have focused more on cognitive
 2                  function and attention-related behavioral problems and less on emotional- and mood-
 3                  related behaviors. Pb biomarker levels have been associated with ratings of withdrawn
 4                  behavior, depression, and anxiety in children, and this evidence is supported by findings
 5                  of Pb-induced anxiety, emotionality and depression-like behaviors in animals.

 6                  Emotionality has been indicated by loss of motivation and increased frustration in
 7                  response to errors and reward emission in visual or olfactory discrimination task trials in
 8                  Pb-exposed rats (Beaudin et al.. 2007; Stangle et al.. 2007). In each study, Long-Evans
 9                  rats were exposed to Pb during and after lactation (300 ppm Pb acetate via dam drinking
10                  water PND1-PND17 then either 20 or 300 ppm  PND17-PND30 in own water, with
11                  respective blood Pb levels of 13 and 31 (ig/dL on PND52). In Beaudin et al. (2007).
12                  greater disruption in performance (i.e., failure to enter testing alcove) after committing
13                  errors and having rewards omitted was found in rats with blood Pb levels 13 and
14                  31 (ig/dL tested at age 9-15 weeks. Pb-exposed  rats also had greater response latency
15                  with reward omission as indicated by  the entrance into the testing alcove but failure to
16                  respond within a set period of time. In Stangle et al.  (2007).  increased reactivity to errors
17                  and reward omission was found in rats with blood Pb levels  31 (ig/dL. Blood Pb levels
18                  were measured after a lag period, and peak blood Pb levels in these animals may have
19                  been higher than those reported. In rhesus monkeys, emotional dysregulation was
20                  indicated by tactile defensiveness after exposure to Pb acetate/50% glucose  in 4 cc daily
21                  milk formula from PND8 to ages 1-2 years (producing blood Pb 35-40 (ig/dL) (Moore et
22                  al.. 2008).

23                  In other studies, Wistar rats showed emotionality and depression-like behavior in the
24                  open field test and forced swim test (i.e., Porsolt's Test) with gestational/lactational Pb
25                  exposure (de Souza Lisboa et al.. 2005). The open field test monitors activity levels and
26                  movements  of animals in three dimensions. Depression-like behavior is indicated by
27                  freezing behavior and low levels of activity. Emotionality is indicated by grooming or
28                  freezing. In the forced swim test, animals are placed in a vertical cylinder of water from
29                  which there is no escape and monitored for duration of struggling or attempt to escape.
30                  Animals that stop quickly are ascribed a depression-like phenotype. As in many other
31                  neurobehavioral tests, sex-specific differences in responses were found. Pb-exposed
32                  males showed increased emotionality  in the open field test as indicated by fewer counts
33                  of rearing. Pb-exposed females showed a depression-like phenotype in the forced swim
34                  test as indicated by longer time of immobility (de Souza Lisboa et al.. 2005). While
35                  measured blood Pb levels of rats were low, 5-7  (ig/dL, they were measured after a lag in
36                  exposure (PND70) and produced by oral gavage (10 mg/day) of mothers, a route that
      November 2012                            5-177                  Draft - Do Not Cite or Quote

-------
 1                  may have uncertain relevance to human routes of Pb exposure. Pb-induced immobility in
 2                  the forced swim test also was found with 6-weeek postnatal Pb exposure via drinking
 3                  water but producing blood Pb levels 40 (ig/dL. Reducing internal Pb dose with the
 4                  chelator succimer reversed Pb-induced immobility (Stewart et al. 1996).

 5                  Depression initially may seem like an unexpected effect of immune modulation, but it has
 6                  been linked to an interaction between the CNS and the immune system via alterations in
 7                  cytokines such as IL-6 (Section 5.6.6.1). Dyatlov and Lawrence (2002) found that dietary
 8                  Pb exposure through lactation and a brief period after weaning (500 (JVI, PND1-PND22,
 9                  resultant blood Pb level: 17 (ig/dL) potentiated sickness behavior in mice in response to
10                  bacterial infection. Sickness behavior was evidenced by an increase in serum IL-6 levels
11                  with an accompanying decrease in food and water intake and increase in body weight.
12                  This phenotype was correlated with decreases in the populations of several T cell
13                  subtypes. Pb exposure also potentiated release of IL1(3, which plays an important  role in
14                  inflammatory responses to infection and has been shown to inhibit hippocampal
15                  glutamate release in young but not aged animals. Sickness behavior also was induced in
16                  Pb-exposed  animals with IL-6 and IL-1 administration without infection, further
17                  supporting a role for immunomodulation in mediating sickness behavior.

18                  Pb exposure had mixed effects on anxiety-related responses as measured by the elevated
19                  plus maze, which assesses behavior of rodents in an unfamiliar environment. The  maze is
20                  elevated above the floor and consists of two arms that are enclosed with walls intersected
21                  with two arms that have no walls. The animal is placed in the center of the maze,  and
22                  longer latency to enter an open arm, and lower frequency and duration of entries into an
23                  open arm are indicative of anxiety. In a study of gestational/lactational (GD1-PND24)
24                  exposure to 2.84 mg/mL Pb acetate trihydrate in drinking water, Sprague-Dawley rats did
25                  not differ in anxiety-related responses from controls. Blood Pb levels of Pb-exposed rats
26                  were higher than those relevant to humans, 65.8 (ig/dL at PND25 (Molina et al.. 2011).
27                  Another study exposed female rats postnatally (PND1-PND30) to 2,000 ppm Pb acetate
28                  in drinking water, which yielded lower blood Pb levels, 34 (ig/dL. Pb-exposed rats had an
29                  increase in anxiety-related behavior at PND60, as indicated by a lower percentage of
30                  open arm entries and less time spent in the open arms (Foxetal. 2010).
      November 2012                            5-178                 Draft - Do Not Cite or Quote

-------
      5.3.5.3  Summary of Internalizing Behaviors

 1                   Internalizing behaviors, i.e., withdrawn behavior, symptoms of depression, anxiety, have
 2                   been examined less than externalizing behaviors (i.e., attention-related behavioral
 3                   problems, misconduct). However, several prospective studies found associations of
 4                   higher parental and teacher ratings of internalizing behaviors with higher tooth or lifetime
 5                   average blood Pb levels in children ages 8-13 years (Burns et al.. 1999; Bellinger etal..
 6                   1994b) and higher concurrent and lifetime average blood Pb levels in children ages 3-5
 7                   years (Wasserman et al.. 2001;  1998). Collectively, the lack of biased participation by
 8                   subjects with higher blood Pb levels and associations found with both parent and teacher
 9                   ratings increase confidence that the evidence is not unduly influenced by biased reporting
10                   of behaviors by parents of children with higher blood Pb levels. The prospective studies
11                   found associations with adjustment for several potential confounding factors, including
12                   SES and parental education and caregiving quality. In prospective studies, associations
13                   were found with tooth Pb levels and lifetime average and concurrent blood Pb levels.
14                   While, there is not a clear indication of an individual critical lifestage or time period of
15                   Pb exposure associated with internalizing behaviors, several observations point to an
16                   effect of cumulative childhood  Pb exposure. These results do not preclude an effect of
17                   later childhood or more recent Pb exposures. In prospective studies, associations with
18                   internalizing behaviors were found with a mean lifetime  average blood Pb levels of
19                   14 (ig/dLto age 11-13 years (Burns etal.. 1999) and 9.6  (ig/dLto age 4.5-5 years
20                   (Wasserman et al.. 2001). In children ages 4.5-5 years  (Yugoslavia cohort), in whom
21                   behavioral ratings may be less reliably measured, lifetime average blood Pb level was
22                   associated more strongly associated with the rating of delinquent behavior than ratings of
23                   internalizing behaviors (Wasserman et al.. 2001). Cross-sectional studies provided
24                   supporting evidence of concurrent blood Pb-associated increases in internalizing
25                   behaviors in children ages 3-16 years, and several considered potential confounding by a
26                   similar set of factors as did the  prospective studies. However, associations in populations
27                   with mean concurrent blood Pb levels ~5  (ig/dL were inconsistent.

28                   Evidence in children is supported by observations that dietary Pb exposure (early
29                   postnatal to day 22 or 30) resulted in depression-like behavior and emotionality in
30                   rodents, (Beaudin et al.. 2007; Stangle et al.. 2007; Dyatlov and Lawrence. 2002) and
31                   rhesus monkeys (postnatal to age 1-2 years) (Moore et al.. 2008). with some evidence in
32                   rodents at blood Pb levels relevant to humans (13-31 (ig/dL). Evidence for Pb-induced
33                   anxiety in animals is mixed. Mode of action support is provided by well-documented
34                   evidence  for Pb-induced changes in the HPA axis (Section 5.3.2.3) and dopaminergic and
35                   GABAergic systems (Sections  5.3.2.2 and 5.3.11.4). which are involved in regulating
36                   mood and emotional state.
      November 2012                             5-179                  Draft - Do Not Cite or Quote

-------
      5.3.6  Psychopathological Effects in Adults


      5.3.6.1 Epidemiologic Studies of Psychopathological Effects in Adults

 1                  The potential effects of Pb exposure on mood and psychopathological effects
 2                  (e.g., anxiety, depression, schizophrenia) in adults have been examined less than that in
 3                  children or cognitive function in adults. However, evaluation of mood states is an integral
 4                  part of the neurocognitive test battery of the World Health Organization (WHO), and it
 5                  has been suggested that indices of the Profile Of Mood States may be particularly
 6                  sensitive to toxicant exposures (Johnson et al.. 1987). As with other nervous system
 7                  endpoints in adults, several studies reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b)
 8                  found higher prevalence of self-reported mood disorders and anxiety among Pb-exposed
 9                  workers (n = 43-576, mean concurrent or peak blood Pb levels: 31-79  (ig/dL) in
10                  association with higher blood Pb levels or compared with unexposed controls
11                  (n = 24-181, mean blood Pb levels: 15-38 (ig/dL) (Schwartz etal.. 2005; Maizlish et al..
12                  1995; Parkinson et al.. 1986; Baker etal..  1985; Baker etal.. 1984; Lilisetal.. 1977).
13                  Several studies considered potential confounding by age, sex, education, medical
14                  conditions, smoking, and alcohol use, but only Maizlish et al. (1995) examined other
15                  occupational exposures. Most studies were cross-sectional, which makes uncertain the
16                  temporal sequence between Pb exposure and development of psychopathological effects.

17                  The few studies of adults without occupational Pb exposures participating in NAS and
18                  NHANES demonstrated associations of concurrent blood and bone Pb level with
19                  psychopathological effects. As bone Pb is a major contributor to blood Pb levels in adults
20                  without current occupational Pb exposure, cross-sectional associations with each
21                  biomarker may indicate effects of cumulative Pb exposure. These previous and recent
22                  cross-sectional studies found associations with adjustment for several potential
23                  confounding factors, including age, education, employment status, and alcohol use.
24                  Further, the examination of multiple exposures and outcomes in these  studies reduces the
25                  likelihood of participation conditional on Pb exposure and psychopathological effects.

26                  Analyses of 526 men ages 48-70 years in the NAS indicated associations  of both higher
27                  concurrent blood (mean:  6.3 (ig/dL [SD: 4.16]) and tibia (mean: 21.9 ng/g [SD: 13.5]) Pb
28                  levels with higher prevalence of self-reported depression and anxiety (Rhodes et al..
29                  2003). In a recent analysis of 744 NAS men ages 48-94 years, Rajan et al. (2007) found
30                  associations of symptoms assessed using the Brief Symptom Inventory with patella and
31                  tibia Pb levels. A 14 (ig/g increase in tibia Pb level was associated with an increased odds
32                  of an anxiety score  above the median of 1.18 (95% CI: 0.98, 1.42) and of depression
33                  score above the median of 1.16 (95% CI: 0.97, 1.38). Similar effect estimates were found
      November 2012                            5-180                  Draft - Do Not Cite or Quote

-------
 1                  for patella Pb level. Effect modification by ALAD genotype was not in a consistent
 2                  direction. For most mood symptoms, tibia bone Pb levels were associated with larger
 3                  ORs among the 587 men with the ALAD 1-1 genotype. In contrast, ORs for associations
 4                  between patella Pb levels and symptoms such as depression and positive symptom
 5                  distress index were larger among the 121 ALAD-2 carriers. In the NAS, effect
 6                  modification by ALAD genotype also was inconsistent for associations between tibia Pb
 7                  levels and cognitive performance (Rajan et al., 2008) (Section 5.3.2.7). The relationship
 8                  between ALAD-2 genotype and Pb bioavailability is not clear (Section 5.2.3.3).

 9                  A recent analysis of 1,987 adults ages 20-39 years participating in NHANES 1999-2004
10                  was the largest study of psychopathological effects in adults and included both men and
11                  women of multiple races and ethnicities (Bouchard et al., 2009). However, only
12                  concurrent blood Pb levels were available for analysis. Various symptoms were examined
13                  using the WHO Composite International Diagnostic Interview, which follows DSM
14                  criteria. Adults with concurrent blood Pb levels >0.7 (ig/dL had higher prevalence of all
15                  three self-reported disorders. Adults in the highest quintile of concurrent blood Pb level
16                  (> 2.11 (ig/dL) had the highest OR for major depressive disorder (OR: 2.32 [95% CI:
17                  1.13, 4.75]) and panic disorder (OR: 4.94 [95% CI: 1.32, 18.48]) compared with adults
18                  with blood Pb levels <0.7 (ig/dL with adjustment for age, sex, race, education, and
19                  poverty to income  ratio. A monotonic increase in ORs was not found across the quintiles
20                  of blood Pb levels. For all endpoints, ORs were larger in analyses excluding current
21                  smokers. While associations were found with relatively low concurrent blood Pb levels,
22                  there is uncertainty regarding the magnitude, timing, frequency, and duration of Pb
23                  exposure that contributed to the observed associations.

24                  In analyses of cohorts in California and New England born in the 1950s and 1960s, Opler
25                  et al.  (2008; 2004) reported associations between higher levels of cord plasma 5-ALA
26                  and subsequent diagnosis of schizophrenia spectrum disorder (ascertained using DSM-IV
27                  criteria) in adolescence and adulthood. Because of the lack of direct measurements of Pb
28                  biomarker levels, post hoc analysis, and limited consideration for potential confounding,
29                  firm conclusions are not warranted. Investigators measured 5-ALA levels in stored serum
30                  samples as surrogates for Pb exposure only citing previous observations of a high
31                  correlation (0.90) between categorized 5-ALA levels (cutpoint 9.05 ng/mL) and blood Pb
32                  levels (cutpoint 15 (ig/dL). In the California cohort, 5-ALA level > 9.05 ng/mL was
33                  associated with schizophrenia spectrum disorder with an OR of 2.43 (95% CI: 0.99,
34                  5.96), with adjustment for maternal age at delivery. In pooled analyses of the California
35                  and New England cohorts, 5-ALA level > 9.05 ng/mL was associated with schizophrenia
36                  spectrum disorder with an OR of 1.92 (95% CI: 1.05, 3.52), with adjustment for maternal
37                  age and education. An adjusted OR was not presented for the New England cohort alone,
      November 2012                             5-181                  Draft - Do Not Cite or Quote

-------
                   and it appeared that the association in the pooled dataset was influenced by that found in
                   the California cohort.
     5.3.6.2  Toxicological Studies of Mechanisms of Psychopathological Effects

 3                 An environmental origin of schizophrenia was proposed years ago (Tsuang. 2000). and
 4                 while epidemiologic evidence is inconclusive, toxicological studies have provided
 5                 indirect evidence to explain how Pb exposure may contribute to schizophrenia
 6                 development (Figure 5-11). Pb exposure has been shown to reduce function in the
 7                 NMDA receptor (NMDAR) and decrease hippocampal neurogenesis, which have been
 8                 associated with schizophrenia-related endpoints. Pb may bind a divalent cation site in the
 9                 NMDAR and allosterically inhibit glycine binding (Hashemzadeh-Gargari and Guilarte.
10                 1999). NMDAR antagonists have been shown to exacerbate schizophrenia symptoms in
11                 affected individuals and induce a schizophrenic phenotype in unaffected subjects (Coyle
12                 and Tsai. 2004). Evidence  supports a decrease in hippocampal degenerate gyrus (DG)
13                 neurogenesis as a mode of action for Pb-associated schizophrenia induction.
14                 Developmental Pb exposure inhibits neurogenesis in animal models (Section 5.3.11.9).
15                 Decreased neurogenesis is seen in patients with schizophrenia (Kempermann et al.. 2008;
16                 Reif etal.. 2006) and animal models of schizophrenia (Maedaetal.. 2007). and
17                 clozapine, a treatment for schizophrenia, restores hippocampal DG neurogenesis in
18                 animal models of schizophrenia (TVIaeda et al.. 2007) (Figure 5-11). These DG pathways
19                 are also NMDAR-dependent.
                     Human Populations or Cohorts
                     ALAD and Schizophrenia
                   Associated in Human Cohort
                        (Epidemiologic)
                      Rodent
                    ,  Studies
                     Pb exposure
 NMDAR
Antagonism
  Pharmacologically-
induced Schizophrenia in
    Animal models
     Figure 5-11      Schematic representation of the contribution of Pb exposure to
                       the development of a phenotype consistent with schizophrenia.
     November 2012
     5-182
        Draft - Do Not Cite or Quote

-------
      5.3.6.3  Summary of Psychopathological Effects in Adults

 1                   Studies of Pb exposure and behavior in adults have focused on psychopathological effects
 2                   rather than aggression and criminal behavior. Evidence links occupational Pb exposure
 3                   with self-reported mood disorders and anxiety, although the cross-sectional design and
 4                   potential confounding by other occupational exposures limits the implications. However,
 5                   supporting evidence is provided by a few but large (n = 744 and 1,787) cross-sectional
 6                   studies in nonoccupationally-exposed adults that found associations of concurrent blood
 7                   (Bouchard et al.. 2009) and tibia (Raian et al.. 2008) Pb levels with depression and
 8                   anxiety as assessed with widely-used questionnaires such as the Brief Symptom
 9                   Inventory and the WHO Composite International Diagnostic Interview. Evidence was
10                   provided by the NAS study of men (primarily white) and study of men and women
11                   (various races/ethnicities) participating in NHANES, both of which involve the
12                   examination of multiple exposures and outcomes. Studies in  adults with and without
13                   occupational Pb exposure found associations with adjustment for several confounding
14                   factors, including age, education, employment status, and alcohol use. The cross-
15                   sectional nature of these studies makes uncertain the temporal sequence between Pb
16                   exposure and development of psychopathological effects and the critical level, timing,
17                   frequency, and duration of Pb exposure. Both blood and bone Pb levels in adults reflect
18                   cumulative exposure, and it is uncertain what are the relative contributions of past versus
19                   recent Pb exposures to the observed associations.

20                   The epidemiologic evidence is supported by observations that early postnatal (to just after
21                   lactation) Pb exposure induces depressive- and anxiety-related phenotypes in animals
22                   (Section 5.3.5.2). The mode of action is supported by evidence for Pb-induced changes in
23                   the HPA axis and dopaminergic and GABAergic CNS processes, which mediate anxiety
24                   and depression. While epidemiologic evidence for Pb-associated schizophrenia is
25                   inconclusive, a few toxicological studies have shown that Pb exposure decreases
26                   NMDAR function and hippocampal DG neurogenesis, which are found in animal models
27                   of schizophrenia (agitation, trouble finding food, reduced swimming behavior).

28                   Epidemiologic evidence indicates associations of Pb biomarker levels with depression
29                   and anxiety in children and adults as rated by self, parents, or teachers. Differences in
30                   associations for other behaviors may be related to what endpoints are examined. Studies
31                   in children and young adults have focused on attention-related behavioral problems and
32                   misconduct; studies of older adults did not examine such behaviors. Differential effects in
33                   children and adults also may be expected given the predominance of different
34                   neurophysiological processes  operating at different ages, for example, neurogenesis and
35                   brain development in children and neurodegeneration in adults. Differences in the effects
      November 2012                             5-183                  Draft - Do Not Cite or Quote

-------
 1                  of Pb exposure between children and adults also may be related to differences in Pb
 2                  exposure profiles by age.
      5.3.7  Sensory Organ Function


      5.3.7.1  Epidemiologic Studies of Sensory Organ Function in Children

 3                  Although not as widely examined as cognitive and behavioral outcomes, several studies
 4                  found associations of higher blood Pb level with higher hearing thresholds or poorer
 5                  auditory processing in children, with evidence limited largely to that described in the
 6                  2006 Pb AQCD (U.S. EPA. 2006b). The prospective Cincinnati study with repeat
 7                  measures of blood Pb prenatally to age 5 years provided information on the temporal
 8                  sequence between Pb exposure and hearing effects and potential critical lifestages of
 9                  exposure and had extensive consideration for potential confounding. In this cohort,
10                  poorer auditory processing in 215 children at age 57 months was associated with higher
11                  prenatal maternal, neonatal (10-day), yearly age 1 to 5 year (means: 10.6-18.4 (ig/dL),
12                  and lifetime average blood Pb levels, with the strongest associations found for neonatal
13                  blood Pb level (mean: 4.8 [SD: 3.3] (ig/dL). A 1 (ig/dL higher neonatal blood Pb level
14                  was associated with a 0.20-point (p < 0.01)  and 0.26-point (p <  0.10) lower score on the
15                  total and left ear Filtered Word test (indicating incorrectly identified, filtered, or muffled
16                  words), with adjustment for hearing screen, social class, HOME score, birth weight,
17                  gestational age, obstetrical complications, and maternal alcohol consumption (Dietrich et
18                  al., 1992). Overall, the findings pointed to a stronger effect of Pb exposure in infancy.

19                  Additional support was provided by the large U.S. cross-sectional NHANES II
20                  (n = 4,519) (Schwartz and Otto.  1987) and Hispanic Health and Nutrition Examination
21                  Survey (HHANES, n = 3,262) (Schwartz and Otto. 1991)  studies. The examination  of
22                  multiple exposures and outcomes in these studies reduces  the likelihood of participation
23                  conditional on Pb exposure and hearing function. Each study found an association
24                  between higher concurrent blood Pb level and higher hearing thresholds in children (ages
25                  4-19 years). In HHNANES, an increase in concurrent blood Pb  level (median: 8 (ig/dL)
26                  from 6 to 18 (ig/dL also was associated with a 15% increase in the percentage of children
27                  with a substandard hearing threshold (2,000 Hz). Higher concurrent blood Pb level also
28                  was associated with higher hearing thresholds across several frequencies in a smaller
29                  (n = 155) study of similarly aged (4-14 years) children in Poland with similar blood Pb
30                  levels (median: 7.2 (ig/dL [range: 1.9-28]) (Osman et al.. 1999). In each of the studies in
31                  children, associations persisted in analyses restricted to subjects with concurrent blood Pb
32                  levels <10 (ig/dL. Each of these  studies adjusted for different potential confounding
      November 2012                            5-184                  Draft - Do Not Cite or Quote

-------
 1                   factors, but in stepwise regression analyses, each considered parental education, maternal
 2                   smoking, nutritional factors, and environmental noise. Across studies, associations were
 3                   found with adjustment for factors such as age, sex, ethnicity, family income, concurrent
 4                   or past colds, antibiotic use, degree of urbanization, and Apgar score.

 5                   Mechanistic support for associations with higher hearing thresholds in children was
 6                   provided by a few studies that found associations of blood Pb level with lower brainstem
 7                   auditory evoked potentials in children. Brainstem auditory evoked potentials measure
 8                   nerve electrical activity and are used to assess neurological auditory function. In
 9                   prospective analyses of the Mexico City cohort (n = 100,  113), Rothenberg et al. (2000;
10                   1994b) reported associations with prenatal and postnatal blood Pb levels. At age 5-7
11                   years, the shape of the concentration-response relationship differed between prenatal
12                   maternal and postnatal (ages 1 and 4 years) blood Pb  level. Higher blood Pb level at age 1
13                   year and at age 4 years (mean reported for age 28 months: 10.8 (ig/dL) was associated
14                   with lower interpeak intervals in auditory evoked potentials. Prenatal maternal blood Pb
15                   level showed a biphasic relationship, with lower evoked potentials found with blood Pb
16                   levels 1-8 (ig/dL and higher evoked potentials found with blood Pb levels 8-30  (ig/dL.
17                   Results were adjusted for age, sex, and head circumference. In this cohort, maternal first
18                   trimester blood Pb levels 10.5-32 ug/dL were associated with supernormal retinal ERG
19                   (Rothenberg et al.. 2002b), the impact of which on visual  impairment is not clear.
20                   Associations with lower auditory evoked potentials also were found in small studies
21                   (n = 13, 29) of children with higher concurrent blood Pb levels (i.e., range 6-84 (ig/dL)
22                   than most of the current U.S. population (Holdstein et al.. 1986; Robinson et al.. 1985).

23                   Recent cross-sectional studies aimed to identify the locus  in the auditory  system affected
24                   by Pb exposure in the examination of a population  of children (n = 53, 117, ages 2-18
25                   years) living in Pb glazing communities in Ecuador with higher blood Pb levels than
26                   those relevant to current U.S. children (means 33 and 37 (ig/dL) (Buchanan et al.. 2011;
27                   Counter et al..  2011). Concurrent blood Pb level was  not correlated with the acoustic
28                   stapedius reflex (Counter et al.. 2011) or distortion product otoacoustic emissions
29                   (Buchanan et al.. 2011). indicating lack of effect on the auditory brainstem or inner ear,
30                   respectively. Other loci were not examined, and potential  confounding was not
31                   considered.
      November 2012                             5-185                  Draft - Do Not Cite or Quote

-------
      5.3.7.2  Epidemiologic Studies of Sensory Organ Function in Adults

 1                  Studies of auditory function reviewed in the 2006 Pb AQCD consistently indicated
 2                  associations between blood Pb levels and changes in auditory evoked brainstem
 3                  potentials in occupationally-exposed adults but found less consistent associations with
 4                  hearing thresholds (U.S. EPA. 2006b). A few recent studies found increases in hearing
 5                  thresholds in Pb-exposed workers. A recent analysis of the NAS provided evidence in
 6                  nonoccupationally-exposed men for associations of tibia Pb levels with hearing loss.

 7                  Among 448 men in the NAS, higher tibia Pb level (mean: 22.5 ug/g) at mean age 64.9
 8                  years, measured up to 20 years after initial hearing testing, was associated with a faster
 9                  rate of increase in hearing threshold for frequencies of 1, 2, and 8 kHz and a pure tone
10                  average. Men were free of hearing loss at baseline and had hearing tested repeatedly
11                  (median 5 times per subject) over a median of 23 years (Park et al., 2010). Blood Pb was
12                  not examined. In cross-sectional analyses adjusted for age, race, education, body mass
13                  index, pack-years of cigarettes, diabetes, hypertension, occupational noise (based on a
14                  job-exposure estimate), and presence of a noise notch (indicative of noise-induced
15                  hearing loss), higher patella Pb level (mean 32.5  ug/g, measured within 5 years of hearing
16                  test) was associated with a higher hearing threshold for frequencies greater than 1 kHz. A
17                  21 ug/g (interquartile range) increase in patella Pb level was associated with pure tone
18                  average hearing loss with an OR of 1.48 (95% CI: 1.14, 1.91) in adjusted analyses.
19                  Similar, but slightly weaker associations were found for tibia bone Pb levels. In the NAS,
20                  bone Pb levels were measured after the initial hearing measurement but reverse causation
21                  is unlikely since bone Pb is an indicator of cumulative Pb exposure, and tibia Pb has a
22                  half-life on the order of decades (Section 4.3). Bone Pb levels increase with age, and
23                  although age was included as a model covariate, residual confounding by age is possible.

24                  Recent cross-sectional studies added evidence for associations between higher concurrent
25                  blood Pb levels and higher hearing thresholds in adults with occupational Pb exposures.
26                  A hospital-based case-control study examined workers from diverse occupations and
27                  examined potential confounding by other occupational exposures. Cases included
28                  workers referred for hearing testing (average hearing thresholds above 25 dB), and
29                  controls comprised workers with normal hearing thresholds who were having routine
30                  occupational health examinations (Chuang et al.. 2007). Geometric mean blood Pb levels
31                  were 10.7 ug/dL for the  121 cases and 3.9 ug/dL for the 173 controls. In models that
32                  adjusted for age, smoking, alcohol consumption, years of noise exposure, as well as Mn,
33                  As, and Se levels in blood, higher blood Pb levels were associated with higher hearing
34                  threshold (0.5-6 kHz). The potential selection bias arising from the nonrandom
35                  population sample may limit implications of these findings. Other studies found
36                  associations of higher concurrent blood Pb level with increased hearing thresholds or
      November 2012                            5-186                  Draft - Do Not Cite or Quote

-------
 1                  hearing loss in Pb-exposed workers (n = 183-259) but had limited or no consideration for
 2                  potential confounding (Forst et al.. 1997) and/or examined workers with mean blood Pb
 3                  levels >50 ug/dL (Hwang et al.. 2009; Wu et al.. 2000).
      5.3.7.3  lexicological Studies of Sensory Organ Function

                     Effects on Auditory Function

 4                   The 2006 Pb AQCD (U.S. EPA. 2006b) described impaired auditory function in
 5                   nonhuman primates exposed to lifetime Pb beginning in gestation or birth to ages
 6                   8-13 years (resulting in blood Pb levels 33-170 (ig/dL during or just after Pb exposure)
 7                   (Rice. 1997: Lilienthal and Winneke. 1996). Pb-related effects persisted after Pb
 8                   exposure was terminated, and blood Pb levels had returned to baseline. Recent studies
 9                   provided similar evidence with lower Pb exposures and blood Pb levels. These
10                   observations in animals are consistent with the epidemiologic associations described
11                   above (Sections 5.3.7.1 and 5.3.7.2) but were related to higher Pb exposures than those
12                   relevant for humans. Monkeys with lifetime Pb exposure from birth to age 13  years were
13                   found to have decrements in auditory function, as evidenced by elevated thresholds and
14                   increased latencies in brainstem auditory evoked potentials. Further, half of the pure tone
15                   detection thresholds were above the  control range at certain frequencies (Rice. 1997). In
16                   addition to indicating hearing loss, brainstem auditory evoked potentials can indicate
17                   impaired synaptic maturation and incomplete neuron axon myelination leading to
18                   impaired neuronal conduction (Gozdzik-Zolnierkiewicz and Moszynski. 1969). Thus, the
19                   findings from Rice (1997) and those described in the preceding sections for children may
20                   indicate that Pb exposure impairs auditory nerve conduction. Studies in animals with
21                   blood Pb levels >300 (ig/dL found that the cochlear nerve was especially sensitive to Pb
22                   exposure (Gozdzik-Zolnierkiewicz and Moszvnski. 1969).

23                   In a recent study, Laughlin et al. (2009) studied rhesus monkeys exposed to Pb acetate
24                   prenatally to birth or postnatally from birth through weaning at age 5.5 months (maternal
25                   drinking water, 3 months prior to mating until weaning, resulting in bone Pb levels at age
26                   11 years of 7 and 13  (ig/g for prenatal and postnatal groups, respectively, and blood Pb
27                   levels during Pb exposure of 35 and 46 (ig/dL, respectively). Auditory threshold testing
28                   and threshold task testing was conducted at 13 years of age after blood Pb levels had
29                   returned to those found in controls. At birth, animals were cross fostered, creating  a
30                   control group, a prenatal Pb group, and a postnatal Pb group; however, Pb-exposed
31                   animals were analyzed as a single group. Pb exposure induced small, statistically
32                   nonsignificant elevations in auditory thresholds in animals. Auditory threshold task-
33                   related behavioral testing was also impaired in Pb-exposed animals. This study has


      November 2012                             5-187                  Draft - Do Not Cite or Quote

-------
 1                  multiple limitations that could have contributed to that lack of statistically significant
 2                  aberrations, including limited power with the examination of 5 animals per group, the
 3                  inability of some of the monkeys to engage or focus on the task at hand which resulted in
 4                  fewer available measurements, differences between the sexes in inattention, and mixing
 5                  of the postnatal Pb and prenatal Pb-exposed animals into one group.


                    Effects on Vision

 6                  The 1986 and 2006 Pb AQCDs (U.S. EPA. 2006b) detailed the  effects of Pb exposure
 7                  during perinatal development and adulthood on the visual system of animals, including
 8                  reduced visual acuity and supporting mechanisms of action such as alterations in the
 9                  retina (Fox et al.. 1997; Fox and Sillman. 1979). CNS visual processing areas (Costa and
10                  Fox. 1983). and subcortical neurons involved in vision (Cline et al.. 1996). For example,
11                  environmentally-relevant doses of Pb (10~3 (iM) in tadpoles inhibited the growth of
12                  neurons in the subcortical retinotectal pathway, the main efferent from the retina (Cline et
13                  al.. 1996). Pb-related aberrations in electrical responses in retinal cells, as measured by
14                  electroretinograms (ERGs), have been found in rodents, nonhuman primates, and
15                  children. Recent research expands upon the extant evidence by examining effects in
16                  animals with  lower Pb exposures or blood Pb levels.

17                  Extensive work in nonhuman primates with Pb exposure during development or over a
18                  lifetime (peak blood Pb levels 50-115 (ig/dL) showed dysfunction in temporal visual
19                  function (responses to different frequencies of light flicker) under high luminance but no
20                  change in spatial function (Rice. 1998). A recent study found no effects of Pb exposure
21                  on spatial acuity as assessed with the modified Teller preferential looking paradigm
22                  (Laughlin et al.. 2008) in rhesus monkeys exposed to Pb  acetate postnatally (PND8-age
23                  26 weeks via commercial milk formula, producing blood Pb levels of 35-40 ug/dL). In
24                  monkeys, effects on vision were tested with higher Pb exposures than those relevant to
25                  humans. Low-level developmental Pb exposure was found to result in sensorimotor
26                  deficits in adult zebrafish (Rice et al.. 2011). Fish that were exposed as embryos (2 to 24
27                  hours post-fertilization) to water containing 0.03 uM PbCl2 had impaired response to
28                  visual stimulation (a rotating bar) under low light conditions. These zebrafish also failed
29                  to respond normally to mechanosensory stimulation (0.01 and 0.03 uM PbCl2), showing a
30                  significantly impaired startle response.

31                  Animal toxicological evidence also shows that the lifestage of exposure and the dose of
32                  Pb contribute to the complex and variable effects of Pb on the retina, as assessed by ERG
33                  (summarized in Table 5-13).  The biological relevance of these variable findings is
34                  uncertain. Female rats exposed postnatally to 200 or 2,000 ppm Pb acetate exposure via
35                  dam drinking water from birth through lactation, resulting in blood Pb levels of 19 and
      November 2012                            5-188                  Draft - Do Not Cite or Quote

-------
 1                  59 ug/dL at weaning, respectively, had subnormal scotopic ERGs (decreased A- and B-
 2                  wave amplitudes) with decreased sensitivity and temporal resolution when assessed at
 3                  90 days of age (Fox et al., 1991). Similar results were obtained in multiple studies
 4                  conducted in in vitro models (Otto and Fox. 1993; Fox and Farber. 1988; Fox and Chu.
 5                  1988). Monkeys exposed to relatively high levels of Pb continuously from the prenatal
 6                  period to age 7 years (350 or 600 ppm Pb acetate, resulting in blood Pb levels of 40 and
 7                  50 ug/dL, respectively) had persistently increased maximal retinal ERG amplitude (B-
 8                  wave only, supernormality) and increased mean ERG latency when assessed 2 years after
 9                  Pb exposure was terminated when blood Pb levels were <10  ug/dL (Lilienthal et al.
10                  1988).
     November 2012                             5-189                  Draft - Do Not Cite or Quote

-------
Table 5-13 Summary of Pb-related effects observed on the visual system.

Study
Foxet al.
(2008)


Lilienthal et
al. (1988)

Fox et al.
(1997)

Rothenberg
etal.
(2002b)
F = Females,
" — " Denotes

Species
Long-
Evans
Rat


Rhesus
Monkey

Long-
Evans
Rat

Human
children

Pb Exposure Maximal Blood
Sex Protocol/Dose Pb Level (ug/dL)
F Prenatal-PNDIODW
Low, 27 ppm 1 2
Moderate, 55 ppm 24
High, 109 ppm 46
M Prenatal-lifetime, DW
& F 350 ppm -50
600 ppm -115
F PND1-PND21, DW
200 ppm DW 1 9
2,000 ppm DW 59
M Prenatal maternal a 10. 5
^ p 1st trimester

ERG
Abnormality
Supernormal
Supernormal
Subnormal
Supernormal
Supernormal
Subnormal
Subnormal
Supernormal
Retinal Retinal Retinal Cell
Progenitor cell Cellular Dopamine Layer
proliferation Apoptosis Levels Thickness
Yes Not affected Dose-dependent J, f
Yes Not affected Dose-dependent J, f
No Yes Dose-dependent J, J,

— — — —
— Yes — I
— Yes — I
— — — —
M = Males, PND = postnatal day, DW = Drinking water
not measured.



November 2012
5-190
Draft - Do Not Cite or Quote

-------
 1                  Pb exposure beginning in the gestational period (Pb acetate in drinking water from
 2                  2 weeks before mating to PND10) also induced supernormal ERGs in adult Long-Evans
 3                  rats, but only with low (27 ppm) and moderate (55 ppm) doses that produced blood Pb
 4                  levels 10-12 (ig/dL and 21-24 (ig/dL (Fox et al.. 2008) (Figure 5-12 and Table 5-13). This
 5                  exposure window represents the developmental period for the retina of the rat and is
 6                  analogous to gestational human retinal development. Subnormal ERGs were induced by
 7                  the high 109 ppm dose (Figure 5-12). which produced blood Pb levels 40-46  (ig/dL.
 8                  Results of this rodent study demonstrated persistent supernormal scotopic rod
 9                  photoreceptor-mediated ERGs in animals with blood Pb levels relevant to humans. These
10                  findings were consistent with the associations observed between supernormal ERG and
11                  prenatal maternal blood Pb levels > 10.5 ug/dL in male and female children (Rothenberg
12                  et al., 2002b). The functional relevance of findings is uncertain as supernormal scotopic
13                  ERGs may be recorded without other overt ophthamalogic changes and are rarely seen in
14                  the clinical setting (Terziivanov et al., 1982).

15                  Animal studies indicate that the dose of Pb and the exposure lifestage not only
16                  differentially affect functional tests, i.e., ERG but also differentially affect retinal cell
17                  numbers and morphology. Concomitant with Pb-induced supernormal ERGs, Fox et al.
18                  (2008) found that 27 and 55 ppm gestational plus early postnatal Pb exposure increased
19                  neurogenesis of rod photoreceptors and rod bipolar cells without affecting apoptosis of
20                  Miiller glial cells and increased the number of rods in central and peripheral retina (Table
21                  5-13). Concomitant with subnormal ERGs, higher-level gestational plus early postnatal
22                  Pb exposure (109 ppm, blood Pb level 40-46 (ig/dL) decreased the number of rods in the
23                  central and peripheral retina (Fox et al.. 2008). Early postnatal (PND1-PND21) Pb
24                  exposure (200 or 2,000 ppm, producing blood Pb levels 19 and 59 (ig/dL) induced
25                  scotopic ERG subnormality in adult rats, decreased the number of rods in the central  and
26                  peripheral retina, and decreased the retinal Zn concentration (Fox et al.. 1997) (Table
27                  5-13). Similar observations were made in separate work in mice. Low and moderate
28                  doses of Pb from gestation to PND10 (27 or 55 ppm Pb acetate in dam drinking water,
29                  producing blood Pb levels of 12 and 25 ug/dL, respectively) induced greater and
30                  prolonged rod bipolar cell neurogenesis and greater thickness and cell number of the
31                  outer and inner neuroblastic layers of the retina (Giddabasappa et al., 2011). As in rats, at
32                  higher doses of Pb (109 ppm Pb acetate, resulting in blood Pb levels of 56  ug/dL), there
33                  was no increased rod neurogenesis in mice.  Nitric oxide has been shown to regulate
34                  retinal progenitor cell proliferation in chick embryos (Magalhaes et al.. 2006). Thus,
35                  these authors postulated that impaired NO production may contribute to aberrant retinal
36                  cell proliferation (Giddabasappa et al..  2011).
      November 2012                            5-191                 Draft - Do Not Cite or Quote

-------
                                                                       •§   ; 0:0
                                                                       I   1J»
                                                                       I?
                                                                       ?-i I.WO
                                                     High iino
      Note: *p <0.05. Low Pb = 27 ppm, blood Pb level 10-12 |jg/dL, Moderate Pb = 55 ppm, blood Pb level 21-24 |jg/dL, High
      Pb = 109 ppm, blood Pb level 40-46 |jg/dL. Relative to controls (gray bars), low (white bars) or moderate (blue bars) Pb exposure
      from gestation through postnatal day 21 induced supernormal electroretinograms (ERGs) whereas high Pb exposure (black bars)
      induced subnormal ERGs.
      Source: Fox et al. (2008)

      Figure 5-12      Retinal a-wave and b-wave ERG amplitude in adult rats after
                         prenatal plus early postnatal Pb exposure.

 1                   Mechanistic understanding of the effect of Pb on the visual system includes the capability
 2                   of Pb to displace divalent cations, inhibit physiological enzymes, regulate cell
 3                   proliferation and apoptosis, perturb normal neuroanatomical formation (cytoarchitecture
 4                   in the brain), and affect neurotransmitters. The effects of Pb on the retina were shown to
 5                   be mediated by its inhibition of cGMP phosphodiesterase (PDE) (Srivastava  et al.. 1995;
 6                   Fox and Farber. 1988). Independent of Pb exposure, pharmacological inhibition of cGMP
 7                   PDE has been linked with visual problems including alterations in scotopic ERGs (Laties
 8                   and Zrenner. 2002). Postnatal Pb exposure of animals (peak blood Pb levels:  19,
 9                   59 (ig/dL) or in vitro Pb exposure of rods isolated from these animals elevated cGMP
10                   which contributed to elevated rod calcium concentration (Fox and Katz. 1992) and
11                   subsequently induced apoptotic cell death in a concentration-dependent manner.

12                   Pb has been shown to affect a plethora of neurotransmitters in the brain, and  it has
13                   recently been shown to affect neurotransmitters in the retina. With the aforementioned
14                   gestational to PND10 exposure, Pb induced concentration-dependent decreases in adult
15                   rat retinal synthesis of dopamine, which has functions in retinal growth and development
16                   and signal transduction in rods and cones (Fox et al.. 2008) (Figure 5-13). As discussed in
17                   the 2006 Pb AQCD (U.S. EPA. 2006b). other mode of action support for the effects of Pb
18                   on the visual system is provided  by observations of Pb-induced decreased Na+/K+ATPase
19                   activity which have been reported in vitro and in vivo.  Also, structural changes from
20                   chronic Pb exposure (birth to age 6 years) included cytoarchitectural changes in visual
21                   projection areas of the brain of monkeys; maximum blood Pb level in the low and high
22                   dose group reached 20 (ig/dL and 220 (ig/dL, respectively (Reuhl etal.. 1989).
      November 2012
5-192
Draft - Do Not Cite or Quote

-------
    6

u e 4
                     2.0
                 I   1.6
                 si
                 >'! u
                 1— 0.8
                     0.4
                      0
                                           I Dark-adapted
                                           I Light-adapted
                         Control    Low    Moderate   High
                                    GLE
                                             Control   Low   Moderate  High
                                                       GLE
                         Control    Low   Moderate   High
                                    GLE
                                             Control   Low   Moderate   High
                                             	 GLE
     Note: *p O.05. GLE = Gestational Pb exposure to postnatal day 10. Low Pb = 21 ppm, blood Pb levels 10-12 ug/dL, Moderate
     Pb = 52 ppm, blood Pb levels 21-24 ug/dL, High Pb = 109 ppm, blood Pb levels 40-46 ug/dL. A. DA = dopamine,
     B. DOPAC = dopamine metabolite, C. HVA = dopamine metabolite, D. DOPAC/DA = ratio of dopamine metabolite to dopamine. Pb
     exposure decreased dopamine, DOPAC, HVA, and DOPAC/DA in a concentration-dependent manner in light-adapted animals (blue
     bars). In dark adapted animals (black bars), Pb exposure decreased dopamine, DOPAC, and DOPAC/DA but not always in
     concentration-dependent manner.
     Source: Fox et al. (2008)
     Figure 5-13      Retinal dopamine  metabolism in adult control and gestationally
                        Pb-exposed (GLE) rats.
     5.3.7.4  Summary of Sensory Function
i
2
3
4
5
6
7
8
9
   Children
   Several studies indicated that higher blood Pb levels are associated with decrements in
   auditory function in children ages 3-19 years, as evidenced by increases in hearing
   thresholds. Results from the prospective Cincinnati cohort study (n = 215) provide key
   evidence for associations of neonatal, yearly age 1 to 4 year, and lifetime average blood
   Pb levels with increased hearing thresholds at age 57 months (Dietrich et al.. 1992). and
   large (n = 3,000-4,000) cross-sectional analyses of children participating in NHANES
   and HHANES provide supporting evidence for concurrent blood Pb levels (Schwartz and
   Otto. 1991. 1987). The examination of multiple exposures and outcomes in NHANES
   and HHANES reduces the likelihood of participation conditional on Pb exposure and
     November 2012
                              5-193
Draft - Do Not Cite or Quote

-------
 1                   hearing function. In the Cincinnati cohort, mean blood Pb levels were 4.8 (ig/dL for
 2                   neonatal and 17.4 (ig/dL for lifetime average. In HHANES, the median concurrent blood
 3                   Pb levels was ~8 (ig/dL. Across studies, associations were found with adjustment for
 4                   factors such as age, sex, ethnicity, family income, concurrent or past colds, antibiotic use,
 5                   degree of urbanization, and Apgar score. Potential confounding by parental education,
 6                   nutritional factors, environmental noise, and maternal smoking also was considered.

 7                   Mechanistic evidence was provided by observations of associations between blood Pb
 8                   level with lower auditory evoked potentials in children, particularly associations found in
 9                   the prospective analysis of children in Mexico City with prenatal, age 1 year, and age
10                   4-year blood Pb levels (Rothenberg et al.. 2000). Biological plausibility is provided by
11                   evidence in animals indicating increased thresholds and increased latencies in brainstem
12                   auditory evoked potentials in nonhumans primates with multi-year postnatal Pb exposure
13                   beginning at birth (Rice. 1997; Lilienthal and Winneke. 1996). although auditory
14                   assessment were made in adult animals ages 8-13 years. Pb exposure limited to the
15                   gestational period or to the postnatal period to age 5 months was found to have weaker
16                   effects (Laughlin et al.. 2009). In animals, auditory effects were found with higher blood
17                   Pb levels (i.e., 33-170 (ig/dL) than those relevant to humans; thus, it is difficult to
18                   ascertain support for observations in children.

19                   Maternal first trimester blood Pb levels 10.5-32  ug/dL were associated with supernormal
20                   retinal ERGs in children in Mexico City at ages  5-7 years (Rothenberg et  al., 2002b). The
21                   animal evidence showed ERGs in different directions depending on lifestage of Pb
22                   exposure and blood Pb level. Supernormal ERGs were found in adult rats with prenatal
23                   plus early postnatal (PND10) Pb exposure that produced blood Pb levels of 12 and
24                   24 ug/dL (Fox et al., 2008).  The implications of supernormal ERGs on visual  impairment
25                   is not clear, and the biological relevance of the nonlinear concentration-response is not
26                   clear. For these reasons, the  evidence for the effects of Pb on retinal ERGs was not a
27                   major consideration in drawing conclusions about the effects  of Pb exposure on sensory
28                   function. In these animals, Pb exposure also increased rod cell neurogenesis and
29                   decreased dopamine. Toxicological studies demonstrated a range of other effects on the
30                   visual system including impaired visual function, and potential mechanisms such as
31                   alterations in morphology and cell architecture, signaling, enzyme inhibition,
32                   neurotransmitter levels, neuroanatomical development, cell proliferation,  and  retinal cell
33                   apoptosis.
      November 2012                             5-194                 Draft - Do Not Cite or Quote

-------
                    Adults

 1                  In adults, increased hearing thresholds or hearing loss were associated with bone Pb
 2                  levels in 448 NAS men who were unlikely to have had occupational Pb exposures (Park
 3                  et al., 2010) and with concurrent blood Pb levels in adults with current occupational Pb
 4                  exposure. In the NAS, the examination of multiple exposures and outcomes reduces the
 5                  likelihood of participation conditional on Pb exposure and hearing function. Among NAS
 6                  men, higher tibia Pb  levels were associated with a faster rate of increase in hearing
 7                  thresholds over a 23  year follow-up with adjustment for age, race, education, body mass
 8                  index, pack-years of cigarettes, diabetes, hypertension, occupational noise, and presence
 9                  of a noise notch. Tibia Pb levels were measured up to 20 years after initial hearing
10                  testing, and while Pb in tibia has a half-life on the order of decades, there is uncertainty
11                  regarding the temporal sequence with changes in hearing thresholds. Temporality also is
12                  difficult to establish in the cross-sectional occupational studies. A hospital-based case
13                  control study found an association between higher concurrent blood Pb levels and higher
14                  hearing thresholds among workers with relevant blood Pb levels (means 10.7 and
15                  3.9 ug/dL in workers with and without hearing problems, respectively) (Chuang et al.,
16                  2007). Among other  factors, results were adjusted for other occupational exposures.
17                  Biological plausibility is provided by evidence in animals with lifetime Pb exposure but
18                  with higher blood Pb levels (i.e., 33-107 (ig/dL) than those relevant to humans. Adult
19                  monkeys were found to have supernormal ERGs with developmental or lifetime Pb
20                  exposure that produced blood Pb levels of 50 and 115 (ig/dL (Rice. 1998): however,
21                  Pb-associated visual  system effects in human adults are not well characterized.
      5.3.8  Motor Function

22                   Some studies in children have assessed fine motor function, i.e., response speed,
23                   dexterity, as part of a battery of neurodevelopmental testing, and most have found
24                   associations with blood Pb level. Fewer studies have examined gross motor function,
25                   i.e., postural balance, action tremor, agility, but also have found associations with blood
26                   Pb level in children. Poorer motor function also was found in Pb-exposed workers.

27                   Key evidence from the prospective studies of the Cincinnati and Yugoslavia cohorts
28                   demonstrated associations of blood Pb levels with poorer motor function with either
29                   adjustment for or consideration for several potential confounding factors related to SES,
30                   parental caregiving quality and education, smoking exposure, birth outcomes, sex, and
31                   child health. In the Cincinnati cohort, higher earlier childhood average blood Pb levels
32                   (0-5 year average or 78 month, exact levels not reported) were associated with poorer
33                   fine (n = 195) (Ris et al., 2004) and gross motor function (n = 91) (Bhattacharya et al..
      November 2012                             5-195                  Draft - Do Not Cite or Quote

-------
 1                   2006) assessed in adolescence (ages 12, 15-17 years). In this cohort recruited from birth,
 2                   follow-up participation was high and not conditional on blood Pb levels. The fine motor
 3                   function results were adjusted for maternal IQ, SES, HOME score, and adolescent
 4                   marijuana consumption. Collectively, these findings suggest the persistence of effects of
 5                   earlier childhood Pb exposure to later childhood; however, later childhood or concurrent
 6                   blood Pb levels were not examined. Assessments in Cincinnati cohort children at age 6
 7                   years indicated associations of concurrent (mean:  11.66 (ig/dL), lifetime average (mean:
 8                   12.3 (ig/dL), and neonatal (mean: 4.8 (ig/dL) but not prenatal maternal (mean: 8.4 (ig/dL)
 9                   blood Pb levels with poorer upper limb dexterity, fine motor composite score (n = 245)
10                   (Dietrich et al., 1993a). and poorer postural balance (n = 202) (Bhattacharya et al., 1995).
11                   These results were adjusted for HOME score and race. Additional covariates included
12                   maternal IQ, SES, and sex for fine motor functions (Dietrich et al., 1993a) and height,
13                   BMI, birth weight, bilateral ear infection, and foot area for postural balance
14                   (Bhattacharya et al., 1995). Blood Pb levels were associated with fine and gross motor
15                   function in unadjusted and adjusted analyses, increasing confidence that confounding by
16                   the examined covariates did not unduly bias the observed associations. Prospective
17                   analysis of the Yugoslavian cohort indicated an association of lifetime average blood Pb
18                   level (exact levels not reported) with decrements in fine but not gross motor function at
19                   age 4.5 years in 283 children  (Wasserman et al.. 2000). Although only 50% of the cohort
20                   was examined, and  participation was greater among children with lower SES and HOME
21                   score, participation  was not conditional on higher blood Pb levels.

22                   Supporting evidence was provided by most cross-sectional studies of motor function,
23                   which found that concurrent blood Pb level was associated with poorer fine motor
24                   function in children in Asia and Canada ages 3-16 years (n = 61-814) (Palaniappan et al..
25                   2011; Min et al., 2007; Despres et al., 2005). In exception, Surkan et al. (2007) found that
26                   higher concurrent blood Pb level was associated with better fine motor function as
27                   indicated by faster finger tapping speed among 534 children in New England. This study
28                   examined lower blood Pb  levels than other studies (mean: 2.2 (ig/dL) but a similar age
29                   range (6-10 years) and set of potential confounding factors (age, sex, caregiver IQ, SES,
30                   race). Concurrent blood Pb level (mean: 5.0 (ig/dL) was associated with greater sway
31                   oscillation, alternating arm movements, and action tremor in 110 Inuit children (ages 4-6
32                   years) in Quebec, Canada (Despres et al.. 2005) with consideration for potential
33                   confounding by factors such as HOME score, maternal education and several nutrient
34                   levels. The population of Inuit children was selected from subsistence fishing
35                   communities, who have higher exposure to Hg and polychlorinated biphenyls. Several
36                   indices of fine and gross motor function were associated with blood Pb level, with
37                   adjustment for these other exposures. Min et al. (2007) found impaired fine motor
38                   function in 61 children in Korea with a mean concurrent blood Pb level of 2.9 (ig/dL;
39                   however, the results were not adjusted by SES-related variables.

      November 2012                             5-196                 Draft - Do Not Cite or Quote

-------
 1                   An association of Pb exposure with poorer motor function in adults was found in
 2                   Pb-exposed workers (Iwata et al.. 2005). although implications are limited by the cross-
 3                   sectional design, high concurrent blood Pb levels (mean: 40 ug/dL), and lack of
 4                   consideration for potential confounding by other occupational exposures. Among 121 Pb
 5                   smelter workers in Japan, higher blood Pb level was associated with greater sagittal sway
 6                   with eyes open (p <0.05) and eyes closed (p <0.01) and transversal sway with eyes closed
 7                   (p <0.05) with adjustment for age, height, smoking status, and drinking status. The
 8                   authors calculated a benchmark dose level (Budtz-Jorgensen et al.. 2001; NRC. 2000) of
 9                   14.3 ug/dL from a linear concentration-response model. A supralinear concentration-
10                   response function was found to fit the data slightly better than was a linear function.

11                   Pb exposure has shown mixed effects on endurance, balance, and coordination in animals
12                   as measured by rotarod performance and treadmill testing. Lower concentration
13                   gestational plus early postnatal (to PND10) Pb exposure (27 ppm, producing peak blood
14                   Pb level < 10 (ig/dL at PNDO-PND10) resulted in significantly poorer rotarod
15                   performance (i.e., falling off more quickly) than did higher exposure (109 ppm, blood Pb
16                   level: 33-42 (ig/dL) in male  (but not female) adult mice, indicative of a nonlinear
17                   concentration-response relationship (Leasure et al.. 2008). Other rotarod experiments
18                   examining various speeds of rotarod rotation and higher Pb exposures producing blood
19                   Pb levels >60 (ig/dL, some administered by routes with uncertain relevance to humans,
20                   yielded mixed results  (Kishietal.  1983; Grant etal.. 1980; Overmann. 1977). Herring
21                   gull chicks injected with a single i.p. bolus dose of Pb (100 mg/kg Pb acetate, a dose
22                   selected to represent exposure in the wild) had slower development of motor skills versus
23                   control birds, as assessed by the treadmill test (Burger and Gochfeld. 2005).

24                   In summary, epidemiologic evidence demonstrates associations of higher blood Pb levels
25                   with poorer fine and gross motor function in children ages 3-17 years. Little evidence is
26                   available in adults. Prospective analyses of the Cincinnati and Yugoslavia cohorts
27                   (n = 91-283) that considered several potential confounding factors such as SES and child
28                   health found associations with earlier childhood blood Pb levels (i.e., age 78 month, 0-5
29                   year average) in adolescents (Bhattacharya et al.. 2006; Ris et al.. 2004) and neonatal,
30                   lifetime average,  and concurrent blood Pb levels in children ages 4-6 years (Wasserman
31                   et al.. 2000; Bhattacharya et al.. 1995; Dietrich et al.. 1993a). In the Cincinnati cohort,
32                   neonatal blood Pb levels were lower than concurrent or lifetime average blood Pb levels
33                   at age 6 years (means 4.8, 11.7, 12 (ig/dL, respectively). In cross-sectional studies that
34                   examined similar potential confounding factors, results were inconsistent in populations
35                   with lower blood Pb levels (means <5 (ig/dL) (Surkan et al.. 2007; Despres et al.. 2005).
      November 2012                             5-197                  Draft - Do Not Cite or Quote

-------
      5.3.9  Seizures in Animals

 1                  Previous studies did not consistently show that Pb exposure induced seizures in animals.
 2                  Pb-induced seizures were found in male Wistar rats exposed to Pb acetate postnatally
 3                  (250-1,000 ppm in drinking water PND30-PND60, resulting in blood Pb levels of
 4                  -20-42 (ig/dL), as indicated by a decrease in the elapsed time required to develop the first
 5                  myoclonic jerk and tonic-clonic seizure (Arrieta et al.. 2005). Also, the dose of the
 6                  seizure-inducing agent pentylenetetrazol (PTZ) required to induce seizures significantly
 7                  decreased in all Pb dose groups. Other studies showed no effect of Pb exposure on
 8                  seizures (Schwark et al.. 1985; Alfano and Petit 1981). In a study of early postnatal
 9                  Pb acetate exposure (2,000 ppm in drinking water PND1-PND25), Pb had variable  effects
10                  on induction of seizures in Sprague Dawley rats examined at PND25 or PND50,
11                  depending on the convulsant-inducing agent administered (Chen and Chan. 2002).  Chen
12                  and Chan (2002) hypothesized that the variable effects may be due to the selective  effects
13                  on inhibitory and excitatory neurotransmission by age and blood Pb level, which were
14                  47 (ig/dL and 11 (ig/dL at PND25 and PND50, respectively.

15                  Recent investigation expanded on the work by Arrieta et al. (2005) by showing that Pb
16                  exposure may induce seizure activity in another rodent species, BALB/c mice. Adult
17                  (ages 2-3 months) male BALB/c mice were exposed to Pb acetate for 30 days via
18                  drinking water (range of blood Pb levels 50-400 ppm Pb groups: 6.4-18 (ig/dL)
19                  (Mesdaghinia et al.. 2010). Except for 50 ppm Pb exposure, all other Pb concentrations
20                  significantly reduced the thresholds efface and forelimb clonus, myoclonic twitch,
21                  running and bouncing clonus, and tonic hindlimb extension. In a study of adult male
22                  Wistar rats, Pb administration by bolus injection (200 mg/kg Pb acetate or 50 mg/kg
23                  Pb nitrate, single injection, 2 days, blood  Pb levels >20 (ig/dL) also induced epileptic
24                  form activity or seizures (Krishnamoorthy et al.. 1993).
      November 2012                            5-198                 Draft - Do Not Cite or Quote

-------
      5.3.10 Neurodegenerative Diseases


      5.3.10.1      Alzheimer's Disease

 1                  Higher bone Pb level in NAS men (Wang et al.. 2007a: Weisskopf etal.. 2004; Wright et
 2                  al.. 2003) but not blood Pb level in adults in Sweden (Nordberg et al.. 2000) has been
 3                  associated lower scores on the MMSE, which is widely used as a screening tool for
 4                  dementia. Direct evidence regarding the effects of Pb exposure on Alzheimer's disease is
 5                  limited to studies reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b). which did not find
 6                  higher occupational exposure to Pb  (Graves etal.. 1991) or Pb level in the brains
 7                  (Haraguchi et al.. 2001) in Alzheimer's disease cases than healthy controls. Overall, the
 8                  latter studies have sufficient limitations (e.g., case-control design that may be subject to
 9                  reverse causation, lack of blood or bone Pb measures, limited consideration for potential
10                  confounding) such that evidence is inconclusive regarding the effect of Pb exposure on
11                  Alzheimer's disease.

12                  Despite inconclusive  epidemiologic evidence, toxicological evidence indicates that Pb
13                  exposure in early life promotes Alzheimer's disease-like pathologies in the brains of aged
14                  adult animals. Alzheimer's disease is characterized by amyloid-beta peptide (Ab)
15                  plaques, hyper-phosphorylation of the tau protein, neuronal death and synaptic loss. In
16                  the last decade, the developmental origins of adult health and disease paradigm and the
17                  similar Barker hypothesis have indicated that early life exposures can produce aberrant
18                  effects in adults. Bolin et al. (2006)  demonstrated the connection between developmental
19                  exposure to Pb in the rat and inflammation-associated DNA damage with
20                  neurodegenerative loss in the adult brain. Wu and colleagues (2008a) had similar findings
21                  in a study examining  infantile Pb exposure of monkeys. These results suggest the need to
22                  directly examine the long-term effects of developmental exposure to toxicants rather than
23                  relying on adult exposure alone to predict potential health risks in adults (Dietert and
24                  Piepenbrink. 2006).

25                  The fetal basis of amyloidogenesis has been examined extensively by the Zawia
26                  laboratory in both rodents and nonhuman primates. Mechanistically, Ab plaques  originate
27                  from the  cleavage of the amyloid precursor protein (APP). In male rodents exposed to Pb
28                  as infants (200 ppm Pb acetate PND1-PND20 in dam drinking water, resulting in pup
29                  PND20 blood Pb level of 46 (ig/dL and cortex 0.41 (ig/g wet weight of tissue) or as
30                  adults, infancy Pb exposure induced APP gene expression in the aged animal brains. A
31                  bimodal response was observed, with a significant increase in APP expression above that
32                  in control animals first manifesting in infancy and again in old age (82 weeks) (Basha et
33                  al.. 2005). A concomitant bimodal response was observed in specificity protein 1 (Spl), a
      November 2012                            5-199                 Draft - Do Not Cite or Quote

-------
 1                  transcription factor involved in gene expression in the early development of an organism
 2                  and known to be related to APP expression. Ab was also significantly elevated in the
 3                  aged animals developmentally exposed to Pb. Adult-only (18-20 weeks) exposure to Pb
 4                  did not alter APP or Sp 1 expression or Ab production.

 5                  Consistent with findings in rodents, Wu et al. (2008a) found that Pb exposure in infancy
 6                  (PND1-PND400, 1.5 mg/kg/day in infant formula) resulted in significantly higher gene
 7                  expression of APP and Spl and significantly higher protein expression of APP and Ab in
 8                  aged female monkey cortex tissue (23 year-old Macaca fasclculans) from a cohort of
 9                  animals established in the 1980s by Rice (1992a. 1990). At PND400, the monkeys had
10                  blood Pb levels of 19-26 (ig/dL. In old age when amyloid plaques had manifested, blood
11                  Pb levels and brain cortex Pb levels had returned to control levels. Together, the rodent
12                  and nonhuman primate evidence concurs, and indicates that developmental Pb exposure
13                  and not adult-only exposure induces elevations in neuronal Alzheimer's Disease-related
14                  plaque proteins in aged animals.

15                  Mechanistic understanding of Ab production and elimination after Pb exposure was
16                  examined in human SH-SY5Y neuroblastoma cells exposed to Pb concentrations of 0,  5,
17                  10, 20, and 50 (iM for 48 hours. Pb was found to affect two separate pathways to increase
18                  Ab. Pb exposure induced both the overexpression of APP and repression of neprilysin, a
19                  rate-limiting enzyme involved in Ab metabolism or removal (Huang et al.. 201 la).
20                  Further mechanistic understanding of how Ab peptide formation is affected by Pb
21                  exposure was provided by Behl et al. (2009). The choroid plexus is capable of removing
22                  beta-amyloid peptides from the brain extracellular matrix.  Pb was shown to impair this
23                  function, possibly via the metalloendopeptidase, insulin-degrading enzyme (IDE), which
24                  metabolizes Ab (Behl et al., 2009). In another study, lactational Pb exposure of Long -
25                  Evans hooded rat pups induced perturbations in DNA binding of SP 1 via its Zn finger
26                  protein motif (Basha et al.. 2003). This effect of Pb was ameliorated by exogenous Zn
27                  supplementation.

28                  An additional study with gestational plus lactational Pb exposure (1,000-10,000 ppm,
29                  dam drinking water, resultant offspring blood Pb levels: 40-100 (ig/dL) showed that the
30                  rodent hippocampus as early as PND21 contained neurofibrillary changes, commonly
31                  used a marker for Alzheimer's disease. These changes manifested with hyper-
32                  phosphorylated Tau, which comprises neurofibrillary tangles, and increased tau and beta
33                  amyloid hippocampal protein levels (Li et al., 2010b).

34                  In summary, recent studies showed that Pb exposure of rats and monkeys during infancy
35                  or during gestation/lactation induced  significant increases in neuronal plaque associated
36                  proteins such as Ab-peptide, activation of Ab-supporting transcription factors, and
37                  hyperphosphorylation of tau, all of which are pathologies found in humans with


      November 2012                            5-200                 Draft - Do Not Cite or Quote

-------
 1                  Alzheimer's disease. These pathologies were not found with adult-only Pb exposure of
 2                  animals, further demonstrating that early life Pb is a critical window for Pb-induced
 3                  Alzheimer's-like pathologies in animals. The few epidemiologic studies have not linked
 4                  higher Pb exposure with Alzheimer's disease. These case-control studies lacked
 5                  assessment of blood or bone Pb levels or consideration for potential confounding. The
 6                  animal evidence indicates that epidemiologic studies assessing concurrent brain Pb levels
 7                  or occupational Pb exposure may not have examined the etiologically relevant exposure
 8                  period. However, the observations that were made in experimental animals with high Pb
 9                  exposure and blood Pb levels (>40 (ig/dL) may have uncertain relevance to humans.
10                  Further, animals were not behaviorally assessed for dementia.
      5.3.10.2      Amyotrophic Lateral Sclerosis

11                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported mixed epidemiologic findings for an
12                  association between Pb and ALS based on case-control studies, several of which relied on
13                  indirect methods of assessing Pb exposure. Case-control studies that measured blood Pb
14                  levels produced contrasting results. A study of 16 ALS cases (mean blood Pb level:
15                  12.7 (ig/dL) and 39 controls (mean blood Pb level: 10.8 (ig/dL) found a small difference
16                  in the mean concurrent blood Pb level (Vinceti etal.. 1997). A larger study of 109 cases
17                  and 256 controls that examined concurrent blood and bone Pb levels in a New England-
18                  area population found higher odds of ALS among subjects with concurrent blood Pb
19                  levels > 3 (ig/dL (e.g.,  OR: 14.3 [95% CI: 3.0, 69.3] for n = 55 blood Pb levels 3-4 (ig/dL
20                  compared with blood Pb levels  10 (ig/g
22                  patella Pb and > 8 (ig/g tibia Pb), but lacked precision. For example, compared with
23                  subjects with tibia Pb level <8 (ig/g, the OR for tibia Pb levels 8-14 (ig/g was 1.6 (95%
24                  CI: 0.5, 5.6). Results were adjusted for age, education, and hours/day inactive. Potential
25                  confounding by smoking was not considered. Also in this population,  an estimate of
26                  cumulative Pb exposure based on occupational history was found to be associated with
27                  ALS (Kamel et al.. 2002). The stronger findings for blood Pb level were surprising given
28                  that bone Pb level is a better biomarker of cumulative  Pb exposure. One explanation for
29                  these findings is reverse causation. Blood was collected from people who already had
30                  ALS, and reduced physical activity among those with  ALS could lead to more bone
31                  turnover and greater release of Pb from bones into circulation in ALS  cases than controls.

32                  Since the 2006 Pb AQCD, a few additional studies of  ALS have been  conducted with the
33                  same New England-area case-control group. Kamel et al. (2005) reported that the
34                  association between blood Pb level and ALS was not modified by the  ALAD genotype,
35                  and Kamel et al. (2008) found that higher tibia and patella Pb levels were associated with


      November 2012                            5-201                  Draft - Do Not Cite or Quote

-------
 1                  longer survival time among 100 of the original 110 ALS cases with adjustment for age,
 2                  sex, and smoking. Results were not altered by the additional adjustment for education,
 3                  BMI, or concurrent physical activity. Higher blood Pb levels were associated weakly with
 4                  longer survival time. These paradoxical findings that point to a protective effect of Pb are
 5                  not easily explainable but find coherence with results for Pb-induced increased survival
 6                  time in an ALS mouse model (see below). On one hand, the cases with longer survival
 7                  time may have higher bone Pb  levels because they reflect a longer period of cumulative
 8                  exposure. On the other hand, with longer survival time, there could be greater progression
 9                  of disease and less mobility. Decreased mobility would tend to increase bone resorption,
10                  lower bone Pb levels, and increase blood Pb levels over time. The latter hypothesis is a
11                  less likely explanation for findings in this New England cohort because higher bone Pb
12                  levels were more strongly associated with longer survival time than was blood Pb level.

13                  Another case-control study examined concurrent blood Pb levels and ALS among 184
14                  cases (33 were either progressive muscular atrophy or primary lateral sclerosis, mean
15                  blood Pb level: 2.41 (ig/dL) and 194 controls (mean blood Pb level: 1.76 (ig/dL) (Fang et
16                  al.. 2010). The cases were recruited from the National Registry of U.S. Veterans with
17                  ALS, and controls were recruited from among U.S. Veterans without ALS and frequency
18                  matched by age, gender, race, and past use of the Veterans Administration system for
19                  health care. A doubling of concurrent blood Pb level was associated with ALS with an
20                  OR of 2.6 (95% CI: 1.9, 3.7) with adjustment for age and a collagen protein as an
21                  indicator of bone formation. Associations did not differ substantially by indicators of
22                  bone turnover but were slightly higher among ALAD 1-1 carriers. The association with
23                  blood Pb level was similar in analyses that excluded the progressive muscular atrophy
24                  and primary lateral sclerosis cases. The similar results by degree of bone turnover suggest
25                  that reverse causation is not likely explaining the association between blood Pb level and
26                  ALS. However, as in other ALS case-control studies, the directionality of effects is
27                  difficult to establish. This study did not have measures of bone Pb to assess the
28                  association with biomarkers of longer-term Pb exposure.

29                  Although epidemiologic studies have provided inconsistent evidence for associations of
30                  Pb biomarker levels with ALS in adults, toxicological studies have found that Pb
31                  exposure affects neurophysiologic changes associated with ALS. For example, chronic
32                  postnatal Pb exposure from weaning onward (200 ppm Pb acetate in drinking water,
33                  resultant blood Pb level: 27 (ig/dL) reduced astrocyte reactivity and induced increased
34                  survival time in the superoxide dismutase transgenic (SOD1 Tg) mouse, which has SOD
35                  mutations found in humans with familial ALS (Barbeito et al.. 2010). In this model, Pb
36                  exposure did not significantly increase the onset of the ALS disease. These findings
37                  provide coherence with the association observed between bone Pb level and longer
38                  survival time in patients diagnosed with ALS (Kamel et al.. 2008).
      November 2012                            5-202                  Draft - Do Not Cite or Quote

-------
 1                  Research outside of the Pb field has suggested different mechanisms for ALS initiation
 2                  versus ALS progression, i.e., motor neuron function versus astrocyte and microglia
 3                  function (Yamanaka et al., 2008; Boillee et al., 2006). Astrocyte vascular endothelial
 4                  growth factor (VEGF) was examined for its involvement in the effects of Pb on
 5                  increasing survival time in the ALS mouse model. Lower VEGF expression has been
 6                  linked with risk of ALS in humans and ALS-like symptoms in animals. Baseline VEGF
 7                  levels were elevated in astrocytes from the ventral spinal cord of untreated SOD1 Tg
 8                  mice versus untreated nontransgenic mice. VEGF was not induced in the astrocytes of
 9                  Pb-treated nontransgenic mice. In comparison, Pb-exposed SOD1 Tg mice, which had
10                  longer survival time, also had significant elevations in astrocyte VEGF (Barbeito et al.,
11                  2010). These findings for Pb-induced effects on astrocytes in a mouse model for ALS
12                  may provide a mechanistic explanation for Pb effects on survival time in ALS.

13                  Others reported that VEGF administration to the SOD1 Tg mice significantly reduced
14                  glial reactivity, a marker or neuroinflammation (Zheng et al.. 2007). Using a cell-based
15                  co-culture system of neurons and astrocytes isolated from Pb-exposed SOD1 Tg mice,
16                  Barbeito et al. (2010) found that an up-regulation of VEGF production by astrocytes was
17                  protective against motor neuron death in the SOD1 Tg mouse cells. Thus, in vivo and in
18                  vitro results indicate that chronic Pb exposure resulted in increased survival time in an
19                  ALS mouse model and was correlated with higher spinal cord VEGF levels, which made
20                  astrocytes less cytotoxic to surrounding motor neurons (Barbeito et al.. 2010).

21                  In summary, there is inconsistent evidence of association between indicators of Pb
22                  exposure (history of occupational exposure, Pb biomarker levels) and ALS prevalence
23                  and survival time in humans. Because of the potential for reverse causality and bias due
24                  to survival time in the case-control studies, and the lack of objective assessment of
25                  occupational exposure, firm conclusions are not warranted. While several studies have
26                  considered potential confounding by age,  education, and physical activity, few have
27                  considered smoking. Toxicological evidence also points to Pb exposure increasing
28                  survival in a mouse model of ALS and has suggested explanations including Pb-induced
29                  increases in VEGF expression and subsequent reduction in glial activity and protection of
30                  motor neurons against inflammation.
      November 2012                            5-203                  Draft - Do Not Cite or Quote

-------
      5.3.10.3      Parkinson's Disease

 1                   Previous Pb AQCDs reviewed a few studies, some ecological (Rybicki etal.. 1993;
 2                   Aquilonius and Hartvig. 1986) and some case-control relying on questionnaire data or
 3                   occupational history (Gulson et al.. 1999; Gorelletal.. 1997; Tanner etal.. 1989) that
 4                   indicated associations between exposure to heavy metals, particularly Pb, and risk of
 5                   Parkinson's disease. The limited number of previous studies, weak study designs, and
 6                   lack of examination of Pb biomarkers did not permit firm conclusions. Recent studies
 7                   maintain several of these limitations but have indicated associations with bone Pb levels.

 8                   A recent large case-control study (330 cases, 308 controls) examined a population in the
 9                   Boston, MA area with virtually no occupational exposures to Pb (Weisskopf et al., 2010).
10                   Subjects in the highest quartile of tibia Pb level (>16.0 (ig/g) had higher odds of
11                   Parkinson's disease compared to those in the lowest quartile (< 5 (ig/g) (OR: 1.91 [95%
12                   CI: 1.01, 3.60]) with adjustment for age, race, pack-years smoking, education, and
13                   recruitment site. Cases and controls were recruited from several different sources
14                   including movement disorder clinics and the NAS, which could have introduced biased
15                   participation by Pb exposure or reduced representativeness to the target population. In the
16                   NAS, cases were ascertained from self-report, which may introduce measurement error.
17                   However, when analyses were  restricted to cases recruited from movement disorder
18                   clinics and their spouse, in-law, or friend as controls, the results were even stronger (OR:
19                   3.21 [95% CI: 1.17, 8.83]). Although the use of spouse, in-law, and friend controls can
20                   introduce bias, this is expected to be toward the null as these groups are likely to share
21                   many exposures.  Manganese (Mn) exposure has been associated with Parkinsonian
22                   symptoms and could potentially confound associations between Pb and Parkinson's
23                   disease. Weisskopf et al. (2010) did not adjust results for Mn exposure. However, unlike
24                   occupational exposure to Pb, general environment exposure to Pb is less likely to be
25                   correlated with environmental Mn exposure. Thus, it is less likely that the observed
26                   associations with Pb were confounded by co-occurring Mn exposure.

27                   Coon et al. (2006) conducted a smaller case-control study of 121 Parkinson's disease
28                   patients and 414 controls frequency-matched by age, sex, and race, all receiving health
29                   care services from the Henry Ford Health System in Michigan. Subjects in the highest
30                   quartile of both tibia (OR:  1.62 [95% CI: 0.83, 3.17] for levels > 15 (ig/g) and calcaneus
31                   (OR: 1.50 [95% CI: 0.75, 3.00] for levels > 25.29 (ig/g) bone Pb levels had higher odds
32                   of Parkinson's disease compared to those in the lowest quartiles (0-5.91 (ig/g for tibia and
33                   0-11.70 (ig/g for calcaneus). Subjects in the highest quartile of whole-body lifetime Pb
34                   level (> 80.81 (ig/g, estimated using PBPK modeling) had the highest OR: 2.27 (95% CI:
35                   1.13, 4.55) versus the lowest quartile, 0-40.04 (ig/g. These results were adjusted for age,
36                   race, sex, pack-years smoking, regular coffee consumption, and regular alcohol use, but
      November 2012                             5-204                  Draft - Do Not Cite or Quote

-------
 1                  Mn exposure was not considered. It was not clear what the extent of occupational
 2                  exposure to Pb was among the participants; however, a previous Henry Ford Health
 3                  System study had linked occupational Pb exposure to Parkinson's disease (Gorell et al.,
 4                  1997). Thus, it is uncertain whether the observed associations were confounded by
 5                  co-occurring Mn exposure.

 6                  In summary, a small number of recent case-control studies expand on previous evidence
 7                  by finding associations of tibia and calcaneus bone Pb levels, biomarkers of cumulative
 8                  Pb exposure, with Parkinson's disease in adults. The associations observed with
 9                  biomarkers of cumulative Pb exposure increase confidence that associations are not
10                  explained by reverse causality. However, firm conclusions are not warranted. While
11                  associations were adjusted for potential confounding by age, sex, race, and education, Mn
12                  co-exposure was not considered.
      5.3.10.4      Essential Tremor

13                  The few available case-control studies of essential tremor have found associations with
14                  concurrent blood Pb levels. The 2006 Pb AQCD (U.S. EPA. 2006b) described case-
15                  control studies that found associations between concurrent blood Pb levels and essential
16                  tremor in New York City metropolitan area populations (Louis et al., 2005; Louis et al.,
17                  2003). In the larger study, mean (SD) blood Pb levels were 3.3 (2.4) (ig/dL in the 100
18                  essential tremor cases and 2.6 (1.6) (ig/dL in the 143 controls (Louis et al., 2003). In the
19                  other study, mean (SD) blood Pb levels were 3.5 (2.2) (ig/dL in the 61 essential tremor
20                  cases and 2.6 (1.5) (ig/dL in the 101 controls (Louis et al., 2005). In Louis et al. (2005).
21                  the magnitude of association was larger among the 35 ALAD-2 carriers than among 129
22                  adults with the ALAD-1 genotype. History of occupational Pb exposure was similarly
23                  rare in cases and controls (2%).

24                  Recently, Dogu et al. (2007) reported on 105 essential tremor cases selected from a
25                  movement disorder clinic in Turkey and 105 controls (69 spouses and 36 other relatives)
26                  living in the same district. With adjustment for age, sex, education, smoking status,
27                  cigarette pack-years, and alcohol use, a 1 ug/dL higher blood Pb level (measured at the
28                  time of study recruitment) was associated with essential tremor with an OR of 4.19 (95%
29                  CI: 2.59, 6.78). This OR was much larger than that obtained in the New York area study
30                  (OR:  1.19 [95% CI:  1.03, 1.37]) (Louis et al.. 2003). The magnitude of association in
31                  Dogu et al. (2007) is even more striking because so many of the controls were spouses
32                  who are expected to share many environmental exposures as cases. Most of the essential
33                  tremor cases were retired at the time of the study; however past occupational history was
34                  not examined.
      November 2012                            5-205                 Draft - Do Not Cite or Quote

-------
 1                  In summary, a small body of studies indicates associations between blood Pb level
 2                  measured at the time of the study and prevalence of essential tremor in adults. However,
 3                  because of the case-control design, reverse causation cannot be excluded as a potential
 4                  explanation for the observed associations since loss of physical activity and subsequent
 5                  bone resorption may lead to an increase in blood Pb level. Further, the level, timing,
 6                  frequency, and duration of Pb exposure associated with essential tremor are uncertain.
 7                  History of occupational Pb exposure was not consistently examined, and potential
 8                  confounding by Mn exposure was not examined.
      5.3.10.5      lexicological Studies of Cell Death Pathways

 9                  A common element of the neurodegenerative diseases described above is neuronal cell
10                  death. Studies reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b) documented that Pb
11                  exposure induced cell death or apoptosis in various models including rat brain (Tavakoli-
12                  Nezhadetal., 2001), retinal rod cells (He et al., 2003; He et al.. 2000), cerebellar neurons
13                  (Oberto et al.. 1996). and PC12 cells (Sharifi and Mousavi. 2008). Recent studies
14                  produced similar findings, in most cases, in animals with higher blood Pb levels than
15                  those relevant to humans. Long-term (40 days) exposure to 500 ppm Pb in drinking water
16                  was found to increase pro-apoptotic Bax protein levels and the number of apoptotic cells
17                  in the hippocampus in young (exposure starting at 2-4 weeks of age) and adult (exposure
18                  starting at 12-14 weeks of age) male rats with blood Pb levels 98 (ig/dL (Sharifi et al..
19                  2010). Apoptosis was verified by light and electron microscopy. Another study followed
20                  the developmental profile of changes in various apoptotic factors in specific brain regions
21                  of animals exposed to 2,000 ppm Pb acetate during lactation (to PND20) via drinking
22                  water of dams (Chao et al.. 2007). At the end of lactation, male offspring blood Pb level
23                  was 80 (ig/dL. The data showed that hippocampal mRNA for various apoptotic factors
24                  including caspase-3, Bcl-x, and Brain-derived neurotrophic factor (BDNF) was
25                  significantly upregulated on PND12,  PND15 and PND20. The cortex of these male pups
26                  showed upregulation of Bcl-x and BDNF on PND15 and PND20. The cerebellum did not
27                  have elevated apoptotic mRNA levels in this model. Thus, in this study, Pb-induced
28                  apoptosis varied by age  and brain region in male offspring.

29                  Pb exposure also has been shown to induce apoptosis of spinal cord cells during spinal
30                  cord development in chicks treated with 150 or 450 (ig Pb acetate in ovo at
31                  embryonic day 3 or 5  and visualized six days later (Muller et al.. 2012). TUNEL positive
32                  cells, indicating DNA fragmentation induced by apoptosis, were at significantly higher
33                  levels in Pb-exposed animals and were visualized in all layers of the developing spinal
34                  cords. Also, levels of glial fibrillary acidic protein (GFAP), a factor important in neuronal
35                  migration and cellular differentiation during nervous system development, was
      November 2012                            5-206                  Draft - Do Not Cite or Quote

-------
 1                  significantly attenuated in spinal cords of Pb-exposed chicks. Liu et al. (201 Ob) examined
 2                  apoptotic effects in 30 day-old male rats that were treated with Pb acetate once daily for
 3                  6 weeks via intragastric infusion. Four groups: control, low (2 mg/kg BW), medium (20
 4                  mg/kg BW), and high (200 mg/kg BW) had blood Pb levels of 1.0-7.5 ug/dL;
 5                  4.5-11 ug/dL; 9-42 ug/dL; and 48-73 ug/dL, respectively. Pb induced hippocampal
 6                  neuronal apoptosis (TUNEL positive staining, statistically significant at all Pb doses)
 7                  with hippocampal XIAP (significant at high dose only) and Smac (statistically
 8                  nonsignificant trend) downregulation at the termination of the 6  week treatment. In
 9                  another study, Pb exposure (500 ppm Pb acetate in drinking water for 8 weeks) of adult
10                  male rats induced regional-specific changes in brain apoptotic proteins poly(ADP-ribose)
11                  polymerase, Bel-2, and caspase-3 with a greater effect observed in the hippocampus and
12                  cerebellum and a lesser effect observed in the brainstem and the frontal cortex (Kiran
13                  Kumar etal.. 2009).

14                  In summary, a small body of epidemiologic studies found Pb-associated increases in
15                  essential tremor and Parkinson's Disease in adults. However, limitations such as the
16                  potential for reverse causation to explain cross-sectional associations observed with blood
17                  Pb level, and the potential for confounding by Mn exposure preclude firm conclusions.
18                  However, toxicological evidence supports an effect of Pb on neurodegeneration by
19                  demonstrating that Pb exposure during various lifestages, early postnatal or adulthood,
20                  induces neuronal apoptosis in animals. Several of these observations were made with
21                  routes of Pb exposure (i.e., i.p.) that may not be relevant to those in humans.
      5.3.11 Modes of Action for Pb Nervous System Effects
      5.3.11.1      Effects on Brain Physiology and Activity

22                  The 2006 Pb AQCD (U.S. EPA. 2006b) reviewed a small body of available
23                  epidemiologic studies demonstrating associations of Pb biomarkers with
24                  electrophysiologic and physical changes in the brains of young adults (Yuan et al.. 2006;
25                  Cecil et al.. 2005) and children (Meng etal.. 2005; Trope etal.. 2001) as assessed by
26                  magnetic resonance imaging (MRI) or spectroscopy (MRS). The implications of previous
27                  findings were limited by the small sample sizes (n = 12-45) and limited consideration for
28                  potential confounding. Recent studies examining MRI data were limited largely to the
29                  Cincinnati cohort as adults (ages 19-24 years). In addition to supporting associations of
30                  childhood blood Pb levels with physiological changes in the brain of adults, these recent
31                  analyses expanded on previous studies by including larger sample sizes, aiming to
32                  characterize important lifestages of Pb exposures, and evaluating potential links between
      November 2012                            5-207                 Draft - Do Not Cite or Quote

-------
 1                   changes in brain activity and functional neurodevelopmental effects. While there are
 2                   overall few studies in few populations, by showing physical and physiologic changes in
 3                   areas of the brain associated with neurodevelopmental function, the evidence provides
 4                   biological plausibility for the associations observed between Pb biomarker levels and
 5                   cognitive function and behavior.

 6                   In prospective analyses of the Cincinnati cohort as adults (ages 20-23 years, n = 35, 42),
 7                   Cecil et al. (2005) and Yuan et al. (2006)  conducted functional MRI during a verb
 8                   generation language task and found that higher age 3-78 month average blood Pb level
 9                   (mean 14.2 (ig/dL) was associated with decreased activation in the left frontal gyrus and
10                   left middle temporal gyrus, regions implicated in semantic language function. Yuan et al.
11                   (2006) considered birth weight, marijuana consumption, sex, SES, gestational age, and
12                   IQ as potential confounding factors. Whereas previous analyses of the Cincinnati cohort
13                   focused on activity in specific regions of the brain, Cecil et al. (2011) examined brain
14                   metabolites. Higher age 3-78 month average blood Pb levels (mean: 13.3 (ig/dL) were
15                   associated with lower levels of N-acetylaspartate (NAA) and creatine (Cr) in the basal
16                   ganglia and lower levels of choline in white matter in 159 adults, ages 19-23 years. These
17                   results were adjusted  for age and FSIQ; however, several other unspecified factors were
18                   considered. Lower levels of NAA, Cr, and choline are linked to decreased neuronal
19                   density and alteration in myelin. A recent prospective analysis of 31 men in the NAS
20                   cohort similarly reported an association between biomarkers of cumulative, long-term Pb
21                   exposure  and changes in brain metabolites in older adults. Weisskopf et al. (2007b) found
22                   higher tibia and patella Pb levels to be associated with a higher myoinositol/Cr ratio in
23                   the hippocampus measured more than 10  years after bone  Pb and adjusted for age.
24                   Myoinositol/Cr ratio may be indicative of glial activation and is a signal reportedly found
25                   in the  early stages of HIV-related dementia and Alzheimer's disease.

26                   Other studies in the Cincinnati cohort as young adults found that childhood average blood
27                   Pb levels  were associated with altered brain architecture. Among 91 adults ages 20-26
28                   years, Brubaker et al.  (2009) found associations of age 3-78 month average blood Pb
29                   levels (mean:  13.3 (ig/dL) with diffusion parameters that were indicative of less
30                   organization of fibers throughout white matter. Results were adjusted for maternal IQ,
31                   prenatal alcohol and tobacco exposure, and adult marijuana use. In regions of the corona
32                   radiata, higher blood Pb levels were associated with less myelination axonal integrity. In
33                   regions of the corpus  callosum, higher blood Pb levels were associated with greater
34                   myelination and axonal integrity. The differential impact among neural elements may be
35                   related to the stage of myelination development present at various time periods.

36                   Another study of 157 Cincinnati cohort adults ages 19-24 years provided evidence of
37                   region-specific reductions in gray matter volume in association with age 3-78 month
      November 2012                             5-208                  Draft - Do Not Cite or Quote

-------
 1                   average blood Pb levels (mean: 13.3 (ig/dL) with adjustment for sex (Cecil et al., 2008).
 2                   The most affected regions included frontal gray matter, specifically the anterior cingulate
 3                   cortex, and the ventrolateral prefrontal cortex (i.e., areas related to executive functions,
 4                   mood regulation, and decision-making). Further, fine motor factor scores were positively
 5                   correlated with gray matter volume in the cerebellar hemispheres; adding blood Pb level
 6                   as a variable to the model attenuated this correlation. These findings suggested that
 7                   changes observed with MRI may mediate the association between blood Pb levels and
 8                   decrements in motor function. The functional relevance of these structural changes in the
 9                   brain also is supported by observations from other studies that link changes in brain
10                   architecture and activity to changes in cognitive function (e.g., visuoconstruction, visual
11                   memory,  eye-hand coordination) (Schwartz et al.. 2007) and behavioral problems
12                   (impulsivity, aggression, violence) (Yang et al.. 2005; Raine et al., 2000). In a subsequent
13                   comparison of blood Pb levels measured at various lifestages in 157 Cincinnati cohort
14                   adults ages 19-24 years, Brubaker et al. (2010) found that blood Pb levels at older ages
15                   (annual means from 3-6 years, means: 9.6-16.3 (ig/dL) were associated with greater
16                   losses in gray matter volume than were age 3-78 month average or maximum blood Pb
17                   levels (mean: 23.1 (ig/dL). Both Cecil et al. (2008)  and Brubaker et al. (2010) found that
18                   Pb-associated reductions in gray matter were more pronounced in males than females in
19                   the Cincinnati cohort.

20                   Studies of Pb-exposed workers (n = 15-532) also found associations of concurrent blood
21                   (means: 17-63.5 (ig/dL) and tibia (mean 14.5  (ig/dL) Pb levels with changes in brain
22                   structure and physiology, supporting the effects of chronic Pb exposure. Pb-associated
23                   changes included white matter lesions, smaller brain volumes, less total gray matter, and
24                   lower levels of brain metabolites such as NAA and Cr (Hsieh et al.. 2009b: Jiang et al..
25                   2008; Bleecker et al.. 2007b; Stewart et al., 2006) with adjustment for similar factors as
26                   associations for cognitive function. Other occupational exposures were not examined. In
27                   a few of these occupational groups, Pb-associated brain changes were linked to poorer
28                   performance in cognitive function tests (Caffo et al.. 2008; Bleecker et al.. 2007b).

29                   Higher concurrent blood Pb level also was associated with lower NAA/Cr ratio in small
30                   cross-sectional studies that included children (n = 6, 16, ages 4-21 years), although
31                   neither study considered potential confounding (Meng et al.. 2005; Trope et al.. 2001).
32                   All subjects had normal MRIs with no evidence of structural abnormalities. Thus, the
33                   biological relevance of the observed physiological changes is unclear. Additionally, the
34                   representativeness of findings is uncertain because results were based on comparisons of
35                   subjects with relatively high blood levels (23-65 (ig/dL) to those with blood Pb levels
36                   <10 (ig/dL.
      November 2012                             5-209                  Draft - Do Not Cite or Quote

-------
 1                  In summary, results in a few populations indicate associations of childhood blood or adult
 2                  tibia Pb levels with changes in brain structure and physiology in adults assessed by MRI
 3                  or MRS. Associations were found in children, but implications are limited because of
 4                  small samples sizes, lack of consideration of potential confounding, and high blood Pb
 5                  levels of the children examined. Evidence from the prospective Cincinnati cohort studies
 6                  improves characterization of the temporal sequence between Pb exposure and changes in
 7                  brain structure and physiology. Several studies linked these changes to functional
 8                  changes in cognitive performance or motor skills. Because of the small samples sizes of
 9                  several studies and limited consideration for potential confounding, firm conclusion
10                  regarding the effects of Pb exposed on changes in brain structure and physiology is not
11                  warranted. However, the evidence provides biological plausibility for the associations
12                  observed between Pb biomarker levels and cognitive function and behavioral problems.
      5.3.11.2      Oxidative Stress

13                  Because the brain has the highest energy demand and metabolism of any organ, energy
14                  homeostasis is of utmost importance. Energy imbalance can increase the susceptibility of
15                  the highly energetic brain tissue to stressors and cell death. Pb has been shown to induce
16                  energy imbalance by inhibiting various enzymes involved in energy production or
17                  glucose metabolism including glyceraldehydes-3 phosphate dehydrogenase, hexokinase,
18                  pyruvate kinase, and succinate dehydrogenase (Verma et al.. 2005; Yun and Hover. 2000;
19                  Regunathan and Sundaresan. 1984; Sterling et al., 1982). Mitochondria produce ATP or
20                  energy through oxidative phosphorylation. Aberrant mitochondrial function can decrease
21                  the energy pool and contribute to ROS formation via electron transport chain disruption.
22                  ATP depletion can also affect synaptic and extracellular neurotransmission. The
23                  mitochondrial Na+/K+ATPase is important in maintaining the inner mitochondrial
24                  membrane potential A*Pm (delta psifsub m]) and the functioning of the mitochondria.

25                  Gestational Pb exposure was found to impair mitochondrial function and energy
26                  production in neuronal cells from mice and produce concomitant increases in
27                  mitochondrial and cellular ROS production. The effect of Pb exposure on these
28                  mitochondrial parameters were examined in the brains of mice after prenatal Pb exposure
29                  (1,000 ppm Pb acetate in dam drinking water, resulting in offspring blood Pb levels of
30                  4 (ig/dL and cerebella Pb levels of 7.2 (ig/g dry weight) (Baranowska-Bosiacka et al.,
31                  201 Ib). Cerebellar granular cells were harvested from control  and Pb-exposed animals at
32                  PND8. These neuronal cells were cultured for 5 days in vitro, at which point various
33                  mitochondrial parameters were measured. With Pb exposure, ROS were significantly
34                  increased in both the cortical granule cells and in the mitochondria. Intracellular ATP
35                  concentration and adenylate energy charge values were significantly decreased in cells of


      November 2012                            5-210                  Draft - Do Not Cite or Quote

-------
 1                  Pb-exposed mice versus controls. Neuronal Na+/K+ATPase activity was significantly
 2                  lower in cortical granule cells from Pb-exposed mice versus cells from controls.
 3                  Mitochondrial mass was unaffected with Pb treatment, but mitochondrial membrane
 4                  potential was significantly decreased with Pb exposure. Energy imbalances also were
 5                  found in Wistar rats (PND15) of each sex injected daily for 2 weeks with Pb acetate
 6                  (15mg/kg BW, i.p., resulting in a mean blood Pb level of 30 (ig/dL; control blood Pb
 7                  level 3 (ig/dL) (Baranowska-Bosiacka et al., 201 la). ATP and ADP were significantly
 8                  decreased in various brain regions with Pb exposure, with the cerebellum and
 9                  hippocampus more strongly affected than the forebrain cortex. Also, expression of the
10                  pro-inflammatory P2XR receptor was enhanced in the glial fraction, indicating the
11                  astrocyte pool may be involved in the pathological changes found in Pb-exposed
12                  immature rat brains. Mitochondrial energy imbalances also were found in Pb-exposed
13                  crayfish that were placed under hypoxic conditions which induced a decrease in
14                  metabolism (Morris et al.,  2005).

15                  In rats, Pb exposure has been shown to induce oxidative stress, in some cases, with
16                  concomitant functional CNS changes. Exposure of adult male rats to 4,000 ppm
17                  Pb acetate in drinking water for 6 weeks increased brain levels of lipid peroxides (LPO)
18                  and lowered levels of antioxidants including nitric oxide (NO), total antioxidant capacity
19                  (TAC), glutathione (GSH), glutathione-S-transferase (GST), and superoxide dismutase
20                  (SOD). Whole blood Pb levels were positively correlated with brain tissue LPO levels
21                  and negatively correlated with NO levels. Evidence also indicated a role for oxidative
22                  stress in mediating the effects of Pb on cognition as evidenced by changes  in synaptic
23                  plasticity (Hamed et al., 2010). These effects of Pb on oxidative stress parameters were
24                  attenuated with co-exposure to green tea extract (1.5%), which reduced brain (1.9 to
25                  1.2 ppm) and blood Pb levels of rats (0.773 to 0.654 ppm). In a study of adult male
26                  Wistar albino rats, Pb acetate treatment by i.p. (20 mg/kg, 5 days) elevated lipid
27                  peroxidation, neuronal damage, and brain tissue DNA fragmentation and decreased
28                  antioxidant GSH levels and antioxidant enzyme activity, (Abdel Moneim et al.. 201 la).
29                  These effects were attenuated with co-administration of the polyunsaturated fatty acid
30                  flaxseed oil (oral gavage l,000mg/kg body weight for 5 days, 1 hour prior to Pb dosing).
31                  Flaxseed oil co-treatment also significantly attenuated the blood Pb level of Pb-exposed
32                  animals (~31 (ig/dL the day after the last Pb injection to -12 (ig/dL) and control animals,
33                  indicating that flaxseed oil may alter Pb toxicokinetics in animals. Another study
34                  provided indirect evidence of Pb-induced oxidative stress with observations that
35                  Pb-induced (2,000 ppm Pb acetate in drinking water PND1-PND67) impairments in long-
36                  term potentiation (LTP), paired-pulse reactions, and input/output functions in the DG of
37                  male and female Wistar rats were significantly attenuated with treatment with the
38                  antioxidant quercetin (30 mg/kg BW,  PND60-PND67) at PND67 (Hu et al.. 2008a).
      November 2012                            5-211                  Draft - Do Not Cite or Quote

-------
 1                  Quercetin-treated animals had significantly less hippocampal Pb than did the animals
 2                  exposed only to Pb.

 3                  Oxidative stress may be involved in neurodegenerative pathologies including Alzheimer's
 4                  disease. Hydrogen peroxide-induced oxidative stress has been shown to induce
 5                  intracellular accumulation of Ab in human neuroblastoma cells (Misonou et al.. 2000).
 6                  Oxidative stress-induced DNA damage can be measured as the ratio of the adduct
 7                  8-hydroxy-2'-deoxyguanosine to 2-deoxyguanosine (8-oxo-dG/2-dG). 2-dG is a product
 8                  of oxidative cleavage and is oxidized to form 8-oxo-dG. Pb-induced changes in the
 9                  8-oxo-dG to 2-dG ratio were examined recently as a mechanism underlying
10                  neurodegeneration. Similar to Ab levels, changes in the 8-oxo-dG to 2-dG ratio showed a
11                  biphasic relationship in the brains of rats exposed to 2,000 ppm Pb acetate via drinking
12                  water of dams from PND1-PND20 (blood Pb level 46 ng/dL) (Bolin et al.. 2006). The
13                  8-oxo-dG to 2-dG ratio decreased early in exposure (PND5) but increased at age
14                  20 months.  No increase was found in animals exposed to Pb from age 18 to 20 months
15                  (blood Pb level: 60 (ig/dL). Activity of the base-excision DNA repair enzyme oxoguanine
16                  glycosylase was unaffected by Pb exposure. Similar findings were reported in a monkey
17                  study (Wu et al., 2008a). The ratio of 8-oxo-dG to 2-dG in the brains of aged monkeys
18                  (23 years) was significantly elevated above that in controls only with Pb exposure in
19                  infancy (PND1-PND400, infant formula, blood Pb levels: 19-26 (ig/dL) but not as aged
20                  adults (Wu  et al.. 2008a). Thus, evidence in rats and monkeys suggests a possible role for
21                  oxidative stress in Pb-induced neurodegenerative effects and indicates that early life but
22                  not adult-only Pb exposure induces oxidative DNA damage and amyloidogenesis.
      5.3.11.3      Nitrosative Signaling and Nitrosative Stress

23                  The NO system is increasingly being recognized as a signaling system in addition to its
24                  more classical role as a marker of cellular stress. Pb exposures during the gestational-
25                  early postnatal period (GD6-PND21) (Chetty etal.. 2001) and during the postnatal period
26                  only (Fan et al.. 2009a) were found to reduce hippocampal levels of NO or neuronal NO
27                  synthase. In the hippocampus, NO mediates LTP, which is considered to be a major
28                  cellular mechanism underlying learning and memory. Thus, observations of Pb-induced
29                  changes in hippocampal NO may provide a mechanistic explanation for the effects of Pb
30                  on cognitive function decrements. Fan et al. (2009a) found reduced hippocampal NOS
31                  and NO in weanling male rats after either 4 or 8 weeks of Pb exposure resulting in blood
32                  Pb levels of 47 and 66 (ig/dL, respectively. In the same study, dietary supplementation
33                  with taurine or glycine concomitant with 8 weeks of Pb exposure induced significant
34                  increases in hippocampal NOS, whereas Pb plus dietary supplementation with vitamin C,
35                  methionine, tyrosine, or vitamin Bl decreased hippocampal NOS. In this study,


      November 2012                            5-212                  Draft - Do Not Cite or Quote

-------
 1                  co-exposure of specific nutrients also prevented Pb-induced impairments in learning as
 2                  evidenced by lack of increased escape latency in the Morris water maze. Dietary
 3                  supplementation with tyrosine, methionine, or ascorbic acid after 4 weeks of Pb exposure
 4                  in weanling males (blood Pb levels upon cessation of exposure and after 4-week lag:
 5                  47.6 and 8.1 (ig/dL, respectively) reversed Pb-induced decrements in NO/NOS. Zn
 6                  supplementation given after Pb exposure had no effect on the NO system.
      5.3.11.4      Synaptic Changes

 7                  Previous toxicological studies pointed to an effect of developmental Pb exposure on
 8                  synapse development, which mechanistically may contribute to multiple Pb-related
 9                  aberrant effects, including changes in LTP and facilitation. Facilitation of a neuronal
10                  terminal is defined as the increased capability to transmit an impulse down a nerve due to
11                  prior excitation of the nerve. Earlier work showed that developmental  Pb exposure
12                  resulted in altered density of dendritic hippocampal spines (Kiraly and Jones. 1982; Petit
13                  and Leboutillier. 1979). aberrant synapse elimination (Lohmann and Bonhoeffer. 2008).
14                  and abnormal long-term and short-term plasticity (MacDonald et al.. 2006). In a recent
15                  study, Li et al. (2009c) focused on inflammatory endpoints and synaptic changes after
16                  gestational plus lactational Pb exposure (1,000-10,000 ppm Pb acetate via drinking water
17                  of dams, producing offspring blood Pb levels 40-100 (ig/dL, respectively at PND21).
18                  Hippocampal TNF-a was significantly elevated with Pb exposure, and proteins that
19                  comprise the SNARE complex were  all changed with Pb exposure. The SNARE complex
20                  of synaptic proteins includes SNAP-25, VAMP-2 and Syntaxin la and is essential in
21                  exocytotic neurotransmitter release at the synapse. Thus, Li et al. (2009c) found
22                  significant differences in hippocampal  synaptic protein composition and increased pro-
23                  inflammatory cytokine levels in the brains of Pb-exposed offspring.

24                  Recent research using the Drosophila larval neuromuscular junction model showed that
25                  compared with unexposed controls, Pb-exposed larvae had significant increases in
26                  intracellular calcium and significant delays in calcium decays back to baseline levels at
27                  the pre-synaptic neuronal bouton (as stimulated with multiple action potentials, also
28                  called AP trains). Pb-exposed larvae  had reduced activity of the plasma membrane
29                  Ca2+ATPase, which is responsible for extravasations of calcium from the synaptic
30                  terminal (He et al.. 2009). Intracellular calcium in Pb-exposed larvae was no different
31                  from that in  controls under resting conditions or in neurons with stimulation by a single
32                  action potential. Pb media concentrations in these experiments were 100 or 250 (iM with
33                  the body burden of Pb from the lower dose calculated to be 13-48 (iM per larvae. After
34                  stimulation of the axon, facilitation of the excitatory post-synaptic potential, which is
35                  dependent on residual terminal calcium, was significantly elevated in Pb-exposed larvae


      November 2012                            5-213                  Draft - Do Not Cite or Quote

-------
 1                  versus control (He et al., 2009). The data from this synapse study demonstrate that
 2                  developmental Pb exposure affected the plasma membrane Ca2+ATPase, induced changes
 3                  in the intracellular calcium levels during impulse activation, and produced changes in
 4                  facilitation of the neuronal networks of Drosophila. Thus, the neuromuscular junction is a
 5                  potential site of Pb interaction.

 6                  Neurotransmission is an energy-dependent process as indicated by the presence of
 7                  calcium-dependent ATP releases at the synaptic cleft. At the synapse, ATP is
 8                  metabolized by ectonucleotidases. Acute exposure (96 hours) of male and female
 9                  zebrafish to Pb acetate (2 (ig/dL) in their water induced significant decreases in ATP
10                  hydrolysis in brain tissue (Senger et al.. 2006). This dose is deemed to be
11                  environmentally relevant. With chronic exposure (30 days), Pb acetate promoted the
12                  inhibition of ATP, ADP and AMP hydrolysis. These findings were consistent with
13                  findings in rodents (Baranowska-Bosiacka et al.. 201 la). The authors hypothesized that at
14                  30 days, this Pb-induced change in nucleotide hydrolysis was likely due to post-
15                  translational modification because expression of enzymes responsible for the hydrolysis,
16                  NTPDasel and 5'-nucleotidase, were unchanged (Senger et al.. 2006). Thus, Pb has been
17                  shown to affect nucleotidase activity in the CNS of zebrafish, possibly contributing to
18                  aberrant neurotransmission.

19                  Another enzyme important in synaptic transmission at cholinergic junctions  in the CNS
20                  and at neuromuscular junctions is acetylcholinesterase (AChE). After 24 hours of
21                  exposure to Pb acetate (2 (ig/dL water), AChE activity was significantly inhibited in
22                  zebrafish brain tissue (Richetti et al., 2010). AChE activity returned to baseline by 96
23                  hours and maintained baseline activity after 30 days of exposure. Thus, Pb was shown to
24                  affect synaptic homeostasis of AChE in the brains of zebrafish only transiently.

25                  Pb has been shown to act as an antagonist of the NMDA receptor (NMDAR). The
26                  NMDAR is essential for proper presynaptic neuronal activity and function. Primary
27                  cultures of mouse hippocampal cells exposed to 10 or  100 (iM Pb during the period of
28                  synaptogenesis had loss of two proteins necessary for presynaptic vesicular release,
29                  synaptophysin (Syn) and synaptobrevin (Syb) but no change in a similar protein
30                  synaptotagmin (Syt) (Neal etal. 2010a). This deficit was found in both GABAergic and
31                  glutamatergic neurons. Pb also induced an increase in the number of presynaptic contact
32                  sites. But, these sites may have been nonfunctional as they lacked the protein receptor
33                  complexes necessary for proper vesicular exocytosis. Another factor involved in
34                  maturation and signaling of presynaptic neurons is brain-derived neutrotrophic factor
35                  (BDNF), which is synthesized and released by postsynaptic neurons regulated by the
36                  NMDAR. In hippocampal cells, both pro-BDNF and BDNF release were significantly
37                  attenuated with Pb exposure (Neal etal., 2010a). Further,  exogenous BDNF
      November 2012                             5-214                  Draft - Do Not Cite or Quote

-------
 1                  administration rescued the aforementioned Pb-related effects on presynaptic proteins.
 2                  Thus, this cell culture model showed that Pb-related presynaptic aberrations are
 3                  controlled by NMDAR-dependent BDNF effects on synaptic transmission.

 4                  Glutamate is another neurotransmitter that is released from presynaptic neurons and via
 5                  interactions with the NMDAR causes postsynaptic neuron depolarization. A recent study
 6                  of Wistar albino rats exposed to Pb postnatally from birth to age 12 weeks (drinking
 7                  water 3 x 104 (ig/dL Pb acetate, resulting in blood Pb levels of 17 (ig/dL at age 6 weeks)
 8                  showed decreased learning ability, decreased hippocampal glutamate at 6, 8, 10 and
 9                  12 weeks of age, as well as significant decrements in the hippocampal glutamate
10                  synthesis-related enzymes aspartate aminotransferase  and alanine aminotransferase (Niu
11                  et al.. 2009).
      5.3.11.5      Blood Brain Barrier

12                  Two barrier systems exist in the body to separate the brain or the CNS from the blood.
13                  These two barriers are the blood brain barrier (BBB) and the blood cerebrospinal fluid
14                  barrier (BCB). The BBB, formed by tight junctions at endothelial capillaries forming the
15                  zonulae occludens (occludins, claudins, and cytoplasmic proteins), separates the brain
16                  from the blood and its oncotic and osmotic forces, allowing for selective transport of
17                  materials across this barrier.

18                  Pb exposure during various developmental windows has been shown to increase the
19                  permeability of the BBB of animals (Dvatlov et al., 1998; Struzvnskaet al.. 1997b;
20                  Moorhouse et al.. 1988; Sundstrom et al.. 1985). Possibly due the underdevelopment of
21                  the BBB early in life, prenatal and perinatal Pb exposure has been found to result in
22                  higher brain Pb accumulation than have similar exposures later in life (Moorhouse et al..
23                  1988). The choroid plexus and cerebral endothelial cells that form the BBB and BCB
24                  tight junctions have been shown to accumulate Pb more than other cell types and regions
25                  of the CNS. Studies reviewed in earlier Pb AQCDs showed that the chemical form of Pb
26                  and its capability to interact with proteins and other blood components affects its
27                  capability to penetrate the BBB (U.S. EPA. 2006b). Pb also has been shown to
28                  compromise the BCB and decrease the cerebrospinal fluid level of transthyretin, which
29                  binds thyroid hormone in the cerebrospinal fluid. Low thyroid hormone levels in
30                  pregnant women have been linked with IQ deficits in their children (Lazarus. 2005).

31                  Recent research with male weanling rats exposed to Pb acetate via drinking water showed
32                  leaky cerebral vasculature, an indication of a compromised BBB, as detected
33                  histologically with lanthanum nitrate staining of the brain parenchyma (Wang et al..
34                  2007b). Cerebral vasculature leakiness was ameliorated or resembled that of controls
      November 2012                            5-215                 Draft - Do Not Cite or Quote

-------
 1                  after iron supplementation. The cerebral vasculature leakiness may by explained by
 2                  observations of significant Pb-induced decreases in the BBB tight junction protein
 3                  occludin in the hippocampus, brain cortex, and cerebellum in these weanling animals.
 4                  Occulin levels were rescued to control levels with iron supplementation. This loss of
 5                  integrity at the junctional protein level was affirmed with additional experiments using
 6                  the rat brain vascular endothelial cell line RBE4, in which 10 (iM Pb acetate exposure for
 7                  2, 4, 8, 16 and 24 hours resulted in decreases in junctional proteins occludin and claudin
 8                  5 as well as scaffold proteins ZO1 and ZO2 (Balbuena et al.. 2011). Because gene
 9                  expression for these junctional and scaffold proteins did not show decrements, it was
10                  determined that these protein decrements were due to post-translational modifications.

11                  Pb exposure also was found to contribute to leakiness of the BBB by decreasing the
12                  resistance across the junction (Balbuena et al.. 2010). An in vitro co-culture system
13                  employing endothelial cells (RBE4 or bovine brain microvascular endothelial cells) and
14                  astrocytes (primary Sprague-Dawley neonatal pup astrocytes, GD21) served as the barrier
15                  between Pb-containing media and neurons. Pb acetate exposure (1 and 10 (iM) for 14
16                  hours significantly impaired transendothelial electrical resistance (TEER), a marker of
17                  BBB integrity, in a concentration-dependent manner.

18                  Long-term Pb exposure of adult mice was found to increase regional edema and BBB
19                  permeability (Lopez-Larrubia and Cauli. 2011). Adult male rats exposed to Pb acetate in
20                  drinking water for 4 or 12 weeks (50 or 500 ppm, resulting in blood Pb levels of 12 and
21                  55 (ig/dL, respectively) were assessed by diffusion weighted imaging for changes in
22                  apparent diffusion coefficent (ADC), a measure of tissue water diffusivity that changes
23                  under pathological conditions  like cerebral edema. With 12 weeks of exposure, 50 ppm
24                  Pb increased the ADC values in the cerebellum and mesencephalic reticular formation,
25                  and 500 ppm Pb exposure significantly increased ADC in the corpus callosum and
26                  caudate putamen. With 4 weeks of exposure, 500 ppm Pb significantly increased the
27                  water ADC in the hippocampus, mesencephalic reticular formation, and cerebellum but
28                  not in other brain areas. The brain areas  with elevated ADC also showed increased BBB
29                  permeability as measured with evans blue dye.
      November 2012                            5-216                 Draft - Do Not Cite or Quote

-------
      5.3.11.6      Cell Adhesion  Molecules

 1                  Classic cell adhesion molecules including neural cell adhesion molecule (NCAM) and the
 2                  cadherins are junctional or cell surface proteins that are critical for cell recognition and
 3                  adhesion. While direct effects of Pb on cell adhesion molecules have not been described,
 4                  the calcium-dependency of these molecules suggests that interaction from competing
 5                  cations like Pb can potentially contribute to nervous system barrier function disruption,
 6                  neurite outgrowth, synaptic plasticity, learning and memory (Prozialeck et al., 2002).
      5.3.11.7       Effects on Glial Cells

 7                   Astroglia and oligodendroglia are supporting cells in the nervous system that maintain the
 8                   extracellular space in the brain and provide structural support to neurons, deliver
 9                   nutrients to neurons, and promote myelination. Glial cells provide immune surveillance
10                   in the brain and contribute to inflammation-mediated pathologies. In Wistar rats, Pb
11                   treatment (15 mg/kg of Pb acetate, i.p.) during early postnatal maturation was observed to
12                   produce chronic glial activation with inflammation and neurodegeneration (Struzynska et
13                   al., 2007). Among the cytokines detected in the brains of these Pb-treated rats were
14                   IL-1P, TNF-a and IL-6.  Glial cells have been shown to serve as Pb sinks in the
15                   developing and mature brain by sequestering Pb (Tiffany-Castiglioni etal.. 1989). This
16                   glial sequestration of Pb was accompanied by a decrease in brain glutamine
17                   concentrations at doses of 0.25 ±1.0 (iM Pb acetate and a reduction in glutamine
18                   synthetase activity in the astroglia; astroglia take up released glutamate and convert it to
19                   glutamine. Pb has been shown to induce hypomyelination and demyelination (Coria et
20                   al.. 1984) mediated through the oligodendrocytes with younger animals found to be more
21                   susceptible to the  effects of Pb (Tiffany-Castiglioni etal.. 1989). Pb accumulation in
22                   young glial cells may contribute to a lifelong exposure of neurons to Pb as Pb is released
23                   from the sink over time. Thus, Pb accumulation in glial cells can contribute to continual
24                   damage of surrounding neurons (Holtzman et al.. 1987).


                     Glial transmitters

25                   Evidence indicates that glial transmission is affected with Pb exposure and that the
26                   NMDAR may be  involved in this aberrant glial transmission. To determine the
27                   contribution of the gliotransmitter serine to Pb-mediated changes in LTP, Sun et al.
28                   (2007) exposed rats to Pb acetate from gestation through lactation to PND28 via maternal
29                   drinking water and collected hippocampal sections. CA1 section LTPs were examined
30                   using in vitro patch clamp monitoring. Chronic Pb  exposure impaired the magnitude of
      November 2012                            5-217                  Draft - Do Not Cite or Quote

-------
 1                  hippocampal LTPs, but the magnitude of long-term depression was restored with
 2                  supplementation with D-serine (Sun et al.. 2007). which is known to be regulated by the
 3                  NMDAR (Bear and Malenka. 1994). The use of 7-chlorokynurenic acid, an antagonist of
 4                  the glycine binding site of the NMDAR, which also is the binding site of D-serine,
 5                  effectively abolished the rescue of LTP by D-serine. NMDAR-independent LTP
 6                  hippocampal neurotransmission was inhibited in slices of Pb-exposed mossy-CA3
 7                  synapses and was not rescued by exogenous D-serine supplementation.
      5.3.11.8      Neurotransmitters

 8                  Pb has been shown to compete with calcium for common binding sites and second
 9                  messenger activation. When Pb activates a calcium-dependent system in the nervous
10                  system, it can contribute to aberrant neurotransmitter regulation and release because this
11                  system intimately relies on calcium signaling for its homeostasis. Pb also has been shown
12                  to interfere  with other physiological divalent cations. Pb-related alterations in
13                  neurotransmission are discussed in further detail below.


                    Monoamine Neurotransmitters and Stress

14                  The monoamine neurotransmitters include dopamine (DA), serotonin (5HT), and
15                  norepinephrine (NE). Combined exposures of maternal stress and Pb exposure can
16                  synergistically enhance neurobehavioral impairments in offspring of exposed animals and
17                  can sometimes potentiate an effect that would otherwise be sub-threshold. Virgolini et al.
18                  (2008a) found enhanced DA and NE release in male rats and enhanced NE release in
19                  female rats  after developmental Pb exposure (50 or 150 ppm via drinking water,
20                  2 months prior to mating through lactation, resulting in blood Pb levels of 11 (ig/dL and
21                  35 (ig/dL, respectively) and combined maternal and offspring stress. In most cases, stress
22                  potentiated  the effects of Pb exposure on offspring NE and DA concentrations. Regional
23                  5HT levels  were unaffected in offspring with Pb exposure alone. Pb (50 and 150 ppm)
24                  combined with stress (maternal and/or offspring stress) significantly potentiated 5HT
25                  levels in the frontal cortex in females and in the nucleus accumbens (NAC) and striatum
26                  in male offspring. The concentration of 5-Hydroxyindoleacetic acid (5HIAA), the main
27                  metabolite of 5HT, was significantly increased in the striatum of male offspring with
28                  150 ppm  Pb exposure alone. With 50 ppm Pb, stress potentiated  striatal and frontal cortex
29                  5HIAA in males. Potentiation of 5HIAA levels in females was significant in the NAC
30                  with 50 ppm Pb exposure; stress alone also significantly increased 5HIAA levels in the
31                  NAC of females with no Pb exposure. Pb-induced changes in brain neurochemistry with
      November 2012                             5-218                  Draft - Do Not Cite or Quote

-------
 1                  or without concomitant stress exposure are complex with differences varying by brain
 2                  region, neurotransmitter type, and sex of the animal.


                    Monoamine Neurotransmitters and Auditory Function

 3                  Earlier work showed that perinatal Pb exposure of rats induced increased tyrosine
 4                  hydroxylase, increased DA, and increased cerebral cortex catecholamine
 5                  neurotransmission (Devi et al.. 2005; Leret et al.. 2002; Bielarczyk et al.. 1996). Earlier
 6                  publications detailing important time windows, durations, and doses of Pb exposure
 7                  indicated varying effects on monoamine neurotransmitters. In recent work, these
 8                  neurotransmitters, among others, have been implicated in Pb effects on auditory function
 9                  in various integration centers of the brainstem including the lateral superior olive (LSO)
10                  and the superior olivary complex (SOC). Among various functions, the SOC is vital for
11                  sound detection in noisy settings. A recent study in mice found significant decreases in
12                  immunostaining of LSO and SOC brainstem sections for monoamine vesicular
13                  transporter VMAT2, 5HT, and dopamine beta-hydroxylase (DbH, a marker for NE) after
14                  gestational-lactational Pb exposure (10 or 100 (iM Pb acetate from the  formation of
15                  breeding pairs to PND21). This exposure period corresponds to the period of auditory
16                  development in the mouse. Statistically significant decreases in VMAT2 and DbH were
17                  found in mice with blood Pb levels of 8.0 and 42.2 (ig/dL; however, decrements in 5HT
18                  were statistically significant only in mice with 8.0 (ig/dL blood Pb level. Immunostaining
19                  for tyrosine hydroxylase and transporters including VGLUT1, VGAT,  VAChAT
20                  indicated that they were unaffected by developmental Pb exposure. These data provide
21                  evidence that specific regions of the brainstem involved in auditory integration are
22                  affected by developmental Pb exposure via effects on monoamine neurotransmitters
23                  (Fortune and Lurie. 2009). The  Pb-induced effects on the monoamine system of the
24                  auditory portion of the brainstem provide possible mechanistic explanation for the
25                  epidemiologic and toxicological evidence for Pb-associated decrements in auditory
26                  processing (Section  5.3.7).


                    Dopamine

27                  The 2006 Pb AQCD (U.S. EPA. 2006b) detailed evidence for Pb-related decreased
28                  dopaminergic cell activity in the substantia nigra and ventral segmental areas. Earlier
29                  studies with postnatal or adult Pb exposure reported changes in DA metabolism, as
30                  indicated by changes in DA and DOPAC, a DA metabolite. Expanding upon these
31                  findings, a recent study measured DA and DOPAC in various brain regions of year-old
32                  male C57BL/6 mice to examine if gestational plus lactational Pb exposure affected DA
33                  metabolism (Leasure et al.. 2008). Exposure of males to 27 and 109 ppm Pb acetate
      November 2012                            5-219                 Draft - Do Not Cite or Quote

-------
 1                  induced significant elevations in DOPAC concentration and the DOPAC to DA ratio in
 2                  the forebrain. In the forebrain, DA was significantly decreased with the lower dose and
 3                  significantly elevated with the higher dose compared to controls. In the striatum, DOPAC
 4                  was significantly elevated with both doses, but DA concentration was only significantly
 5                  elevated with the higher dose. The striatum ratio of DOPAC to DA was not significantly
 6                  different from that in controls. These recent data expand upon the monoamine literature
 7                  base which indicates that Pb exposure of rats during the gestational/lactational,
 8                  lactational, or postweaning period producing blood Pb levels 9-34 (ig/dL induces
 9                  increased sensitivity of the dopamine receptors (D2 and D3) (Gedeon etal. 2001; Cory-
10                  Slechtaetal.. 1992). produces higher DA levels (Devi etal.. 2005;  Leret et al.. 2002). and
11                  enhances catecholamine neurotransmission in the cerebral cortex, cerebellum, and
12                  hippocampus (Devi et al.. 2005).

13                  The interaction of DA and the NO system  in the striatum was studied after prenatal Pb
14                  exposure (Nowak et al.. 2008). Blood Pb levels were not reported in this study, but
15                  similarly treated Wistar rat pups had blood Pb levels at parturition in range of
16                  50-100 (ig/dL (Grant etal.  1980). 7-nitroinidazole (7-NI), a selective inhibitor of nNOS,
17                  enhanced amphetamine-evoked DA release in the rat striatum (Nowak et al.. 2008).
18                  Prenatal Pb exposure attenuated the facilitatory effect of 7-NI on DA release in the
19                  striatum. This interaction was ROS-independent; using spin trap measurements,
20                  investigators found no significant concentration changes in hydroxyl radical with Pb
21                  exposure (Nowak et al.. 2008). Thus, the neuronal NO system appears to be involved in
22                  specific aspects of Pb-related dopaminergic changes.

23                  In various animal models, the loss of retinal DA, dopamine turnover (DOPAC:DA ratio),
24                  or Zn was associated with abnormal rod-mediated scotopic ERGs. These effects may
25                  explain observations of Pb-associated subnormal or supernormal retinal ERGs observed
26                  in animals and children (Rothenberg et al., 2002b; Lilienthal etal..  1994; Lilienthal et al..
27                  1988; Alexander and Fishman. 1984) (Sections 5.3.7.1 and 5.3.7.3). although the
28                  biological relevance of the variable effects of Pb exposure on subnormal versus
29                  supernormal ERGs is not clear.


                    NMDA Receptors

30                  The glutamate receptor, NMDAR, has been shown to contribute to  synaptic plasticity,
31                  and Pb exposure at different developmental stages has been shown to contribute to
32                  aberrations in LTP or long term depression (LTD) in the hippocampus via reduced
33                  NMDA current, among other mechanisms (Liu et al.. 2004). The 2006 Pb AQCD (U.S.
34                  EPA. 2006b) indicated that  Pb attenuated the stimulation of glutamate release, which in
35                  turn, affected LTP. Further, the effects of Pb exposure on decreasing the magnitude of
      November 2012                            5-220                 Draft - Do Not Cite or Quote

-------
 1                  LTP and increasing the threshold of the LTP in the hippocampus were found to be
 2                  biphasic or nonlinear. NMDAR subtypes have been shown to be significantly decreased
 3                  with developmental Pb exposure (Guilarte and McGlothan, 1998). Recent evidence
 4                  indicated Pb-related decreases in the gene expression and protein level of NMDAR
 5                  subunits NR1, NR2A,  and NR2B in weanling male rats exposed to 4 x  104 (ig/dL
 6                  Pb acetate in drinking water for 8 weeks. Several of these responses were attenuated with
 7                  methionine-choline co-exposure (Fan et al., 2010). Other recent mechanistic studies
 8                  found that pretreatment of primary fetal brain neuronal rat cultures with glutamic acid, a
 9                  NMDAR agonist, reversed Pb-induced reductions in NMDAR subunits (Xu and Raj anna.
10                  2006) whereas pretreatment with the NMDA antagonist MK-801 exacerbated Pb-induced
11                  NMDAR deficits (Xu and Rajanna. 2006). Further strengthening the link among Pb
12                  exposure, NMDAR function, and learning, Guilarte et al. (2003) demonstrated that rats
13                  exposed to 1,500 ppm  Pb during gestation and lactation then reared in isolation, had
14                  reduced expression of hippocampal NMDA receptor subunit 1, reduced induction of
15                  BDNF mRNA, and learning impairment. These effects were attenuated in Pb-exposed
16                  rats reared in an enriched environment with toys.


                    Other Glutamate Receptors

17                  The metabotropic glutamate receptor (mGluR) is another well-recognized target of Pb
18                  toxicity. In vitro (GD18 fetal rat cultures,  0.01, 100, 1 (iM Pb chloride in culture media)
19                  and in vivo studies (gestational and lactational Pb acetate exposure; 500, 2,000,
20                  5,000 ppm in dam drinking water, with respective weanling blood Pb levels of 18, 57,
21                  186 (ig/dL) showed that Pb exposure induced mGluR5 mRNA and protein decrements in
22                  a concentration-dependent manner (Xu et al.. 2009c). Recent evidence indicates a role for
23                  mGluR5 in synaptic plasticity, LTP, and LTD; thus, the Pb-related attenuation of mGlu5
24                  expression may represent a mechanism by which Pb impairs learning and memory.
      5.3.11.9      Neurogenesis

25                  Studies continue to show that Pb exposure decreases neurogenesis (i.e., proliferation of
26                  neuronal cells) in the hippocampus, which is important in LTP, spatial learning, neuronal
27                  outgrowth, and possibly mood disorders such as schizophrenia. Coherence for these
28                  findings is provided by evidence for Pb-induced decreases in NMDAR, which mediates
29                  the integration of new neurons into existing neuronal pathways in the adult hippocampal
30                  DG. Earlier work by Schneider et al. (2005) showed that postnatal Pb exposure (PND25
31                  to PND50 or PND55, 1,500 ppm Pb acetate in chow, resulting in blood Pb level of
      November 2012                            5-221                  Draft - Do Not Cite or Quote

-------
 1                  20 (ig/dL) of male Lewis rats induced significant decrements in BrdU incorporation (an
 2                  indicator of DNA replication) at PND5 0-PND5 5.

 3                  Recent publications affirm this previous finding with different sex of animals, dosing and
 4                  exposure time windows. Postnatal Pb exposure of Wistar rat pups from PND1-PND30
 5                  (2,000 ppm Pb acetate, resulting in blood Pb levels of 34 and 6.5 (ig/dL at PND21 and
 6                  PND80, respectively) induced a statistically significant decrement in the number of new
 7                  cells (BrdU positive cells) in the DG at PND80 (Fox etal.. 2010) (Figure 5-14). In
 8                  another study, lifetime Pb exposure beginning in gestation (1,500 ppm Pb acetate in chow
 9                  from 10 days before mating to PND50 or PND78, resulting in blood Pb levels 26 (ig/dL)
10                  of female Long-Evans rats induced significant decrements in hippocampal granule cell
11                  neurogenesis in adult rats (Verina et al., 2007). Also, Pb-exposed animals had significant
12                  decreases in brain volume in the stratum oriens (SO) region of the hippocampus,
13                  specifically in the mossy fiber terminals of the SO. Pb-exposed animals also showed a
14                  significant decrease in the length-density of immature or newly-formed neuron in the
15                  outer portion of the DG. These findings show that Pb exposure at doses relevant to
16                  humans induced significant decreases in adult hippocampus granule cell neurogenesis
17                  and morphology, potentially providing mechanistic explanations for Pb-induced neuronal
18                  aberrations and downstream effects such as learning and memory. Exposure of zebrafish
19                  embryos to Pb (50-700 (iM Pb acetate in embryo medium from 0 to 6 days post hatch)
20                  caused significant apoptosis of brain cells (increased TUNEL positive brain cells) and
21                  decreased brain levels of some (gfap and huC) but not all (crestin and neurogeninl) genes
22                  involved in neurogenesis (Dou and Zhang. 2011).
      November 2012                            5-222                  Draft - Do Not Cite or Quote

-------
                        '
                                                         6000
                                                         5000
                                                         4000
                                                         3000
                                                         2000
                                                         1000
                                                           0
                                                                   Con
Volume (mm3) i
» e •» -
3 cn O t



                                                                   Con
      Source: Reprinted with permission of Elsevier Science, Fox et al. (2010)
      Note: Light micrograph pictures of Brd-U positive cells (proliferating cells undergoing DMA replication), black dots, in control (A) and
      Pb-exposed rats (B). Counts of Brd-U positive cells (C) and Volume of hippocampus dentate gyrus (D) in control (white bars) and
      Pb-exposed animals (black bars). *p <0.05 vs. control. Rats were exposed to 2,000 ppm Pb acetate from postnatal day 1 -30 (blood
      Pb levels 34 ug/dL) and were examined at postnatal day 80.

      Figure 5-14     Neurogenesis (production of new cells) in the rat hippocampal
                        dentate gyrus after early postnatal Pb exposure.
      5.3.11.10    Neurite Outgrowth
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
As described in the 2006 Pb AQCD (U.S. EPA. 2006b). Pb was shown to decrease
neurite outgrowth in vitro and mediate such effects via protein kinase mediated pathways
(MAPK/ERK); earlier work had documented decreased primary DA neuron outgrowth
with 0.001 (iM Pb exposure (Lidsky and Schneider. 2004). A recent study showed that
gestational exposure of female Wistar rats to 500-4,000 (iM Pb chloride (resulting in
offspring blood Pb levels up to 12 (ig/dL) significantly decreased offspring hippocampal
neurite outgrowth and reduced the expression of hippocampal polysialylated neural cell
adhesion molecule (PSA-NCAM), NCAM, and sialytransferase (Hu et al.. 2008b). PSA-
NCAM is transiently expressed in newly formed neurons during the period of neurite
outgrowth from embryogenesis until the early postnatal period and is down-regulated in
the brains of adults except in areas known to exhibit synaptic plasticity (Seki and Arai.
1993). NCAM is important for memory formation, plasticity and synapse formation, and
its suppression by early-life Pb exposure may represent a mechanism mediating
Pb-associated impairments in cognitive function.
     November 2012
                           5-223
Draft - Do Not Cite or Quote

-------
      5.3.11.11     Epigenetics

 1                  Many investigators are beginning to show that environmental chemical exposures and air
 2                  pollution exposure are associated with epigenetic changes in humans (Baccarelli and
 3                  Bollati. 2009: Pavanello et al. 2009: Tarantini et al.. 2009: Bollati et al.. 2007).
 4                  Epigenetic changes involve changes in DNA expression without changes in the DNA
 5                  sequence, and these changes may be heritable. Epigenetic changes include histone
 6                  modification, DNA methylation, miRNA changes, or pathways that affect these
 7                  processes. Differential epigenetic modification has the potential to contribute to disease
 8                  by silencing or activating genes in an aberrant manner. For example, a recent study
 9                  identified differential methylation of a specific locus in monozygotic twins discordant for
10                  schizophrenia (Dempster et al., 2011): Pb was not examined in this study.

11                  DNA methyltransferases catalyze the transfer of a methyl group to DNA and are
12                  important in epigenetics (i.e., silencing of genes like tumor suppressors) and imprinting.
13                  DNA methyltransferase activity was significantly decreased  in cortical neurons from
14                  aged monkeys at ages 20-23 years after infancy  exposure (PND1-PND400, blood Pb
15                  level 19-26 (ig/dL) and fetal mouse brain cells exposed to Pb in culture (0.1 (iM Pb) (Wu
16                  et al.. 2008b). Changes in DNA methyltransferases (Dnmtl,  DnmtSa) were noted in
17                  control monkey brains as they aged and these changes were exacerbated by early
18                  postnatal Pb exposure (Bihaqi et al.. 2011). Another enzyme involved in DNA
19                  methylation, methyl CpG binding protein 2 MECP2, showed a similar trend as the
20                  Dnmts. Profiles of the histone modifying gene H34mc2 increased with age  in control
21                  animals. This age-related increase was significantly attenuated in Pb-exposed animals.
22                  The cerebral cortex tissue used in this experiment was obtained from female primates
23                  who had received 1.5 mg/kg Pb acetate via diet per day from birth until 400 days of age
24                  (resulting in blood Pb levels 19-26 (ig/dL at age 400 days) (Rice. 1990).
25                  Methyltransferases catalyze biological methylation reactions using cofactor S-adenosyl
26                  methionine (SAM) as the methyl donor.  In rats,  SAM exposure after gestational-
27                  lactational Pb exposure (1,500 ppm Pb acetate via drinking water of dams followed by
28                  20-22 days of daily 20 mg/kg BW SAM exposure of offspring) improved hippocampal
29                  LTP and Morris water maze performance at PND44-PND54 (Cao et al.. 2008). Thus, the
30                  improved cognition and synaptic plasticity observed with co-exposure to Pb and the
31                  methyl donor SAM suggest that methylation reactions may be involved in Pb-associated
32                  effects on cognition.
      November 2012                            5-224                 Draft - Do Not Cite or Quote

-------
      5.3.11.12    Cholesterol and Lipid Homeostasis

 1                  Various pathological conditions are associated with elevated plasma free fatty acids or
 2                  elevated cholesterol. Adult male rats exposed to Pb acetate (200, 300, or 400 ppm) in
 3                  their drinking water for 12 weeks had increased cholesterogenesis and phospholipidosis
 4                  in brain tissue (Ademuyiwa et al., 2009). Pb-induced changes in brain cholesterol showed
 5                  an inverse U concentration-response relationship, with the largest increase in brain
 6                  cholesterol observed with 200 ppm Pb followed by 300 ppm Pb. Animals exposed to
 7                  400 ppm Pb did not have significant changes in brain cholesterol. In a separate study, Pb
 8                  treatment (single dose  100 (imol/kg, i.v.) was shown  to depress the activity of
 9                  cholesterol-7-a-hydroxylase, an enzyme involved in biosynthesis of bile acid, which
10                  mediates elimination of cholesterol from the body (Kojima et al., 2005). In Ademuyiwa
11                  et al. (2009). Pb exposure significantly increased brain triglycerides by 83% at 300 ppm
12                  and by 108% at 400 ppm. At 200 ppm, Pb exposure induced a statistically nonsignificant
13                  decrease in brain triglycerides. Pb exposure across all three dose groups induced
14                  significantly increased brain phospholipids. Interestingly, plasma free fatty acids were
15                  significantly elevated in a concentration-dependent manner; plasma triglycerides and
16                  cholesterol were unaffected by Pb exposure. The molar ratio of brain cholesterol to
17                  phospholipids, an indicator of membrane fluidity, was significantly increased at 200 and
18                  300 ppm Pb exposure,  indicating increased membrane fluidity. Brain Pb in all dose
19                  groups was below the limit of detection (0.1 ppm). Blood Pb levels at 0, 200, 300, and
20                  400 ppm were 7, 41, 61, and 39 (ig/dL, respectively,  higher than those relevant to
21                  humans. In summary, a recent study found that adult  12-week Pb exposure significantly
22                  increased brain cholesterol, triglycerides, and phospholipids as well as significantly
23                  increased plasma free fatty acids in rats. These  effects were sometimes more prominent at
24                  the lower 200 ppm  Pb dose. The impacts of these Pb-related changes in phospholipidosis
25                  and cholesterogenesis in the brain on downstream nervous system effects are not well
26                  characterized.
      5.3.12 Lifestage of Pb Exposure and Neurodevelopmental Deficits

27                  Environmental exposures during critical lifestages can affect key physiological systems
28                  that orchestrate plasticity (Feinberg. 2007). Exposures during the prenatal and/or early
29                  postnatal period may be especially detrimental for neurodevelopmental effects because of
30                  active neuronal growth and/or synaptogenesis/pruning structure that occur during these
31                  periods (Rice and Barone. 2000; Landrigan et al.. 1999). However, brain development
32                  has been shown to continue throughout adolescence. MRI studies in children and adults
33                  ages 3-30 years have shown that total cerebral volume peaks at age  10.5 and 14.5 years in
34                  females and males, respectively (Giedd et al., 2009; Lenroot and Giedd. 2006). The

      November 2012                            5-225                  Draft - Do Not Cite or Quote

-------
 1                   volume of the cerebellum was found to peak two years later. Lateral ventricular volume
 2                   and white and gray matter volume also were found to increase throughout adolescence.
 3                   Gray matter volume peaked 1 to 3 years earlier in females than males. These observations
 4                   that brain development is active throughout childhood and in adolescence indicate the
 5                   potential for neurodevelopment to be altered later in childhood.

 6                   Epidemiologic studies consistently show that blood Pb levels measured during various
 7                   lifestages and time periods, including the prenatal period, early childhood, later
 8                   childhood, and averaged over multiple years, are associated with cognitive function
 9                   decrements and increases in behavioral problems. These observations of Pb-associated
10                   elevated risk of neurodevelopmental deficits in children are well supported by findings in
11                   animals that prenatal and/or postweaning Pb exposure alters brain development via
12                   changes in synaptic architecture (Section 5.3.11.4) and neuronal outgrowth
13                   (Section 5.3.11.10) and leads to impairments in memory and learning (Section 5.3.2.3)
14                   and increases in impulsivity (Section 5.3.3.1). In monkeys, Pb exposures during multiple
15                   lifestages and time periods, including lifetime, lactational, or postlactation to adulthood,
16                   resulted in impaired cognitive function, although not on all tests (Rice. 1992b. 1990; Rice
17                   and Karpinski. 1988). On one test of executive function in the same monkeys at ages 5-6
18                   years, impairments were found with lifetime Pb exposure starting from birth or starting
19                   after weaning but not infancy-only exposure (Rice and Gilbert. 1990b). The latter
20                   observations indicate that gestational or early infancy Pb exposures are not necessary to
21                   induce cognitive function decrements in juvenile animals.

22                   Unlike other organ systems, the unidirectional nature of CNS development limits the
23                   capability of the developing brain to compensate for cell loss, and environmentally-
24                   induced cell death can result in a permanent reduction in cell numbers (Bayer, 1989).
25                   Hence, when normal development is altered, the early effects have the potential to persist
26                   into adult life even in the absence of concurrent exposure,  magnifying the potential public
27                   health impact. Some epidemiologic evidence indicates associations of earlier childhood
28                   blood or tooth Pb levels with cognitive function decrements, increases in inattention, and
29                   increases in misconduct in adolescents  or adults (Mazumdar et al..  2011; Fergusson et al.
30                   2008; Wright et al.. 2008; Ris et al.. 2004; Bellinger et al.. 1994a: Stiles and Bellinger.
31                   1993). These epidemiologic studies did not  examine adult blood Pb levels, thus the
32                   relative influence of adult Pb exposure  cannot be ascertained. In the Boston cohort,
33                   stronger associations observed for age 2 year blood Pb level than concurrent blood Pb
34                   level  with FSIQ decrements at ages 57 months and 10 years indicated an effect of earlier
35                   rather than later childhood Pb exposures (Bellinger et al.. 1992; 1991). The persistence of
36                   effects of early exposures is supported by findings of impaired learning in adults
37                   monkeys that had juvenile Pb exposure (Rice. 1992b. 1990). A few available recent
38                   toxicological studies also found that infancy Pb exposure but not adult-only Pb exposure
      November 2012                             5-226                 Draft - Do Not Cite or Quote

-------
 1                  led to neurodegenerative amyloid plaque formation in the brains of aged rodents and
 2                  monkeys (Section 5.3.10.1).

 3                  With repeated assessments of children prenatally to later childhood and early adulthood,
 4                  the prospective cohort studies have provided data to compare the neurodevelopmental
 5                  effects associated with blood Pb levels measured at different lifestages and time periods.
 6                  In the collective body of evidence, cognitive function decrements in children have been
 7                  associated with prenatal, early childhood, childhood average, and concurrent blood Pb
 8                  levels, without clear indication of a single critical lifestage or duration of Pb exposure
 9                  related to risk of neurodevelopmental effects in children. In prospective studies, the
10                  identification of critical developmental periods with regard to risk of neurodevelopmental
11                  decrements from Pb exposure has been complicated by the high degree of correlation of
12                  the blood Pb levels of children  over time and the confounding of age and peak blood Pb
13                  levels (Lanphear et al., 2005; Dietrich et al.. 1993a; Needleman et al..  1990).

14                  As described in detail in the 2006 Pb AQCD (U.S. EPA. 2006b). several studies with
15                  varying lengths of follow-up demonstrated associations of prenatal blood Pb levels
16                  (maternal and umbilical cord) with decrements in cognitive function throughout
17                  childhood and into early adulthood (Section 5.3.2). These findings are consistent with the
18                  observations of active CNS development occurring during prenatal development as
19                  described above. Substantial fetal Pb exposure may occur from mobilization of maternal
20                  skeletal Pb stores, which may be related to past maternal Pb exposures (Gulson et al..
21                  2003; Hu and Hernandez-Avila. 2002). Pb can cross the placenta to affect the developing
22                  fetal nervous system (Rabinowitz. 1988). Among 94-211 mother-child pairs in Albany,
23                  NY, maternal-cord blood Pb level  correlations of 0.53-0.81 were reported, depending on
24                  the stage of pregnancy, indicating the influence of maternal blood Pb levels on newborn
25                  blood Pb levels (Schell et al.. 2003). Depending on the magnitude of child exposure, the
26                  contribution of maternal blood  Pb  levels on child blood Pb levels may wane early, and by
27                  age 9 months, child blood Pb levels may be influenced mainly by child Pb exposures
28                  (Section 4.4.1). Thus, associations of neurodevelopmental outcomes assessed after
29                  infancy with postnatal blood Pb levels may reflect effects of postnatal Pb exposures.

30                  In most studies of very young children, ages <2 years, decrements in MDI score were
31                  associated with higher prenatal (maternal or cord) and concurrent blood Pb levels (Table
32                  5-14). Among studies that examined blood Pb  levels at multiple time periods, several
33                  found larger blood Pb-associated decrements in MDI for prenatal blood Pb than
34                  concurrent blood Pb (Hu et al.. 2006;  Gomaaet al.. 2002; Bellinger etal.. 1987; Dietrich
35                  etal.. 1986). In the Yugoslavia cohort, per log increase in blood Pb level, the MDI
36                  decrement at age 2 years was larger for concurrent blood Pb than for prenatal  cord blood
37                  Pb (Wasserman et al.. 1992). Concurrent blood Pb levels were higher than prenatal cord
      November 2012                            5-227                  Draft - Do Not Cite or Quote

-------
                 blood Pb levels. The collective evidence indicates that both prenatal and postnatal child

                 Pb exposures may contribute to neurodevelopmental effects in children from infancy to

                 age 2 years, with some indication that prenatal Pb exposure has a stronger effect.
Table 5-14   Associations of cognitive function with blood Pb levels measured at
                 various lifestages and time periods  in prospective studies.
Study3
Study Population and
Methodological Details
(Within studies, effect estimates are
presented in order of lifestage or
time period of blood Pb
measurement, with largest effect
estimate in bold)	
Blood Pb
Levels
(ug/dL)
                                                            Outcome
                                                                         Effect Estimate
                                                                         (95%  Cir
Cognitive Assessment at Age 2 Years and Younger
Bellinger et    249 children followed from birth
al. (1987)     (1979-1981) to age 3 yr, Boston
             area, MA
             Moderate participation rate, high
             follow-up retention. Participants had
             higher cord blood Pb levels, higher
             SES, maternal IQ, HOME score.
             Regression model adjusted for the
             maternal age, race, IQ, education,
             years cigarette smoking,3rd
             trimester alcoholic drinks/ week,
             study period mean SES, HOME,
             child sex, birth weight, gestational
             age, birth order.
                                 Prenatal
                                 (cord)
                                 Mean (SD):
                                 6.6 (3.2)
              Overall
              Bayley MDI
              among
              Ages 6, 12,
              18, and 24
              mo
                                                                         vs. prenatal blood Pb level <3 ug/dL:
                                                                         Prenatal 6-7 |jg/dL:
                                                                         -3.8 (-6.3,-1.3)
                                                                         Prenatal > 10 |jg/dL:
                                                                         -4.8 (-7.3, -2.3)
                                                                         Concurrent reported to not to be
                                                                         associated with overall MDI, no
                                                                         quantitative data reported.
Dietrich et     280 children followed prenatally to
al. (1986)     age 6 mo, Cincinnati, OH
             No information on participation rate.
             Log linear regression model adjusted
             for birth weight, gestation, sex. Also
             considered potential confounding by
             SES, HOME score, prenatal
             smoking and alcohol use, maternal
             Fe binding.
                                 Prenatal
                                 (maternal)
                                 Mean (SD):
                                 8.0 (3.8)
                                 Concurrent
                                 Mean (SD):
                                 4.5 (2.9)
              Bayley MDI
              Age 6 mo
                                                                         Prenatal: -0.6 (-1.1, -0.09)
                                                                         Concurrent: -0.23 (-0.58, 0.12)
Huetal.      146 children born 1997-1999
(2006)        followed prenatally to age 2 yr,
             Mexico City, Mexico
             Moderate follow-up participation.
             Eligible similar to non-eligible. Log
             linear regression model adjusted for
             sex, maternal age, current weight,
             height-for-age Z score, maternal IQ,
             concurrent blood Pb (in models
             examining blood Pb at other
             lifestages). Considered potential
             confounding by other unspecified
             factors.
                                               Prenatal
                                               (maternal 1st
                                               trimester):
                                               Mean(range):
                                               7.1 (1.5-43.6)
                                               Prenatal avg:
                                               NR
                                               Earlier
                                               childhood at
                                               12 mo:

                                               Mean (SD):
                                               5.2 (3.4)
                                               Concurrent
                                               Mean (SD):
                                               4.8 (3.7)
                                               Bayley MDI
                                               Age 24 mo
                          Per log increase in blood Pb:
                          Prenatal 1st trimester: -4.1 (-8.1, -0.17)
                          Prenatal avg: -3.5 (-7.7, 0.63)
                          Age 12 month: -2.4 (-6.2, 1.49)
                          Concurrent: -1.0 (-3.9, 1.9)
November 2012
                                   5-228
                           Draft - Do Not Cite or Quote

-------
Study3
Study Population and
Methodological Details
(Within studies, effect estimates are
presented in order of lifestage or
time period of blood Pb
measurement, with largest effect
estimate in bold)	
Blood Pb
Levels
(ug/dL)
Outcome
              Effect Estimate
              (95% Cl)b
Gomaa et     197 children followed prenatally to
al. (2002)      age 2 yr, Mexico City, Mexico
              Moderate participation but high
              retention. No selective attrition. Log
              linear regression model adjusted for
              maternal IQ, maternal age, sex,
              parental education, marital status,
              breastfeeding duration, child
              hospitalization status. Did not
              consider potential confounding by
              parental caregiving quality.
                                    Prenatal
                                    (cord)
                                    Mean (SD):
                                    6.7 (3.4)
                                    Concurrent
                                    Mean (SD):
                                    8.4 (4.6)
               Bayley MDI
               Age 24 mo
              Per log increase in blood Pb:
              Prenatal: -2.1 (-3.9, -0.39)
              Concurrent: -0.09 (-0.58, 0.42)
Wasserman   392 children followed prenatally to
et al. (1992)   age 24 mo, Kosovo, Yugoslavia (K.
              Mitrovica, Pristina)
              High follow-up participation, no
              selective attrition.  K. Mitrovica near
              smelter. Log linear regression model
              adjusted for sex, birth order, birth
              weight, ethnicity, HOME, maternal
              education, age, and IQ.
                                    Prenatal
                                    (cord)
                                    Mean (SD):
                                    14.4(10.4)
                                    Concurrent
                                    Means:
                                    K. Mitrovica:
                                    35.4,
                                    Pristina: 8.5
               Bayley MDI
               At age 24
              Per log increase in blood Pb:
              Concurrent: -4.1 (-6.2, -2.0)
              Prenatal: -3.2 (-7.2, 0.86)
              359 children, followed prenatally to
              age Syr, Cleveland, OH
              Prospective. Recruitment at birth
              hospital. High follow-up participation,
              more white, higher IQ,  nonalcoholic
              mothers not followed. 50% born to
              alcoholic mothers. Linear regression
              adjusted for age, race, sex, birth
              order, parental education, maternal
              IQ, Authoritarian Family Ideology,
              HOME.
                                    Means (SD):
                                    Prenatal
                                    cord:
                                    6.0(2.1)
                                    6 mo:
                                    10.1  (3.3)
                                    Concurrent:
                                    16.7(6.5)
               Bayley MDI   Variance estimates:
               Age2yr      Prenatal: 0.0003, t =-0.21d
                            Age 6 mo: 0.00, p = 0.95d
                            Concurrent: 0.00,  p = 0.95d
Cognitive Function Assessments at School Age
Canfield et    172 children born 1994-1995
al. (2003a)    followed from age 6 mo to 5 yr,
              Rochester, NY
              Recruitment from study of dust
              control. 73% nonwhite. High follow-
              up participation, no selective
              attrition. Linear regression model
              adjusted for maternal race, IQ,
              education, and prenatal smoking
              status, household income, HOME
              score, child sex, Fe status, birth
              weight.
                                    Means (SD):
                                    Infancy avg
                                    (6-24 mo):
                                    7.0 (3.8)
                                    Peak:
                                    11.1 (7.1)
                                    Concurrent:
                                    5.8(4.1)
                                    Lifetime (to
                                    age 5 yr)
                                    avg:
                                    7.4 (4.3)
               FSIQ
               Stanford-
               Binet
               Age 5 yr
              Infancy avg: -0.53 (-0.95, -0.13)
              Peak: -0.26, (-0.47, -0.05)
              Concurrent: -0.61 (-0.99, -0.24)
              Lifetime avg: -0.57 (-0.93, -0.20)
Wasserman   332 children followed prenatally to
et al. (1994)   age 3-4 yr, Kosovo, Yugoslavia (K.
              Mitrovica, Pristina)
              High follow-up participation. More
              participants were male, Albanian,
              and had lower maternal IQ and
              HOME. Log linear regression model
              adjusted for HOME score, maternal
              age, intelligence, and education,
              language, birth weight, child sex.
                                    Prenatal
                                    (cord)
                                    Mean (SD):
                                    14.4(10.4)
                                    Concurrent
                                    means:
                                    K. Mitrovica:
                                    39.9
                                    Pristina:
                                    9.6
                                    Overall mean
                                    NR
               FSIQ
               McCarthy
               General
               Cognitive
               Index
               Age 3-4 yr
              Per log increase in blood Pb:
              Prenatal: -7.1 (-11.8, -3.1)
              Age 2 yr: -10.4 (-15.2, -5.7)
              Concurrent:-9.4 (-14.2, -4.6)
November 2012
                                     5-229
                             Draft - Do Not Cite or Quote

-------
Study3
Study Population and
Methodological Details
(Within studies, effect estimates are
presented in order of lifestage or
time period of blood Pb
measurement, with largest effect
estimate in bold)	
Blood Pb
Levels
(ug/dL)
Outcome
              Effect Estimate
              (95% Cl)b
Bellinger et    148 children followed from birth
al. (1992)      (1979-1981) to age 10yr, Boston,
              MA area
              Moderate follow-up participation.
              Participants had higher SES and
              HOME scores. Linear regression
              model adjusted for HOME score
              (age 10 and 5 yr), maternal race, IQ,
              and marital status, SES,  child sex,
              birth order, and stress, # residence
              changes. Also considered potential
              confounding by family stress,
              maternal age, psychiatric factors,
              and child serum ferritin levels.
                                    Age 6 mo
                                    Mean (SD):
                                    6.7 (7.0)
                                    Earlier
                                    childhood
                                    Age 2 yr
                                    Mean (SD):
                                    6.5 (4.9)
                                    Concurrent
                                    Mean (SD):
                                    2.9 (2.4)
               FSIQ
               Wechsler
               Intelligence
               Scale for
               Children-
               Revised
               Age 10 yr
              Age 6 mo:-0.13 (-0.42, 0.16)
              Age 2 yr: -0.58 (-0.99, -0.18)
              Concurrent -0.46 (-1.5, 0.56)
Dietrich et     253 children followed from birth
al. (1993b)     (1979-1985) to age 6.5 yr,
              Cincinnati, OH
              High follow-up participation.
              Participants had slightly higher age 1
              yr blood Pb levels. Linear regression
              model adjusted for HOME score,
              maternal IQ and prenatal cigarette
              smoking, child birth weight, birth
              length, sex. Also considered
              potential confounding by perinatal
              complications, prenatal maternal
              substance abuse, nutritional status.
                                    Prenatal
                                    (maternal)
                                    Mean (SD):
                                    8.3 (3.7)
                                    Age 5 yr
                                    Mean (SD):
                                    11.8(6.3)
                                    Concurrent
                                    NR
                                    Lifetime (to
                                    age 6.5 yr)
                                    avg:
                                    NR
               FSIQ
               Wechsler
               Intelligence
               Scale for
               Children-
               Revised
               Age 6.5 yr
              Prenatal: 0.15 (-0.26, 0.56)
              Concurrent: -0.33 (-0.60, -0.06)
              Lifetime avg: -0.13 (-0.35, 0.09)
Baghurst      494 children followed from birth
et al. (1992)   (1979-1982) to age 7 yr, Port Pirie,
              Australia
              Moderate follow-up participation.
              Participants had higher SES and
              breastfeeding, less maternal
              smoking. Log linear regression
              model adjusted for sex, birth weight,
              birth order, feeding method,
              breastfeeding duration, parental
              education  and smoking, maternal
              age and IQ, SES, HOME, parents
              living together.
                                    Prenatal
                                    (maternal)
                                    Mean 2nd
                                    quartile:
                                    7.4
                                    Earlier
                                    childhood
                                    Age 2 yr
                                    Mean 2nd
                                    quartile:
                                    16.6
                                    Lifetime (to
                                    age 7 yr) avg
                                    Mean 2nd
                                    quartile:
                                    15.7
               FSIQ
               Wechsler
               Intelligence
               Scale for
               Children-
               Revised
               Age 7-8 yr
              Prenatal: 0.26 (-0.67, 1.5)
              Age 2 yr: -2.0 (-3.8, -0.21)
              Lifetime avg:-1.6 (-3.7, 0.52)
Schnaas et    150 children followed from prenatally
al. (2006)      (1987-1992) to age 6-10 yr, Mexico
              City, Mexico
              Low follow-up participation.
              Participants had higher SES, FSIQ,
              higher blood Pb level before age 5
              yr, lower at older ages.  Log linear
              mixed effects regression model
              adjusted for SES, maternal IQ,
              HOME score, child sex, birth weight,
              indicator of first FSIQ measurement,
              random slope for subject.  Most
              covariates assessed in  pregnancy or
              within child age 6  mo.
                                    Geometric
                                    Mean (range)
                                    Prenatal
                                    (maternal
                                    28-36 week
                                    gestation):
                                    7.8 (2.5-
                                    24.6)
                                    Age 5 yr: 9.3
                                    (3.8-18.0)
                                    Age 6-10 yr
                                    avg: 6.2
                                    (2.2-18.6)
               FSIQ
               Wechsler
               Intelligence
               Scale for
               Children-
               Revised
               Ages 6-10
               yr
              Per log increase in blood Pb:
              Prenatal: -4.0 (-6.4, -1.7)
              Age 5 yr: -0.32 (-4.3, 3.4)
              Age 6-10 yr avg: -2.5 (-4.1, -0.81)
November 2012
                                     5-230
                             Draft - Do Not Cite or Quote

-------
Study3
Study Population and
Methodological Details
(Within studies, effect estimates are
presented in order of lifestage or
time period of blood Pb
measurement, with largest effect
estimate in bold)	
Blood Pb
Levels
(ug/dL)
                                                                 Outcome
                                                                               Effect Estimate
                                                                               (95% Cl)b
Ris et al.      195 children followed prenatally
(2004)        (1979-1985) to age 15-17 yr,
              Cincinnati, OH
              Prospective. High follow-up
              participation, no selective attrition.
              Mostly African-American. Linear
              regression model adjusted for SES,
              maternal IQ, average HOME,
              adolescent marijuana use, and
              obstetrical complications. Also
              considered potential confounding by
              birth outcomes, maternal age,
              prenatal smoking, alcohol,
              marijuana, and narcotics use, #
              previous abortions, stillbirths,
              gravidity, parity, caregiver education,
              public assistance, child age, sex,
              health, and Fe status
                                                   Prenatal,
                                                   Earlier
                                                   childhood
                                                   Age 6.5 yr,
                                                   Earlier
                                                   childhood
                                                   avg (Age
                                                   3-78 mo):
                                                   NR
                                                   Learning/IQ
                                                   composite
                                                   Wechsler
                                                   Intelligence
                                                   Scale for
                                                   Children
                                                   indices
                                                   normal
                                                   scores
                                                   Age 15-17
                                                   yr
                             Prenatal: -0.08 (-0.18, 0.03)
                             Age 6.5 yr: -0.08 (-0.17, 0.003)
                             Age 3-78 mo avg: -0.03 (-0.18, 0.03)
Lanphear     1,333 children pooled from Boston,
et ai. (2005)   Cincinnati, Cleveland, Mexico City,
              Port Pirie, Rochester, and
              Yugoslavia cohorts.
              Uniform analysis of cohorts from
              diverse locations and SES. Blood Pb
              levels and FSIQ measured at
              different ages. Several sensitivity
              analyses to examine heterogeneity
              of results by cohort, model
              specification, and confounding. Log
              linear regression model adjusted for
              HOME score, birth weight, maternal
              IQ, maternal education. Also
              considered potential confounding by
              child sex,  birth order, maternal age,
              marital status, prenatal smoking
              status, prenatal alcohol use.
                                    Median
                                    (5th-95th)
                                    Early
                                    childhood
                                    (mean ages
                                    6-24 mo):
                                    12.7
                                    (4.0-34.5)
                                    Peak:
                                    18.0
                                    (6.2-47.0)
                                    Lifetime avg
                                    (to ages
                                    4.8-1 Oyr):
                                    12.4
                                    (4.1-34.8)
                                    Concurrent:
                                    9.7 (3.5-33.2)
               FSIQ
               Various
               tests
               Ages 4.8-10
               yr
                                                                               Mean ages 6-24 mo: -0.14 (-0.23, -0.06)
                                                                               Peak:-0.20 (-0.29,-0.11)
                                                                               Concurrent: -0.23 (-0.29, -0.11)
                                                                               Lifetime avg: -0.15 (-0.22, -0.09)
Pocock et     Meta-analysis of 5 prospective
al. (1994)     studies (over 1,100 children) from
              Port Pirie and Sydney, Australia,
              Cincinnati, Cleveland, Boston
              Meta-analysis of combining
              covariate-adjusted effect estimates
              from individual studies.
                                    Earlier
                                    childhood
                                    (2 yr) range
                                    in means:
                                    6.8-21.2
                                    Around birth
                                    and
                                    Postnatal:
                                    NR
               FSIQ
               Various
               tests
               Ages 5-10
               yr
                                                                               Around birth: 0.26 (-1.5, 2.0)
                                                                               Age 2 yr: -2.7 (-4.1, -1.2)
                                                                               Postnatal mean: -1.3 (-2.9, 0.37)
MDI = Mental Development Index, FSIQ = full-scale IQ, NR = Not reported
"Results are presented first for MDI in children up to age 3 years, then for FSIQ in school-aged children. Within studies, effect
estimates are presented in order of increase lifestage or time period of blood Pb measurement, with the largest effect estimate in
bold.
bEffect estimates are standardized to a  1 ug/dL increase in blood Pb level in analyses of blood Pb as a linear continuous variable.
°Effect estimates represent comparisons between children in different categories of blood Pb level, with children in the lower
blood Pb category serving as the reference group.
Sufficient data were not provided in order to calculate 95% Cl.
November 2012
                                      5-231
                             Draft - Do Not Cite or Quote

-------
 1                   Prenatal and early postnatal (age 6 month) blood Pb levels also were associated with
 2                   cognitive function in children examined at school-age (ages 4-17 years) (Table 5-14).
 3                   However, most of these studies also found cognitive function decrements in association
 4                   with postnatal blood Pb levels, and results did not identify an individual postnatal time
 5                   period of blood Pb measurement associated with cognitive function decrements.
 6                   Increases in concurrent blood Pb levels were associated with larger decrements in FSIQ
 7                   in the Cincinnati and Yugoslavia cohorts at ages 3-6.5 years than were prenatal blood Pb
 8                   levels (Wasserman et al.. 1994; Dietrich et al..  1993b). In the Cincinnati cohort as
 9                   adolescents ages 15-17 years, increases in both prenatal and higher earlier childhood (age
10                   6.5 years) were associated with decrements in a learning/memory composite score (Ris et
11                   al.. 2004). In the Boston and Port Pirie cohorts, increases in age 2 year blood Pb levels
12                   were associated with larger FSIQ decrements at ages  7 and 10 years, respectively, than
13                   were concurrent or lifetime average blood Pb levels (Baghurst et al.. 1992; Bellinger et
14                   al., 1992). However, in the Port Pirie cohort, an association also was found lifetime (to
15                   age 7 years) average blood Pb level. Among children ages 6-10 years in Mexico City, per
16                   unit increase, prenatal blood Pb levels were associated with larger FSIQ decrements than
17                   individual blood Pb levels between ages 1-5 or the age 6-10 year average (Schnaas et al..
18                   2006). In contrast, results from the Rochester cohort indicated that increases in lifetime
19                   average and concurrent blood Pb level were associated with larger FSIQ decrements at
20                   ages 5 years than were increases in peak blood Pb level (Canfield et al., 2003a).
21                   Collectively, the epidemiologic findings indicate that blood Pb levels measured at various
22                   postnatal time periods, earlier childhood, childhood average, later childhood, and
23                   concurrent blood Pb levels are associated with  decrements in cognitive function when
24                   assessed in school-aged children.

25                   Consistent with individual  studies, analyses combining studies pointed to associations of
26                   FSIQ in school-aged children with blood Pb levels measured at various lifestages and
27                   time periods. The analysis pooling data from seven prospective studies found that
28                   increases in infancy  average (age 6-24 months), peak, concurrent, and lifetime average
29                   peak blood Pb  levels were associated with decreases in FSIQ in children ages
30                   4.8-10 years (Table 5-14). Investigators reported that the model with concurrent blood Pb
31                   level explained the largest proportion of variance in FSIQ (R2) (Lanphear et al., 2005). In
32                   a meta-analysis of results from five cohort studies (Pocock et al.. 1994). a  larger decrease
33                   in FSIQ was estimated for an increase in peak (around age 2 years) blood Pb level than
34                   for blood Pb level  measured around birth or after age 2 years. Deciduous tooth Pb levels
35                   have been associated with decrements in cognitive function and increases in attention-
36                   related behavioral  problems in children and young adults (Table 5-8. Table 5-9. and
37                   Table 5-11). These results indicate cumulative  Pb exposure over several years may
38                   contribute to neurodevelopmental effects in children. In school-aged children, concurrent
39                   blood Pb levels reflect past Pb exposures that are mobilized from bone remodeling to

      November 2012                             5-232                  Draft - Do Not Cite or Quote

-------
 1                  blood and recent exposures. Thus, associations with concurrent blood Pb levels also may
 2                  reflect an effect of cumulative past and more recent Pb exposures.

 3                  Studies conducted in the Cincinnati cohort examined diverse neurodevelopmental effects
 4                  and found that prenatal and neonatal blood Pb levels were associated with impaired
 5                  auditory processing in children ages 5 years and increases in parental ratings of
 6                  delinquent behavior in adolescents ages 15-17 years (Dietrich etal. 2001; Dietrich et al.,
 7                  1992) but not cognitive function or motor function decrements in children at age 6 years
 8                  (Bhattacharya et al., 2006; Bhattacharya et al.. 1995; Dietrich et al., 1993b; Dietrich et
 9                  al.. 1993a). These findings suggest that the critical lifestage of Pb exposure may vary
10                  among nervous system effects.

11                  Some studies have aimed to improve the characterization of important lifestages and time
12                  periods of Pb exposure by examining children in whom blood Pb levels are not strongly
13                  correlated over time (i.e., children whose blood Pb level ranking changed over time)
14                  (Hornung et al.. 2009; Schnaas et al.. 2006; Chen etal.. 2005; Tongetal.. 1998; Bellinger
15                  et al.. 1990). Collectively, most results indicated FSIQ decrements in association with
16                  concurrent blood Pb levels but did not conclusively demonstrate stronger findings for
17                  early or concurrent blood Pb levels (Table 5-15).
     November 2012                            5-233                  Draft - Do Not Cite or Quote

-------
Table 5-15    Comparisons of blood Pb-FSIQ associations in groups of children
                 with different temporal trends in  blood Pb levels.
Study
Hornung
etal.
(2009)






Bellinger
etal.
(1990)



long et
al.
(1998)

Study Population and
Methodological Details
397 children followed from
birth (1979-1 984) to age 6 yr,
Rochester, NY and
Cincinnati, OH.
High follow-up participation.
No selective attrition. Linear
regression model adjusted
for city, HOME score, birth
weight, maternal IQ,
maternal education.
170 children followed
prenatally to age 57 mo,
Boston area, MA
High follow-up participation.
No report on characteristics
of subjects followed. Log
linear regression adjusted for
HOME score, SES, maternal
IQ, maternal age, sex,
ethnicity
375 children followed from
birth (1979-1 982) to age
11-1 Syr, Port Pirie, Australia
Moderate follow-up
Darticioation. Particioants
Blood Pb Levels
(ug/dL)
Geometric mean
(5th-95th):
Earlier childhood
age 2 yr:
8.9 (3.0-23.8)
Concurrent; 6.0
(1.9-17.9)



High prenatal (cord)
>10
Low concurrent: <3
Medium concurrent:
3-10
High concurrent >10


Means
Earlier childhood
age 2 yr: 21 .2
Age 11-1 Syr: 7.9
Outcome
FSIQ
Wechsler Intelligence
Scale for Children-
Revised (WISC-R)
Age 6 yr




Change in McCarthy
General Cognitive Index
(GCI)z-score
Between ages 24 and 57
mo



Change in cognitive
function z-scores
between ages 2 and 1 1 yr
n— ., .i_, . it n r\\
Effect Estimate(95% Cl)
Difference in FSIQ at age 6 yr
with an age
6:2 yr blood Pb level ratio = 0.5 as
the reference
Ratio age 6:2 yr
blood Pb level = 2.0
-7.0 (-10, -4.0)



High prenatal/Low concurrent
0.42 (-0.1 5, 0.99)
High prenatal/Medium concurrent
0.15 (-0.14, 0.44)
High prenatal/High concurrent
-0.1 5 (-0.56, 0.26)


<10.2 ug/dL decline:
0.03 (-0.1 5, 0.21)
10.2-1 6.2 ug/dL decline:
0.04 (-0.1 5, 0.23)
          had lower early blood
          Pb-cognitive function
          association. Log linear
          regression model adjusted
          for sex, birth weight, birth
          rank, feeding style,
          breastfeeding duration,
          maternal IQ, maternal age,
          SES, HOME score, parental
          smoking, parents living
          together. ANOVA to assess
          association of change in IQ
          with change in blood Pb
          across time intervals
                   Bayley MDI,
                   age 2 yr
                   McCarthy GCI,
                   age4yr
                   WISC-R,
                   ages 7,
                   and 11-13yr
                        >16.2 ug/dL decline:
                        -0.01 (-0.20,0.18)
Chen et    780 children participating in
al.        the TLC trial from age
(2005)     12-33 mo to age 7 yr,
          Baltimore, MD; Cincinnati,
          OH; Newark, NJ;
          Mostly African-American.
          50% given chelation at ages
          12-33 mo, blood Pb levels
          20-44 ug/dL. No information
          on participation rate.
          Regression-based path
          analysis adjusted for city,
          race, sex, language, parental
          education and employment,
          single parent, age at blood
          Pb measurement, caregiver
          IQ. Considered potential
          confounding by chelation but
          not parental caregiving
          quality.
Mean (SD):
Age 2 yr: 26.2 (5.1)
Age Syr: 12.0(5.2)
Age 7 yr: 8.0 (4.0)
Low age 2 yr: <24.9
Low age 7 yr: <6.2
WISC-lllatage7yr
Difference in FSIQ vs. Low age 2,
Low age 7 as the reference


Low age 2, High age 7:
-0.27 (-0.48, -0.05)
High age 2, Low age 7:
0(-0.21, 0.20)
High age 2, High age 7:
-0.28 (-0.47,-0.10)
November 2012
           5-234
                 Draft - Do Not Cite or Quote

-------
 1                   Schnaas et al. (2006) followed children in Mexico City prenatally to age 10 years and
 2                   found maternal blood Pb levels at 28-36 weeks of pregnancy to be weakly correlated with
 3                   repeated measures of blood Pb between ages 1 and 10 years (Pearson r < 0.23). In models
 4                   analyzing different lifestages of blood Pb level individually and in a mixed effects model
 5                   that included prenatal and multiple postnatal blood Pb measures, maternal 28-36 week
 6                   blood Pb level was associated with a larger decrement in FSIQ (Table 5-15). In the model
 7                   with multiple lifestages of blood Pb level, analysis of variance inflation factors indicated
 8                   a lack of collinearity among the serial blood Pb measures.

 9                   Pooling the Cincinnati and Rochester cohorts (n = 397), Hornung et al. (2009) created a
10                   new indicator of Pb exposure: the ratio of blood Pb level at 6 years of age to that at
11                   2 years of age. As illustrated in Figure 5-15. the three groups of children representing the
12                   three temporal trends in blood Pb levels: no change ages 2-6 years (ratio = 1), higher
13                   blood Pb at age 6 years than 2 years (ratio =  1.25), higher blood Pb  levels at age 2 years
14                   than 6 years (ratio = 0.5), have similar areas under the curve, indicating that cumulative
15                   blood Pb levels were similar in  the three groups. Thus, differences in FSIQ are more
16                   likely to be attributable to differences in temporal trends. The lowest FSIQ at age 6 years
17                   was found in children with an age 6:age 2 year blood Pb ratio  1.25,  i.e., children who had
18                   an increase in blood Pb level from 2 to 6 years of age (Figure 5-15 and Table 5-15).
      November 2012                             5-235                  Draft - Do Not Cite or Quote

-------
                           18
                           16
                           14
                           12
               6-year:2-year
                ratio = 0.5
                 IQ = 89.0
6-year:2-year
 ratio = 1.25
     = 83.7
                                                   3          4
                                                   Age (years)
      Note: In combined Cincinnati and Rochester cohorts, FSIQ was compared among three different patterns of blood Pb level changes
      over time: peak at 2 years and age 6:2 year ratio = 0.5 (blue diamonds), peak at 5 years and age 6:2 year ratio = 1.25 (black
      triangles), and constant blood Pb level and age 6:2 year ratio = 1 (white squares). All three patterns have a similar cumulative blood
      Pb level (10 ug/dL) as indicated the areas under the curve. Children whose blood Pb levels peaked at age 5 years had the lowest
      FSIQ at age 6 years.
      Source: Hornung et al. (2009)
      Figure 5-15      Estimated FSIQ for three patterns of temporal trends in blood Pb
                         level from ages 2 to  6 years in the Rochester and Cincinnati
                         cohorts.
 i
 2
 o
 3
 4
 5
 6
 7
 8
 9
10
n
12
13
U
15
In the Boston cohort with comparable blood Pb levels to those in the Rochester cohort,
Bellinger et al. (1990) found that at age 57 months, FSIQ, as assessed by McCarthy GCI,
was similar between children in the high (> 10 (ig/dL) and low (<3 (ig/dL) prenatal cord
blood Pb groups. Additionally, children with high prenatal and high concurrent blood Pb
level (> 10 (ig/dL) had a decrease (-0.15 standard deviation [95% CI: -0.46, 0.26]) in
FSIQ from age 24 to 57 months. In contrast, children with high prenatal blood Pb but low
concurrent blood Pb level (<3 (ig/dL) had an increase (0.42 standard deviation [95% CI:
-0.15, 0.99]) in FSIQ from age 24 to 57 months (Table 5-15). These findings indicated
that by age 5 years,  children with higher prenatal blood Pb levels appeared to recover the
Pb-associated decrements in cognitive function unless concurrent blood Pb levels
remained high. The  investigators also demonstrated that optimal sociodemographic
characteristics (e.g., higher HOME score, SES, maternal IQ and age, female) also
protected against decrements in cognitive function associated with higher postnatal blood
Pb levels. Collectively, these results suggest that cognitive development is not fixed early
in childhood and can be affected negatively or positively by postnatal influences.
      November 2012
                            5-236
   Draft - Do Not Cite or Quote

-------
 1                   Chen et al. (2005) also found a stronger influence of higher concurrent blood Pb levels on
 2                   FSIQ at age 7 years among children participating in a multi-city chelation trial. Children
 3                   with higher concurrent blood Pb levels (> median 7.2 (ig/dL) had lower IQ at age 7 years,
 4                   regardless of whether blood Pb level at age 2 years was low or high (less than or greater
 5                   than the median of 24.9 (ig/dL, respectively). Blood Pb levels at ages 2 and 7 years were
 6                   weakly correlated (r = 0.27). Because these children had been treated with chelators due
 7                   to high blood Pb levels (20-44 (ig/dL) at ages 12 to 33 months, the findings may have
 8                   limited generalizability to the general population of children currently living in the U.S.

 9                   In contrast with the aforementioned studies, Tong et al. (1998) found that higher early-
10                   life  blood Pb level was associated with a larger decrease in FSIQ than was concurrent
11                   blood Pb level in the Port Pire, Australia cohort at age 11-13 years (Table 5-15). This
12                   conclusion was based on the analysis of groups of children with different degrees of
13                   decline in blood Pb levels between ages 2 and  11-13 years. Although the mean blood Pb
14                   level in the study population declined overall from 21.2 (ig/dL at age 2 years to  7.9 (ig/dL
15                   at age 11-13 years, the magnitude of decline varied among children.  The change in FSIQ
16                   between ages 2 and 11-13 years did not significantly differ between children with the
17                   largest decline (>16 (ig/dL)  in blood Pb level and children with the smaller decline
18                   (<10 (ig/dL) (Table 5-15). These findings indicated an influence of higher blood Pb
19                   levels early in life despite declines in blood Pb with age and a persistence of Pb effects.
20                   The results do not preclude an independent association with concurrent blood Pb level.

21                   A common limitation of studies that examined different temporal trends in blood Pb
22                   levels is the higher blood Pb levels of the study populations compared to those currently
23                   measured in most U.S. children. Additionally, in several study populations, children
24                   experienced large changes in blood Pb levels overtime, for example, 50% decline or 25%
25                   increase in four years in Hornung et al. (2009). It is unclear whether these findings would
26                   apply to children currently in the U.S. within the same age range who would be  expected
27                   to have smaller changes in blood Pb levels over time.

28                   To conclude, the collective body of epidemiologic evidence does not strongly identify an
29                   individual critical lifestage or duration of Pb exposure with regard to neurodevelopmental
30                   effects in children. Cognitive function decrements and behavioral problems have been
31                   associated with prenatal, early childhood, lifetime average, and concurrent blood Pb
32                   levels as well as with childhood tooth Pb levels. The identification of critical lifestages of
33                   Pb exposure is complicated  further by the fact that blood Pb levels in older children,
34                   although affected by recent exposure, are also influenced by Pb stored in bone due to
35                   rapid growth-related bone turnover in children relative to adults. Thus, associations of
36                   neurodevelopmental effects  with concurrent blood Pb level in children may reflect the
37                   effects of past and/or recent Pb exposures (Section 4.3.5.1). Evidence indicates that
      November 2012                             5-237                  Draft - Do Not Cite or Quote

-------
 1                  prenatal blood Pb levels are associated with mental development in very young children
 2                  
-------
 1                  3-5 (ig/dL (Bellinger. 2008; Canfield. 2008; Hornung. 2008; Tellez-Roio. 2008). Except
 2                  for the pooled analysis, the lower strata of blood Pb levels comprised >30% of the study
 3                  population, indicating that the blood Pb-FSIQ relationships calculated for the lower strata
 4                  likely are not outliers or unrepresentative of the overall study population. Studies in the
 5                  Boston and Rochester cohorts each examined different ages of blood Pb and FSIQ but
 6                  lower blood Pb levels compared to other studies (means ~6 (ig/dL), and found nonlinear
 7                  blood Pb-FSIQ concentration-response relationships with respect to children whose peak
 8                  blood Pb levels did not exceed 10  (ig/dL (Bellinger and Needleman. 2003; Canfield et al..
 9                  2003a; Bellinger etal..  1992). Pooled analyses of seven prospective studies, involving a
10                  wider range of blood Pb levels, 5th-95th percentiles 2.4-33.1 (ig/dL, found that nonlinear
11                  models (e.g., log-linear, piecewise linear) fit the relationship between blood Pb level and
12                  FSIQ better than a linear model did (Lanphear et al., 2005; Rothenberg and Rothenberg.
13                  2005).
     November 2012                            5-239                  Draft - Do Not Cite or Quote

-------
 Study              Blood Pb Timing   Outcome     Blood Pb
      (2003)



Canfieldetal.(2003)




Lanphearetal. (2005)
                   Age2yr
                   Concurrent
                   Concurrent
 Cross-sectional
 Tellez-Rojoetal. (2006)   Concurrent
 Lanphearetal. (2000)
 Kordasetal. (2006)
                   Concurrent
                   Concurrent
""-""« stratum (ug/dL)
FSIQ All subjects
Peak<10a*
FSIQ All subjects
Peak< 10=
FSIQ Peak>10=
Peak<10a
Peak>7.5=
Peak<7.5=
BayleyMDI >10b
<10b
5-1 Ob
<5b
Reading score All subjects
<10
<7.5
<5
<2.5
Math score All subjects
<10




1230 -4-








532 -4-







                                                                   Change in cognitive function test score perl ug/dL
                                                                         increase in blood Pb level (95% Cl)
Note: Results are presented first for prospective studies then for cross-sectional studies. FSIQ = full-scale IQ, MDI = mental
development index. Effect estimates (concentration-response) are presented for a 1 ug/dL increase in blood Pb level.  Black symbols
represent effect estimates among all subjects or subjects in the higher blood Pb stratum. Blue symbols represent effect estimates in
lower blood Pb strata. aStrata were  defined by the peak blood Pb level measured  in child at any point during follow up. b95% Cl
estimated from reported p-value.


Figure 5-16      Comparison of associations between  blood Pb  level and cognitive
                     function among various blood Pb strata.
November 2012
                                               5-240
Draft - Do Not Cite or Quote

-------
Table 5-16  Additional characteristics and quantitative results for studies
            presented in Figure 5-16.


Study
Bellinger et
al. (1992)
Bellinger
and
Needleman
(2003)
Bellinger
(2008)






Canfield et
al. (2003a)
Canfield
(2008)









Jusko et al.
(2008)








Lanphear et
al. (2005)
Hornung
(2008)










Study Population and
Methodological Details
148 children followed from
birth (1979-1 981) to age 10
yr, Boston area, MA.
Prospective. Recruitment at
birth hospital. Participation by
59% of original cohort but
88% of eligible. Participants
had higher SES. Linear
regression model adjusted for
HOME score (age 10 and 5),
child stress events, race,
maternal IQ, SES, sex, birth
order, marital status,
# residence changes before
age 5 yr.
172 children born 1994-1995
followed from age 6 mo to
age 5 yr, Rochester, NY.
Prospective. Recruitment
from study of dust control.
73% nonwhite. Moderate
follow-up participation but no
selective attrition. Mixed
effects models adjusted for
child sex, Fe status, and birth
weight, maternal race,
education, IQ, and prenatal
smoking, household income,
HOME score.
174 children born 1994-1995
followed from age 6 mo to
age 6 yr, Rochester, NY.
Prospective. Same cohort as
above. High follow-up
participation. Linear
regression model adjusted for
sex, birth weight, transferrin
saturation, maternal race, IQ,
education, and prenatal
smoking status, HOME score
(6 yr), family income.
1 ,333 children pooled from
Boston, Cincinnati,
Cleveland, Mexico City,
Port Pirie, Rochester, and
Yugoslavia cohorts.
Prospective. Large, uniform
analysis pooling diverse
cohorts. Included 84% of
eligible children. Linear
regression model adjusted for
HOME score, birth weight,
maternal IQ and education.
Considered child sex, marital
status, birth order, prenatal
alcohol consumption and
smoking, maternal age.
Blood Pb
Timing and
Levels
(ug/dL)
Earlier
childhood
(age 2 yr)
Mean (SD)
All subjects:
6.5 (4.9)
Peak <10:
3.8
(range: 1-9.3)





Concurrent
Mean (SD)
All subjects:
5.8 (4.1)

Peak<10:
3.3
(range:
0.5-8.4)

IV/I i n i m 1 1 m
IVIII III 1 IUI 1 1
below limit of
e ec ion
Peak
(6 mo-6 yr)
Mean (SD):
11.4(7.3)








Concurrent
Mean (95th)
Peak> 10:
13.9(35.4)
Peak<10:
4.3 (8.0)
Peak > 7.5:
12.9(34)
Pesk ^7 5'
3.2(6.0)'






Outcome
FSIQ
WISC-R
Age 1 0 yr










FSIQ
Stanford-
Binet
Age 5 yr









FSIQ
WPPSI-R
Age 6 yr








FSIQ
Various
tests
Ages
4.8-1 Oyr










Blood Pb stratum
(ug/dL)
All 148 subjects
48 subjects peak <10










All 172 subjects
101 subjects peak <10









Peak 20-30, n not given
Peak 10-20, n not given
96 subjects peak <1 0








1 ,089 subjects peak > 10
244 subjects peak <10
1 ,203 subjects peak > 7.5
1 03 subjects peak <7.5










Effect Estimate
/OCO/ ^l\^
(95% Cl)
-0.58 (-1 .0, -0.2)b
-1.56 (-2.9, -0.2)b










-0.61 (-0.99, -0.24)
-1.79 (-3.00, -0.60)









-0.15°
-0.32°
-1.2°








-0.1 3 (-2.3, -0.03)
-0.80 (-1.74, 0.14)
-0.1 6 (-0.24, -0.08)
-2.94 (-5. 16, -0.71)









November 2012
5-241
Draft - Do Not Cite or Quote

-------


Study
Tellez-Rojo
et al. (2006)
Tellez-Rojo
(2008)








Lanphear et
al. (2000)








Kordas et al.
(2006)















Study Population and
Methodological Details
384 children followed from
birth (1994-1995, 1997-1999)
to age 2 yr, Mexico City,
Mexico.
Cross-sectional.
Recruitment from prenatal
clinic or birth hospital.
Participants had higher
maternal education, lower
blood Pb level. Linear
regression model adjusted for
sex, birth weight, maternal IQ.
Considered maternal age and
other unspecified factors.
4,853 children ages 6-16 yr,
NHANES 1988-1994.
Cross-sectional. Large study
of multiple exposures and
outcomes. Linear regression
model adjusted for sex,
race/ethnicity, poverty index
ratio, reference adult
education level, serum ferritin
level, serum cotinine level.
Did not consider parental IQ
or caregiving quality.
532 children in 1st grade,
Torreon, Mexico
Cross-sectional.
Recruitment at prenatal clinic.
Residence near metal
foundry. High participation.
Linear regression model
adjusted for sex, age,
hemoglobin, family
possessions, forgetting
homework, house ownership,
crowding, maternal
education, birth order, family
structure, urine As, tester,
school. Did not consider
parental IQ or caregiving
quality.
Blood Pb
Timing and
Levels
(ug/dL)
Concurrent
Mean (SD)
> 10:
MR
Nrx
<10'
4.28 (2.25)
5-10:
6.9 (1 .4)
<5:
2.9(1.1)



Concurrent
Overall
II /!__._ /Q CV
Mean (oh).
1.9(0.1)
Subgroups:
NR






Concurrent
Overall
Mean (SD):
11.4(6.1)
Subgroups:
NR













Outcome
Bayley
MDI
Age 2 yr









Reading
Score
WRAT
Ages 6-1 6
vr
/ '






Math
score
Achieve-
ment test
1 st grade













Blood Pb stratum
(ug/dL)
90 subjects > 10
294 subjects <10
1 01 subjects 5-1 0
1 93 subjects <5








All 4,853 subjects
4681 subjects <10
4,526 subjects <7.5
4,043 subjects <5
2,467 subjects <2.5






All 532 subjects
293 subjects <10















Effect Estimate
/OCO/ ^l\^
(95% Cl)
0.07 (-10,9.2)"
-1.04 (-1.8, -0.30)b
-0.94 (-2.1, 0.2)b
-1.71 (-3.0, -0.42)b








-0.70 (-1 .03, -0.37)
-0.89 (-1 .52, -0.26)
-1 .06 (-1 .82, -0.30)
-1.06 (-2.00, -0.12)
-1 .28 (-3.20, -0.64)






-0.1 7 (-0.28, -0.06)
-0.42 (-0.92, 0.08)














     FSIQ = Full-scale IQ, WISC-R = Wechsler Intelligence Scale for Children-Revised, WPPSI-R = Wechsler Preschool and Primary
     Scale of Intelligence-Revised, MDI = Mental Developmental Index, NR = Not reported, WRAT = Wide Range Achievement Test.
     "Effect estimates are derived from linear models and are presented for a 1 ug/dL increase in blood Pb level.
     b95% CIs calculated from reported p-value.
     °Results not included in Figure 5-16 because nonparametric analysis did not produce 95% CIs for various strata of blood Pb levels.
1
2
3
4
5
6
7
A few cross-sectional studies demonstrated larger Pb-associated decreases in cognitive
function with concurrent blood Pb levels <5 (ig/dL. Tellez-Rojo et al. (2006) estimated a
larger decrement in age 2 year Bayley MDI per unit increase in blood Pb level for
children with concurrent blood Pb levels <5 (ig/dL compared with children with blood Pb
levels 5-10 (ig/dL, and >10 (ig/dL (Figure 5-16 and Table 5-16). However, it is not clear
what the implications of age 2 year MDI results may be on cognitive function at later
ages. Among children ages 5-16 years participating in NHANES 1989-1994, Lanphear et
al. (2000) found larger decrements in reading and math skills and memory per unit
increase in blood Pb level in children with concurrent blood Pb levels <2.5 (ig/dL than
     November 2012
                             5-242
Draft - Do Not Cite or Quote

-------
 1                   children with levels <5 (ig/dL, <7.5 (ig/dL, <10 (ig/dL, and all subjects. However, higher
 2                   Pb exposures earlier in childhood may have contributed to associations.

 3                   Several (Min et al.. 2009: Jusko et  al.. 2008: Schnaas et al.. 2006) but not all
 4                   (Palaniappan et al., 2011) recent studies found a nonlinear blood Pb-cognitive function
 5                   relationship in nonparametric regression analyses using splines or lowess with smoothing
 6                   parameters that did not produce quantitative results for each blood Pb group. Similar to
 7                   the pooled analyses of the seven prospective cohorts, these relationship were evaluated
 8                   for a wide range of blood Pb levels. In the Rochester and Mexico City cohorts, the blood
 9                   Pb-FSIQ relationship was more negative for children with lower blood Pb levels,
10                   specifically for peak blood Pb levels <10 (ig/dL (range: 2.1-45.7 (ig/dL) in the Rochester
11                   cohort at age 6 years and (Jusko  et al.. 2008) and for prenatal maternal week 28-36 blood
12                   Pb levels <6(ig/dL (5th-95th percentile 2.5-24.6 (ig/dL) in the Mexico City cohort at ages
13                   6-10 years (Schnaas  et al.. 2006). In a formal test of nonlinearity, Schnaas et al. (2006)
14                   found the nonlinear blood Pb term  to fit the data better than a linear term. Among 267
15                   children ages 4 years (blood Pb range: 1.3-23.8 (ig/dL) who had high prenatal alcohol
16                   and drug exposure, Min et al. (2009) reported a p-value of 0.19 for a restricted cubic
17                   spline term for blood Pb level and described the covariate-adjusted concurrent blood Pb
18                   level-FSIQ curve to be more negative at blood Pb levels <7 (ig/dL. Among 814 children
19                   in India ages 3-7 years, Palaniappan et al. (2011) mostly found linear associations
20                   between concurrent blood Pb level (range: 2.6-40.5 (ig/dL) and indices of cognitive
21                   function. The exception was visual-motor skills, for which a greater blood Pb-associated
22                   decline was found with blood Pb levels >30 (ig/dL. The linearity versus nonlinearity of
23                   the blood Pb-FSIQ concentration-response relationship within a lower, more narrow
24                   range of blood  Pb levels has not been examined in detail.

25                   Few studies of adults have examined whether the relationship between blood or bone Pb
26                   level and cognitive function is described better with a linear or nonlinear function. In
27                   analyses of adults in NHANES, only log-linear models were used to fit the data (Krieg et
28                   al..201Q: Krieg and Butler. 2009: Krieg et al.. 2009). Nonlinearity in the BMS and NAS
29                   cohorts was examined with the use of quadratic terms, penalized splines, or visual
30                   inspection of bivariate plots (Bandeen-Roche  et al.. 2009: Weisskopf et al.. 2007b: Shih
31                   et al.. 2006). There was some evidence for nonlinearity in prospective analyses of the
32                   NAS cohort (Figure 5-8 and Figure 5-9), but not all results indicated greater declines in
33                   cognitive function per unit increase in bone Pb level in the lower bone Pb groups. Wang
34                   et al. (2007a) found that among NAS men with an HFE variant, there was a larger decline
35                   in MMSE score per unit increase in tibia Pb level at higher tibia Pb levels, 20-25 ug/g
36                   (Figure 5-9). In the BMS cohort, linear relationships were indicated for various tests of
37                   cognitive function by a statistically nonsignificant quadratic term (Shih et al.. 2006) or
38                   spline (Bandeen-Roche et al.. 2009) for tibia Pb level.
      November 2012                             5-243                  Draft - Do Not Cite or Quote

-------
 1                  Although not specific to Pb exposure, attenuation of concentration-response relationships
 2                  at higher exposure or dose levels has been reported in the occupational literature, and
 3                  explanations have included greater exposure measurement error and saturation of
 4                  biological mechanisms at higher levels and larger proportions of at-risk populations at
 5                  lower exposure levels (Stavner et al.. 2003). Hypotheses for nonlinearity in the
 6                  relationship between Pb and cognitive function have included a lower incremental effect
 7                  of Pb due to covarying  risk factors such as low SES, poor caregiving environment, higher
 8                  exposure to other environmental factors (Schwartz.  1994). different mechanisms
 9                  operating at different exposure levels, and confounding by omitted or misspecified
10                  variables. The contribution of these factors to the supralinear relationship between blood
11                  Pb levels and cognitive function in children has not  been examined in many
12                  epidemiologic studies to date. Several studies found that risk factors such as SES,
13                  parental education, and parental caregiving quality explain a greater proportion of
14                  variance in cognitive function than does blood Pb level (Wasserman et al.. 1997; Greene
15                  et al.. 1992). Recently,  among 57,678 fourth grade children across North Carolina,
16                  Miranda et al. (2009) found that lower parental education and enrollment in a
17                  free/reduced fee lunch program accounted for larger decrements in EOG scores than did
18                  blood Pb level across the various quantiles of EOG score distribution (Figure 5-7).

19                  Few studies have examined effect modification of the blood Pb level-cognitive function
20                  relationship by covarying risk factors such as sociodemographic factors, and the limited
21                  evidence is inconclusive. None of these studies examined effect modification within
22                  specific strata of blood Pb levels. In the Boston cohort at age 57 months, a greater
23                  Pb-associated FSIQ decrement was reported in the group that was female and had higher
24                  HOME score, SES, and maternal IQ (Bellinger et al.. 1990). However, the Boston cohort
25                  overall  had higher SES and parental education, and the group that included higher SES
26                  may not be comparable to other cohorts. In the Port  Pirie, Australia cohort, larger blood
27                  Pb-associated FSIQ decrements were found in groups with lower SES (Tong et al., 2000)
28                  but not  lower HOME score (McMichael et al.. 1992). Overall, evidence does not clearly
29                  indicate whether the blood Pb-IQ relationship is modified by factors such as SES or other
30                  sociodemographic characteristics or whether these differences can explain the observed
31                  nonlinear concentration-response relationship.

32                  Results from the pooled analysis by Rothenberg and Rothenberg (2005) do not indicate
33                  that residual confounding by covariates explains the nonlinear blood Pb-FSIQ
34                  relationship. Modeling  maternal IQ, HOME score, and maternal education as spline
35                  functions (df = 2) did not significantly improve model fit either with a linear blood Pb
36                  term or log blood Pb term, which indicated that the improved model fit with log-
37                  specification of blood Pb level was not influenced by the modeling of covariates as linear
38                  or nonlinear functions.
      November 2012                             5-244                  Draft - Do Not Cite or Quote

-------
 1                  Bowers and Beck (2006) postulated that a supralinear concentration-response function
 2                  necessarily will be found in a model with a log-normally distributed independent variable
 3                  and a normally distributed outcome variable. However, as discussed in the
 4                  2006 Pb AQCD, this modeling strategy was not employed in the epidemiologic analyses
 5                  showing a supralinear concentration-response function. FSIQ scores generally were not
 6                  forced into a normal distribution. Normalized FSIQ scores were not the basis for
 7                  individual findings from four of the seven studies included in the pooled analysis by
 8                  Lanphear et al. (2005) or the results pooling the seven cohorts (Hornung et al. 2006).
 9                  Further, a log-linear model (a linear relationship between IQ and the log of blood Pb)
10                  provided a better fit of the pooled data.

11                  Results from prospective analyses in the Boston and Rochester cohorts for associations
12                  between blood Pb level as a continuous variable and FSIQ in groups of children in the
13                  lower segment of the population blood Pb distribution have not identified a threshold in
14                  the range of blood Pb levels examined. In the Boston cohort, higher age 2 year blood Pb
15                  levels were associated with lower FSIQ at age 10 years in children with blood Pb levels
16                  1-9.3 (ig/dL whose peak blood Pb levels never exceeded 10 (ig/dL (Bellinger. 2008;
17                  Bellinger and Needleman. 2003). Schwartz (1994) explicitly assessed evidence for a
18                  threshold in the Boston cohort data by regressing FSIQ and blood Pb level on age,  race,
19                  maternal IQ, SES, and HOME score and fitting  a nonparametric smoothed curve to the
20                  residuals of each regression model (variation in FSIQ or blood Pb level not explained by
21                  covariates). A 7-point decrease in FSIQ was found over the range of blood Pb residuals
22                  below 0 (corresponding to the mean blood Pb level of 6.5 (ig/dL), indicating an
23                  association between blood Pb level and FSIQ down to a blood Pb level of 1 (ig/dL. In the
24                  Rochester cohort, higher peak blood Pb levels were associated with lower FSIQ at  ages 3
25                  and 5 years in children with peak blood Pb levels < 10 (ig/dL (Canfield et al., 2003a). A
26                  threshold also was not identified for the association between concurrent blood Pb level
27                  and MDI score at age 2 years among children in Mexico City with blood Pb levels  in the
28                  range of 0.8-9.8  (ig/dL (Tellez-Roio. 2008: Tellez-Roio et al.. 2006).

29                  In conjunction with downward trends in population blood Pb distributions (Figure 4-16).
30                  more sensitive quantification methods have improved the detection limits for blood Pb
31                  measurements (e.g., in NHANES, from 0.6 (ig/dL in 1999-2002 to 0.025 (ig/dL in
32                  2003-2004). Consequently, the examination of groups of children (ages 8-11 years) with
33                  lower blood Pb levels, overall range <1 to 16  (ig/dL, has indicated Pb-associated
34                  cognitive function decrements or increases in attention-related behavioral problems, at
35                  lower blood Pb levels  (Cho et al.. 2010: Kim et  al.. 2009b: Miranda etal.. 2009: 2007a).
36                  In the studies examining concurrent blood Pb levels, the potential contribution of higher
37                  past Pb exposures obscures assessment of a threshold. However, Miranda et al. (2009)
38                  examined blood  Pb levels measured between ages 6 and 36 months during 1995-1999
      November 2012                            5-245                 Draft - Do Not Cite or Quote

-------
 1                   and found lower 9th grade EOG scores in 57,678 children in North Carolina with early
 2                   childhood blood Pb levels of 2 (ig/dL compared with children with blood Pb levels of
 3                   1 (ig/dL. Other studies did not identify a threshold for Pb-associated cognitive function
 4                   decrements in children using nonparametric regression analyses, but results have weaker
 5                   implications because the blood Pb levels in the examined populations of children were
 6                   higher than those in the current U.S. population (e.g., minimum 2.1, 5th percentiles 2.5
 7                   and 4.0 (ig/dL) (Jusko et al.. 2008; Schnaas et al.. 2006; Lanphear et al.. 2005).

 8                   Analyses of blood Pb level as a categorical variable did not clearly address the
 9                   identification of a threshold for the blood Pb-cognitive function relationship; however,
10                   such analyses are not as sensitive as those of blood Pb level as a continuous variable. In
11                   the analysis of large numbers (>600) of children participating in NHANES with blood Pb
12                   levels <1 (ig/dL, Braun et al. (2008; 2006) found higher odds of parental reports of
13                   conduct disorder and ADHD among children ages 4-15 years with concurrent blood Pb
14                   levels ~1.0 (ig/dL compared with children with blood Pb levels <0.8 (ig/dL. However,
15                   higher past Pb exposures may have contributed to associations found with concurrent
16                   blood Pb levels because of the older age of some subjects and the birth of study
17                   adolescents in the 1970s during the use of leaded gasoline. Other analyses of blood Pb
18                   level categories indicated that cognitive function decrements were limited to children
19                   ages 7-8 years with age 30-month blood Pb levels >10  (ig/dL (Chandramouli et al., 2009)
20                   or children ages 6-10  years with concurrent blood Pb levels 5-10 (ig/dL (Surkan et al..
21                   2007).

22                   Some toxicological studies found nonlinear relationships between Pb exposure and
23                   effects related to impaired learning and memory in animals. These results are distinct
24                   from epidemiologic results as toxicological studies often show that lower and higher Pb
25                   exposures have effects in opposite directions (U- or inverse U-shaped curves). Results
26                   summarized across multiple studies in multiple species demonstrated that lower Pb
27                   exposures increased FI response rates relative to controls, and higher Pb exposures
28                   decreased FI response rates (Cory-Slechta. 1994). Increased FI response rates indicate
29                   impaired learning by reflecting the impaired ability of animals to respond according to a
30                   fixed schedule of reinforcement (Section 5.3.3.1). Consistent with previous findings,
31                   Rossi-George et al. (2011) found that 50 ppm gestational plus lactational Pb exposure
32                   when combined with stress increased FI responses of 2-month old rats whereas 150 ppm
33                   Pb exposure with stress did not affect FI responses. Nonlinear effects of Pb on learning
34                   are less consistently observed with longer duration exposures (e.g., 8-11 months) (Rossi-
35                   George et al.. 2011; Cory-Slechta. 1990). These nonlinear effects of Pb on  impaired
36                   learning were supported by evidence in animals indicating that lower and higher Pb
37                   exposures differentially activate underlying mechanisms. Gilbert et al. (1999) found
38                   reduced LTP in adult rats exposed to 1,000 and  5,000 ppm but not 10,000 ppm Pb acetate
      November 2012                             5-246                 Draft - Do Not Cite or Quote

-------
 1                  in drinking water (from GDI6). LTP is one indication of synaptic plasticity
 2                  (Section 5.3.11.4). which is considered to contribute to learning and memory. Learning
 3                  and memory have also been affected by glutaminergic neurotransmission via its NMDA
 4                  receptor (Section 5.3.11.8). and reduced glutamate release in the hippocampus was found
 5                  in adult rats exposed to Pb acetate from GDI5-GDI6 with blood Pb levels 27-40 (ig/dL
 6                  but not with blood Pb levels of 62-117 (ig/dL (Laslev and Gilbert. 2002).

 7                  Dopaminergic neurotransmission is involved in many CNS processes including
 8                  cognition, behavior, and motor function. The shape of the Pb-DA concentration-response
 9                  relationship varied among toxicological studies. Some studies found that lower Pb
10                  exposures (~50 ppm) did not affect or increased DA activity relative to controls and
11                  higher Pb exposure (109-250 ppm) (Leasure et al.. 2008; Virgolini et al.. 2005; Lewis and
12                  Pitts. 2004). Others found higher Pb exposures (109 or 150 ppm) to increase or impair
13                  DA activity (Leasure et al.. 2008; Virgolini et al.. 2005). These differential responses of
14                  DA may be related to the diverse CNS effects of DA in different regions of the brain. For
15                  example, the increased forebrain dopamine turnover with 27 ppm gestational/lactational
16                  Pb acetate exposure was accompanied  by less spontaneous activity in male mice
17                  compared with male  mice exposed to 109 ppm Pb (Leasure et al.. 2008).

18                  In vitro results indicated differential effects on calcineurin enzyme activity, with inhibited
19                  activity resulting from higher Pb exposure (>2 x 10"4 (iM) and stimulated activity from
20                  lower Pb exposure (Kern and Audesirk. 2000). While calcineurin activity has been found
21                  to modulate learning, LTP, and behavior in animals, studies have found lower calcineurin
22                  activity to be associated with both impaired and improved effects related to learning
23                  (Zeng et al.. 2001). Thus, it is uncertain whether altered calcineurin activity contributes to
24                  the nonlinear relationships observed between Pb exposure and impaired learning in
25                  animals. At lower concentrations, Pb may displace calcium at its binding sites on
26                  calmodulin and by acting as a calmodulin agonist at the catalytic A subunit of calcineurin
27                  and stimulate calcineurin activity. At higher Pb exposure, Pb may bind directly to a
28                  separate calcium-binding B subunit, override the calmodulin-dependent effect and turn
29                  off the activity of calcineurin. Lasley and Gilbert  (2002) found that 2,000 ppm but not
30                  5,000 or 10,000 ppm Pb acetate exposure of rats (in drinking water starting at
31                  GD15-GD16) inhibited glutamate release by acting as a calcium mimetic.

32                  Some toxicological studies have found nonlinear relationships for non-cognitive
33                  outcomes in animals. U-shaped Pb concentration-response relationships were found for
34                  spontaneous motor activity level and latency to fall from rotarod (Leasure et al.. 2008).
35                  Inverted U-shaped relationships were found for hippocampal neurogenesis (Fox et  al..
36                  2008; Gilbert et al.. 2005). Evidence also points to differences in hormone production by
37                  Pb exposure concentration. In male mice with long-term Pb exposure (PND21-9 months
      November 2012                             5-247                  Draft - Do Not Cite or Quote

-------
 1                  of age), basal corticosterone levels were significantly lower with 50 ppm Pb than with
 2                  150 ppm Pb or controls (Cory-Slechta et al.. 2010). Visual system effects in animals also
 3                  have shown to be affected differentially by lower versus higher Pb exposure (GD1-
 4                  PND10, pup blood Pb levels 12, 24, and 46 (ig/dL). Inverted U-shaped concentration-
 5                  response curves were observed for rod photoreceptor numbers or neurogenesis
 6                  (Giddabasappa et al.. 2011) and retinal thickness (Fox et al.. 2010). These dichotomous
 7                  histological findings may give insight to the complex Pb-associated changes in ERG
 8                  wave amplitudes that vary by exposure window and dose (Section  5.3.7.3).

 9                  To conclude, several studies found a supralinear blood Pb-cognitive function
10                  concentration-response relationship in children but not adults based on comparisons of
11                  effect estimates in lower and higher strata of blood Pb level and nonparametric
12                  regression. Explanations for this supralinear relationship have not been well characterized
13                  by epidemiologic studies. Evidence from the prospective studies in the Boston and
14                  Rochester cohorts has not identified a threshold for Pb-associated cognitive function
15                  decrements in the range of blood Pb levels examined. Increases in childhood blood  Pb
16                  levels in the range of <1.0-9.8 ug/dL (means: 2.9 and 3.8  ug/dL) were associated with
17                  cognitive function decrements at ages 3  to 10 years in children whose peak blood Pb
18                  levels did not exceed 10 ug/dL (Bellinger. 2008: Canfield. 2008: Bellinger and
19                  Needleman. 2003: Canfield et al.. 2003a). Further, a recent study found an association
20                  between higher ages 6-36 month blood Pb levels (1995-1999) and lower 4th grade EOG
21                  scores in 57,678 children in North Carolina with blood Pb levels 1-16 ug/dL (Miranda et
22                  al.. 2009). Concurrent blood Pb levels in the range of 0.8-9.8 ug/dL were associated with
23                  MDI decrements in children age 2 years in Mexico City (Tellez-Rojo, 2008: Tellez-Rojo
24                  et al.. 2006). The  lack of a reference population with blood Pb levels reflecting pre-
25                  industrial Pb exposures limits the ability to identify a threshold. Analysis of ancient bones
26                  in pre-industrialized societies suggests that "background" blood Pb levels in preindustrial
27                  humans was approximately 0.016 (ig/dL (Flegal and Smith, 1992). approximately 65-fold
28                  lower than that of the current U.S. population and lower than the levels at which
29                  neurodevelopmental effects have been examined. Thus, the current evidence does not
30                  preclude the possibility of a threshold for neurodevelopmental effects in children existing
31                  with lower blood  levels than those currently examined. While distinct from  supralinear
32                  relationships observed in epidemiologic studies, toxicological studies showed that lower
33                  Pb  exposures (e.g., 50 ppm in drinking water) induced learning and memory impairments
34                  in animals compared to control exposures or higher Pb exposures (e.g., 150 ppm).
35                  Additional toxicological evidence suggests that differentially activated mechanisms at
36                  lower and higher Pb exposures and  reduced LTP and hippocampal glutamate release with
37                  lower Pb exposures may provide explanation for impaired learning observed with lower
38                  Pb  exposures in animals.
      November 2012                            5-248                  Draft - Do Not Cite or Quote

-------
      5.3.14 Confounding in Epidemiologic Studies of Nervous System Effects

 1                  In addition to Pb exposure, many factors influence cognitive function and behavior in
 2                  children, including parental IQ and education, SES of the family, quality of the
 3                  caregiving environment, and other environmental exposures (Wasserman and Factor-
 4                  Litvak. 2001). These other risk factors often are correlated with blood, tooth, and bone Pb
 5                  levels, thus, a major challenge to observational studies examining associations of Pb
 6                  biomarker levels with cognitive function and behaviors in children has been the
 7                  assessment and control for potential confounding factors. By definition, a confounder is
 8                  associated with both the independent variable and the outcome and consequently has the
 9                  potential to bias the association between the independent variable of interest and the
10                  outcome. Most epidemiologic studies of Pb biomarkers in children have examined
11                  potential confounding by parental IQ and SES-related variables such as parental
12                  education, household income, and the Hollingshead Four-Factor Index of Social Position,
13                  which incorporates education and income of both parents. Fewer but still several studies
14                  have examined confounding by quality of the caregiving environment (i.e., HOME
15                  score), birth weight, and smoking exposure. A relatively smaller number of studies have
16                  considered nutritional status, other environmental exposures, parental substance abuse, or
17                  parental psychopathology. Studies have varied with respect to the number of potential
18                  confounding factors examined, with some studies considering multiple SES-related
19                  variables and other studies focusing on a smaller set. The extent of confounding by a
20                  particular factor likely varies across studies, depending on the population examined.
21                  Thus, the impact of adjustment for specific covariates on the Pb effect estimate also
22                  likely varies across studies.

23                  Various methods have been used to control for potential confounding, including
24                  examining a population relatively homogeneous in SES, examining populations in which
25                  factors are not correlated, conducting multivariate regression, characterizing the change
26                  in the blood Pb level effect estimate with adjustment for a covariate, and examining
27                  associations in different strata of a covariate. The evidence derived from each of these
28                  control strategies is discussed below. No single method is without limitation and
29                  adjustment for SES is difficult as it is  highly correlated with Pb exposure and there is no
30                  single measure that represents SES. Residual confounding also is likely by factors  not
31                  considered. The combination of evidence from prospective studies that considered
32                  several well-characterized potential confounding factors plus evidence that Pb exposure
33                  induces impairments in cognitive function in animals, in particular, visual-spatial
34                  memory and executive  function, which are also found to be affected in children, increase
35                  confidence that the associations observed between blood Pb and tooth Pb levels and
36                  cognitive function in children represent a relationship with Pb exposure.
      November 2012                            5-249                  Draft - Do Not Cite or Quote

-------
 1                  In the Boston prospective study, potential confounding by SES was controlled for by
 2                  study design and statistical adjustment for SES. The study subjects were from middle- to
 3                  upper-middle-class families, a majority with married, college-educated parents. Hence,
 4                  the potential for confounding by SES in this study was considerably less compared to
 5                  other studies examining similar outcomes. In this cohort, higher prenatal and concurrent
 6                  blood Pb levels were associated with FSIQ decrements at age 57 months, and higher age
 7                  2 year blood Pb levels were associated decrements in FSIQ and executive function at age
 8                  10 years (Stiles and Bellinger. 1993; Bellinger et al.. 1992; Bellinger et al. 1990). In
 9                  contrast, blood Pb levels were weakly associated with cognitive function decrements in
10                  the Sydney, Australia cohort of middle-SES children (i.e., 20% mothers with greater than
11                  high school education) (Cooney et al.. 1991). However, a relationship between Pb
12                  exposure and cognitive function decrements is supported by a similar magnitude of blood
13                  Pb-associated FSIQ decrement found in the Boston and low-SES Rochester cohort
14                  (majority of mothers with less than college education and annual income <$15,000), with
15                  adjustment for similar covariates (Figure 5-2 and Table 5-3) (Canfield et al.. 2003a). For
16                  some outcomes, larger effects were estimated in the Boston cohort than in other cohorts.

17                  Blood Pb levels also were associated with cognitive function decrements in populations
18                  in which blood Pb levels were not correlated with SES-related factors (Factor-Litvak et
19                  al.. 1999; Bellinger et al.. 1987). In the Yugoslavia cohort, blood Pb levels  at age 4 years
20                  were higher in groups with higher maternal education, maternal IQ, and HOME score in
21                  one city near Pb sources and were lower in the distant city. Among all children, higher
22                  blood Pb level was associated with  lower FSIQ and learning and memory scores and with
23                  higher ratings of internalizing behaviors (Factor-Litvak et al.. 1999). In the Boston
24                  cohort,  parental education, social class, and HOME score were similar among low
25                  (<3 (ig/dL), medium (6-7 (ig/dL), and high (> 10 (ig/dL) cord blood Pb level groups.
26                  Further, adjusting for these and other demographic variables, Bellinger et al. (1987)
27                  found that children in the high cord blood Pb group had a 4.8-point lower Bayley MDI
28                  score at age 2 years than did children in the low cord blood Pb group.

29                  The primary method used by epidemiologic studies to control for potential  confounding,
30                  in particular recent studies of children with blood Pb levels more similar to current U.S.
31                  levels, has been multivariate regression. Some studies modeled a set of covariates based
32                  on a priori evidence, whereas others selected specific covariates based on their
33                  association with the outcome in a model with all potential covariates and/or a greater than
34                  10% change in the blood Pb  level effect estimate. Studies also varied in the number of
35                  potential confounding factors included in models. Some included multiple SES-related
36                  variables, whereas others analyzed one or two factors. Regardless of the method  used to
37                  select model covariates or the number of covariates included, studies consistently found
38                  associations of higher blood  Pb level with cognitive function decrements and behavioral
      November 2012                             5-250                  Draft - Do Not Cite or Quote

-------
 1                   problems. The evidence suggests that confounding by particular factors may vary across
 2                   populations and increases confidence that the associations observed with
 3                   neurodevelopmental effects in children represent a relationship with Pb exposure

 4                   The consistency of association across populations with different SES and co-exposures
 5                   and across studies examining different covariates was reinforced in pooled and meta-
 6                   analyses (Marcus etal. 2010; Lanphear et al., 2005; Schwartz. 1994). Pooling data from
 7                   seven international prospective cohorts, Lanphear et al. (2005) found similar FSIQ
 8                   decrements per log increase in blood Pb level (-2.6 to +8.6% difference) by excluding
 9                   one study at a time. These results indicated a relatively robust pooled estimate despite
10                   between-study differences in population characteristics, including SES. In a meta-
11                   analysis, Schwartz (1994) found a relatively narrow range of blood Pb-FSIQ effect
12                   estimates among studies despite  large between-study differences in the correlation
13                   between blood Pb level and SES. A wider range of effect estimates would be expected if
14                   omitted SES factors confounded the association. A recent meta-analysis of the
15                   association between blood Pb level  and conduct problems in earlier and recent studies of
16                   children (Marcus etal. 2010) found that adjustment for SES and HOME score did little
17                   to attenuate the association.

18                   Among the several studies that provided both unadjusted and adjusted effect estimates,
19                   most indicated that blood Pb level was a statistically significant predictor of cognitive
20                   function (e.g., FSIQ, executive function, learning, memory) in children ages 5-10 years
21                   before and after adjusting for potential confounders. Although most effect estimates
22                   changed by 20-50% in multivariate  models, they  remained within the 95% CI of the
23                   unadjusted estimate (Min et al.. 2009; Kordas et al.. 2006; Schnaas et al.. 2006; Canfield
24                   et al.. 2003a; Dietrich et al.. 1993a;  Bellinger etal., 1992).  Such observations were made
25                   in previous analyses of the Boston and Rochester cohort with mean blood Pb levels 6.5
26                   and 5.8 (ig/dL, respectively, with adjustment for SES, maternal IQ and education, and
27                   HOME score (Canfield et al.. 2003a; Bellinger et al.. 1992). These analyses also adjusted
28                   for or considered potential confounding by nutritional factors. Recent studies  of children
29                   of a similar age range and mean  blood Pb levels also found statistically significant
30                   associations (as indicated by correlation and/or regression coefficients) between blood Pb
31                   level and cognitive function before and after adjustment for similar covariates; however,
32                   these populations had high prevalence of prenatal alcohol or drug use which may limit
33                   the representativeness of their results (Min et al..  2009; Chiodo et al.. 2007).

34                   Blood Pb level also was a statistically significant predictor of cognitive function after
35                   adjustment for covariates such as maternal education and IQ, SES, and HOME score  in
36                   with children with higher mean blood Pb levels, 8-14 (ig/dL (Kordas et al.. 2006; Schnaas
37                   et al.. 2006; Tong and Lu. 2001; Dietrich et al.. 1993b). Exceptions include  multiple
      November 2012                             5-251                  Draft - Do Not Cite or Quote

-------
 1                   analyses of the Cleveland cohort, in which blood Pb level was estimated to have a weak
 2                   and imprecise or null effect after adjustment for potential confounding factors (Greene et
 3                   al.. 1992; Ernhart et al.. 1989; Ernhart et al.. 1988). Analyses of the Cleveland cohort
 4                   considered similar potential confounding factors as other studies, with the exception of
 5                   Greene et al. (1992) who also adjusted for pica and home conditions. However, these
 6                   latter variables each accounted for only a small partial correlation with FSIQ. HOME
 7                   score was the major factor accounting for the attenuation of the effect of Pb in the
 8                   Cleveland cohort. An analysis of the Yugoslavia cohort, which adjusted for most of the
 9                   same covariates as several of the Cleveland analyses reported larger magnitude blood
10                   Pb-cognitive function associations in covariate-adjusted models (Factor-Litvak et al..
11                   1999). The collective findings in children indicate potential confounding by the SES-
12                   related and demographic factors examined in the literature base but also demonstrate that
13                   blood Pb level is an independent predictor of cognitive function decrements in children
14                   with adjustment for these factors.

15                   A challenge to  separating the effects of Pb  exposure from those related to  SES and
16                   parental caregiving quality is their frequently high correlation with blood Pb levels. In
17                   such cases, it is difficult to know how much variation in the outcome to attribute to each
18                   of the risk factors (Needleman and Bellinger. 2001). For example, due to the high
19                   correlation between blood Pb level and SES, a model that includes SES may
20                   underestimate the Pb effect because some of the variance in outcome due to Pb is
21                   mistakenly attributed to the variance due to SES. This misattribution may  be exacerbated
22                   when multiple correlated variables are included in the same model (i.e., overcontrol). The
23                   relationships observed  for Pb biomarker levels with SES and parental caregiving quality
24                   may indicate that they are proxies or determinants of Pb exposure rather than a
25                   confounder of the association of interest. Lower SES in urban children is closely linked
26                   to residence in  older, poorer condition housing that, in turn, may increase exposure of
27                   children to environmental Pb and risk of cognitive deficits (Clark et al.. 1985). In such
28                   cases where Pb exposure is a mediator of the SES effect, statistical adjustment for SES
29                   will result in overcontrol of the Pb effect (Bellinger. 2004a). This type of overcontrol
30                   could explain results from the New Zealand cohort, which were adjusted for residence in
31                   older wooden housing, which is associated with higher exposure to Pb paint and
32                   accumulated dust and soil and higher child tooth Pb levels (Fergusson et al.. 1988a. b).
33                   However, even in models with older wooden housing, Pb remained a statistically
34                   significant predictor of poorer reading skills and teacher ratings of school performance.
35                   SES has been shown to be an effect modifier of the Pb-child cognitive function
36                   relationship.  Larger blood Pb-associated decreases in cognitive function were found with
37                   lower SES in some studies (Ris et al.. 2004; Tong et al.. 2000; Bellinger et al.. 1990) and
38                   higher SES in a meta-analysis (Schwartz. 1994). In cases of effect modification, potential
39                   confounding by SES is less likely.

      November 2012                             5-252                  Draft - Do Not Cite or Quote

-------
 1                  In summary, the collective epidemiologic evidence consistently demonstrates
 2                  associations of higher blood and tooth Pb levels with cognitive function decrements and
 3                  behavioral problems in children. These associations have been observed in diverse
 4                  populations in the U.S., Mexico, Europe, Asia, and Australia. Associations have been
 5                  observed across studies that use different methods to control for confounding and adjust
 6                  for different potential confounding factors but commonly, maternal IQ and education,
 7                  SES, and HOME score. Several studies have found associations with additional
 8                  adjustment for smoking exposure, child birth outcomes, and nutritional factors. No single
 9                  method to control for potential confounding is without limitation, and there is potential
10                  for residual confounding by unmeasured factors. However, the consistency of findings
11                  among different populations and study methods with consideration of several well-
12                  characterized potential confounding factors as described above increases confidence that
13                  the associations observed between blood Pb level and neurodevelopmental effects in
14                  children represent a relationship with Pb exposure. Biological plausibility is provided by
15                  the coherence with extensive evidence in animals with Pb exposures that produce blood
16                  Pb levels relevant to humans and that is not subject to confounding by factors such as
17                  social class and correlated environmental factors. Further, Pb exposure has been shown to
18                  induce impairments in visual-spatial memory, rule learning and reversal, and response
19                  inhibition, which also have been associated with blood or tooth Pb levels in children.
20                  Additional support for the epidemiologic evidence is provided by extensive toxicological
21                  evidence describing modes of action for Pb-induced cognition and behavioral problems,
22                  including  changes in neurogenesis, synaptic pruning, and neurotransmitter function in the
23                  hippocampus, prefrontal cortex, and nucleus accumbens of the brain (Section 5.3.11).
      5.3.15 Public Health Significance of Associations between Pb Biomarkers and
             Neurodevelopmental Effects

24                  As described in Section 5.3.2.1. most studies found that a 1 (ig/dL increase in blood Pb
25                  level was associated with decrements in FSIQ in school-aged children in the range of <1
26                  to 2  points, depending on the model and blood Pb level range examined (Figure 5-2 and
27                  Table 5-3). Similarly, a 1 (ig/dL increase in blood Pb level typically was associated with
28                  lower scores on tests of executive function (Table 5-8) and academic performance (Table
29                  5-9). and higher ratings of behavioral problems (Figure 5-10. Table 5-11. Table 5-12) on
30                  the order of less than 1 standard deviation. Such findings prompt consideration of the
31                  public significance of blood Pb level-associated effects on cognitive function and
32                  behavioral problems in children, specifically, whether the magnitudes  of change have
33                  consequences on the health and life-success of individuals. According to the WHO,
34                  "Health is a state of complete physical, mental and social well-being and not merely the
35                  absence of disease or infirmity" (WHO. 1948). By this definition, even decrements in

      November 2012                            5-253                  Draft - Do Not Cite or Quote

-------
 1                   health status that are not severe enough to meet diagnostic criteria might be undesirable if
 2                   they reflect a decrement in the well being of an individual. Deficits in health indices or
 3                   life-success may not be observable except at the population level. The American
 4                   Thoracic Society discussed the need to consider the prevalence of exposures in the
 5                   population and exposure to other risk factors in evaluating whether shifts in the
 6                   population-level risk are adverse (ATS. 2000). Neurodevelopmental deficits measured in
 7                   childhood may set affected children on trajectories more prone toward lower educational
 8                   attainment and financial well-being. Thus, early deficits in children may have lifetime
 9                   consequences. There also may be groups in the population at increased risk of
10                   neurodevelopmental deficits from Pb exposure. For example, some evidence points to
11                   larger blood Pb-associated decrements in cognitive function in children with lower SES
12                   (Ris et al..  2004; Tong et al.. 2000; Bellinger et al.. 1990). whereas a meta-analysis found
13                   a larger effect estimate for studies with higher SES (Schwartz. 1994).

14                   It has been argued that blood Pb-associated decrements in IQ points less than 3 or 4
15                   points are meaningless given that such changes are within the standard error of a single
16                   test (i.e., the statistic that defines the range within which the true value  of an individual is
17                   likely to lie) (Kaufman. 2001). However, this argument incorrectly assumes that
18                   conclusions drawn from individual-level data apply to populations. Evidence does not
19                   indicate that the standard error is nonrandom, i.e., biased in one direction. Hence, there is
20                   no reason to expect that children with higher blood Pb levels systematically test lower
21                   than their true IQ value and that children with lower blood Pb levels test higher than their
22                   true IQ value. Thus, in a population of children, on a given assessment, some children
23                   will test lower than their true value and others will test higher than their true value. In
24                   such cases, between-group differences will be measureable on a population basis. Error in
25                   the measurement of IQ in an individual will contribute nondifferential error on a
26                   population-level and bias the association to the null.

27                   The issue of individual-level versus population-level risk also pertains to the relevance of
28                   the magnitude of decrease in cognitive function or increase in behavioral problems per
29                   unit increase in blood Pb level. Although fractional changes in IQ, memory,  or attention-
30                   related behavioral problems may not be consequential for an individual, they may be
31                   consequential on a population level, especially in the two tails of the distribution
32                   (Bellinger. 2007. 2004b). For example, interventions that shift the population mean,  in a
33                   beneficial direction, by an amount that is without clinical consequence  for an individual
34                   have been shown to produce substantial  decreases in the percentage of individuals with
35                   values that are clinically significant (Bellinger. 2007. 2004b). In  statistical exercises not
36                   specific to  Pb or analysis of data collected from individuals, Weiss (1990. 1988)
37                   predicted that a downward shift of five points in mean IQ, if the amount of dispersion in
38                   the distribution remained the same, should be accompanied by a  doubling of the numbers
      November 2012                             5-254                  Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
of individuals with scores two or more standard deviations below the mean. With a
reduction in population mean IQ from 100 to 95, the percentage of individuals predicted
to score above 130 (two standard deviations above the mean) decreases from 2.3% to
0.99%. Weiss (1988) stated that the implication of such a loss transcends the  current
circumscribed definitions of risk. In addition to implications on the loss of intellectual
ability, the loss of a few IQ points potentially could result in the loss of academic
opportunities in children. For example, schools or programs for the gifted have used IQ
cut-offs (e.g., score of 130) to screen or accept applicants. In another hypothetical
analysis presented in the 2006 Pb AQCD (U.S. EPA. 2006b). based on a blood Pb-IQ
effect estimate of-0.9 points/(ig/dL (the median for blood Pb levels <10 (ig/dL), the
fraction of the population with an IQ <80 was estimated to more than double  from 9%
with a blood Pb level of 0 (ig/dL to 23% with a blood Pb level of 10 (ig/dL (Figure 5-17).
The proportion with an IQ <70, which often requires community support to live (WHO.
1992). is predicted to increase from a little over 2% with a blood Pb level  of 0 (ig/dL to
about 8% with a blood Pb level of 10 (ig/dL [(Figure 5-17) and (U.S. EPA.2006b)1.
These theoretical exercises estimate that for an individual in the low range of the IQ
distribution, a Pb-associated decline of a few points might be sufficient to drop that
individual into the range associated with increase risk of educational, vocational, and
social failure.
           0.40
           0.35
           0.30
         f 0.25
         ~ 0.20
         c 0.15
         B 0.10
         O
         ™ 0.05
           0.00
                      234567
                           Blood lead (pg/dL)
                                                      10
                                                     3   4   S   6   7
                                                     Blood Lead (pg/dL)
                                                                                                   10
      Note: The results presented in the figure are based on a theoretical analysis of changes in population IQ using a concentration-
      response estimate of-0.9 IQ points/ug/dL, which was the median estimate from studies reviewed in the 2006 Pb AQCD for blood Pb
      levels <10 ug/dL .
      Source: 2006 Pb AQCD (U.S. EPA, 2006b).

      Figure 5-17     Hypothetical effect of increasing blood Pb level on the proportion
                        of the population with IQ <70 and <80 points.
      November 2012
                           5-255
Draft - Do Not Cite or Quote

-------
 1                  The hypothetical predictions presented in Weiss (1990. 1988) and the 2006 Pb AQCD
 2                  (U.S. EPA. 2006b) have been supported by findings from analyses using data from
 3                  studies of blood Pb levels and FSIQ decrements in children. Among children in 1st and
 4                  2nd grades from towns around Boston, Needleman et al. (1982) found that a 4-point
 5                  downward shift in the study population mean IQ estimated for tooth Pb levels >24 ppm
 6                  was associated with a 3-fold increase in the percentage of children with an IQ of <80 and
 7                  a decrease in the percentage achieving an IQ >125 from 5% to 0%.

 8                  The aforementioned hypothetical analyses and those using data collected from
 9                  individuals that estimate Pb-associated changes in the population IQ distribution assume
10                  that the magnitude of change is equal across segments of the IQ distribution. Few studies
11                  of Pb and cognitive function have examined whether the effect of Pb varies across the
12                  distribution of cognitive function. However, in a recent study of fourth graders across the
13                  entire state of North Carolina, Miranda et al. (2009) found that higher blood Pb level
14                  measured once in each child between age 9 months and 3 years was associated with
15                  larger decreases in fourth grade EOG scores in the lower segment of the EOG
16                  distribution. An increase in blood Pb level from 1 to 10 (ig/dL was estimated to decrease
17                  EOG score by 0.8 points in the 95th percentile of EOG scores but by 2.3 points in the 5th
18                  percentile of EOG score (Figure 5-7). These findings by Miranda et al. (2009) based
19                  analysis of a large database representative of fourth graders in North Carolina indicate
20                  that a shift in the population mean from increased Pb exposure may increase the
21                  proportions of children at the lower end of the cognitive function over that estimated by
22                  theoretical analyses.

23                  In summary, the public health significance of evidence demonstrating associations
24                  between increases in blood Pb levels and decrements in IQ of children in the range of a
25                  few points is supported by hypothetical predictions that a shift in the population mean
26                  increases the proportion of individuals in the lower range of cognitive function and
27                  decreases proportion of individual in the upper range of cognitive function. These
28                  changes in the population distribution also were found in children in 1st and 2nd grade in
29                  Massachusetts in whom higher tooth Pb level was associated with decrements in IQ
30                  (Needleman et al.. 1982). Further support for the public health significance is provided by
31                  findings that the blood Pb-associated decrement in cognitive function may be larger in
32                  children in the lower range of cognitive function (Miranda et al., 2009) and in specific
33                  groups within the populations such as those with lower SES (Ris et al.. 2004; Tong et al..
34                  2000; Bellinger et al.. 1990). On a population-level, small Pb-associated decreases in
35                  cognitive function could increase the number of individuals at increased risk of
36                  educational, vocational, and social failure and decrease the number of individuals with
37                  opportunities for academic and later-life success.
      November 2012                             5-256                  Draft - Do Not Cite or Quote

-------
      5.3.16 Summary and Causal  Determination

 1                  The collective body of epidemiologic and toxicological evidence integrated across that
 2                  reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b) and recent studies indicates
 3                  relationships between Pb exposure and a range of nervous system effects. In children,
 4                  effects on cognitive function include FSIQ, learning, memory, executive function, and
 5                  academic performance. Outcomes evaluated related to attention-related behavioral
 6                  problems include inattention, impulsivity, hyperactivity, and ADHD. Effects on conduct
 7                  problems in children comprise aggression, delinquency, and criminal offenses. Effects on
 8                  internalizing behaviors include withdrawn behavior, depression-like symptoms,
 9                  fearfulness, and anxiety. Other nervous system effects evaluated in children are sensory
10                  function and motor function. Relationships for Pb exposure with cognitive function and
11                  sensory function also were evaluated in adults. Other nervous system effects in adults
12                  examined in relation to Pb exposure include psychopathological effects such as
13                  depression-like symptoms, anxiety, and panic disorder. Additionally, effects on
14                  neurodegenerative diseases include Alzheimer's disease, ALS, Parkinson's disease, and
15                  essential tremor. The subsequent sections describe the evaluation of evidence for each of
16                  these outcome groups with respect to causal relationships with Pb exposure using the
17                  framework described in Table II of the Preamble. The application of the key supporting
18                  evidence to the causal framework is summarized in Table 5-17.
      5.3.16.1      Evidence for Cognitive Function in Children

19                  A causal relationship between Pb exposure and cognitive function decrements in children
20                  is supported by evidence from prospective studies in diverse populations consistently
21                  demonstrating associations of higher blood and tooth Pb levels with lower FSIQ and
22                  performance on tests of executive function and academic performance in children ages
23                  4-17 years (Section 5.3.2). coherence with evidence in animals for impairments in
24                  learning, memory, and executive function with relevant Pb exposures, and evidence
25                  describing modes of action (Table 5-17).

26                  Clear support for Pb-associated cognitive function decrements in children, as described in
27                  the 2006 Pb AQCD (U.S. EPA, 2006b). was provided by prospective epidemiologic
28                  studies indicating associations of higher earlier childhood, concurrent, and childhood
29                  average blood and tooth Pb levels with lower FSIQ in children ages 4-17 years (Table
30                  5-17 and Section 5.3.2.1). Across studies, FSIQ was  measured with similar instruments
31                  (i.e., WISC-R, WISC-III, WPPSI, Stanford-Binet) scored on a similar scale with similar
32                  measurement error. Associations were found in most of the prospective  studies,
33                  conducted in the U.S., Mexico,  Europe, and Australia in representative populations, most
      November 2012                            5-257                 Draft - Do Not Cite or Quote

-------
 1                   of which had moderate to high follow-up participation without indication of selective
 2                   participation among children with higher blood Pb levels and lower cognitive function
 3                   (Table 5-3). The prospective studies found associations of blood or tooth Pb levels
 4                   measured earlier in life (i.e., prenatal cord or maternal, age 2 year, age 6 year) and
 5                   averaged over multiple years with FSIQ decrements later in childhood (i.e., ages 4-17
 6                   years), better establishing the temporal sequence between Pb exposure and decrements in
 7                   cognitive function compared with cross-sectional analyses. Another strength of the
 8                   evidence from prospective studies was the consideration for several potential
 9                   confounding factors. As indicated in Table 5-3. results from most cohort studies were
10                   adjusted for maternal IQ and education, child sex and birth weight, SES, and HOME
11                   score. Although not considered as frequently, some studies also found associations with
12                   adjustment for parental smoking and nutritional factors. The consistency and
13                   reproducibility of the blood Pb-FSIQ association in children were substantiated in a
14                   pooled analysis of seven prospective studies by Lanphear et al. (2005) as well as multiple
15                   meta-analyses that combined results across various prospective and cross-sectional
16                   studies (Pocock et al., 1994; Schwartz. 1994; Needleman and Gatsonis.  1990). with
17                   Schwartz (1994) demonstrating the robustness  of evidence to potential publication bias.

18                   Among individual studies, a wide range of blood Pb-FSIQ effect estimates was obtained,
19                   which is not unexpected given the wide range of blood Pb levels examined and modeling
20                   methods used (i.e., linear, log-linear).  The pooled analysis of seven prospective cohorts
21                   demonstrated precision of effect estimates by applying a uniform method across
22                   populations (Lanphear et al.. 2005). A narrow range of estimates was obtained by
23                   excluding one study at a time, -2.4 to -2.9 points per log increase in concurrent blood Pb
24                   level. Results from several individual  studies indicated a supralinear concentration-
25                   response relationship which estimated a greater decrement in cognitive function per  unit
26                   increase in blood Pb level among children in lower strata of blood Pb levels than children
27                   in higher strata of blood Pb levels (Figure 5-16 and Table  5-16). Among the largest effect
28                   estimates were found in the Boston and Rochester cohorts (Canfield et al.. 2003a;
29                   Bellinger et al.. 1992). which had relatively smaller sample sizes but considered several
30                   potential confounding factors as listed above and in Table 5-3. examined lower blood Pb
31                   levels than did other prospective studies, and examined cohorts of different SES. Thus,
32                   their results may be more generalizable. In the  Boston cohort, a l(ig/dL increase in age 2
33                   year blood Pb level was associated with a -1.6  (95% CI: -2.9, -0.2) point change in FSIQ
34                   at age 10 years in 48 children with blood Pb levels 1-9.3 (ig/dL whose peak blood Pb
35                   levels never exceeded 10 (ig/dL (Bellinger.  2008; Bellinger and Needleman. 2003). In the
36                   Rochester cohort, a 1 (ig/dL increase in concurrent blood Pb level was associated a -1.8
37                   (95% CI: -3.0, -0.60) change in FSIQ  at age 5 years in 101 children with concurrent
38                   blood Pb levels 0.5-8.4 (ig/dL, and peak blood Pb levels < 10 (ig/dL (Canfield. 2008;
39                   Canfield et al.. 2003a). In the pooled analysis, a 1 (ig/dL increase in concurrent blood Pb

      November 2012                             5-258                   Draft - Do Not Cite or Quote

-------
 1                   level was associated a -2.9 (95% CI: -5.2, -0.71) change in FSIQ at ages 4.8-10 years in
 2                   103 children with concurrent blood Pb levels 0.9-8.4 (ig/dL and peak blood Pb levels
 3                   <7.5 (ig/dL (Hornung. 2008; Lanphear et al.. 2005). These observations do not provide
 4                   evidence for a threshold in the ranges of blood Pb level examined. Null or weak
 5                   associations were limited to a few cohorts, namely, the Cleveland and Sydney cohorts
 6                   (Greene et al.. 1992: Coonevetal.. 1991: 1989a. b; Ernhart et al. 1988). The Cleveland
 7                   and Sydney studies were not outliers with respect to population mean blood Pb levels or
 8                   the specific confounding factors considered (Table 5-3). and the Cleveland cohort had
 9                   high prevalence of maternal prenatal substance abuse which may limit the
10                   representativeness of results.

11                   Other previous studies estimated smaller magnitude of effects for the blood Pb-FSIQ
12                   association with either linear or log-linear models but examined higher blood Pb levels
13                   (means: 8-16 (ig/dL) without analysis of the concentration response at the lower range of
14                   the study population blood Pb distribution (Table 5-3). Recent cross-sectional studies
15                   supported associations between higher concurrent blood Pb levels and decrements in
16                   FSIQ. Among studies that examined populations with mean blood Pb levels <5 (ig/dL,
17                   some lacked representative populations due to high prevalence of prenatal alcohol and/or
18                   drug exposure (Chiodo et al.. 2007) or had limited consideration for potential
19                   confounding (Zailina et al.. 2008). Kim et al. (2009b) estimated similar effects as the
20                   Boston and Rochester studies although the log-linear model makes  comparisons difficult.
21                   Among children ages 8-11 years  in Korea with mean blood Pb level 1.73 (ig/dL, a
22                   1 (ig/dL increase in concurrent blood Pb level was associated with a 3.2-point decrease
23                   (95% CI: -6.1, -0.23) in FSIQ among children ages 8-11 years in Korea with blood Mn
24                   levels >1.4 (ig/dL in the 10th-90th percentile interval of blood Pb level (0.9-2.8 (ig/dL).
25                   In this study, the potential influence of higher past Pb exposures cannot be excluded.
26                   Among children ages 6-10 years  in New England with a mean concurrent blood Pb level
27                   2.2 (ig/dL, lower FSIQ was found in the group with blood Pb levels 5-10 (ig/dL (Surkan
28                   et al.. 2007). While results from these studies were adjusted for SES and parental IQ or
29                   education, parental caregiving quality was not examined.

30                   A causal relationship between Pb exposure and cognitive function decrements in children
31                   also is supported by previous prospective studies (several of which contributed to the
32                   FSIQ evidence) that found associations of blood or tooth Pb level with decrements in
33                   executive function and academic performance in children ages 4-18 years
34                   (Sections 5.3.2.4. 5.3.2.5. and Table 5-17). The bodies of evidence  for executive function,
35                   and academic performance are smaller than that for FSIQ but consistently indicate
36                   associations with blood or tooth Pb level. Associations with performance on tests of
37                   learning and memory were less consistently found across populations (Section 5.3.2.3).
38                   In most studies, previous and recent, multiple testing was common; however, the
      November 2012                             5-259                  Draft - Do Not Cite or Quote

-------
 1                   consistent pattern of association observed across the ages of blood Pb level and/or
 2                   cognitive test examined increases confidence that the evidence is not unduly biased by an
 3                   increased probability of associations found by chance alone. While recent studies of
 4                   executive function and academic performance adjusted for SES and parental education
 5                   and/or IQ, few examined potential confounding by parental caregiving quality.

 6                   Adding to the evidence for Pb-associated cognitive  function decrements in children were
 7                   recent prospective studies that indicated associations between higher earlier childhood
 8                   blood Pb level and poorer academic performance in school-aged children (Chandramouli
 9                   et al.. 2009; Miranda et al.. 2009). Among 57,678 children in North Carolina, lower
10                   fourth grade EOG scores were found in children with age 3-36 month blood Pb levels
11                   2 (ig/dL compared with children with blood Pb levels 1 (ig/dL, with adjustment for
12                   parental education and enrollment in a free lunch program as an indicator of SES
13                   (Miranda et al., 2009). In addition to finding associations with lower early childhood
14                   blood Pb levels, this study indicated a greater incremental association of blood Pb level
15                   with decrement in EOG score among children in the lower end of the EOG distribution.
16                   Chandramouli et al. (2009) found decrements in school achievement tests in 488 children
17                   ages 7-8 years in U.K. in children with age 30 month blood Pb levels >5  (ig/dL. These
18                   results were adjusted for several potential confounding factors, including SES, parental
19                   education, SES, home facilities score, and family adversity. Recent cross-sectional
20                   studies conducted in the U.S. (Krieg etal.. 2010; Surkan et al.. 2007) found associations
21                   of concurrent blood Pb level with decrements in executive function and academic
22                   performance in children, including the large analysis of >700 children ages 12-16 years
23                   participating in NHANES (Krieg etal.. 2010). Cho  et al. (2010) did not find an
24                   association between concurrent blood Pb level and executive function among children in
25                   Korea ages 8-11 years with a mean blood Pb level 1.9 (ig/dL.

26                   Several studies found associations of higher prenatal, earlier infancy, and concurrent
27                   blood Pb levels with lower Bayley MDI scores in children ages 2 and 3 years (Table 5-4).
28                   Similar to studies of FSIQ, Tellez-Rojo et al. (2006) estimated a larger decrement in age
29                   2 year MDI per unit increase in concurrent blood Pb level for children in Mexico City
30                   with blood Pb levels <5 (ig/dL compared with children with blood Pb levels 5-10 (ig/dL,
31                   and >10 (ig/dL (Figure 5-16 and Table 5-16). MDI is a well-standardized measure of
32                   current infant mental development. However, the test of MDI is not an intelligence test,
33                   and MDI scores, particularly before ages 2-3 years,  are not necessarily correlated with
34                   later measurements of FSIQ in children with normal development.

35                   A causal relationship between Pb exposure and cognitive function decrements in children
36                   is further supported by consistent observations in animals of decrements in learning,
37                   memory, and executive function with relevant dietary Pb exposures. In particular,
      November 2012                             5-260                  Draft - Do Not Cite or Quote

-------
 1                  coherence was found between evidence in children and animals of Pb-associated
 2                  decrements in visual-spatial memory, working memory (Section 5.3.2.3) and rule
 3                  learning and reversal (Section 5.3.2.4). Previous studies in monkeys demonstrated
 4                  impairments in learning, memory, and executive function with dietary Pb exposures
 5                  during infancy only, lifetime after infancy, and lifetime from birth that produced blood
 6                  Pb levels of 19 and 26 ug/dL (Rice. 1992b. 1990: Rice and Karpinski. 1988).

 7                  Several recent toxicological studies added to the evidence for impaired learning and
 8                  memory in animals with lower blood Pb levels, 8-17 (ig/dL after gestational-lactational,
 9                  lactational, or lifetime (with and without gestational) Pb exposure (Cory-Slechta et al..
10                  2010: Niu et al.. 2009: Virgolini et al.. 2008a: Stangle et al.. 2007). Together, the
11                  prospective epidemiologic and toxicological studies provide evidence for the temporal
12                  sequence between Pb exposure and decrements in cognitive function. Additional
13                  biological plausibility for Pb-associated cognitive function decrements was provided by
14                  toxicological evidence for the effects of Pb on modes of action for cognitive function
15                  (Section 5.3.11). Pb has been shown to increase the permeability of the blood-brain
16                  barrier and deposit in the target CNS. Pb has been shown to impair neurogenesis,
17                  synaptic architecture, and neurite outgrowth. The high activity of these processes during
18                  fetal and infant development provides biological plausibility for the effects of childhood
19                  Pb exposure on decrements in cognitive function. Cognitive function is mediated by the
20                  cortical and subcortical structures of the brain that integrate function in the hippocampus,
21                  prefrontal cortex, and nucleus accumbens using dopamine and glutamate as primary
22                  neurotransmitters. Experimental studies have shown that Pb induces changes in dopamine
23                  and glutamate release in these regions and decreases the magnitude of LTP, which is a
24                  major cellular mechanism underlying synaptic plasticity and learning and memory.

25                  With regard to critical lifestages of Pb exposure, toxicological evidence clearly
26                  demonstrates impaired learning and memory in animals exposed to Pb gestationally with
27                  or without lactational exposure that produced blood Pb levels 8-17 (ig/dL. This evidence
28                  is well supported by knowledge that processes such as neurogenesis, synaptogenesis, and
29                  synaptic pruning are very active during this developmental period. However, evidence in
30                  monkeys also indicates impaired cognitive function at ages 5-8 years with Pb exposure
31                  starting after weaning (Rice. 1992b: Rice and Gilbert.  1990a: Rice. 1990: Rice and
32                  Gilbert, 1990b). Epidemiologic studies also found cognitive function decrements
33                  associated with blood Pb levels measured during various lifestages  and time periods.
34                  Distinguishing among the effects of Pb exposures at different time periods is difficult in
35                  epidemiologic studies due to the high correlations commonly found among blood Pb
36                  levels within children over time. Among studies of young children ages 6 months to
37                  3 years, several found larger magnitudes of associations of MDI with prenatal maternal or
38                  cord blood Pb than with postnatal child blood Pb (Hu et al.. 2006: Bellinger etal., 1987:
      November 2012                             5-261                  Draft - Do Not Cite or Quote

-------
 1                   Dietrich et al., 1987a; Vimpani et al. 1985). However, in older children, ages 4-17 years,
 2                   in whom cognitive function is more stable and reliably measured, larger decrements in
 3                   cognitive function were found in association with postnatal blood Pb levels,
 4                   i.e., concurrent, earlier childhood, and cumulative average blood Pb levels as well with
 5                   tooth Pb levels. There was no clear indication of an individual critical lifestage, timing, or
 6                   duration of Pb exposure associated with cognitive function decrements in children.
 7                   Because of the contribution of bone Pb levels to concurrent blood Pb levels in children,
 8                   associations with concurrent blood Pb levels may reflect an effect of past and/or recent
 9                   Pb exposures.

10                   The consideration for potential confounding varied among studies.  Most studies adjusted
11                   for SES-related variables such as the Hollingshead Index, household income, and/or
12                   parental education. Several, in particular the prospective studies, additionally adjusted for
13                   parental cognitive function and caregiving quality commonly evaluated as HOME score.
14                   A few studies considered nutritional factors. Few recent studies considered potential
15                   confounding by parental caregiving quality. The adjustment for SES is difficult as it is
16                   highly correlated with Pb exposure and there is no single measure that represents SES.
17                   Residual confounding also is possible by factors not considered. The combination of
18                   evidence from prospective studies that considered several well-characterized potential
19                   confounding factors plus evidence that Pb exposure induces impairments in cognitive
20                   function in animals, in particular, for similar constructs as those associated with blood Pb
21                   levels in children, increase confidence that the associations observed between blood Pb
22                   levels and cognitive function in children represent a relationship with Pb exposure.

23                   In conclusion, multiple prospective studies conducted in diverse populations consistently
24                   demonstrate associations of higher blood and tooth Pb levels with lower FSIQ, executive
25                   function, and academic performance and achievement. Most studies examined
26                   representative populations and had moderate to high follow-up participation without
27                   indication of selective participation among children with higher blood Pb levels and
28                   lower cognitive function. Associations between blood Pb level and cognitive function
29                   decrements were found with adjustment for several potential confounding factors, most
30                   commonly, SES, parental IQ, parental education, and parental caregiving quality. In
31                   children ages 4-11 years, associations were found with prenatal, early childhood,
32                   childhood average, and concurrent blood Pb levels in populations with mean blood Pb
33                   levels in the range 2-8 (ig/dL. Neither epidemiologic nor toxicological evidence has
34                   identified an individual critical lifestage or duration of Pb exposure within childhood that
35                   is associated with cognitive function decrements. Several epidemiologic studies found a
36                   supralinear concentration-response relationship. Examination of children with blood Pb
37                   levels in the range <1 to  10 (ig/dL, with consideration of early or peak childhood blood
38                   Pb levels, has not identified a threshold for cognitive function decrements in the range of
      November 2012                             5-262                  Draft - Do Not Cite or Quote

-------
 1                   blood Pb levels examined. Evidence in children was clearly supported by observations of
 2                   Pb-induced impairments in learning, memory, and executive function in juvenile animals.
 3                   Several studies in animals indicated learning impairments with prenatal, lactational, and
 4                   lifetime (with or without prenatal) Pb exposures that resulted in blood Pb levels of 8-
 5                   26 (ig/dL. The mode of action for Pb-associated cognitive function decrements is
 6                   supported by observations of Pb-induced impairments in neurogenesis, synaptogenesis
 7                   and synaptic pruning, LTP, and neurotransmitter function in the hippocampus, prefrontal
 8                   cortex, and nucleus accumbens. The associations consistently found for FSIQ and other
 9                   measures of cognitive function in prospective studies of children with adjustment for
10                   SES, parental education, and caregiving  quality and the biological plausibility provided
11                   by evidence in animals for impairments in learning, memory, and executive function with
12                   relevant Pb exposures and evidence describing modes of action is sufficient to conclude
13                   that there is a causal relationship between Pb exposure and decrements in cognitive
14                   function in children.
      5.3.16.2      Evidence for Attention-related Behavioral Problems in Children

15                   A causal relationship between Pb exposure and attention-related behavioral problems in
16                   children is supported by evidence from prospective studies in diverse populations for
17                   associations of blood or tooth Pb levels with inattention, impulsivity, and hyperactivity,
18                   coherence with evidence  in animals with relevant Pb exposures, and evidence describing
19                   modes of action (Table 5-17). Although attention-related behavioral problems have been
20                   examined less extensively than cognitive function, several epidemiologic and
21                   toxicological studies have found associations with Pb. Prospective studies provided key
22                   evidence for associations of childhood blood and tooth Pb levels with increases in
23                   inattention, impulsivity, and hyperactivity in children ages 6-17 years and young adults
24                   ages 19-24 years. These associations were found in populations in the U.S., U.K.,
25                   Australia, and New Zealand (Table 5-17). Most studies had population-based recruitment
26                   from prenatal clinics, hospitals at birth, or schools and had moderate to high participation.
27                   A few prospective studies had increased loss-to-follow-up in certain groups, for example,
28                   lower SES, earlier FSIQ,  or HOME score. This potential  selection bias may have reduced
29                   the generalizability of findings to the original study population, but there was not a strong
30                   indication that participation was biased to those with higher blood Pb levels and greater
31                   behavioral problems. The most compelling evidence was that for inattention, impulsivity,
32                   and hyperactivity assessed with neuropsychological testing or rated by parents or teachers
33                   with widely-used structured questionnaires. A few studies found associations between
34                   blood Pb level and diagnosis of ADHD but in studies that did not  consider potential
35                   confounding by parental caregiving quality.
      November 2012                             5-263                  Draft - Do Not Cite or Quote

-------
 1                   Key evidence provided by prospective studies supported associations of blood and tooth
 2                   Pb levels with inattention and impulsivity as assessed using neuropsychological tests or
 3                   parent or teacher ratings (Section 5.3.3.1). Thus, the collective evidence does not appear
 4                   to be unduly influenced by biased reporting of such behaviors by parents of children with
 5                   higher blood Pb levels. Most studies that examined inattention with the continuous
 6                   performance test found associations with blood Pb level (Figure 5-10 and Table 5-11).
 7                   including previous prospective studies that indicated associations of higher prenatal or
 8                   earlier childhood blood Pb levels or tooth (from ages 5-8 years) Pb levels with increases
 9                   in commission and omission errors or reaction time in adolescents ages 15-17 years in
10                   Cincinnati (Ris et al.. 2004) and young adults  19-24 years in Chelsea/Sommerville, MA
11                   (Bellinger etal.. 1994a). Results from prospective studies also indicated associations with
12                   parental and teacher ratings of inattention and impulsivity in children ages 8-13 years in
13                   Australia, New Zealand, and Boston, MA (Burns etal.. 1999; Fergusson et al.. 1993;
14                   Leviton et al., 1993). The mean blood Pb levels (prenatal cord, early childhood, lifetime
15                   average) of populations examined in the prospective  studies were in the range of
16                   7-14 (ig/dL. The prospective findings better established the temporal sequence between
17                   Pb exposure and inattention than did cross-sectional studies. Although the specific factors
18                   varied by study, prospective studies of inattention and impulsivity considered several
19                   potential confounding factors, including SES, parental IQ, maternal education, HOME
20                   score, self drug use, prenatal drug and alcohol exposure, and birth outcomes.  Evidence
21                   did not strongly indicate associations between concurrent blood Pb levels and ratings of
22                   inattention in younger children ages 4-5 years in Rochester and Yugoslavia (Canfield et
23                   al.. 2003b: Wasserman et al.. 2001). These groups of younger children had lower blood
24                   Pb levels, mean 6.5 (ig/dL; however, inattention may be less reliably rated in younger
25                   children.

26                   Consistent with previous prospective studies, recent cross-sectional studies found
27                   associations of higher concurrent blood Pb level with increases in inattention as measured
28                   by CPT, impulsivity using the stop task, and higher ratings of inattention and impulsivity
29                   in children ages 8 to 12 years with mean concurrent blood  Pb levels of 1 to 4  (ig/dL (Cho
30                   etal.. 2010; Nicolescu etal.. 2010; Nigg et al.. 2008). However, the contribution of
31                   higher Pb exposures earlier in childhood cannot be excluded. Further, while these recent
32                   studies considered potential confounding by parental education, they had less consistent
33                   consideration for other SES-related factors or parental caregiving quality than did
34                   prospective studies.  Some considered parental history of psychopathology; however, its
35                   relationship with parental caregiving quality is not well characterized (Cho etal.. 2010;
36                   Nicolescu et al.. 2010). Chiodo et al. (2007; 2004)  found associations between concurrent
37                   blood Pb level and increases in inattention as measured by the CPT and by independent
38                   examiner ratings.  Mean blood Pb levels were ~5 (ig/dL, and results were adjusted for
39                   SES, parental education, and HOME score. However, the study population lacked

      November 2012                             5-264                  Draft - Do Not Cite or Quote

-------
 1                   representativeness because of the high prevalence of prenatal alcohol and drug use.
 2                   Recent cross-sectional studies that included younger children (ages 3-5 years) also found
 3                   associations between concurrent blood Pb level and higher inattention as rated by
 4                   teachers or study examiners (Plusquellec et al. 2010; Roy et al.. 2009a): however, ratings
 5                   in young children may be less reliably measured.

 6                   A causal relationship between Pb exposure and attention-related behavioral problems also
 7                   is supported by a recent prospective study that found higher teacher ratings of
 8                   hyperactivity among children ages 7-8 years in the U.K. with age 30 month blood Pb
 9                   levels >10 (ig/dL (Chandramouli et al.. 2009). Previous findings were limited to cross-
10                   sectional and case-control studies (Section 5.3.3.2). A strength of the recent prospective
11                   study was the adjustment for several potential confounding factors, including maternal
12                   education and smoking, SES, home facilities score, and family adversity index. Among
13                   the recent cross-sectional studies, associations were found with consideration for
14                   potential confounding by SES and maternal education; however, parental caregiving
15                   quality was examined infrequently. Recent cross-sectional studies also found associations
16                   of concurrent blood Pb  level with higher parental ratings of a composite index of ADHD-
17                   related behaviors, including the large U.S. representative analysis of 2,588 children
18                   participating in NHANES which used DSM-IV criteria (Froehlich et al.. 2009). With the
19                   exception of findings from the Rochester cohort (Canfield et al., 2003b). studies
20                   generally found associations between blood Pb level and attention-related behavioral
21                   problems with adjustment for child IQ or other measure of cognitive function (Cho et al..
22                   2010: Chandramouli et al.. 2009: Nigg et al.. 2008: Silvaetal.. 1988). These  findings add
23                   support for higher Pb exposures having effects on attention-related behavioral problems,
24                   independent of effects on cognitive function.

25                   In the few available studies, concurrent blood Pb levels were associated with ADHD
26                   prevalence in children (Section 5.3.3.4). Because of the cross-sectional or case-control
27                   design of studies and lack of consideration for potential confounding by parental
28                   caregiving quality or attention-related problems, the ADHD evidence is not a major
29                   consideration in drawing conclusions about the relationship between Pb exposure and
30                   attention-related behavioral problems.

31                   Further support for a causal relationship between Pb exposure and attention-related
32                   behavioral problems is provided by observations of impulsivity in animals in tests of
33                   response inhibition (e.g., discrimination reversal learning, FR/waiting for reward) with
34                   relevant dietary Pb exposures. In particular, coherence is found with observations in
35                   children of associations between blood Pb levels and performance on the stop signal task,
36                   which also measures response inhibition. Impulsivity in rodents and monkeys was found
37                   with gestational and lactational dietary Pb exposures that resulted in blood Pb levels  of 10
      November 2012                             5-265                  Draft - Do Not Cite or Quote

-------
 1                   to 31 (ig/dL (Table 5-17 and Section 5.3.3.1). In animals, the effects of Pb exposure on
 2                   sustained attention were inconsistent as assessed with a signal-detection test with
 3                   distracting stimuli. In monkeys ages 9-10 years, lifetime dietary Pb exposure from birth
 4                   producing blood Pb levels 15 and 25 (ig/dL induced distractibility as assessed by poorer
 5                   performance on discrimination reversal tasks with distracting stimuli (Gilbert and Rice.
 6                   1987). Relevant dietary Pb exposures (i.e., producing blood Pb level of 10 (ig/dL) also
 7                   were found to increased activity in male (not female) mice with but only with
 8                   amphetamine co-treatment (Leasure et al.. 2008): thus, the findings may  not be directly
 9                   comparable to observations of Pb-associated increases in hyperactivity in children.

10                   Additional support for a causal relationship between Pb exposure and attention-related
11                   behavioral problems is provided by evidence describing modes of action. Attention-
12                   related behavioral problems have been linked with changes in the prefrontal cerebral
13                   cortex, cerebellum, and hippocampus, and Pb exposure has been found to affect
14                   development and neuronal processes in these regions. For example, Pb has been found to
15                   affect dopaminergic neurons of the  frontal cortex and striatum of the brain by altering
16                   dopamine release and receptor density. Other lines of evidence supporting the mode of
17                   action for the effects of Pb exposure on attention-related behavioral problems include
18                   Pb-induced changes in neurogenesis, synapse formation, and synaptic plasticity.

19                   In conclusion, although examined less extensively than cognitive function, several
20                   prospective studies demonstrate associations of earlier childhood (e.g., age 6 years) and
21                   lifetime  average blood Pb levels or tooth (from ages 5-8 years) Pb levels with inattention,
22                   impulsivity, and hyperactivity in children 7-17 years and young adults ages 19-24 years
23                   as assessed using objective neuropsychological tests and rated by parents and teachers.
24                   Most of these prospective studies examined representative populations without indication
25                   of participation conditional on blood Pb levels and behavioral problems.  The results from
26                   prospective studies were adjusted for potential confounding by SES  and parental
27                   caregiving quality, with a few studies also considering substance abuse and nutritional
28                   status. Blood Pb-associated increases in attention-related behavioral problems were found
29                   in populations with earlier childhood (age 6 years) or lifetime average  (to age 11-13
30                   years) mean blood Pb levels of 7 and  14 (ig/dL and groups with earlier childhood (age
31                   30 months) blood Pb levels >10 (ig/dL. Several cross-sectional studies found associations
32                   between concurrent blood Pb level (means 1-4 (ig/dL) and attention-related behavioral
33                   problems in children ages 8-12 years but had less consistent adjustment for SES and
34                   parental caregiving quality. Biological plausibility for observations in children is
35                   provided by several findings in animals for increases in impulsivity or impaired response
36                   inhibition with relevant prenatal, lactational, and lifetime Pb exposures that resulted in
37                   blood Pb levels of 10 to 31 (ig/dL. The mode of action for Pb-associated attention-related
38                   behavioral problems is supported by observations of Pb-induced impairments in
      November 2012                             5-266                  Draft - Do Not Cite or Quote

-------
 1                  neurogenesis, synaptic pruning, and dopamine transmission in the prefrontal cerebral
 2                  cortex, cerebellum, and hippocampus. The consistency of epidemiologic evidence,
 3                  particularly for inattention, impulsivity, and hyperactivity in prospective studies, and the
 4                  biological plausibility provided by evidence for Pb-induced impulsivity in animals and
 5                  for underlying modes of action is sufficient to conclude that there is a causal relationship
 6                  between Pb exposure and attention-related behavioral problems in children.
      5.3.16.3      Evidence for Conduct Problems in Children and Young Adults

 7                   Epidemiologic evidence indicates that a causal relationship is likely to exist between Pb
 8                   exposure and misconduct in children and young adults. Key evidence is provided by
 9                   recent prospective studies finding associations of higher earlier childhood blood and
10                   tooth (from ages 6-8 years) Pb levels with criminal offenses in young adults in Cincinnati
11                   and Christchurch, New Zealand, ages 19-24 years, as assessed through government
12                   records (Fergusson et al., 2008; Wright et al., 2008). In the Cincinnati cohort, a 1 (ig/dL
13                   increase in age 6 year blood Pb level (mean 6.8 (ig/dL) was associated with an increased
14                   risk of criminal arrests at age 19-24 years with an RR of 1.05 (95% CI: 1.01, 1.09).
15                   Additional support was provided by most previous and recent prospective studies that
16                   found associations of blood or tooth Pb levels with higher parent and teacher ratings of
17                   delinquent behavior, aggression, antisocial activities, and destructive behavior in children
18                   ages 8-17 years from diverse locations and SES (i.e., U.K., Cincinnati, Port Pirie,
19                   Australia) (Table 5-12 and Table 5-17). Associations were found with lifetime average
20                   blood Pb levels in boys ages 11-13 years in Port Pirie, Australia with a mean blood Pb
21                   level 14 (ig/dL (Burns etal.. 1999) and with age 30 month blood Pb levels >10 (ig/dL in
22                   children ages 7-8 years in the U.K. (Chandramouli et al.. 2009). The moderate to high
23                   follow-up participation and associations found with parent and teacher ratings of conduct
24                   problems do not provide strong evidence for biased participation or reporting of conduct
25                   problems for children with higher blood Pb levels. Studies of criminal offenses and
26                   ratings of conduct problems found associations with adjustment for several potential
27                   confounding factors such as SES, smoking, drug, and alcohol exposure, and parental
28                   caregiving quality.

29                   Supporting evidence was provided by the large cross-sectional analysis of 2,867
30                   adolescents ages 8-15 years participating in NHANES, which found that compared with
31                   children with concurrent blood Pb levels <0.8 (ig/dL, children with concurrent blood Pb
32                   levels 0.8-1.0 (ig/dL had higher odds of conduct disorder as assessed by parents with
33                   adjustment for age, sex, race, poverty to income ratio, and smoking exposure (Braun et
34                   al., 2008). Parental caregiving quality was not examined. These  associations observed in
35                   adolescents with relatively low concurrent blood Pb levels could have been influenced by
      November 2012                             5-267                  Draft - Do Not Cite or Quote

-------
 1                  higher past Pb exposures. Further supporting the consistency of association between
 2                  blood Pb levels and conduct problems in children, a recent meta-analysis found that
 3                  evidence was robust to heterogeneity in study design, definition and assessment method
 4                  of conduct problems, potential confounding variables examined, and range of blood Pb
 5                  levels (Marcus etal.. 2010). Evidence of Pb-induced aggression in animals was
 6                  inconsistent, with increases in aggression found in some studies of adult animals with
 7                  gestational plus lifetime Pb exposure but not juvenile animals.

 8                  Associations of conduct problems in children and young adults with earlier childhood,
 9                  earlier childhood average, and lifetime average blood Pb levels, tooth Pb levels, and bone
10                  Pb levels point to the effects of early childhood or cumulative Pb exposures. Most
11                  prospective  studies did not analyze Pb biomarker levels at multiple  lifestages or time
12                  periods and  thus did not provide information on potential associations with more recent
13                  blood Pb measurements or differences in association among Pb biomarkers at various
14                  time periods. With respect to blood Pb levels, an association with criminal offenses was
15                  found in young adults ages 19-24 years with a mean age 6 year blood Pb level of
16                  6.8 (ig/dL, and associations with ratings of conduct problems were found in children ages
17                  7-8 years with age 30 month blood Pb levels >10 (ig/dL and boys ages 11-13 years with a
18                  mean lifetime  average blood Pb level of ~14 (ig/dL (Table 5-17).

19                  In conclusion, the few prospective studies consistently indicate that earlier childhood (age
20                  30 months)  or lifetime average (to age 11-13 years) blood Pb levels or tooth (from ages
21                  6-8 years) Pb levels are associated with criminal offenses in young adults ages 19-24
22                  years and with higher parent and teacher ratings of conduct problems in children ages
23                  7-17 years. These associations were found without indication of strong selection bias  and
24                  with adjustment for SES, parental education and IQ, parental caregiving quality, family
25                  functioning, smoking, and substance abuse. Supporting evidence is  provided by cross-
26                  sectional evidence of children participating in NHANES and a meta-analysis of
27                  prospective  and cross-sectional studies. Evidence for Pb-induced aggression in animals is
28                  mixed. The  consistent epidemiologic evidence from prospective and cross-sectional
29                  studies for criminal offenses and ratings of misconduct but lack of clear evidence for
30                  aggression in animals is sufficient to conclude that a causal relationship is likely to exist
31                  between Pb  exposure and conduct problems in children and young adults.
      5.3.16.4      Evidence for Internalizing Behaviors in Children

32                   Epidemiologic and toxicological evidence indicates that a causal relationship is likely to
33                   exist between Pb exposures in children and internalizing behaviors, including withdrawn
34                   behavior, fearfulness, and symptoms of depression and anxiety. Internalizing behaviors
      November 2012                             5-268                  Draft - Do Not Cite or Quote

-------
 1                   have been examined to a lesser extent than cognitive function and attention-related
 2                   behaviors. However, supporting evidence is provided by a few previous prospective
 3                   studies that found associations of higher lifetime average blood or tooth Pb levels with
 4                   higher parent and teacher ratings of withdrawn behavior or symptoms of depression in
 5                   school-aged children, 8-13 years (Burns etal.. 1999; Bellinger et al..  1994b) (Table
 6                   5-17). These prospective studies followed children from birth and had moderate follow-
 7                   up participation to later childhood. Participation was not conditional on early childhood
 8                   blood Pb levels, and associations were found with both parent and teacher ratings of
 9                   internalizing behaviors, reducing the likelihood of undue influence by biased
10                   participation and ratings of internalizing behaviors by parents of children with higher Pb
11                   exposures. Internalizing behaviors were assessed with widely-used structured
12                   questionnaires such as the Child Behavior Checklist but not assessed as clinically-
13                   diagnosed conditions such as depression.

14                   The analysis of the Port Pirie, Australia cohort had the most extensive consideration of
15                   potential confounding. Among only the 163 girls, ages 11-13 years, Burns et al. (1999)
16                   found that 1 (ig/dL increase in lifetime average blood Pb level was associated with an
17                   increased odds of an anxious/depressed parental rating above the median of 1.04 (95%
18                   CI: 1.0, 1.09) with the adjustment for several SES-related variables and factors related to
19                   parental caregiving including HOME score, family functioning score, and current
20                   maternal psychopathology. In a Boston-area cohort, Pb level in deciduous teeth collected
21                   around age 6 years was associated with a higher teacher rating of a composite of anxious
22                   and social withdrawn behaviors in children ages 8 years with adjustment for receiving
23                   public assistance at birth and maternal  education but not parental caregiving quality
24                   (Bellinger etal..  1994b). In the Yugoslavia cohort, higher lifetime average blood Pb
25                   levels were associated with higher maternal ratings of anxious-depressed and withdrawn
26                   behaviors in 191 children ages 4-5 years with a mean blood Pb level ~8 (ig/dL, with
27                   stronger associations found with delinquent behaviors (Wasserman et al.. 2001).
28                   Behavior ratings may be less reliable in these younger children. With respect to critical
29                   lifestages and durations of Pb exposure, evidence from prospective studies for
30                   associations with tooth Pb levels and lifetime average blood Pb levels indicates an effect
31                   of cumulative childhood exposure on increasing internalizing behaviors in children.

32                   Cross-sectional studies, including several recent studies, indicated associations between
33                   concurrent blood Pb levels and internalizing behaviors in children ages 3-16 years, but
34                   most did not consider potential confounding by parental caregiving quality
35                   (Section 5.3.5.1). Previously, Chiodo et al. (2004) found that among children age 7 years
36                   in Detroit, MI (mean blood Pb level: 5  (ig/dL), HOME score,  SES, maternal education,
37                   and prenatal alcohol and drug exposure did not influence associations between  blood Pb
38                   level and internalizing behaviors; however, the population lacks representativeness
      November 2012                             5-269                  Draft - Do Not Cite or Quote

-------
 1                   because of the high prevalence of prenatal alcohol or drug exposure (Table 5-17). A
 2                   recent study reported a lack of association between concurrent blood Pb level and
 3                   internalizing behaviors in Inuit children age 5 years in Quebec, Canada with a mean
 4                   blood Pb level of ~5 (ig/dL with consideration of potential confounding by HOME score,
 5                   caregiver education and IQ, blood Hg levels, and prenatal smoking and alcohol exposure
 6                   (Plusauellecetal.. 2010).

 7                   Supporting evidence for a relationship between Pb exposure and internalizing behaviors
 8                   is provided by the coherence of epidemiologic findings in children with evidence in
 9                   rodents that dietary prenatal plus lactational or lactational only Pb exposure resulted in
10                   depression-like and loss of motivation behavior in rodents, in some cases with blood Pb
11                   levels relevant to humans (13-17 (ig/dL) (Beaudin et al.. 2007; Dyatlov and Lawrence,
12                   2002). Other studies found Pb-induced increases in emotionality, depression, and tactile
13                   defensiveness in animals with blood Pb levels >30 (ig/dL after gestational and/or
14                   lactational Pb exposure (Section 5.3.5.2). Biological plausibility for Pb-associated
15                   increases in internalizing behaviors also is provided by evidence that describes mode of
16                   action, including Pb-induced changes in the HPA axis (Section 5.3.2.3) and dopaminergic
17                   and GABAergic systems (Sections 5.2.2.2. 5.3.11.4. and 5.3.11.8). which are found to
18                   affect mood and emotional state.

19                   In conclusion, prospective studies in a few populations demonstrate associations of
20                   higher lifetime average blood (mean: ~14 (ig/dL) or childhood tooth (from ages 6-8
21                   years) Pb levels with higher parent and teacher ratings of internalizing behaviors such as
22                   depression, anxiety, and withdrawn behavior in children ages 8-13 years. The lack of
23                   selective participation by blood Pb level and associations found with parental and teacher
24                   ratings do not provide strong indication of biased reporting of behaviors for children with
25                   higher blood Pb levels. While results were adjusted for maternal education and SES-
26                   related variables, consideration for potential confounding by parental caregiving quality
27                   was inconsistent. The biological plausibility for the effects of Pb on internalizing
28                   behaviors is provided by consistent findings in  animals with dietary prenatal plus
29                   lactational or lactational only Pb exposure, with some evidence at blood Pb levels
30                   relevant to humans. Additional toxicological evidence supports modes of action,
31                   including Pb-induced changes in the HPA axis  and dopaminergic and GABAergic
32                   systems. The evidence from prospective studies in a few populations and the supporting
33                   toxicological evidence with some uncertainty related to potential confounding by parental
34                   caregiving quality in studies of children is sufficient to conclude that a causal relationship
35                   is likely to exist between Pb exposure and internalizing behaviors  in children.
      November 2012                             5-270                  Draft - Do Not Cite or Quote

-------
      5.3.16.5       Evidence for Sensory Function in Children

 1                   A causal relationship is likely to exist between Pb exposure and sensory function
 2                   decrements in children based on evidence from a previous prospective study indicating
 3                   associations of increased hearing thresholds with early childhood and lifetime average
 4                   blood Pb levels in 215 children age 5 years (Dietrich et al.. 1992) and previous large
 5                   (n = >3,000) cross-sectional NHANES and HHANES studies for concurrent blood Pb
 6                   levels in children ages 4-19 years (Schwartz and Otto. 1991.  1987). The high follow-up
 7                   participation from birth in the Cincinnati cohort and the examination of multiple
 8                   exposures and outcomes in NHANES and HHANES reduce the likelihood of biased
 9                   participation by children with high blood Pb levels.

10                   The epidemiologic evidence in children is strengthened by the consideration for several
11                   potential confounding factors. In the Cincinnati cohort at age 5 years, higher prenatal
12                   maternal, neonatal (10 day, mean 4.8 (ig/dL), and lifetime average (mean: 17.4 (ig/dL)
13                   blood Pb levels were associated with higher hearing thresholds with adjustment for SES,
14                   HOME score, several birth outcomes, and maternal alcohol consumption and
15                   consideration for factors such as maternal smoking and child health (Dietrich et al..
16                   1992). In NHANES and HHANES, higher concurrent blood Pb levels (median: 8 (ig/dL)
17                   were associated with increased hearing thresholds with adjustment for age, sex, race,
18                   family income, parental education, and nutritional factors (Schwartz and Otto.  1991.
19                   1987).

20                   Additional support for a relationship between Pb exposure and sensory function
21                   decrements in children is provided by evidence supporting modes of action. A previous
22                   prospective study in children in Mexico City (n = 100, 113) found associations of
23                   prenatal maternal (1-8 (ig/dL) and age 1 and 4 year blood Pb levels (age 2 year mean:
24                   10.8 (ig/dL) with lower auditory evoked potentials (Rothenberg et al.. 2000). Increased
25                   thresholds and increased latencies in brainstem auditory evoked potentials were also
26                   found in nonhuman primates ages 8-13 years with long-term  (multiple years) postnatal Pb
27                   exposure beginning at birth (Rice. 1997; Lilienthal and Winneke. 1996). Pb exposure
28                   from gestation through age 5 months was found to have weaker effects (Laughlin et al..
29                   2009). In animals, auditory effects were examined with higher Pb exposures than those
30                   relevant to the current U.S. general population (i.e., resultant blood Pb levels
31                   33-170 (ig/dL); thus, it is difficult to assess coherence with observations in children.

32                   Toxicological studies demonstrated a range of effects on the visual system including
33                   impaired visual function, and potential mechanisms such as alterations in morphology
34                   and cell architecture, signaling, enzyme inhibition, neurotransmitter levels,
35                   neuroanatomical development, cell proliferation, and retinal cell apoptosis. An
36                   epidemiologic study in children (Rothenberg et al.. 2002b) and toxicological studies in

      November 2012                             5-271                  Draft - Do Not Cite or Quote

-------
 1                   rats found Pb-associated supernormal retinal ERGs (Fox et al.. 2008). Animal studies
 2                   also showed subnormal ERGs depending on lifestage of Pb exposure and blood Pb level.
 3                   Because the implications of supernormal ERGs on vision are not clear, the retinal ERG
 4                   findings are not a major consideration in drawing conclusions about Pb exposure and
 5                   sensory function decrements in children.

 6                   In conclusion, evidence from a prospective study and cross-sectional studies in a few
 7                   populations indicates associations of higher blood Pb level with increases in hearing
 8                   thresholds and decreases in auditory evoked potentials with adjustment for potential
 9                   confounding by SES in most studies and by child health and nutritional factors in some
10                   studies. The high participation rates, particularly in the prospective study with follow-up
11                   from birth, reduce the likelihood of biased participation by children with higher blood Pb
12                   levels. Across studies,  associations were found with blood Pb levels measured at various
13                   time periods, including prenatal maternal, neonatal  (10 day, mean 4.8 (ig/dL), lifetime
14                   average (to  age 5 years, mean 17.4 (ig/dL), and concurrent (ages 4-19 years) blood Pb
15                   levels (median 8 (ig/dL). Findings in monkeys ages 8-13 years indicate increases hearing
16                   thresholds and latencies for auditory evoked potentials with lifetime postnatal dietary Pb
17                   exposure, albeit with higher blood Pb levels (i.e., 33-107 (ig/dL) than those relevant to
18                   humans. The evidence in children, particularly that from a prospective study, but
19                   uncertainties related to effects on auditory function in animals with relevant Pb
20                   exposures, is sufficient to conclude that a causal relationship  is likely to exist between Pb
21                   exposure and decrements in sensory function in children.
      5.3.16.6      Evidence for Motor Function in Children

22                   A causal relationship is likely to exist between Pb exposure and motor function
23                   decrements in children based on evidence from previous prospective epidemiologic
24                   studies and supporting toxicological evidence. In the Cincinnati cohort, higher neonatal,
25                   concurrent, and lifetime average blood Pb levels were associated with poorer fine and
26                   gross motor function in 245 children ages 6 years (Dietrich et al., 1993a). and higher age
27                   0-5 year average and 78 month blood Pb levels were associated with poorer fine motor
28                   function in 195 children ages 15-17 years (Ris et al., 2004). In the Yugoslavian cohort,
29                   higher lifetime average blood Pb level was associated with decrements in fine but not
30                   gross motor function at age 4.5 years in 283 children (Wasserman et al., 2000). These
31                   studies had high follow-up participation from birth  or infancy reducing the likelihood of
32                   biased participation by children with higher blood Pb levels. Motor function was assessed
33                   using varied but widely-used, structured tests. The evidence from the Cincinnati and
34                   Yugoslavia cohorts is substantiated by the consideration of several potential  confounding
35                   factors such as SES, parental caregiving quality, child health, and in  adolescents,


      November 2012                             5-272                 Draft - Do Not Cite or Quote

-------
 1                   marijuana use (Ris et al., 2004). In the prospective studies, mean concurrent and
 2                   childhood average blood Pb levels mostly ranged from 11 to 28 (ig/dL, higher than those
 3                   in most of the current U.S. population. Recent cross-sectional studies examining lower
 4                   concurrent blood Pb levels, means 2-5 (ig/dL, produced contrasting associations (Surkan
 5                   et al., 2007; Despres et al., 2005). An association between concurrent blood Pb level and
 6                   poorer motor function was found with adjustment for several potential confounding
 7                   factors including SES, parental caregiving quality, and blood levels of Hg and
 8                   polychlorinated biphenyls in 110 Inuit children living in subsistence fishing communities
 9                   (Despres et al., 2005). Higher blood Pb level was associated with improved motor
10                   function in a more representative population of 534 children ages 6-10 years in New
11                   England (Surkan et al. 2007).

12                   Epidemiologic evidence is supported by observations of poorer performance on the
13                   rotarod balance test in male  (not female) mice with relevant blood Pb levels,
14                   i.e., 10 (ig/dL after dietary Pb exposure from gestation to PND10 (Leasure et al.. 2008).
15                   Other toxicological studies produced mixed results for effects on endurance, balance, and
16                   coordination (Section 5.3.8) but are less relevant to humans because of the higher
17                   concentrations of Pb exposure examined, i.e., those producing blood Pb levels
18                   >30 (ig/dL.

19                   In conclusion, evidence from prospective and cross-sectional studies in a few populations
20                   indicates associations of decrements in fine and gross motor function with higher blood
21                   Pb levels measured earlier in childhood (ages 0-5 year average, age 78 months) in
22                   children ages 15-17 years or lifetime average blood Pb levels in children ages 4.5 years
23                   with adjustment for several potential confounding factors, including SES, parental
24                   caregiving quality, and child health. The prospective studies had high follow-up
25                   participation from birth or early infancy, reducing the likelihood of biased participation
26                   by children with higher blood Pb levels. The biological plausibility for associations
27                   observed in children is provided by a study that found poorer balance in male mice with
28                   relevant gestational to early  postnatal (PND10) Pb exposures. The evidence in children,
29                   particularly from a few prospective studies, and the coherence with limited available
30                   findings in mice is sufficient to conclude that a causal relationship is likely to exist
31                   between Pb exposure and decrements in motor function in children.
      5.3.16.7      Evidence for Cognitive Function in Adults

32                   Epidemiologic and toxicological evidence indicates that a causal relationship is likely to
33                   exist between Pb exposure and cognitive function decrements in adults based primarily
34                   on recent prospective and cross-sectional studies that indicate associations with bone Pb
      November 2012                             5-273                  Draft - Do Not Cite or Quote

-------
 1                  level (Table 5-17). Key evidence for bone Pb levels comprised prospective analyses of
 2                  the BMS and NAS, with support provided by the cross-sectional Nurses' Health Study.
 3                  The multiple risk factors and health outcomes examined in these studies reduces the
 4                  likelihood of biased participation and/or follow-up by adults with higher Pb exposure and
 5                  lower cognitive function. While the NAS and Nurses' Health Study examined primarily
 6                  white men and white women, respectively, the BMS examined a more ethnically diverse
 7                  population of men and women, increasing the generalizability of findings. There was
 8                  variability in associations across the various domains of cognitive function tested within
 9                  studies; however, bone Pb levels were associated with decrements in most
10                  neuropsychological tests performed. In many studies, bone Pb levels were associated
11                  with poorer executive function, visuospatial skills, learning, and memory.

12                  Recent evidence from prospective analyses of the NAS and BMS cohorts expanded upon
13                  previous cross-sectional evidence by improving characterization of the temporal
14                  sequence between Pb exposure and cognitive function declines in adults (n = 405-943,
15                  mean ages 60 and 69 years) by demonstrating that higher tibia (means 19, 20 (ig/g) or
16                  patella (mean 25 (ig/g) bone Pb levels measured at baseline were associated with
17                  subsequent declines in cognitive function over 2- to 4-year periods (Bandeen-Roche et
18                  al.. 2009; Weisskopf et al.. 2007b). The specific potential confounding factors considered
19                  differed between studies; both studies adjusted for age and education. Additional
20                  adjustment was made for income in the  BMS and current alcohol use and current
21                  smoking in the NAS.

22                  Evidence from most cross-sectional analyses supported associations between higher bone
23                  Pb level and decrements in cognitive  function in adults. A strength of cross-sectional
24                  studies overall was the adjustment for the same potential confounding factors described
25                  above  and also dietary factors, physical  activity, medication use, and comorbid conditions
26                  (Rajan et al.. 2008; Weuve et al.. 2006). Cross-sectional studies generally demonstrated
27                  larger decrements in cognitive function  in adults in association with tibia or patella Pb
28                  levels than with concurrent blood Pb levels.  Results from the NAS and Nurses' Health
29                  Study did not clearly indicate a difference in association with cognitive performance
30                  between tibia and patella Pb levels (Weuve et al..  2009; Weisskopf et al.. 2007b). In
31                  NHANES analyses, higher concurrent blood Pb levels were associated with lower
32                  cognitive function in particular age and  genetic variant subgroups but not consistently
33                  across the various cognitive tests conducted  (Krieg etal. 2010; Krieg and Butler. 2009;
34                  Krieg et al.. 2009). NHANES did not have bone Pb measures for comparison.

35                  Because bone Pb  is a major contributor to blood Pb levels, blood Pb level also can reflect
36                  longer term exposures, including higher past exposures,  especially in adults without
37                  occupational exposures. Thus, in the NHANES results, it is difficult to characterize the
      November 2012                             5-274                  Draft - Do Not Cite or Quote

-------
 1                   relative contributions of recent and past Pb exposures to the associations observed
 2                   between concurrent blood Pb level and cognitive function. The discrepant findings for
 3                   blood and bone Pb levels indicate that cumulative Pb exposure that likely included higher
 4                   past exposures may be a better predictor of cognitive function in adults than is concurrent
 5                   blood Pb level.

 6                   Additional support for the effects of cumulative or past Pb exposure is provided by
 7                   analyses of a few child cohorts as adults, which indicate that childhood tooth (from ages
 8                   5-8 years) and blood (e.g., age 10 years) Pb levels are associated with decrements in
 9                   cognitive function in adults ages 19-30 years (Mazumdar et al.. 2011; Bellinger et al..
10                   1994a). An uncertainty in the evidence for bone Pb levels is potential residual
11                   confounding by age. Increasing age is highly correlated with increasing bone Pb level
12                   (Section 4.3.5.2). and distinguishing Pb-related declines in cognitive function from age-
13                   related declines with model adjustment is difficult. One explanation for the more variable
14                   findings in adults than  in children may be that cognitive reserve may compensate for the
15                   effects of Pb exposure  on learning new information.  Compensatory mechanisms may be
16                   overwhelmed with age and with higher long-term or cumulative Pb exposure represented
17                   by higher bone Pb levels.

18                   Higher blood and bone Pb levels were associated with cognitive function decrements in
19                   adults with current or former occupational Pb exposures. Some studies examined current
20                   workers with blood Pb level means 26 or 31 ug/dL (Table 5-17). Among adults with
21                   current occupational Pb exposures, both concurrent and cumulative exposures may affect
22                   cognitive function. Several of these studies considered potential confounding by a similar
23                   set of factors as did studies of adults without occupational Pb exposures but did not
24                   examine other occupational exposures. In the prospective study of adults with former
25                   occupational Pb exposure, peak tibia Pb  levels were  associated more strongly with
26                   cognitive performance than were concurrent blood Pb levels (Khalil et al., 2009a). Thus,
27                   in the absence of higher current Pb exposures, cumulative Pb exposures may have a
28                   greater effect on cognitive function in adults.

29                   Additional support for a relationship between Pb exposure and cognitive function
30                   decrements in adults is provided by the coherence with evidence in adult animals that
31                   lifetime Pb exposure of animals starting  from gestation, birth, or after weaning induces
32                   learning impairments (Table 5-17 and Section 5.3.2.3). Biological plausibility also is
33                   provided by evidence describing the effects of Pb on modes of action underlying
34                   cognitive function. Cognitive function is mediated by actions of the  neurotransmitters
35                   dopamine and glutamate in the hippocampus, prefrontal cortex, and  nucleus accumbens.
36                   Experimental  studies have shown that Pb induces changes in neurotransmitter release in
37                   these regions. Studies also have shown Pb-induced decreases in the magnitude  of LTP.
      November 2012                             5-275                  Draft - Do Not Cite or Quote

-------
 1                   In conclusion, in adults without occupational exposure, recent prospective studies in the
 2                   NAS and BMS cohorts indicate associations of higher baseline tibia (means 19, 20 (ig/g)
 3                   or patella (mean 25 (ig/g) Pb levels with declines in cognitive function in adults (>age 50
 4                   years) over 2- to 4-year periods. While the specific covariates differed between studies,
 5                   these tibia Pb-associated cognitive function decrements were found with adjustment for
 6                   potential confounding factors such as age, education, SES, current alcohol use, and
 7                   current smoking. Supporting evidence is provided by cross-sectional analyses of the
 8                   NAS, BMS, and the Nurses' Health Study, which found stronger associations with bone
 9                   Pb level than concurrent blood Pb level. Cross-sectional studies also considered more
10                   potential confounding factors, including dietary factors, physical activity, medication use,
11                   and comorbid conditions. The multiple exposures and health outcomes examined in many
12                   studies reduces the likelihood of biased participation by adults with higher Pb exposure
13                   and lower cognitive function. The collective evidence indicates associations in cohorts of
14                   white men and women and a cohort of more ethnically diverse men and women. The
15                   specific timing, frequency, duration, and magnitude of Pb exposures contributing to the
16                   associations observed with bone Pb levels are uncertain. Also uncertain is the potential
17                   for residual confounding by age. The effects of recent Pb exposures on cognitive  function
18                   decrements were indicated in Pb-exposed workers by associations found with blood  Pb
19                   levels, although these studies did not consider potential confounding by other workplace
20                   exposures. The biological plausibility for the effects of Pb exposure on cognitive  function
21                   decrements in adults is provided by findings that lifetime Pb exposures from gestation,
22                   birth, or after weaning induce learning impairments in adult animals and by evidence for
23                   the effects of Pb altering neurotransmitter function in hippocampus, prefrontal cortex,
24                   and nucleus accumbens. The associations between bone Pb level and cognitive function
25                   decrements consistently found in the few prospective and cross-sectional studies of adults
26                   without occupational Pb exposure, the coherence with animal findings, and toxicological
27                   evidence supporting modes of action but uncertainties related to potential residual
28                   confounding by age in epidemiologic studies are  sufficient to conclude that a causal
29                   relationship is likely to exist between long-term cumulative Pb exposure and cognitive
30                   function decrements in adults.
      5.3.16.8      Evidence for Psychopathological Effects in Adults

31                   Evidence indicates that a causal relationship is likely to exist between Pb exposure and
32                   psychopathological effects in adults, based on the cross-sectional associations found
33                   between concurrent blood Pb level (Bouchard et al. 2009) or bone Pb level (Rajan et al..
34                   2008) and self-reported depression, anxiety, and panic disorder in adults participating in
35                   NHANES and NAS, respectively and supporting toxicological evidence. Higher prenatal
      November 2012                             5-276                 Draft - Do Not Cite or Quote

-------
 1                   blood 5-ALA level was associated with schizophrenia in adults in California (Opler et al..
 2                   2008; 2004). but because of the lack of assessment of blood or bone Pb levels, results
 3                   were not a major consideration in conclusions. In both NHANES and NAS, the high
 4                   participation rates and examination of multiple exposures and outcomes reduces the
 5                   likelihood that findings are influenced by biased participation or reporting of symptoms
 6                   by subjects with higher Pb exposures. Depression and anxiety were assessed with widely-
 7                   used structured questionnaires such as the Profile of Mood States, but there is uncertainty
 8                   regarding the effects of Pb exposure on clinically-diagnosed conditions.

 9                   Epidemiologic evidence for associations between blood or bone Pb level and
10                   psychopathological effects is strengthened by the consideration of several potential
11                   confounding factors. Among  adults ages 20-39 years  participating in NHANES, 369
12                   adults with concurrent blood  Pb level > 2.11 (ig/dL had the highest OR for self-reported
13                   major depressive disorder (OR: 2.32 [95% CI: 1.13, 4.75]) and panic disorder (OR: 4.94
14                   [95% CI: 1.32, 18.48]) compared with the 449 adults  with blood Pb levels <0.7 (ig/dL
15                   with adjustment for age, sex,  race, education, and poverty to income ratio (Bouchard et
16                   al.. 2009). Among 526 NAS men ages 48-70 years, a  27  (ig/g increase in tibia Pb level
17                   was associated with a combined index of self-reported anxiety, depression, and phobic
18                   anxiety with an OR of 2.08 (95% CI: 1.06, 4.07) with adjustment for age, grams/day
19                   alcohol ingested, education, and employment status (Rhodes et al.. 2003). Because of the
20                   cross-sectional design of studies, the temporal sequence between Pb exposure and
21                   psychopathological symptoms in adults is uncertain. This uncertainty is somewhat
22                   reduced with results for tibia  Pb, since it is an indicator of cumulative Pb exposure. For
23                   results with blood and bone Pb level, there is uncertainty regarding the critical level,
24                   timing, frequency, and duration of Pb exposure associated with psychopathological
25                   effects.

26                   The epidemiologic evidence for Pb-associated psychopathological effects is supported by
27                   the coherence with findings in rodents that dietary prenatal/lactational or lactational Pb
28                   exposure resulted in depression-like and loss of motivation behavior in rodents, with
29                   some evidence at blood Pb levels relevant to humans  (13-17 (ig/dL) (Beaudin et al..
30                   2007; Dvatlov and Lawrence. 2002). Other studies found Pb-induced increases in
31                   depression-like behavior in animals with higher blood Pb levels (Section 5.3.5.2). Further
32                   support for Pb-associated increases in psychopathological effects in adults is provided by
33                   evidence that describes modes of action, including Pb-induced changes in the HPA axis
34                   (Section 5.3.2.3) and dopaminergic and GABAergic systems (Sections 5.2.2.2 and
35                   5.3.11.8). which are found to affect mood and emotional state.

36                   In conclusion, cross-sectional studies in a few populations demonstrate associations of
37                   higher concurrent blood or tibia Pb levels with self-reports of depression and anxiety in
      November 2012                             5-277                  Draft - Do Not Cite or Quote

-------
 1                  adults. The examination of multiple exposures and outcomes in the available studies does
 2                  not provide strong indication of biased reporting of psychopathological effects by adults
 3                  with higher Pb exposures. In adults, Pb-associated increases in depression and anxiety
 4                  were found with adjustment for age, SES, and in the NAS, daily alcohol intake. The
 5                  biological plausibility for epidemiologic evidence is provided by observations of
 6                  depression-like behavior in animals with dietary prenatal/lactational or lactational Pb
 7                  exposure, with some evidence at blood Pb levels relevant to humans and by toxicological
 8                  evidence supporting modes of action, including Pb-induced changes in the HPA axis and
 9                  dopaminergic and GABAergic systems. The associations of blood and bone Pb level with
10                  self-reported psychopathological effects found in the few studies of adults without
11                  occupational Pb exposure, the biological plausibility provided by the coherence of
12                  findings in animals and underlying modes of action, but uncertainties related to residual
13                  confounding of bone Pb results by age in epidemiologic studies are sufficient to conclude
14                  that a causal relationship is likely to exist between Pb exposure and psychopathological
15                  effects in adults.
      5.3.16.9       Evidence for Sensory Function Decrements in Adults

16                   The small body of epidemiologic and toxicological evidence is suggestive of a causal
17                   relationship between Pb exposure and sensory function decrements in adults. Key
18                   evidence in humans is provided by the recent analysis of NAS males in which a 15 (ig/g
19                   higher tibia Pb level at mean age 64.9 years was associated with a 0.05 dB/year (95% CI:
20                   0.017, 0.083) increase in hearing threshold for a pure tone average frequency (Park et al..
21                   2010). Results were adjusted for baseline age, race, education, occupational noise, BMI,
22                   pack-years smoking, noise notch, diabetes, and hypertension. Although the
23                   generalizability of results in this primarily white population of men is limited, high
24                   follow-up participation and the examination of multiple exposures and outcomes in this
25                   cohort reduces the likelihood that findings are biased by selective participation of men
26                   with higher Pb exposures. Bone Pb levels were measured up to 20 years after the initial
27                   hearing measurement; however, tibia Pb level is considered an indicator of cumulative Pb
28                   exposure since the half-life of Pb in bone is on the order of decades (Section 4.3). Bone
29                   Pb levels increase with age, and although age was included as a model covariate, residual
30                   confounding by age is possible. Supporting evidence was provided by a recent case-
31                   control study of adults attending a hospital for occupational health exams. Despite
32                   limitations of a nonrandom population and uncertain comparability of controls, the
33                   examination of multiple metals reduces the likelihood of biased participation by higher
34                   Pb exposure. Higher concurrent blood Pb level was associated with hearing loss with
      November 2012                             5-278                  Draft - Do Not Cite or Quote

-------
 1                  adjustment for several factors (Table 5-17) including blood levels of Mn, As, and Se
 2                  (Chuang et al.. 2007).

 3                  These epidemiologic findings are supported by the coherence of findings for Pb-induced
 4                  increases in hearing thresholds in animals, albeit at higher blood Pb levels than those
 5                  relevant for humans. Monkeys that were exposed to Pb in drinking water from gestation
 6                  through testing at age 13 years and had blood Pb levels 33-107 (ig/dL were found to have
 7                  elevated thresholds and increased latencies in brainstem auditory evoked potentials (Rice.
 8                  1997; Lilienthal  and Winneke. 1996). A recent study found lack of persistence of effects
 9                  in monkeys tested at age 13 years that had shorter duration exposure, gestation through
10                  age 5.5 months (Laughlin et al.. 2009).

11                  In conclusion, the evidence provided by the analysis of NAS men for associations of
12                  higher tibia Pb level with a greater rate of elevations in hearing threshold over 20 years
13                  and the biological plausibility provided by the evidence for Pb-induced decreases in
14                  auditory evoked potentials in animals but at higher blood Pb  levels than those relevant to
15                  humans, is suggestive of a causal relationship between Pb exposure and sensory function
16                  decrements in adults.
      5.3.16.10     Evidence for Neurodegenerative Diseases in Adults

17                   Epidemiologic and toxicological studies have found associations between indicators of
18                   Pb exposure and neurodegenerative diseases such as Parkinson's disease and essential
19                   tremor, but evidence is inconclusive for Alzheimer's disease and ALS. Despite the
20                   evidence for some neurodegenerative diseases, because of limitations as described below,
21                   the evidence is inadequate to determine that there is a causal relationship between Pb
22                   exposure and neurodegenerative diseases.

23                   The few case-control studies of essential tremor found higher concurrent blood Pb levels
24                   in cases than controls (Section 5.3.10.4). A common limitation of these studies was the
25                   potential for reverse causation. Reduced physical activity among cases could result in
26                   greater bone turnover and greater release of Pb from bones into blood in cases than
27                   controls. Some case-control studies found adults with Parkinson's  disease to have higher
28                   bone Pb levels, which are not likely to increase with decreases  in physical activity
29                   ("Weisskopf et al.. 2010; Coon et al.. 2006). While some of these studies of Parkinson's
30                   disease and essential tremor considered potential confounding by factors such as age, sex,
31                   race, education, and alcohol consumption, they did not consider Mn co-exposure.
32                   Epidemiologic findings for Parkinson's disease are supported by limited available mode
33                   of action evidence for Pb-induced decreased dopaminergic cell activity in the substantia
34                   nigra, which contributes to the primary symptoms of Parkinson's disease.

      November 2012                             5-279                  Draft - Do Not Cite or Quote

-------
 1                  The few case-control studies of Alzheimer's disease did not find higher prevalence of
 2                  occupational Pb exposure or higher brain Pb levels in cases but did not measure Pb in
 3                  blood or bone (Section 5.3.10.1). Toxicological studies indicated that infancy exposure
 4                  during lactation but not adult-only Pb exposures of monkeys and rats induced pathologies
 5                  that underlie Alzheimer's disease, including the formation of amyloid plaques and
 6                  neurofibrillary tangles in the brains of aged animals (Section 5.3.10.1). While these
 7                  results suggest the need to consider early-life Pb exposure in epidemiologic studies, some
 8                  indicate that effects may be attributable to the high Pb exposure concentrations tested,
 9                  i.e., producing blood Pb levels >40 (ig/dL in rats (Li et al., 2010; BashaetaL 2005).
10                  Studies of ALS have not consistently found higher blood Pb levels among ALS cases and
11                  controls (Section 5.3.10.2). and a recent study found that higher tibia and patella Pb
12                  levels were associated with longer survival time among ALS cases (Kamel et al., 2008).

13                  In conclusion, while evidence is inconclusive for ALS and Alzheimer's disease,  a few
14                  case-control studies each found higher blood Pb levels in adults with essential tremor and
15                  higher bone Pb levels in adults with Parkinson's disease. Because of the inconclusive
16                  evidence for some diseases and limitations such as reverse causation for essential tremor
17                  and the lack of consideration for potential confounding by Mn exposure for both essential
18                  tremor and Parkinson's disease, the evidence is inadequate to determine that there is a
19                  causal relationship between Pb exposure and neurodegenerative diseases.
      November 2012                             5-280                  Draft - Do Not Cite or Quote

-------
Table 5-17    Summary of Evidence Supporting Nervous System Causal
                  Determinations.
Attribute in
Causal
Framework3
Key Supporting Evidence13 References'3
Pb Biomarker Levels
Associated with
Effects0
Cognitive Function Decrements in Children - Causal
Consistent           Evidence from prospective studies for
associations from     decrements in FSIQ in association
multiple, high         with prenatal, earlier childhood, peak,
quality              concurrent, lifetime average blood Pb
epidemiologic        levels and tooth Pb levels in children
studies with          ages 4-17 yr in multiple U.S. locations,
relevant blood Pb     Mexico, Europe, Australia
levels

                    Most studies adjust for confounding by
                    SES, maternal IQ and education,
                    HOME score. Several adjust for birth
                    weight, smoking. A few, nutritional
                    factors.

                    Pooled analysis of seven cohorts
                    indicates precise effect estimates,
                    -2.36 to -2.94 FSIQ points per log
                    increase in blood Pb level,  excluding
                    one study at a time

                    Meta-analyses demonstrate the
                    consistency of association

                    Evidence from prospective studies for
                    lower scores on tests of executive
                    function and academic performance in
                    association with earlier childhood or
                    lifetime average blood Pb levels or
                    tooth Pb levels in children ages 5-18
                    yrin multiple U.S. locations, U.K, New
                    Zealand. Associations less consistent
                    for learning and memory.

                    Supporting evidence from cross-
                    sectional studies of children ages 3-16
                    yr, but most did not consider potential
                    confounding by parental caregiving
                    quality. Includes large NHANES III
                    analysis.

                    Studies had population-based
                    recruitment, most with moderate to
                    high follow-up participation not
                    conditional on blood or tooth Pb level.

                    Outcomes assessed using widely-
                    used, structured questionnaires.

                    Several studies indicate supralinear
                    C-R relationship, with larger
                    decrements in cognitive function per
                    unit increase in blood Pb at lower
                    blood Pb levels in children  ages
                    5-1 Oyr
      Canfield et al. (2003a),
      Bellinger etal. (1992).
      Jusko et al. (2008).
      Dietrich et al.  (1993b),
      Schnaaset al. (2006).
      Wasserman et al. (1997),
      long et al. (1996)
      Section 5.3.2.1
      Lanphear et al. (2005)
      Pococket al. (1994).
      Schwartz (1994)

      Bellinger etal. (1991).
      Canfield et al. (2004).
      Risetal. (2004),
      Stiles and Bellinger (1993).
      Miranda et al. (2009: 2007a).
      Fergusson et al. (1997. 1993),
      Leviton et al. (1993).
      Chandramouli et al. (2009)
      Sections 5.3.2.3. 5.3.2.4. 5.3.2.5

      Surkan et al. (2007).
      Kim et al. (2009b).
      Roy etal. (2011).
      Lanphear etal. (2000),
      Froehlich et al. (2007)
      Table 5-3. Table 5-5:
      Table 5-8. Table 5-9
      Canfield et al. (2003a).
      Bellinger etal. (1992).
      Jusko et al. (2008).
      Kordasetal. (2006).
      Lanphear et al. (2005)
              Blood Pb (various
              lifestages and time
              periods): means 3-
              14ug/dL
              Blood Pb (various
              lifestages and time
              periods):
              Means 4.8-8 ug/dL,
              Groups with early
              childhood blood Pb
              >2 and >5 ug/dL
              Tooth Pb (ages 6-8 yr):
              means 3.3, 6.2 ug/g

              Concurrent blood Pb :
              Means 1.7-11.4 ug/dL,
              Groups with blood Pb
              >10 ug/dL
              Groups with peak blood
              Pb <10 ug/dL: concurrent
              mean 3.3 ug/dL, age 2
              year mean  3.8 ug/dL
November 2012
5-281
Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Epidemiologic
evidence supported
by consistent
toxicological results
with relevant
exposures
Impaired learning in juvenile and adult
animals as indicated by performance
in Y maze, Operant Schedules of
Reinforcement with relevant dietary
Pb exposure.
Impaired learning, memory, executive
function in juvenile and adult animals
as indicated  by poorer performance on
spatial delayed alternation and
discrimination reversal learning tasks
with dietary Pb exposures.
Stangleetal. (2007).
Niuetal. (2009).
Cory-Slechta et al. (2010).
Virgolini et al. (2008a).
Altmannetal. (1993).
Section 5.3.2.3

Gilbert and Rice (1987).
Rice and Gilbert (1990b).
Rice (1992b).
Rice and Gilbert (1990a),
Rice and Karpinski (1988).
Sections 5.3.2.3 and 5.3.2.4
 Blood Pb (after prenatal/
lactation, lactation only,
postlactation,
prenatal/lifetime Pb
exposure): 8-31 ug/dL
Blood Pb (after lifetime
Pb exposure after
lactation): 15-26 ug/dL
Evidence clearly
describes mode of
action

  Impaired neuron
  development
  Synaptic changes
  LTP
  Neurotransmitter
  changes
Decreased neurogenesis in
hippocampus DG, which is involved in
LTP and learning, with lactational,
postlactational (25 days), lifetime from
gestation dietary Pb exposures.
Decreased NMDAR, which is involved
in integration of new neurons into
existing neuronal pathways with
postlactational (8 weeks) and lifetime
from gestation dietary Pb exposures.
Decreased neurite outgrowth in
animals with gestational Pb exposure

Decreased synaptic development with
gestational-lactational dietary Pb
exposures.
Changes in synaptic protein
composition with gestational-
lactational Pb exposure.
Decreased ATP, AchE, which mediate
neurotransmission with gestational Pb
exposure

Decreased magnitude, increased
threshold of LTP with gestational-
lactational Pb exposure.

Decreased dopamine in substantia
nigra with gestational-lactational
dietary Pb exposure.
Increased sensitivity of dopamine
receptor with gestational-lactational,
lactational, or postlactational Pb
exposure.
Increased catecholamine transmission
in cerebral cortex, cerebellum,
hippocampus with gestational-
lactational Pb exposure.
Decreased glutamate and expression
of glutamate receptor, NMDAR in vitro
and in rats with gestational-lactational
Pb exposure.
                                     Section 5.3.11.9 and 5.3.11.10
                                                          Section 5.3.11.4
                                                          Section 5.3.11.8
                                                          Section 5.3.11.8
November 2012
                               5-282
                     Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Attention-related Behavioral Problems in Children (e.g., inattention, impulsivity, hyperactivity, ADHD)
- Causal
Consistent            Evidence from prospective studies for
associations from     inattention, impulsivity, and
multiple, high         hyperactivity in association with
quality               prenatal (maternal or cord), earlier
epidemiologic         childhood, and  lifetime avg blood Pb
studies with           and tooth Pb levels in children ages 7-
relevant blood Pb     17 yr and young adults 19-24 yr in
levels                U.S., U.K., Australia, New Zealand.
                     Most studies adjusted for SES,
                     maternal education, and parental
                     caregiving quality. Some also
                     considered parental IQ, smoking, birth
                     outcomes. A few considered
                     substance abuse, nutritional factors.
                     Studies had population-based
                     recruitment with moderate to high
                     follow-up participation not conditional
                     on blood or tooth Pb level.
                     Associations found with
                     neuropsychological tests (CPT) and
                     teacher and parent ratings using
                     widely-used, structured
                     questionnaires.

                     Associations with inattention ratings
                     inconsistent in prospective studies
                     examining children with lower blood
                     Pb levels, but children were younger,
                     <5yr

                     Supporting evidence from cross-
                     sectional studies for associations of
                     concurrent blood Pb level with
                     inattention, impulsivity, and
                     hyperactivity, and total ADHD rating in
                     children ages 8-15 yr.
                     Cross-sectional studies had less
                     extensive consideration for potential
                     confounding, particularly parental
                     caregiving quality.
                                      Risetal. (2004),
                                      Fergusson et al. (1993).
                                      Bellinger et al. (1994a).
                                      Chandramouli et al. (2009).
                                      Leviton et al. (1993)
                                      Burns et al. (1999) with the most
                                      extensive consideration for potential
                                      confounding


                                      Sections 5.3.3.1. 5.3.3.2. 5.3.3.3
                                   Blood Pb:
                                   means 6.8 ug/dL
                                   (prenatal cord),
                                   14ug/dL
                                   (lifetime avg to age 11 -
                                   13yr), Group with age
                                   30 mo >10 ug/dL


                                   Tooth Pb (age 5-8 yr)
                                   means: 3.3,  6.2, 14 ug/g
                                      Wasserman et al. (2001)
                                      Canfield et al. (2003b)
                                      Section 5.3.3.1
                                      Choetal. (2010).
                                      Nicolescu et al. (2010).
                                      Froehlichetal. (2009).
                                      Silva et al. (1988)
                                      Sections 5.3.3.1. 5.3.3.2. 5.3.3.3
                                   Concurrent blood Pb
                                   means 6.5 ug/dL
                                   Concurrent Blood Pb
                                   means 1.9-11.1 ug/dL
Epidemiologic
evidence supported
by toxicological
results with relevant
exposures
Impulsivity indicated by premature
responses, increased perseveration,
decreased pause time between events
on tests of response inhibition in
rodents and monkeys with relevant
dietary postnatal Pb exposures.
                     Increased distractibility found in adult
                     monkeys with relevant lifetime dietary
                     Pb exposure as assessed by poorer
                     performance on discrimination
                     reversal learning tests with distracting
                     stimuli.
                     Relevant postnatal dietary Pb
                     exposure not found consistently to
                     affect sustained attention in rats as
                     assessed using signal detection test
                     with distracting stimuli.
Stangleetal. (2007).
Brockel and Cory-Slechta (1998),
Rice (1985).
Brockel et al. (1999b)
Section 5.3.3.1
                                      Gilbert and Rice (1987)
                                      Section 5.3.3.1
                                      Brockel and Cory-Slechta (1999a)
                                      Stangle et al. (2007)
                                      Section 5.3.3.1
 Blood Pb: 15, 25 ug/dLin
monkeys after infancy
only exposure, 11,
29 ug/dL in rats after 40-
day postweaning
exposure, 10, 26 ug/dL
after 3, 7 mo postweaning
exposure

Blood Pb after lifetime
(after birth) exposure: 15,
25 ug/dL
                                   Blood Pb: 16,28 ug/dL
                                   after 34-day postweaning
                                   exposure, 13, 31 ug/dL
                                   after lactational exposure.
November 2012
                               5-283
                     Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Evidence clearly
describes mode of
action
Same as above for cognitive function
in children
Internalizing Behaviors in Children (e.g., withdrawn behavior, symptoms of depression, anxiety, tearfulness)
- Likely Causal
Associations found
in high-quality
epidemiologic
studies with
relevant exposures
Evidence from prospective studies for
higher ratings of internalizing
behaviors in children ages 8-13 yr in
Boston and Port Pirie cohorts in
association with tooth or lifetime
average blood Pb levels.
Results adjusted for SES, birth
outcomes, parental education. Port
Pirie also adjusted for HOME, current
maternal  psychopathological
symptoms.
Burns etal. (1999).
Bellinger etal. Q994b)
Section 5.3.5.1
Blood Pb lifetime (to age
11-13 yr) average mean:
~14ug/dL
Tooth Pb (age 6 yr) mean:
3.4 ug/g
                    Associations also found in children
                    age 4-5 yr in Yugoslavia in association
                    with lifetime average blood Pb level;
                    stronger association found for
                    delinquent behavior.
                    Results adjusted for similar covariates
                    as above plus maternal history of
                    smoking, residence type.
                    Studies had population-based
                    recruitment with moderate follow-up
                    participation. Participation not
                    conditional on tooth/blood Pb levels
                    and behavior.
                    Associations found with teacher and
                    parent ratings on widely used,
                    structured questionnaires.
                                     Wasserman et al. (2001)
                                   Blood Pb lifetime (to age
                                   4-5 yr) average mean:
                                   9.6 ug/dL
                    Cross-sectional studies found
                    associations with concurrent blood Pb
                    level but had limited consideration for
                    potential confounding and/or
                    nonrepresentative populations (e.g.,
                    prenatal drug exposure).
                                                         Section 5.3.5.1
Epidemiologic
evidence supported
by toxicological
evidence with
relevant exposures
Postnatal dietary Pb exposure
increased emotionality, loss of
motivation in response to reward
omission in juvenile female rats
Postnatal dietary Pb exposure
increased sickness behavior due to
bacteria infection in juvenile mice.
No specific mode of action examined
with Pb exposure
Stangleetal. (2007).
Beaudinetal. (2007)
Dyatlov and Lawrence (2002)
Section 5.3.5.2
Blood Pb at PND45 after
PND1-PND30 exposure:
13, 31 ug/dL
Blood Pbat PND22 after
PND1-PND22 exposure:
17ug/dL
November 2012
                              5-284
                     Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Conduct Problems in Children and Young Adults (e.g., criminal offenses, delinquent behavior, aggression, antisocial
behavior)
 - Likely Causal
Consistent results
from high-quality
epidemiologic
studies with
relevant blood or
tooth Pb levels
Evidence from prospective studies for
criminal offenses in young adults ages
19-24 yr in Cincinnati and New
Zealand in association with earlier
childhood blood or tooth Pb levels.

Evidence from prospective studies for
higher parent and teacher ratings of
aggression, antisocial behavior,
delinquent behavior in  children ages 8-
17 yr in  U.S.,  U.K., Australia in
association with earlier childhood or
lifetime average blood  Pb and tooth
Pb levels.

Associations also found in children
age 4-5 yr in Yugoslavia in association
with  lifetime average blood Pb level;
stronger association found for
delinquent behavior.

Studies had moderate  to high follow-
up participation, not conditional on
blood Pb level.
Most studies considered potential
confounding by SES, parental
education and IQ, other SES factors,
parental caregiving/functioning,
smoking, substance abuse.

Supporting evidence for parental
report of conduct disorder in
association with concurrent blood Pb
in cross-sectional study of children
ages 8-15 yr participating in
NHNANES. Examination of multiple
exposures and outcomes reduces
likelihood of selection bias.

Consistency supported by meta-
analysis indicating similar effect
estimates by study design, potential
confounding factors considered

Teacher and parental ratings derived
from widely-used, structured
questionnaires.
Wright et al. (2008).
Fergusson et al. (2008),
Section 5.3.4.1
                                                          Dietrich et al. (2001).
                                                          Burns etal. (1999).
                                                          Chandramouli et al. (2009).
                                                          Bellinger etal. (1994b),
                                                          Section 5.3.4.1
                                                          Wasserman et al. (2001)
                                                          Braun et al. (2008)
                                                          Section 5.3.4.1
                                                          Marcus et al. (2010)
                                                          Section 5.3.4.1
Age 6 yr blood Pb mean
6.8 ug/dL
Tooth Pb (age 6-8 yr)
mean: 6.2 ug/g

Blood Pb:  lifetime (to age
11-13 yr) avg mean:
14 ug/dL, age 30 month
group with blood Pb
>10ug/dL
Tooth Pb (age 6 yr) mean
3.4 ug/g

Mean lifetime (to age 4-5
yr) average: 9.6 ug/dL
                                   Groups with concurrent
                                   blood Pb >0.8 ug/dL.
                                   Blood Pb range of study
                                   means (concurrent or
                                   lifetime avg): 1.0-26 ug/dL
Inconsistent
evidence in animals
for aggression at
relevant exposures
Aggression observed in adult
hamsters with gestational-lifetime
dietary Pb exposure. Other evidence
in adult animals with similar duration
exposure inconsistent.
Aggression generally not found in
juvenile animals with lactational Pb
exposure.
No specific mode of action examined
with Pb exposure
Delville (1999)
Section 5.3.4.2
Blood Pb after
gestational-lifetime
exposure: 10-15 ug/dL
November 2012
                               5-285
                     Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Sensory Function Decrements in Children - Likely Causal
Consistent findings   Prospective study indicated
from high-quality      associations of prenatal (maternal),
epidemiologic        neonatal, yearly age 1 to 5 yr, lifetime
studies with          avg blood Pb levels with poorer
relevant blood Pb     auditory processing in children at age
levels               Syr in Cincinnati.
                    Information was not provided on
                    participation rates.
                    Results were adjusted for SES,
                    HOME, birth weight, gestational age,
                    obstetrical complication, maternal
                    smoking. Several other factors
                    considered.

                    Supporting evidence from cross-
                    sectional studies for increased hearing
                    thresholds in children ages 4-19 yr
                    participating in  NHANES and HHANES
                    in association with  higher concurrent
                    blood Pb levels.
                    Examination of multiple exposures and
                    outcomes in NHANES and HHANES
                    reduces likelihood of selection bias.
                    Studies considered potential
                    confounding by age, sex,  race,
                    income, parental education, nutritional
                    factors.
                                     Dietrich et al. (1992)
                                     Section 5.3.7.1
                                     Schwartz and Otto (1991. 1987)
                                     Section 5.3.7.1
                                  Blood Pb means:
                                  neonatal (10 day)
                                  4.8 ug/dL, yearly age 1 to
                                  5 year 10.6-17.2 ug/dL,
                                  lifetime (to age 5 yr) avg
                                  NR
                                  Concurrent blood Pb
                                  median in HHANES:
                                  8 ug/dL, NHANES: NR
Limited
toxicological results
at relevant
exposures
Increased hearing thresholds in
monkeys age 13 years with lifetime
dietary Pb exposure.
Supernormal or subnormal retinal
ERGs in rats depending on timing and
dose of Pb exposure. Uncertain
biological relevance.
Rice (1997).
Section 5.3.7.3
                                                         Foxetal. (1997).
                                                         Section 5.3.7.3
Blood Pb after lifetime
(from birth) exposure: 33-
170ug/dL

Blood Pb: 12, 24 ug/dL
with gestational-
lactational exposure,
19 ug/dL with lactational
exposure
Evidence describes
mode of action
Decreased auditory evoked potentials
with lifetime Pb exposure of monkeys
ages 13 years.
Rod cell proliferation, retinal cell
apoptosis, dopamine changes
                                                         Section 5.3.7.3
November 2012
                              5-286
                    Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Motor Function Decrements in Children - Likely Causal
Consistent findings   Evidence from prospective studies for
from high-quality      fine and gross motor function
epidemiologic        decrements in children ages 4.5-17 yr
studies with          in Cincinnati, Yugoslavia in association
relevant blood Pb     with neonatal, earlier childhood,
levels               concurrent, lifetime avg blood Pb
                    levels.
                    High follow-up participation, no
                    selective attrition in Cincinnati cohort,
                    higher loss-to-follow-up in Yugoslavia
                    cohort with lower maternal IQ, HOME.
                    Both studies adjusted for maternal IQ,
                    parental education, SES, HOME score

                    Supporting evidence from cross-
                    sectional studies in children ages 3-7
                    yr in  India, Canada. In Inuit Canadian
                    children, potential confounding  factors
                    varied by outcome but included
                    HOME, maternal education, weight,
                    prenatal alcohol exposure, Hg,
                    polychlorinated biphenyls.

                    No decrease in motor function found in
                    children ages 6-10 yr in New England
                    with lower concurrent blood Pb levels
                    with adjustment for age,  race, sex,
                    caregiver education, SES.
                    Studies used various, widely-used
                    tests to assess outcomes.
                                    Risetal. (2004).
                                    Dietrich et al. Q993a),
                                    Bhattacharya et al. (2006),
                                    Wasserman et al. (2000)


                                    Section 5.3.8
                                  Cincinnati: blood Pb
                                  means: neonatal
                                  4.8 ug/dL, age 6 yr
                                  11.6 ug/dL, lifetime (to
                                  age 15-17 yr) avg
                                  12.3ug/dL
                                  Yugoslavia: NR
                                    Despres et al. (2005).
                                    Palaniappan et al. (2011)


                                    Section 5.3.8
                                    Surkan et al. (2007)
                                  Concurrent blood Pb
                                  means: 4.1 ug/dL
                                  Canada, 11.5 ug/dL India
                                  Concurrent blood Pb
                                  mean: 2.2 ug/dL
Limited
toxicological
evidence at
relevant exposures
Poorer balance (fell off rotarod more
quickly) in adult mice with gestational-
lactation dietary Pb exposure
Leasure et al. (2008)
Section 5.3.8
Peak blood Pb after
gestational -lactational
exposure ~10 ug/dL
                    postnatal exposures
                                                                       postnatal Pb exposure
                                                                       >60 ug/dL
November 2012
                              5-287
                    Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Cognitive Function Decrements in Adults - Likely Causal
Consistent results    Prospective analyses in MAS cohort of
from high-quality     white men and BMS cohort of men
epidemiologic        and women of diverse ethnicities
studies with          found cognitive function decrements
relevant bone Pb     over 2 to 4 years in association with
levels               bone Pb levels.
                    Baseline participation rates differed
                    but high follow-up participation, not
                    conditional on bone Pb levels.
                    Different potential confounding factors
                    considered. Adjustment for age and
                    education in both cohorts, SES in
                    BMS, smoking and alcohol use in
                    MAS.

                    Supporting cross-sectional evidence
                    from MAS, BMS,  and also women  in
                    Nurses' Health Study with adjustment
                    for additional potential confounding
                    factors,  including dietary factors,
                    medications, physical activity,
                    comorbid conditions.

                    Associations with blood Pb level found
                    in men and women  participating in
                    NHANES in certain genetic variant
                    groups with adjustment for age, sex,
                    education, income, race/ethnicity,
                    alcohol use, computer/video game
                    familiarity.

                    Several studies found associations
                    with  blood Pb levels and bone Pb
                    levels in former and current
                    Pb-exposed workers. Most studies
                    adjusted for age and education. Some
                    also  adjusted for depression and/or
                    alcohol use, but none considered
                    other occupational exposures.

                    Outcomes assessed using various but
                    widely used, structured instruments.

                    Uncertainty regarding potential
                    residual confounding  of bone Pb
                    results by age.
                                     Weisskopfetal. (2007b),
                                     Bandeen-Roche et al. (2009)
                                     Table 5-10 and Section 5.3.2.7
                                     Nurses Healthy Study:
                                     Weuve et al. (2009)
                                     Section 5.3.2.7
                                     Table 5-10
                                     Krieg et al. (2009).
                                     Krieg et al. (2010).
                                     Krieg and Butler (2009)
                                     Section 5.3.2.7
                                     Khaliletal. (2009a).
                                     Dorsey et al. (2006).
                                     Bleeckeret al. (2007a)
                                     Stewart et al. (2002)
                                     Section 5.3.2.7
                                   Baseline tibia Pb means:
                                   18.8, 20 ug/g, patella
                                   mean 25 ug/g
                                   Concurrent tibia Pb
                                   Mean: 10.5 ug/g
                                   Concurrent blood Pb
                                   means: 3-4 ug/dL
                                   Concurrent blood Pb: 12
                                   (former workers)-
                                   31 ug/dL.
                                   Peak tibia Pb: mean
                                   26.2 ug/g, median
                                   57 ug/g.
Epidemiologic
evidence supported
by consistent
toxicological results
with relevant
exposures
Impaired learning, memory, and
executive function in adult monkeys
with lifetime dietary Pb exposures after
weaning.
Impaired learning in animals with
lifetime dietary Pb exposures starting
in gestation.
Rice (1992b),
Rice and Gilbert (1990a).
Rice (1990)
Section 5.3.2.3
See above for cognitive function in
children.
Blood Pb after post-
weaning exposure to age
7-1 Oyr means 19,
26 ug/dL
Evidence clearly
describes mode of
action
Same as above for cognitive function
in children
November 2012
                              5-288
                     Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Psychopathological Effects in Adults (e.g., self ratings of depression, anxiety, panic disorder)
- Likely Causal
Consistent findings
from high-quality
epidemiologic
studies with
relevant blood and
bone Pb levels
A few cross-sectional studies indicate
associations of higher concurrent
blood or tibia Pb level with increased
reporting of depression, anxiety, panic
disorder in adults without occupational
Pb exposures.
Studies examine multiple exposures
and outcomes.

Association with blood Pb level among
1,987 adults participating in NHANES;
adjustment for age, sex, race,
education, poverty to income ratio.

Associations with blood and bone Pb
level among MAS men (mostly white)
with high follow-up participation and
adjustment for education, age,
employment, pack-years smoking,
alcohol use.

Higher  ratings of disorders also in
Pb-exposed workers with higher blood
Pb levels.
Studies used widely used,  structured
instruments to assess outcomes but
not diagnosed conditions.
Bouchard et al. (2009),
Rhodes etal. (2003).
Rajanetal. (2008)
Section 5.3.6.1
                                                         Bouchard et al. (2009)
                                                         Section 5.3.6.1
                                                         Rhodes etal. (2003),
                                                         Rajanetal. (2007)
                                                         Section 5.3.6.1
                                                         Section 5.3.6.1
                                  Group with concurrent
                                  blood Pb>2.11 ug/dL
                                  Concurrent blood Pb
                                  mean: 6.3 ug/dL,
                                  concurrent tibia Pb mean:
                                  21.9 ug/g
                                                                                           Concurrent or peak blood
                                                                                           Pb: means 31 -79 ug/dL
Supporting
toxicological
evidence at
relevant Pb
exposures
Same as above for internalizing
behaviors in children
Sensory Function Decrements in Adults - Suggestive
Limited but high-
quality
epidemiologic
evidence with
relevant bone or
blood Pb levels
Prospective study indicates
associations between tibia Pb level
and faster rate of increase in hearing
threshold over 23 yr, among MAS male
adults.
Population comprises only males,
primarily white, but study examines
multiple exposures and outcomes and
has high follow-up participation.
Results adjusted for age, race,
education, BMI, pack-years smoking,
diabetes, hypertension, occupational
noise.

Supporting evidence from case-control
study finding higher blood Pb levels in
workers from various occupations with
hearing loss with adjustment for age,
smoking, alcohol consumption, years
of noise exposure, blood Mn, As, Se
Park etal. (2010)
Section 5.3.7.2
Tibia Pb mean: 22.5 ug/g,
measured near end of
follow-up
                                                         Chuang et al. (2007
                                                         Section 5.3.7.2
                                  Concurrent blood Pb
                                  mean in cases:
                                  10.7ug/dL
Supporting
toxicological
evidence with
relevant exposures
Same as above for sensory function
decrements in children, including
mode of action
November 2012
                              5-289
                    Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
Pb Biomarker Levels
Associated with
Effects0
Neurodegenerative Diseases (e.g..Alzheimer's disease, ALS, Parkinson's disease, Essential tremor)
- Inadequate
The available         The few available case-control studies
evidence is not        found higher blood or bone Pb levels
sufficiently            and history of occupational exposure
informative           in cases with Parkinson's disease and
                     essential tremor. For ALS, association
                     also found with increased survival
                     time.
                     Studies subject to selection and recall
                     bias and reverse causation (blood Pb)
                     that could produce artifactual
                     associations.
                     Some studies consider potential
                     confounding by age, smoking,
                     education, BMI, and activity levels,
                     occupational studies did  not consider
                     Mn co-exposures.
                                     Parkinson's disease:
                                     Gorell et al. (1997).
                                     Gulson et al. (1999).
                                     Tanner et al. (1989).
                                     Weisskopfetal. (2010).
                                     Coon et al. (2006)
                                     Section 5.3.10.3


                                     Essential tremor:
                                     Louis etal. (2005: 2003),
                                     Dogu et al. (2007)
                                     Section 5.3.10.4


                                     ALS:
                                     Kamel et al. (2002).
                                     Kamel et al. (2005).
                                     Vinceti et al. (1997).
                                     Fang et al. (2010)
                                     Section 5.3.10.2
                                   Parkinson's disease:
                                   groups with tibia Pb levels
                                   >15 ug/g
                                   Essential tremor:
                                   concurrent blood Pb
                                   means 3-4 ug/dL
                                                                                            ALS:
                                                                                            groups with blood Pb
                                                                                            >3 ug/dL,
                                                                                            groups with tibia Pb
                                                                                            >8 ug/g
                     Case-control studies did not show
                     associations between occupational
                     history of Pb exposure or brain Pb
                     levels and Alzheimer's disease.
                                     Graves et al. (1991).
                                     Hariguchi et al. (2001)


                                     Section 5.3.10.1
Some evidence
describes mode of
action
Amyloid plaques found in brains of
adult monkeys and rodents with
infancy-only lactational Pb exposures.

In monkeys (ages 20-23 yr), no effect
with adult-only Pb exposure

In rodents, adult blood Pb levels at
testing had  returned to baseline.

Increases in neuronal cell apoptosis
found in vitro with  Pb exposure
                                                          Section 5.3.10.1
                                                          Basha et al. (2005)
                                                          Section 5.3.10.5
                                   Blood Pb with lactational
                                   exposure: 19-26 ug/dL

                                   Blood Pb with lactational
                                   exposure: 46 ug/dL
"Described in detail in Table II of the Preamble.
bDescribes the key evidence and references contributing most heavily to causal determination. References to earlier sections
indicate where full body of evidence is described.
""Describes the blood Pb levels in children with which the evidence is substantiated and blood Pb levels in animals most relevant to
humans.
November 2012
                              5-290
                     Draft - Do Not Cite or Quote

-------
      5.4       Cardiovascular Effects
      5.4.1        Introduction

 1                  The 2006 Pb AQCD (U.S. EPA. 20061)) concluded that both epidemiologic and animal
 2                  toxicological studies support the relationship between increased Pb exposure and
 3                  increased cardiovascular effects, in particular, increased blood pressure (BP) and
 4                  increased incidence of arterial hypertension. Although fewer in number, epidemiologic
 5                  studies demonstrated associations of blood and bone Pb levels with other cardiovascular
 6                  diseases (CVDs) in adults, such as ischemic heart disease, cerebrovascular disease,
 7                  peripheral vascular disease, and CVD-related mortality. As the cardiovascular and renal
 8                  systems are intimately linked, cardiovascular effects can arise secondarily to Pb-induced
 9                  renal injury (Section 5.5). Toxicological studies also provided compelling evidence
10                  supporting the biological plausibility for Pb-associated cardiovascular effects by
11                  characterizing a number of the underlying mechanisms by which Pb exposure can lead to
12                  human cardiovascular health effects. Such studies demonstrated that the Pb content in
13                  heart tissue of animals reflects the increases in blood Pb levels (Lai et al., 1991).
14                  indicating that the cardiovascular morbidity associated with blood Pb levels may
15                  represent the effects of the bioavailable Pb in the target tissue. The strongest evidence
16                  supported the role of oxidative stress in the pathogenesis of Pb-induced hypertension.
17                  Additionally, several toxicological studies characterized other pathways or cellular,
18                  molecular, and tissue events promoting the Pb-induced increase in BP. These
19                  mechanisms included  inflammation, adrenergic and sympathetic activation, renin-
20                  angiotensin-aldosterone system (RAAS) activation, vasomodulator imbalance, and
21                  vascular cell dysfunction.

22                  With regard to the concentration-response relationship, a meta-analysis of human studies
23                  found that each doubling of blood Pb level (between 1 and >40 ug/dL measured
24                  concurrently in most studies) was associated with a 1 mmHg increase in systolic BP and
25                  a 0.6 mmHg increase in diastolic BP (Nawrot et al.. 2002). On a population-wide basis,
26                  the estimated effect size could translate into a clinically significant increase in the
27                  segment of the population with the highest BP. In a moderately-sized population, a
28                  relatively small effect size thus has important health consequences for the risk of
29                  sequelae of increased BP, such as stroke, myocardial infarction, and sudden death. It was
30                  also noted that most of the reviewed studies examining bone Pb levels, biomarkers of
31                  cumulative Pb exposure, also showed increased BP (Cheng etal.. 2001; Huetal.. 1996a)
32                  or increased hypertension with increasing bone Pb level (Lee etal.. 200 la). Across
33                  studies, over a range of bone Pb concentrations  (<1.0 to 96 ug/g), every  10  ug/g increase
34                  in bone Pb was associated with increased odds ratios of hypertension between 1.28 and


      November 2012                             5-291                  Draft - Do Not Cite or Quote

-------
 1                   1.86. Studies observed an average increase in systolic BP of-0.75 mmHg for every
 2                   10 ug/g increase in bone Pb concentration over a range of <1 to 52 ug/g.

 3                   With regard to etiologically-relevant timing of Pb exposure, toxicological evidence
 4                   demonstrated increases in BP after long-term (>4 weeks) Pb exposure. In epidemiologic
 5                   studies, cardiovascular outcomes were most often examined in cross-sectional studies
 6                   with one or a limited number of Pb biomarker measurements, so uncertainty exists as to
 7                   the specific Pb exposure level, timing, frequency, and duration that contributed to the
 8                   observed associations. While associations of adult bone Pb  (particularly tibia Pb) with
 9                   health outcomes in adults are indicative of effects related to past or cumulative exposures,
10                   interpretation of similar associations involving adult blood Pb levels, especially those
11                   measured concurrently with outcomes, is complicated by the higher past exposures
12                   generally observed in U.S. adults populations. Detailed interpretation of Pb in blood and
13                   bone are provided in Sections 4.3 and 4.7.3. Briefly, higher past Pb exposures  in adults
14                   increased their bone Pb stores which contribute to current blood Pb levels through the
15                   normal process of bone remodeling, as well as periods of increased bone remodeling and
16                   loss (e.g., osteoporosis and pregnancy). Due to the long latency period for the
17                   development of increased BP and CVD, associations of cardiovascular effects with low
18                   concurrent blood Pb levels (e.g., population means 1.6-4 ug/dL) in adults may be
19                   influenced by higher past Pb exposures (Section 4.4.1).

20                   Past air Pb concentration and blood Pb data provide context for the cardiovascular
21                   studies. Section 3.2 notes that the peak U.S. use of Pb anti-knock additives in automobile
22                   gasoline occurred between 1968 and 1972 and was finally banned from use in 1996.
23                   Section 3.5 shows that air Pb measured at trends monitors across the U.S. decreased from
24                   1.3 (ig/m3 in 1980 to 0.14 (ig/m3 in 2010. Many of the monitors reporting to the trends
25                   network were more recently influenced only by Pb sources; the mean 2010 3-month
26                   rolling average for non-source monitors was an order of magnitude lower than the 2010
27                   trends site  average. Collective review of blood Pb studies from the late 1960s and 1970s,
28                   including NHANES II (1976-1980)  suggest that blood Pb levels ranged from roughly 10
29                   to 30 (ig/dL (Pirkle et al.. 1994; Billick et al.. 1979; Tepper and Levin. 1975; Fine et al..
30                   1972).
      November 2012                             5-292                  Draft - Do Not Cite or Quote

-------
 1                  This section reviews the published studies pertaining to the cardiovascular effects of Pb
 2                  exposure in humans, experimental animals, isolated vascular tissues, and cultured
 3                  vascular cells. With the large and strong existing body of evidence serving as the
 4                  foundation, emphasis was placed on studies published since the 2006 Pb AQCD.
 5                  Epidemiologic and toxicological studies continued to augment the evidence for increases
 6                  in BP and hypertension development associated with long-term Pb exposure and
 7                  expanded the evidence for the biological pathways of these effects. Epidemiologic studies
 8                  strengthened the evidence for associations between Pb biomarkers and cardiovascular
 9                  effects after adjusting for potential confounding factors such as age, SES, diet, alcohol
10                  use, BMI, comorbidities, and smoking. Emphasis was placed on studies that had
11                  extensive consideration for confounding and prospective study designs. The
12                  epidemiologic evidence was substantiated with results from several available prospective
13                  studies demonstrating the directionality of effects by indicating associations between Pb
14                  biomarkers and the subsequent incidence of cardiovascular health effects.
      5.4.2        Blood Pressure and Hypertension
                     5.4.2.1      Epidemiology

15                   The most commonly used indicator of cardiovascular morbidity was increased BP and its
16                   derived index, hypertension. Hypertension in these studies was defined as diastolic and/or
17                   systolic BP above certain cut-points or use of anti-hypertensive medicines. The BP cut-
18                   points were established by reference to informed medical opinion, but BP cut-points
19                   defining hypertension have been lowered over time, as medical knowledge has improved.
20                   Consequently, different studies using "hypertension" as a cardiovascular outcome may
21                   have assigned different cut-points, depending on the year and location of the study and
22                   the individual investigator. All of the  recent studies in the current review used the same
23                   criteria for hypertension (e.g., systolic BP at or above 140, diastolic BP at or above 90, or
24                   use of anti-hypertensive medications). Studies in the medical literature show that elevated
25                   BP is associated with increased risk of CVD including coronary disease,  stroke,
26                   peripheral artery disease, and cardiac  failure. Coronary disease (i.e., myocardial
27                   infarction, angina pectoris, and sudden death) is the most lethal sequelae  of hypertension
28                   (Ingelsson et al., 2008; Chobanian et al., 2003; Pastor-Barriuso et al., 2003; Prospective
29                   Studies Collaboration. 2002; Kannel.  2000a. b; Neatonetal. 1995).

30                   Earlier, U.S. EPA (1990a) reviewed the then available studies examining Pb exposure
31                   and BP and hypertension outcomes which included evaluation of several studies
32                   conducting analysis of the data in NHANES II (1976-80). They noted that across a range
      November 2012                             5-293                  Draft - Do Not Cite or Quote

-------
 1                   of 7 to 34 (ig/dL, no evident threshold was found below which blood Pb was not
 2                   significantly related to blood pressure. U.S. EPA Q990a) concluded that a small but
 3                   positive association exists between blood Pb levels and increases in blood pressure.
 4                   Quantitatively, the relationship appears to hold across a wide range of blood Pb values
 5                   and, furthermore, an estimated mean increase of about 1.5-3.0 mmHg in systolic blood
 6                   pressure appears to occur for every doubling of blood Pb concentration in adult males and
 7                   something less than 1.0-2.0 mmHg for adult females. U.S. EPA Q990a) further
 8                   concluded that the plausibility of these relationships observed in epidemiologic studies of
 9                   human populations being of a  causal nature is supported by controlled experimental
10                   animal studies demonstrating increased blood pressure clearly attributable to Pb.
11                   Subsequently, the 2006 Pb AQCD (U.S. EPA. 2006b) reviewed the literature examining
12                   Pb exposure and effects of BP and hypertension, published after the 1990 document as
13                   discussed in Section 5.4.1.

14                   Several recent general population and occupational cohort and cross-sectional studies
15                   strengthened the evidence for associations of blood and bone Pb levels with measures of
16                   BP (Figure 5-18 and Table 5-18) and with the prevalence and incidence of hypertension
17                   (Figure 5-19 and Table 5-19).  Further, recent studies expanded evidence, finding
18                   differences in association among racial/ethnic groups, perceived stress, diet, and genetic
19                   variants, and thus, identified populations potentially at increased risk of Pb-associated
20                   cardiovascular effects.

21                   In a cross-sectional analysis, Martin et al. (2006) examined the associations of concurrent
22                   blood and tibia Pb levels with  BP and hypertension in a large, community-based study of
23                   older adults (n = 964, age ranging from 50 to 70 years) in Baltimore, MD. Although
24                   cross-sectional in design, a key strength of this study was the extensive consideration of
25                   potential confounding variables. Four models evaluated associations for BP and
26                   hypertension. The base model  included age, sex, BMI, sodium intake, potassium intake,
27                   total cholesterol, time of day, testing technician, and hypertensive medication use. Other
28                   models added SES, race/ethnicity, or both as covariates.  Blood Pb but not tibia Pb level
29                   was a strong predictor of BP in all models; a 1 (ig/dL increase in concurrent blood Pb
30                   level was associated with an approximately 1 mmHg increase in systolic BP and an
31                   approximately 0.5 mmHg increase in diastolic BP. Tibia Pb but not blood Pb was
32                   associated with hypertension in logistic regression models. The authors applied
33                   propensity analysis to their models to better account for the effect of other risk factors for
34                   hypertension such as race/ethnicity, age, and SES that were strongly associated with tibia
35                   Pb level. The propensity score analysis and model adjustment did not substantially
36                   change the numerical findings and conclusions (e.g., tibia Pb and hypertension were
37                   positively associated independently of race/ethnicity and SES), indicating that neither
38                   SES nor race/ethnicity confounded the association between tibia Pb level and
      November 2012                             5-294                  Draft - Do Not Cite or Quote

-------
1

2
3

4
5

6
7
hypertension. No evidence for effect modification by race/ethnicity was found either.
Martin et al. (2006) concluded that Pb in blood has a short term effect on BP and that Pb
contributes to hypertension risk as a function of cumulative, chronic exposure (as
represented as bone Pb in this population). While different aspects of Pb exposure may
contribute  differentially to  increases in BP and hypertension, it is important to note that
concurrent blood Pb levels in adults also reflect cumulative Pb exposure. Thus, its
association with BP may not reflect a short term effect but may also reflect an effect of
cumulative Pb exposure.
Reference
Martin etal. (2006)
Glenn etal. (2006)

Weaver etal. (2008)
Scinicariello etal. (2010)

Martin etal. (2006)
Glenn etal. (2006)
Peters etal. (2007)

Elmarsafawy etal. (2006)

Weaver etal. (2008)
Peters etal. (2007)
Population
Adults, Baltimore, MD
Korean Pb Workers

Korean Pb Workers
NHANES III - Blacks
NHANES III - Mexicans
Adults, Baltimore, MD
Korean Pb Workers
NASmen, High Stress
NASmen, Low Stress
NASmen, High calcium
NASmen, Low calcium
Korean Pb Workers
NASmen, High Stress
Pb Distribution a
2.9(2.0,4.4)


27.2(19.3,28.3)
1.4(0.6,3.6)
2.0(1.0,3.9)
15.7(10.5,23.5)

18.1(12.2,26.9)
18.1(12.2,26.9)
21.6(12.0)
21.6(12.0)
74.3(67.3,82.0)
26.9(18.4,39.3)
PbBiomarker SBP
Blood Pb — . —
Blood Pb (concurrent) ™«
Blood Pb (longitudinal) — .... —
Blood Pb — .,—
In Blood Pb — 	 —
In Blood Pb — —
Tibia Pb — ' —
Tibia Pb (historical) — —
Tibia Pb ' —' • —
Tibia Pb — '- —
Tibia Pb — '.—
Tibia Pb . — ' —
Patella Pb 	
Patella Pb • — - —
              Martin etal. (2006)
             Scinicariello etal. (2010)
              Martin etal. (2006)
             Perlstein etal. (2007)
              Zhang etal. (2010)
             Adults, Baltimore, MD
             NHANES III-Whites
             NHANES III - Blacks
            NHANES III - Mexicans
             Adults, Baltimore, MD
                NAS men
                NAS men
                N AS men
                N AS men
                N AS men
                N AS men
                N AS men
                N AS men
           NASmen, HFE Wild-type
            NASmen, HFE H63D
            NASmen, HFE C282Y
           NAS men, Any HFE variant
           NASmen, HFE Wild-type
            NASmen, HFE H63D
            NASmen, HFE C282Y
           NAS men, Any HFE variant
 2.9(2.0,4.4)
 1.6(0.8,3.3)
 1.4(0.6,3.6)
 2.0(1.0,3.9)
15.7(10.5,23.5)
  12.4(4.4)
  7.4 (0.6)
  5.4 (0.5)
  3.9(0.3)
  40.9(14)
  29.4(2.2)
  18.9(1.4)
  14.1(1.4)
  18(12,27)
  19(14,26)
  20(14,27)
  19(14,27)
  26(17,34)
  27(19,37)
  25(17,37)
  26(18,37)
 Blood Pb
In Blood Pb
In Blood Pb
In Blood Pb
 Tibia Pb
 Blood Pb
 Blood Pb
 Blood Pb
 Blood Pb
 Tibia Pb
 Tibia Pb
 Tibia Pb
 Tibia Pb
 Tibia Pb
 Tibia Pb
 Tibia Pb
 Tibia Pb
Patella Pb
Patella Pb
Patella Pb
Patella Pb
                                                                                                         DBF
                                                                                                          PP
                                                                                5-6-4-20    2    4    6    £

                                                                                      Change in BP (mmHg 95% Cl) per
                                                                                 lu.g/dL increase in blood Pb or 10 u.g/gbone Pb
     aPb distributions present the median (IQR), which were estimated from the mean and SD assuming a normal distribution.
     bEffect estimates were standardized to 1 ug/dL blood Pb or 10 |jg/g bone Pb.
     Note: In general, results are categorized by specific BP parameter, then  by Pb biomarker. For categories with multiple studies, the
     order of the studies follows the order of discussion in the text. Results display associations (95% Cl) of a 1  ug/dL increase in blood
     Pb level (closed circles) or 10 |jg/g increase in bone Pb (open circles) with systolic BP (SBP; blue), diastolic BP (DBP; red), and
     pulse pressure (PP; purple) in adults.

     Figure 5-18       Associations of blood and bone Pb levels with systolic  BP,
                           diastolic BP, and pulse pressure  in adults.
     November 2012
                                5-295
                           Draft - Do Not Cite or Quote

-------
Table 5-18  Additional characteristics and quantitative data for associations of
             blood and bone Pb with BP measures for studies presented in
             Figure 5-18.
Study
Martin et al.
(2006)




Glenn et al.
(2006)


















Weaver et
al. (2008)



Study Population
/ Methodology
Cross-sectional
964 men and
women, 50-70 yr,
40% African
American, 55%
White, 5% other, in
Baltimore, MD




Longitudinal
575 Pb exposed
workers, age 18-65
yr, in South Korea
(10/1997-6/2001)


















Cross-sectional
652 current and
former Pb workers
in South Korea
(12/1999-6/2001)
Same cohort as
Glenn et al. (2006)

Parameter Pb Data
BP Concurrent Mean
Blood Pb:
Mean (SD): 3.5
(2.3) ug/dL
African American: 3.4
(2.3)
White: 3.5 (2.4)
Tibia Pb:
Mean(SD): 18.8
(12.4) ug/g
African American:
21.5(12.6)
White: 16.7(11.9)
BP Blood Pb mean (SD):
Visit 1 : 20.3 (9.6),
Women
Visit 2: 20.8(10.8),
Women
Visits: 19.8(10.7),
Women
Visit 1: 35.0(13.5),
Men
Visit 2: 36.5 (14.2),
Men
Visit 3: 35.4 (15.9),
Men

Tibia Pb, mean (SD):
Visit 1: 28.2(19.7),
Women
Visit 2: 22.8 (20.9),
Women
Visit 1 : 41 .7 (47.6),
Men
Visit 2: 37.1 (48.1),
Men
Patella Pb, mean
(SD):
Visit 3 49.5 (38.5)
Women
Visit 3 87.7 (11 7.0)
Men
BP Concurrent Blood Pb:
Mean (SD):
30.9(16.7) ug/dL
Concurrent Patella
Pb:
Mean (SD):
75.1 (101.1) ug/g
Statistical Analysis
Extensive analysis of
potential confounding
factors. Multiple linear
regression base model
adjusted forage, sex, BMI,
a nti hypertensive medication
use, dietary sodium intake,
dietary potassium intake,
time of day, testing
technician, serum total
cholesterol. SES, race/
ethnicity also included in
models that are presented in
Figure 5-18. and tabulated
here.)
Multivariable models using
GEE were used in
longitudinal analyses.
Models were adjusted for
visit number, baseline age,
baseline age squared,
baseline lifetime alcohol
consumption, baseline body
mass index, sex, baseline
BP lowering medication use,
alcohol consumption, BMI,
sex, BP lowering medication
use.
















Linear regression model
adjusted forage, sex, BMI,
diabetes, antihypertensive
and analgesic medication
use, Pb job duration, work
status, tobacco and alcohol
use


Effect Estimate
P (95% Cl)
Blood Pb:
SBP: 1.05(0.53,1.58)
DBP: 0.53 (0.25,
0.81)
mmHg per ug/dL
blood Pb
Tibia Pb:
SBP: 0.07 (-0.05,
0.14)
DBP: 0.05 (-0.02,
0.08)
mmHg per ug/g bone
Pb
Model 1 (short-term)
Blood Pb
(longitudinal):
0.09(0.01, 0.16)
Blood Pb
(concurrent):
0.08 (-0.01, 0.1 6)
Model 4 (short and
longer-term)
Blood Pb
(longitudinal):
0.09(0.01, 0.16)
Blood Pb
(concurrent):
0.10(0.01,0.19)
mmHg per 10 ug/dL
blood Pb










SBP
Patella Pb:
0.0059 (-0.008, 0.02)a
Blood Pb:
0.1007(0.02, 0.1 8)a

mmHa oer 1 ua/dL
                                                                         blood Pb or 1 ug/g
                                                                         patella Pb
                                                                         Interaction between
                                                                         blood Pb/patella Pb
                                                                         with ALAD and
                                                                         vitamin D receptor
                                                                         polymorphisms not
                                                                         significant.
November 2012
5-296
Draft - Do Not Cite or Quote

-------

Study
Perlstein et
al. (2007)
















Peters et al.
(2007)










Elmarsafawy
et al. (2006)



Study Population
/ Methodology
Cross-sectional
593 predominantly
white men from
MAS in
Greater Boston,
MA area
(1991-1997)















Longitudinal and
Cross-sectional
513 elderly men
(mean 67 yr) from
MAS in Greater
Boston, MA area










Cross-sectional
471 elderly men
(mostly white,
mean age 67 yr)
from MAS in
Greater Boston,
MA area



Parameter Pb Data
PP Blood Pb:
Overall mean (SD):
6.12(4.03)ug/dL
Mean (SD) quintiles:
Q1: 2.3(0.8) ug/dL
Q2: 3.9 (0.3) ug/dL
Q3: 5.4 (0.5) ug/dL
Q4: 7.4 (0.6) ug/dL
Q5: 12.4(4.4) ug/dL

Tibia Pb:
Median: 19 ug/g
Mean (SD) quintiles:
Q1: 7.4 (3.2) ug/g
Q2:14.1 (1.4) ug/g
Q3: 18.9(1.4) ug/g
Q4: 24.9 (2.2) ug/g
Q5: 40.9 (14) ug/g










BP Tibia Pb:
mean (SD):
21 .5 (13.4) ug/g
Patella Pb:
Mean (SD):
31 .5 (19.3) ug/g









BP Blood Pb:
Mean (SD):
6.6 (4.3) ug/dL
Tibia Pb:
Mean (SD):
21 .6 (12.0) ug/g
Patella Pb:
Mean (SD):
31 7 (183) uq/q
\/r^yy


Statistical Analysis
BP association assessed
using Spearman correlation
coefficients.
PP association (adjusted
mean difference) assessed
using multiple linear
regression model adjusted
forage, height, race, heart
rate, waist circumference,
diabetes, family history of
hypertension, education level
achieved, smoking, alcohol
intake, fasting plasma
glucose, and ratio of total
cholesterol to HDL
cholesterol












Logistic and linear
regression models adjusted
for age, age squared,
sodium, potassium, and Ca2+
intake, family history of
hypertension, BMI,
educational level, pack-years
of smoking, alcohol
consumption, and physical
activity







Linear regression models
adjusted forage, BMI, family
history of hypertension,
history of smoking, dietary
sodium intake, and
cumulative alcohol ingestion
Lack of consideration for
potential confounding by
SES-related variables.


Effect Estimate
o in co/ f*t\
p (95% Cl)
PP
4.2(1.9, 6.5) mmHg
higher in men with
tibia Pb >19 ug/g
(median) compared
with men with tibia Pb
800 mg/day):
0.40.(0.1 1,0.70)
2+
Low Ca group
(<800 mg/day):
0.19(0.01,0.37)
mmHg per ug/g tibia
Pb
November 2012
5-297
Draft - Do Not Cite or Quote

-------
Study Population
Study / Methodology
Zhang et al. Cross-sectional
(20K)a) 61 9 older adult
males (mostly
white, mean
age 67 yr) enrolled
in the MAS in
Greater Boston,
MA area














Scinicariello Cross-sectional
etal. (2Q10) 6,016 NHANES III
(1988-1994)
participants a 17 yr






Parameter Pb Data
PP Wild type HFE
Tibia Pb:
Median (IQR):
8(1 2-27) ug/g
Patella Pb:
Median (IQR):
26(1 7-37) ug/g

C282Y HFE
Tibia Pb:
Median (IQR):
20 (14-27) ug/g
Patella Pb:
Median (IQR):
25(1 7-37) ug/g

H63D HFE
Tibia Pb:
Median (IQR):
19(1 4-26) ug/g
Patella Pb:
Median (IQR):
27(19-37)ug/g
BP Concurrent Blood Pb:
Overall Mean (SE):
2.99 (0.09) ug/dL
Non-Hispanic Whites:
2.87 (0.09)
Non-Hispanic Blacks
3.59 (0.20)
Mexican American
3.33(0.11)

Statistical Analysis
Linear mixed effects
regression models with
repeated measurements
adjusted for age; education;
alcohol intake; smoking;
daily intakes of Ca2+, sodium,
and potassium; total calories;
family history of
hypertension; diabetes;
height; heart rate; high-
density lipoprotein (HDL);
total cholesterol:HDL ratio;
and waist circumference











Multivariable linear
regression of log-
transformed blood Pb level
adjusted for age, sex,
education, smoking status,
alcohol intake, BMI, serum
creatinine levels, serum
Ca2+, glycosylated
hemoglobin, and hematocrit

Effect Estimate
o in co/ f*t\
p (95% Cl)
PP
mmHg
per 13 ug/g Tibia Pb:
Wild Type HFE:
0.38(0,1.96)
H63D HFE:
3.30(0.16, 6.46)
pOQOY HFF-
w^o^ i nrQ.
0.89 (0, 5.24)
Any HFE variant:
2.90(0.31, 5.51)

mmHci
per 19 ug/g Patella
Pb:
Wild Type HFE:
0.26(0, 1.78)
H63D HFE:
2.95 (0, 5.92)
C282Y HFE:
0.55(0, 1.66)
Any HFE variant:
2.83 (0.32,5.37)
SBP
Non-Hispanic whites:
1 .05 (0.32, 1 .78)
Non-Hispanic blacks:
2.55(1.59, 3.51)
Mexican Americans:
0.84 (-0.06, 1 .74)

DBP
                                                                                               Non-Hispanic whites:
                                                                                               -0.14 (-1.1, 0.82)
                                                                                               Non-Hispanic blacks:
                                                                                               1.99(1.13,2.85)
                                                                                               Mexican Americans:
                                                                                               0.74 (-0.005, 1.48)
                                                                                               mmHg per unit
                                                                                               increase in
                                                                                               In [Blood Pb]


                                                                                               Significant
                                                                                               interactions with
                                                                                               blood PbandALAD
                                                                                               observed in relation
                                                                                               to SBP for non-
                                                                                               Hispanic whites and
                                                                                               non-Hispanic blacks
November 2012
5-298
Draft - Do Not Cite or Quote

-------
Study Population
Study / Methodology Parameter Pb Data
Navas-Acien Longitudinal and BP
et al. (2008)" Cross-sectional
Meta-analysis of
studies using bone
Pb as an exposure
metric and BP as
the outcome
(8 studies)










Yazbeck et Cross-sectional BP Midpregnancy Blood
al. (2009)' 971 pregnant Pb:
women, age 18-45 PIH group mean
yr, (SD):
in France 2.2 (1 .4)
No PIH group mean
(SD):
1.9(1.2)





Statistical Analysis
Inverse variance weighted
random-effects meta-
analyses














Multivariable logistic
regression models adjusted
for maternal age; Cd, Mn,
and Se blood levels;
hematocrit; parity; BMI;
pregnancy weight gain;
gestational diabetes;
educational level; SES;
geographic residence; and
smoking status and alcohol
consumption before and
during pregnancy


Effect Estimate
P (95% Cl)
BP Pooled Estimates
mmHg
per 10 ug/g Tibia Pb
Prospective SBP
0.33 (-0.44, 1.11)
Cross-sectional SBP
0.26 (0.02, 0.50)
Cross-sectional DBP
0.02 (-0.15, 0.19)
Hypertension
per 10 ug/g patella
Pb
Cross-sectional
hypertension
OR: 1.04 (1.01, 1.07)
Pooled Estimate
hypertension
OR: 1.04(0.96, 1.12)
Log-transformed
blood Pb at mid-
pregnancy
SBP:
r = 0.08; p = 0.03
DBP:
r = 0.07; p = 0.03
Significant
correlations also
observed after
24 weeks of gestation
and after 36 weeks of
gestation.
a95% CIs estimated from given p-value.
bReference not included in Figure 5-18. because it is a meta-analysis.
""Reference not included in Figure 5-18. because only correlations were reported
November 2012
5-299
Draft - Do Not Cite or Quote

-------
  Martin etal. (2006)   Baltimore, MD
Elmarsafawy etal. (2006)    NAS
 Yazbecketal. (2009)
 Muntner etal. (2005)
Scinicarielloet al. (2010)  NHANES III

                  1998-1994
  Park etal. (2009)
                        Study Location    Population     Blood Pbc(ug/dL)   Comparison

                                                               perlu.g/dL
                                                               perlu.g/dL
                                                               perlu.g/dL
                                                               Reference
                                                                Q2vQl
                                                                Q3vQl
                                                                Q4vQl
                                                               Reference
                                                                Q2vQl
                                                                Q3vQl
                                                                Q4VQ1
                                                               Reference
                                                                Q2vQl
                                                                Q3vQl
                                                                Q4vQl
                                                               Reference
                                                                Q2vQl
                                                                Q3vQl
                                                                Q4vQl
                                                               Reference
                                                                Q2VQ1
                                                                Q3vQl
                                                                Q4vQl
                                                               ALAD2vlb
                                                               Reference
                                                                Q2VQ1
                                                                Q3vQl
                                                                Q4vQl
                                                               ALAD2vlb
                                                               Reference
                                                                Q2vQl
                                                                Q3vQl
                                                                Q4vQl
                                                               ALAD2vlb
                                                               perlu.g/dL
                                     White Men
                                     BlackMen
                                     White Women
                                     BlackWomen
                                      Men<50
                                      Men>50
                                     Women <50
                                     Women >50

LowCalcium
High Calcium
Pregnant Women



Non-Hispanic Whites



Non-Hispanic Blacks



Mexican Americans



Non-Hispanic Whites




Non-Hispanic Blacks




Mexican Americans




Overall
3.5(2.3)
6.6(4.3)
6.6(4.3)

1.2-1.7
1.71-2.30
>2.30
<1.06
1.06-1.63
1.63-2.47
>2.47
<1.06
1.06-1.63
1.63-2.47
>2.47
<1.06
1.06-1.63
1.63-2.47
>2.47
0.7-1.4
1.5-2.3
2.4-3.7
3.8-52.9
2.4-3.7
0.7-1.4
1.5-2.3
2.4-3.7
3.8-52.9
2.4-3.7
0.7-1.4
1.5-2.3
2.4-3.7
3.8-52.9
2.4-3.7
3.52(0.10)
      Elmarsafawy etal. (2006)
       Martin etal. (2006)     Baltimore, MD
        Petersetal. (2007)      Boston, MA
        Petersetal.(2007)     Boston, MA
      Elmarsafawy etal. (2006)   NAS
                                      LowCalcium
                                     High Calcium
                                     High Stress
High Stress
LowCalcium
High Calcium
                                           Tibia Pbd(ug/g)

                                            21.6(12.0)
                                            21.6(12.0)
                                            18.8(12.4)
                                            21.5(13.4)

                                           Patella Pb d (ug/g)

                                            31.5(19.3)
                                            31.7(18.3)
                                            31.7(18.3)
                          perlO|ag/g
                          perlO|ag/g
                          perlO|ag/g
                          perlO|ag/g
perlO|ag/g
perlO|ag/g
perlO|ag/g
                                                                                                            -O*-
                                                                                   23456

                                                                                      Odds Ratio (95% Cl)
Note: Studies are categorized by Pb biomarker. Within each category, studies generally are presented in order of discussion in the
text.
a: The outcomes plotted are hypertension prevalence with the exception of Yazbeck et al. (2009) which measured pregnancy
induced hypertension and Peters et al. (2007) which measured hypertension incidence.
b: ALAD2 vs. 1 indicates comparison between ALAD 2 carriers (e.g., ALAD1-2 and ALAD2-2) and ALAD 1 homozygotes
(e.g., ALAD1-1).
c:  Effect estimates were standardized to a 1 ug/dL increase in blood Pb (closed circles).
d:  Effect estimates were standardized to a 10 ug/g increase in bone Pb (open circles).

Figure 5-19      Odds ratios (95% Cl) for associations of blood (closed circles)
                       and  bone (open circles) Pb with  hypertension  prevalence and
                       incidence9.
November 2012
                                              5-300
                                            Draft - Do Not Cite or Quote

-------
Table 5-19  Additional characteristics and quantitative data for results presented
            in Figure 5-19 for associations of blood and bone Pb with
            hypertension measures.
Study
(same order
as in text)
Martin et al.
(2006)






Weaver et
al. (2008)a



Peters et al.
(2007)




Elmarsafawy
et al. (2006)












Study
Population and
Methodology
Cross-sectional
964 men and
women, 50-70 yr,
40% African
American, 55%
White,
5% other,
in Baltimore, MD





Cross-sectional
652 current and
former Pb
workers in
South Korea
(12/1999-6/2001)


Longitudinal
51 3 elderly men
(mean 67 yr)
from MAS in
Greater Boston,
MA area




Cross-sectional
471 elderly men
(mean 67 yr)
from MAS in
Greater Boston,
MA area












Parameter
Hypertension
(current use of
antihypertensive
medication,
mean SBP
a 140 mmHg or
DBP > 90
mmHg)





Hypertension
(mean SBP
> 140 mmHg,
DBP > 90
mmHg; and/or
use of
antihypertensive
medications; or
physician
diagnosis)
Hypertension
(mean SBP
>140 mmHg,
DBP >90
mmHg; or
physician
diagnosis)




Hypertension
(mean SBP
> 160 mmHg,
DBP > 95
mmHg; and/or
physician
diagnosis with
current use of
antihypertensive
medications)










Pb Data
Blood Pb:
Mean (SD):
3.5 (2.3) ug/dL
Tibia Pb:
Mean (SD):
18.8(1 2.4) ug/g





Blood Pb:
Mean (SD):
31.9(14.8)ug/dL
Patella Pb:
Mean (SD):
37.5 (41 .8) ug/g


Tibia Pb:
mean (SD):
21 .5 (13.4) ug/g
Patella Pb:
Mean (SD):
31 .5 (19.3) ug/g




Blood Pb:
Mean (SD):
6.6 (4.3) ug/dL
Tibia Pb:
Mean (SD):
21 .6 (12.0) ug/g

Patella Pb:
Mean (SD):
31 .7 (18.3) ug/g









Statistical Analysis
Logistic regression
models adjusted for age,
sex, BMI,
antihypertensive
medication use, dietary
sodium intake, dietary
potassium intake, time
of day, testing
technician, and serum
homocysteine
Lack of consideration for
potential confounding by
SES-related variables.
Logistic regression
models adjusted for age,
sex, BMI, diabetes,
antihypertensive and
analgesic medication
use, Pb job duration,
work status, tobacco
and alcohol use


Cox proportional
hazards models
adjusted for age, age
squared, sodium,
potassium, and Ca +
intake, family history of
hypertension, BMI,
educational level,
smoking, alcohol
consumption, baseline
SBP and DBP, and
physical activity
Logistic regression
models adjusted for age,
BMI, family history of
hypertension, history of
smoking, dietary sodium
intake, and cumulative
alcohol ingestion












Effect Estimate
(95% Cl)
Blood Pb level:
OR=1.02(0.87,
1.19)
Tibia Pb:
OR=1.24(1.05,
1.47)





Quantitative results
not reported. None
of the examined Pb
exposure metrics
(blood, patella, and
In patella) were
significantly
associated with
hypertension

Risk of
Hypertension
Incidence
High Stress
RR=2.66(1.43,
4.95) per SD
increase in tibia Pb
RR=2.64(1.42,
4.92) per SD
increase in patella
Pb
Low Ca2+ group
(<800 mg/day):
Blood Pb OR:
1.07(1.00,1.15)
Tibia Pb OR:
1.02(1.00, 1.04)
Patella Pb OR:
1.01 (1.00,1.03)

High Ca2+ group
(>800 mg/day):
Blood PbOR:
1.03(0.97,1.11)
Tibia Pb OR:
1.01 (0.97,1.04)
Patella Pb OR:
1.01 (0.99, 1.03)
Per ug/dL blood Pb
or ug/g tibia or
patella Pb
November 2012
5-301
Draft - Do Not Cite or Quote

-------
   Study
 (same order
  as in text)
     Study
 Population and
 Methodology
  Parameter
       Pb Data
  Statistical Analysis
                                                                   Effect Estimate
                                                                      (95% Cl)
Yazbeck et
al. (2QQ9)a

Cross-sectional
971 pregnant
women, age
18-45yr, in
France

PIH
(SBP> 140
mmHg or DBP
> 90 mmHg
after the
22nd week of
gestation)

Blood Pb:
PIH group mean (SD):
2.2 (1 .4) ug/dL
No PIH group mean
(SD):
1 .9 (1 .2) ug/dL
Q1: <1.20 ug/dL
Q2: 1.20-1.70 ug/dL
Q3: 1.71-2.30 ua/dL
Multivariable logistic
regression models
adjusted for maternal
age, Cd, Mn, and Se
blood levels, parity,
hematocrit, BMI,
gestational diabetes,
educational levels, SES,
geographic residence,
and smoking status
during pregnancy
PIH
Blood Pb
OR=
3.29(1.11,9.74)
per 1 unit increase
in log maternal
blood Pb level
Q1 : Reference
arouo
                                                 Q4: >2.30 ug/dL
                                                                                   Q2: OR 1.84 (0.77,
                                                                                   4.41)
                                                                                   Q3: OR=2.07 (0.83,
                                                                                   5.13)
                                                                                   Q4: OR=2.56 (1.05,
                                                                                   6.22)
Muntner et
al. (2005)
Cross-sectional
9,961  NHANES
(1999-2002)
participants
Hypertension
(current use of
antihypertensive
medication,
SBP> 140
mmHg, or DBP
> 90 mmHg)
Concurrent Blood Pb:
Overall Mean (Cl):
1.64 (1.59-1.68) ug/dL
quartile 1:
<1.06 ug/dL,
quartile 2:
1.06-1.63 ug/dL,
quartile 3:
1.63-2.47 ug/dL, and
quartile 4:
> 2.47 ug/dL
Multivariable logistic
regression models
adjusted forage, sex,
diabetes mellitus, BMI,
cigarette smoking,
alcohol consumption,
high school education,
and health insurance
status
                                                                         Monotonic increase in
                                                                         OR across blood Pb
                                                                         level groups.
Non-Hispanic white:
Q1: Reference
group
Q2: OR=1.12(0.83,
1.50)
Q3:OR=1.03(0.78,
1.37)
Q4: OR=1.10(0.87,
1.41)


Non-Hispanic black
Q1: Reference
group
Q2:OR=1.03(0.63,
1.67)
Q3: OR=1.12(0.77,
1.64)
Q4: OR=1.44 (0.89,
2.32)
                                                                                                 Mexican American
                                                                                                 Q1:  Reference
                                                                                                 group
                                                                                                 Q2:  OR=1.42 (0.75,
                                                                                                 2.71)
                                                                                                 Q2:  OR=1.48 (0.89,
                                                                                                 2.48)
                                                                                                 Q3:OR=1.54(0.99,
                                                                                                 2.39)
                                                                                                 p for trend=0.04
November 2012
                                    5-302
                                             Draft - Do Not Cite or Quote

-------
   Study
 (same order
  as in text)
     Study
 Population and
 Methodology
  Parameter
       Pb Data
  Statistical Analysis
                                                                   Effect Estimate
                                                                      (95% Cl)
Scinicariello
et al. (2010)
Cross-sectional
6,016 NHANES
111(1988-1994)
participants
> 17yr
Hypertension
(current use of
antihypertensive
medication,
SBP> 140
mmHg, or DBP
> 90 mmHg)
Concurrent Blood Pb:
Mean (SE):
 2.99 (0.09) ug/dL
Q1:0.7-1.4ug/dL,
Q2: 1.5-2.3 ug/dL,
Q3: 2.4-3.7 ug/dL,
Q4: 3.8-52.9 ug/dL
                                                Non-Hispanic Whites:
                                                2.87 (0.09)
                                                Non-Hispanic Blacks:
                                                3.59 (0.20)
                                                Mexican American:
                                                3.33(0.11)
Multivariable logistic
regression model
adjusted for
race/ethnicity, age, sex,
education, smoking
status, alcohol intake,
BMI, serum creatinine
levels, serum Ca2+,
glycosylated
hemoglobin, and
hematocrit
Non-Hispanic
whites:
Q1: Reference
group
Q2: POR=1.21
(0.66, 2.24)
Q3: POR=1.57
(0.88, 2.80)
Q4: POR=1.52
(0.80, 2.88)
ALAD1-2/2-2:
POR= 0.76 (0.17,
3.50)
ALAD-1: Reference
group
                                                                                                Non-Hispanic
                                                                                                blacks:
                                                                                                Q1: Reference
                                                                                                Q2: POR=1.83
                                                                                                (1.08, 3.09)
                                                                                                Q3: POR=2.38
                                                                                                (1.40, 4.06)
                                                                                                Q4: POR=2.92
                                                                                                (1.58,5.41)
                                                                                                ALAD1-2/2-2:
                                                                                                POR= 3.40 (0.05,
                                                                                                219.03)
                                                                                                ALAD-1: Reference
                                                                                                group


                                                                                                Mexican Americans:
                                                                                                Q1: Reference
                                                                                                Q2: POR=0.74
                                                                                                (0.24, 2.23)
                                                                                                Q3: POR=1.43
                                                                                                (0.61, 3.38)
                                                                                                Q4: POR=1.27
                                                                                                (0.59, 2.75)
                                                                                                ALAD1-2/2-2:
                                                                                                POR= 0.49 (0.08,
                                                                                                3.20)
                                                                                                ALAD-1: Reference
                                                                                                group


                                                                                                POR for
                                                                                                hypertension with
                                                                                                ALAD2 carriers
                                                                                                across quartiles of
                                                                                                blood Pb level also
                                                                                                reported. ALAD2
                                                                                                carriers associated
                                                                                                with hypertension in
                                                                                                non-Hispanic
                                                                                                whites.
November 2012
                                    5-303
                                             Draft - Do Not Cite or Quote

-------
Study
(same order
as in text)
Parket al.
(2QQ9c)
















Study
Population and
Methodology Parameter
Cross-sectional Hypertension
12,500 NHANES
111(1988-1994)
participants
















Pb Data
NHANES III Concurrent
Blood Pb Mean and SE
3.52(0.10)
White men
<50yr4.02(0.16)
> 50 yr 4.92 (0.1 8)
Rlspk' mpn
DlaUt\ 1 1 Id 1
<50yr4.55(0.15)
> 50 yr 7.57 (0.22)
White women
<50 yr 2.09 (0.07)
> 50 yr 3.53 (0.1 2)
Black women
<50 yr 2.52 (0.09)
> 50 yr 4.49 (0.1 6)





Statistical Analysis
Logistic regression
models adjusted for age,
education, smoking
status, cigarette
smoking, BMI,
hematocrit, alcohol
consumption, physical
activity, antihypertensive
medication use, and
diagnosis of type-2
HJ3 hpfpQ
UICI UClCO











Effect Estimate
(95% Cl)
OR per SD
(0.75 ug/dL) in log
blood Pb:
Overall: 1.12(1.03,
1.23).

White men:
1.06(0.92,1.22)
Black men:
1.17(0.98, 1.38)
White women:
1.16(1.04,1.29)
Black women:
1.19(1.04, 1.38)
Men <50 yr:
0.98 (0.80, 1 .22)
Men >50 yr:
1.20(1.02,1.41)
Women <50 yr:
1 .23 (1 .04, 1 .46)
Women >50 yr:
1 .09 (0.94, 1 .26)
      aNot included in Figure 5-19 because OR data were not reported.
 1
 2
 o
 5
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18

19
20
21
22
In an occupational cohort in South Korea, Glenn et al. (2006) simultaneously modeled
multiple Pb dose measures of individuals collected repeatedly over four years of follow
up. Thus, through the assessment of cross-sectional and longitudinal relationships with
BP, this study provided key insight on potentially important time periods of Pb exposure
and also informed the directionality of association. The initial blood Pb level  was used as
a baseline covariate and the difference in blood Pb level between visits was computed for
each subsequent visit. The bone Pb measures (tibia Pb at visits  1 and 2, patella Pb at visit
3) were used to indicate historical exposure and cumulative dose. Four models were
specified:  Model 1 was conceptualized to reflect short-term changes in BP associated
with recent dose; Model 2 to reflect longer-term changes associated with cumulative dose
controlling for the association of baseline BP with recent dose; Model 3 to reflect longer-
term changes associated with cumulative dose controlling for cross-sectional  influence of
cumulative dose on baseline BP; and Model 4 to reflect both short-term change with
recent dose and longer-term change with cumulative dose. Concurrent blood Pb and
increases in blood Pb between visits were associated with increases in systolic BP in
Model 1 (short-term dose) and Model 4 (short- and longer-term dose). No association
was observed between BP and tibia Pb at baseline while higher tibia Pb was associated
with a decrease in systolic BP in each of the models.

Glenn et al. (2006) was strengthened by the analysis of associations between changes in
blood Pb and changes in BP overtime within individual subjects. These results indicate
that circulating Pb (e.g., blood Pb) may act continuously on systolic  BP and reduction in
blood Pb may contribute to reductions in BP, while cumulative Pb exposure (represented
      November 2012
                                                  Draft - Do Not Cite or Quote

-------
 1                  by bone Pb in this study) may contribute to hypertension incidence by different
 2                  mechanisms over longer time periods and in older subjects. This analysis in relatively
 3                  young subjects (mean [SD] age at baseline 41.4 [9.5] years) with a low prevalence of
 4                  hypertension suggests that at least one of the biological pathways that influences how
 5                  systolic BP responds to Pb operates over a relatively rapid timeframe. This may reflect an
 6                  immediate response to Pb at a biochemical site of action as a consequence of the
 7                  biologically available Pb circulating in blood. A persistent effect of cumulative doses
 8                  over a lifetime may occur via other mechanisms. Bone Pb level may exert influence on
 9                  blood Pb levels and consequently on BP in an aging population with prolonged Pb
10                  exposure. Thus, the findings contribute important information regarding the various short
11                  and long-term exposure relationships with increases in BP and hypertension. It is
12                  important to acknowledge the uncertainty regarding the applicability of these findings
13                  regarding short-term and long-term effects in Pb workers with relatively high current Pb
14                  exposures contributing to blood Pb levels (mean blood Pb levels over time: 20-37 (ig/dL)
15                  to adults in the U.S. general population whose concurrent blood Pb levels are influenced
16                  more by Pb mobilized from bone stores. Further, for bone Pb analysis, the potential  for
17                  bone Pb BP and hypertension findings in older populations to be impacted by residual
18                  confounding by age may be a factor to consider since exposure studies of older cohorts
19                  (NAS/ mean age >60 years; (Wilker et al.. 2011; Kimetal. 1997)) indicate that bone Pb
20                  is correlated with age.

21                  In a separate cross-sectional analysis of the same occupationally exposed group in year
22                  three of follow-up, Weaver et al. (2008) examined associations of concurrent patella Pb
23                  and blood Pb level with systolic BP, diastolic BP, and hypertension and effect
24                  modification by ALAD and vitamin D receptor (VDR) polymorphisms. None of the Pb
25                  biomarkers were associated with diastolic BP.  Patella Pb alone was not significantly
26                  associated with systolic BP. However, blood Pb, either alone or with patella Pb, was
27                  significantly associated with higher systolic BP. The patella Pb-age and blood Pb-age
28                  interactions were not statistically significant. There were no significant associations of
29                  blood Pb or patella Pb with hypertension status or effect modification by age or sex.
30                  Further, interactions between polymorphisms of the VDR and of ALAD with blood Pb
31                  and patella Pb on systolic BP were not statistically significant. Mean blood Pb level was
32                  high (30.9 (ig/dL) compared to non-occupational groups.

33                  Weaver et al. (2010) provided the results of further analysis of this Korean worker cohort,
34                  with a focus on determining the functional form of the concentration-response
35                  relationships. In a log linear model, the coefficient indicated that every doubling of blood
36                  Pb level was associated with a  systolic BP increase of 1.76 mmHg. The J test, a statistical
37                  test for determining which, if either, of two functional forms of the same variable
38                  provides a superior fit to data in non-nested models (Davidson and MacKinnon.  1981).
      November 2012                             5-305                  Draft - Do Not Cite or Quote

-------
 1                   returned a p-value of 0.013 in favor of the natural log blood Pb level over the linear blood
 2                   Pb level specification. This analysis indicates that the systolic BP increase in this cohort
 3                   is better described as a logarithmic function of blood Pb level within the range of the
 4                   study than by a linear function.

 5                   Several analyses in the NAS cohort of predominantly white older men in the greater
 6                   Boston area found associations of blood and bone Pb level with BP and hypertension, and
 7                   they indicated effect modification by calcium intake, perceived stress, and HFE gene
 8                   variants. In a cross-sectional analysis, Perlstein et al. (2007) found a statistically
 9                   significant association between blood Pb and diastolic BP in adjusted models. The
10                   subjects in this study had at least one bone Pb measurement during the years 1991-1997
11                   and were not on antihypertensive medication at the time of the measurement. While tibia
12                   Pb was not significantly associated with BP, it was associated with pulse pressure (PP).
13                   Men with tibia Pb above the median (19 (ig/g) had a higher mean PP (4.2 mmHg [95%
14                   CI: 1.9, 6.5]) compared to men with tibia Pb below the median. The trend toward
15                   increasing PP with increasing quintile of tibia Pb was statistically significant although
16                   none of the confidence intervals for PP  referenced to the lowest quintile of tibia Pb
17                   (<7.4 (ig/g) excluded the null value.

18                   Peters et al. (2007) examined cross-sectionally the modification of the associations of
19                   tibia and patella Pb with BP  and hypertension by self-reported stress (assessed by
20                   questionnaire) in NAS men.  High stress also has been linked  with higher BP, potentially
21                   via activation of sympathetic pathways, ROS, and the HPA axis. Among all  subjects,
22                   higher bone Pb level was associated (statistically nonsignificant) with greater odds  of
23                   hypertension status and higher systolic BP. As indicated in Figure 5-20. the association
24                   between systolic BP and tibia Pb differed between those with high and low self-reported
25                   stress ((3 for tibia Pb x stress interaction = 3.77 [95% CI: 0.46, 7.09]) per SD increase in
26                   tibia Pb. Stress also  was found to modify the patella Pb-BP association ((3 for patella Pb x
27                   stress interaction = 2.60 [95% CI: -0.95, 6.15] per  SD increase in patella Pb). Neither
28                   bone, self-reported stress, nor their interaction was associated significantly with diastolic
29                   BP. Peters et al. (2007) also used Cox proportional hazards models to assess the
30                   interaction of stress  and bone Pb  level in the development of  hypertension among those
31                   free of hypertension at baseline. The results of this analysis showed that increasing tibia
32                   and patella Pb were  associated with greater risk of developing hypertension among those
33                   with high stress compared to those  with lower perceived stress (RR of developing
34                   hypertension among those with high stress: 2.66 [95% CI: 1.43, 4.95] per SD increase  in
35                   tibia Pb and 2.64 [95% CI: 1.42, 4.92] per SD increase in patella Pb). These  results
36                   provide evidence supporting adults with higher stress as a population at increased risk  of
37                   Pb-associated cardiovascular effects. Earlier, Cheng et al.  (2001) examined the NAS
38                   cohort in 474 subjects without hypertension (mean [SD] blood Pb level: 5.87 [4.01]) at
      November 2012                              5-306                  Draft - Do Not Cite or Quote

-------
1
2
3




baseline measurement and analyzed linear models with patella Pb and reported that only
patella Pb level was associated with a significant increase in the rate ratio for
hypertension using a Cox's proportional hazards model.
140-
135-
"5> 130-
E
£ 120-
115-
110
-1



-

<% * ?'
£ilftfr**9~~r~~'
^'^ ' ^^ • • O High perceived stress
9 • Low perceived stress
	 Trend (high stress)
• 	 Trend (low stress)
0 0 10 30 50 70 90
 4
 5
 6
 7
 8
 9
10

11
12
13
14
15
16
17
                                         Tibia lead (ug/g)
Source: Peters et al. (2007)
Figure 5-20      The relationship between tibia Pb and estimated systolic BP
                  (SBP) for those with high self-reported stress versus those with
                  low self-reported stress.

              Elmarsafawy et al. (2006) examined the modification of the relationship between Pb and
              hypertension by dietary calcium, with 467 subjects from the NAS. Responses on a semi-
              quantitative dietary frequency questionnaire with one-year recall were used to estimate
              calcium intake. Effect modification by calcium intake (dichotomized at 800 mg/day) was
              examined using interaction terms in logistic regression models and by conducting
              analyses stratified on the calcium variable. Increasing bone and blood Pb increased the
              odds of hypertension, particularly among subjects with low dietary calcium.

              Zhang et al. (2010a) examined the effect of polymorphisms of the hemochromatosis gene
              (HFE) on the relationship of bone Pb with PP in NAS men. HFE polymorphisms promote
              Fe absorption and have been shown to modify the impact of adult cardiac function.
              Subjects had up to three PP measurements during the 10 year study period. The overall
              results demonstrated a strong relationship between bone Pb and PP in this study, similar
              to an earlier cross-sectional PP study of many of the same subjects (Perlstein et al.. 2007).
              Zhang et al. (2010a) extended these findings by demonstrating larger increases in PP per
November 2012
                                             5-307
Draft - Do Not Cite or Quote

-------
 1                  unit increase in tibia and patella Pb level among those with the H63D variant compared
 2                  to those with the wild-type or the C282Y variant.

 3                  A small number of cross-sectional studies examined and found that blood Pb level was
 4                  associated with hypertension in pregnancy. Yazbeck et al. (2009) examined a
 5                  community-based group of pregnant women in France and unlike most other studies,
 6                  adjusted for potential confounding by blood concentrations of Cd, Mn, and Se. Pregnancy
 7                  induced hypertension (PIH) was defined as systolic BP >140 mmHg and/or diastolic BP
 8                  >90 mmHg during at least two clinic visits after week 22 of gestation. Patients with pre-
 9                  existing chronic hypertension were excluded. The mean (SD) blood Pb levels measured
10                  during pregnancy were 2.2 (1.4 (ig/dL) in PIH cases and 1.9 (1.2) (ig/dL in normotensive
11                  women. An association between blood Pb and PIH was observed (OR 3.29 [95% CI:
12                  1.11, 9.74] per unit increase in log-transformed blood Pb  level). Cd and Se  concentrations
13                  were comparable between PIH and no PIH groups. Adjustment for the metals slightly
14                  attenuated but did not eliminate the association between blood Pb levels and the risk of
15                  PIH. Investigators observed no  significant interactions among blood Pb level, any of the
16                  other elements, and maternal characteristics in predicting the risk of PIH. Interaction
17                  between blood Se and Pb concentrations was not significant, and the putative protection
18                  effects of Se through antioxidative properties were not found in this study.

19                  Wells et al. (20 lib) measured the relationship of cord blood Pb with BP in 285 women at
20                  admission to the Johns Hopkins Hospital in Baltimore, MD, during labor and delivery.
21                  Women with cord blood Pb levels in the highest quartile for the study group
22                  (>0.96 (ig/dL) had significantly higher systolic and  diastolic BP (upon admission and for
23                  maximum BP) compared to women in the first quartile (<0.46 (ig/dL). The level of
24                  uncertainty at these levels of exposure is difficult to estimate. The authors used
25                  Benchmark Dose Software V2.1, developed by the EPA, to estimate the blood Pb level
26                  (benchmark dose or BMD) and the associated lower confidence limit (BMDL) that was
27                  associated with one standard deviation (SD) increase in BP. In this study group, one SD
28                  is approximately equivalent to a 10% increase above the mean for the first quartile  blood
29                  Pb reference group. The BMD approach was used only as a means of quantifying the
30                  relationship of blood Pb with BP in this population. This analysis indicated that the 95%
31                  lower bound confidence limit on the maternal blood Pb level (estimated from cord blood
32                  Pb levels) that was associated with a 1 SD increase  in all blood pressure outcomes was
33                  about 1.4 ug/dL. These reported results are similar to those reported in the
34                  2006 Pb AQCD as well as those found 25 years ago but with blood Pb levels an order of
35                  magnitude lower in the more recent study. However, uncertainty exists as to the specific
36                  Pb exposure level, timing, frequency, and duration that contributed to the observed
37                  associations.
      November 2012                            5-308                 Draft - Do Not Cite or Quote

-------
 1                  Recent analyses using NHANES data continued to indicate associations of Pb biomarkers
 2                  with BP and hypertension. Muntner et al. (2005) previously used the NHANES
 3                  1999-2002 data to indicate that concurrent blood Pb levels were associated with
 4                  hypertension, peripheral artery disease (PAD), and chronic kidney disease. The PAD
 5                  results are discussed later in Section 5.4.3.5. and chronic kidney disease results are
 6                  discussed in Section 5.5.2. Blood Pb increased regularly with age (geometric means [95%
 7                  CIs]:  1.28 (ig/dL [1.23, 1.33] in the 18-39 age group to 2.32 (ig/dL [2.20, 2.44] in the 75
 8                  and older age group). Associations were observed between concurrent blood Pb level and
 9                  hypertension across race/ethnicity groups with significant trends observed for
10                  non-Hispanic blacks and Mexican Americans.

11                  In the NHANES III 1988-1994 population, Scinicariello et al. (2010) found a gene-
12                  environment interaction between blood Pb level and ALAD genotype (the genotypes
13                  have different affinities for Pb) in relation to systolic BP and diastolic BP in a cross-
14                  sectional analysis. These interactions varied across race/ethnicity strata. The strongest
15                  associations were observed among non-Hispanic blacks (Figure 5-18. Table 5-18).  A
16                  statistically significant interaction was observed between concurrent blood Pb level and
17                  ALADl-2/2-2b among non-Hispanic whites and non-Hispanic blacks. Scinicariello et al.
18                  (2010) also found an interaction between ALAD genotype and blood Pb level in the
19                  association with hypertension. Statistically significant associations between concurrent
20                  blood Pb level and hypertension were observed among non-Hispanic blacks and
21                  nonsignificant increases were observed among non-Hispanic whites and Mexican
22                  Americans (with the exception of Mexican Americans in the second  quartile of blood Pb
23                  level) (Figure 5-19. Table 5-19).  In addition, non-Hispanic white ALAD2 carriers in the
24                  highest blood Pb level quartile  3.8-52.9 (ig/dL) had a significantly higher association
25                  with hypertension compared with ALAD 1 homozygous individuals in the highest quartile
26                  of blood Pb. In the same NHANES population, Park et al.  (2009c) predicted bone Pb
27                  levels using a model developed with NAS data. Concurrent blood Pb was associated with
28                  hypertension overall in the NHANES population, with larger associations observed
29                  among black men and women as well as older adults (Figure 5-19. Table 5-19).
30                  Associations also were observed with estimated bone Pb.
                     5.4.2.2     Toxicology

31                   Studies on the effect of Pb (as blood Pb level) on systolic BP in unanesthetized adult rats
32                   consistently reported an increase in BP with increasing blood Pb level as shown in Figure
33                   5-21 (results summarized in Table 5-20). An array of studies has provided evidence that
34                   long-term Pb exposure (>4 weeks), resulting in blood Pb levels relevant to humans,
35                   i.e., below 10 (ig/dL can result in the onset of hypertension (after a latency period) in


      November 2012                            5-309                  Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
experimental animals that persists long after the cessation of Pb exposure (U.S. EPA.
2006b). Tsao et al. (2000) presented evidence for increased systolic and diastolic BP in
rats with blood Pb levels similar to the current U.S. population (mean [SD]: 2.15
[0.92] Lig/dL blood Pb; 140 [7] mmHg systolic BP, 98 [7] mmHg diastolic BP) compared
to untreated controls (mean [SD]: 0.05 [0.05] Lig/dL blood Pb; 127 [7] mmHg systolic
BP, 88 [7] mmHg diastolic BP). As this was the lowest Pb level tested, no evidence of a
threshold was evident. Further, a test for linear trend revealed a statistically significant,
positive trend for increasing BP with increasing blood Pb levels up to 56 Lig/dL
(e.g., mean [SD]: 5.47 [2.1] Lig/dL blood Pb; 143 [6] mmHg systolic  BP, 97 [8] mmHg
diastolic BP), with the effect leveling off at higher blood Pb levels.
                  120
                                                15        20        25
                                                   Blood Pb Level (ng/dL)
                        -Bravoetal. 2007
                        -Ri Hi eta I. 2009
                 -Changetal. 1997  	Changetal. 2005
                 Rizzietal. 2009   	Tsao etal. 2000
-Heydarietal. 2006 	Nakhouletal. 1992
-Zhangetal. 2009  ^—Fiorimet al. 2011
      Note: Crosses represent standard error for blood Pb and BP measurements. If no crossbar is present, error results were not
      reported. Arrows represent higher doses tested.
      Figure 5-21      Changes in BP after Pb exposure (represented as blood Pb level)
                         in unanesthetized adult rats across studies.
      November 2012
                            5-310
        Draft - Do Not Cite or Quote

-------
Table 5-20 Characteristics of studies of blood Pb with BP measures in animals
presented in Figure 5-21.
Reference3
Fiorim et al.
(2011)
Nakhoul et al.
(1992)
Chang et al.
(2005)
Tsao et al.
(2000)
Rizzi et al.
(2009)
Chang et al.
(1997)
Heydari et al.
(2006)
Bravo et al.
(2007)
Zhang et al.
(2009a)
Lifestage; Exposure Exposure Level;
Sex Duration Route
Adult; M 7 days 4 ug/1 00 g followed
by 0.05 ug/1 00 g
daily; intramuscular
Adult; M 8 weeks 100 ppm;
drinking water
Adult; M 8 weeks 20,000 ppm then
removal and
measurements
1-7 mo after; drinking
water
Adult 8 weeks 100 - 20,000 ppm;
drinking water
Adult; M 8 weeks 30 -90 ppm;
drinking water
Adult; M 8 weeks 500 ppm; drinking
water
Adult; M 12 weeks 100 ppm; drinking
water
Adult; M 14 weeks 100 ppm; drinking
water
Adult; M 40 weeks 100 ppm; drinking
water
Mean
[SEM]b
Blood Pb
Level
(ug/dL)
9.98 [1 .7]
5.3 [3]
Range:
4.5 to 83
Range of
means: 2.15
[0.29] to
85.76 [1 .29]b
7.6 [1.3],
19.3 [3.4]
29.1 [0.6]b
26.8 [2.2]
23.7[1.9]b
28.4[1.1]b
ASBP
(mmHg;
lowest blood
Pb level
compared
n with control)"
12 16
7 28
5 13.8
10 13
11 13.3
10 58
6 25.8
12 30
8-10 15.3
Comments

Spontaneously
hypertensive
rat model







      aStudies are presented in order of increasing duration of exposure.
      ""Standard deviation converted to SEM.
      °Difference in systolic BP (SBP) between group means not within one exposure group.
 1
 2
 o
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
Experimental animal studies continued to provide evidence that long-term Pb exposure
results in sustained arterial hypertension after a latency period.  Systolic BP increased in
rats after exposure to 90-10,000 ppm Pb (as Pb acetate in drinking water) for various time
periods that resulted in blood Pb levels between 19.3-240 (ig/dL (Mohammad et al..
2010; Zhang et al.. 2009a: Badavi et al.. 2008; Grizzo and Cordellini. 2008; Rezaetal..
2008: Bravo et al.. 2007: Vargas-Robles et al.. 2007: Hevdari et al.. 2006: Bagchi and
Preuss. 2005). Past studies have shown statistically significant elevations in BP in rats
with lower blood Pb levels. For example, long-term Pb exposure to  spontaneously
hypertensive rats (resulting in mean [SEM] blood Pb level: 5.3  [3] (ig/dL) led to
increased BP (Nakhoul et al.. 1992). Consistent with measurements of systolic BP by tail-
cuff plethysmography, Pb exposure (100 ppm for 14 weeks; mean blood Pb level:
24 (ig/dL) also caused an increase in intra-aortic mean arterial pressure (Bravo et al..
2007). In a study that tested low levels of Pb exposure (30 ppm; mean blood Pb level:
7.6 (ig/dL), a statistically significant increase in systolic BP was not observed despite
      November 2012
                            5-311
Draft - Do Not Cite or Quote

-------
 1                  elevated blood Pb level after 8 weeks of treatment. Nonetheless, there was a trend of
 2                  higher BP with higher blood Pb levels (Rizzi et al. 2009V

 3                  Studies found that Pb-induced increases in BP persisted long after cessation of Pb
 4                  exposure. Bagchi and Preuss (2005) found that elevated systolic BP was maintained for
 5                  210 days after cessation of Pb exposure (10,000 ppm Pb acetate in water, 40 days,
 6                  monitored for one year). However, chelation therapy using Na2CaEDTA returned systolic
 7                  BP to levels comparable to those in rats not treated with Pb (Bagchi and Preuss. 2005).
 8                  Chang et al. (2005) reported a partial  reversibility of effect after cessation of Pb exposure,
 9                  where Pb-induced elevated BP decreased but did not return to control levels 7 months
10                  post Pb exposure. After Pb exposure was removed, blood, heart, aorta, and kidney Pb
11                  levels decreased quickly within the first three months (Chang et al.. 2005). Pb-induced
12                  elevated systolic BP persisted for one month following Pb exposure cessation, followed
13                  by obvious decreases in BP until 4 months after Pb exposure cessation. Between 4 and
14                  7 months after Pb exposure cessation, the still-elevated BP did not decrease further, thus
15                  never returning to control BP levels. Decreases in BP were strongly correlated with
16                  decreases in blood Pb level after exposure cessation.

17                  The aforementioned studies all assessed the  relationship between long-term exposure
18                  (>4 weeks) of rats to Pb and measures of BP. However, recent research also investigated
19                  BP elevation occurring after short-term treatment with Pb (<4 weeks). Studies found
20                  increased systolic BP after 7 days of Pb treatment (daily injections resulting in mean
21                  [SEM] blood Pb levels of 9.98 [1.7] (ig/dL)  (Fiorim et al.. 2011) and after 2 weeks of Pb
22                  exposure (100 ppm via drinking water) (Sharifi et al.. 2004). A study utilizing intra-
23                  arterial pressure measurements found that a  single high-dose Pb injection in rats
24                  (resulting in mean [SEM] blood Pb levels  of 37 [1.7] (ig/dL) increased systolic arterial
25                  pressure after only 60 minutes (Simdes et al.. 2011). The  injection of Pb into the rat may
26                  not allow for extrapolation of these results to humans since this is not a comparable Pb
27                  exposure method. These studies suggest that there is the potential for increase in BP
28                  following short-term Pb treatment. It is possible that the increases in BP following short-
29                  and long-term Pb exposures are occurring through separate mechanisms; however,
30                  studies using both short- and longer-term Pb exposure have correlated increased  BP  with
31                  an activation of the renin-angiotensin system (i.e., increase in angiotensin converting
32                  enzyme (ACE) activity) (Section 5.4.2.3). Several of these aforementioned studies used
33                  the injection route of Pb administration, and the relevance of these bolus doses over  short
34                  periods of time to human routes of short-term exposure is uncertain. However, it is
35                  important to acknowledge  that the results were similar to  those from the study that
36                  examined short-term exposure to Pb via drinking water,
      November 2012                             5-312                  Draft - Do Not Cite or Quote

-------
                    5.4.2.3      Hypertension Modes of Action

 1                  The 2006 Pb AQCD (U.S. EPA. 2006b) examined a number of mechanisms leading to
 2                  Pb-induced hypertension, including oxidative stress, hormonal and blood pressure
 3                  regulatory system dysfunction, vasomodulation, and cellular alterations. As described
 4                  below, recent studies in experimental animals and cells further supported roles for these
 5                  potential mechanisms in mediating hypertension from Pb exposure.


                    Oxidative Stress Response - Reactive Oxygen Species and Nitric Oxide

 6                  Several studies discussed in the 2006 Pb AQCD demonstrated a role for oxidative stress
 7                  in the pathogenesis of Pb-induced hypertension, mediated by the inactivation of nitric
 8                  oxide (NO) and downregulation of soluble guanylate cyclase (sGC) (Dursun et al.. 2005;
 9                  Attri etal.. 2003; Gonicketal..  1997; Vaziri etal.. 1997; Khalil-Manesh et al.. 1994;
10                  Khalil-Manesh et al.. 1993b). Pb-induced reduction of biologically active NO was found
11                  not to be due to a reduction in NO-production capacity (Vaziri and Ding. 2001; Vaziri et
12                  al.. 1999a); instead it was found to result from inactivation and sequestration of NO by
13                  ROS (Malvezzi et al.. 2001; Vaziri etal.. 1999b).  Oxidative stress from Pb exposure in
14                  animals may be due to upregulation of NAD(P)H  oxidase (Ni et al.. 2004; Vaziri et al..
15                  2003). induction of Fenton and Haber-Weiss reactions (Ding etal.. 2001; Ding et al..
16                  2000). and failure of the antioxidant enzymes, CAT and GPx, to compensate  for the
17                  increased ROS (Farmand et al..  2005; Vaziri et al.. 2003). Many biological actions of
18                  NO, such as vasorelaxation, are mediated by cGMP, which is produced by sGC from the
19                  substrate GTP. Oxidative stress also has been found to play a role in Pb-induced
20                  downregulation of sGC (Farmand et al.. 2005; Courtois et al.. 2003; Marques et al..
21                  2001). Thus, the reduction of the vasodilator NO from inactivation and sequestration by
22                  Pb-induced ROS leads to increased vasoconstriction and BP.

23                  Pb-induced oxidative stress also has been found to induce renal tubulointerstitial
24                  inflammation which plays a crucial role in models of hypertension (Rodriguez-Iturbe et
25                  al.. 2005; Rodriguez-Iturbe et al.. 2004). Tubulointerstitial inflammation from treatment
26                  with Pb has been coupled with activation of the redox sensitive NF-KB (Ramesh et al..
27                  2001). Pb-induced hypertension, inflammation, and NF-KB activation can be ameliorated
28                  by antioxidant therapy (Rodriguez-Iturbe et al.. 2004). There is mixed evidence to
29                  suggest that Pb-induced hypertension may also be promoted by activation of PKC leading
30                  to enhanced vascular contractility (Valencia et al.. 2001;  Watts etal.. 1995).

31                  Recent studies continued to provide evidence for the role of ROS and NO metabolism in
32                  Pb-induced hypertension and vascular disease. Increased systolic BP  after Pb exposure
33                  was accompanied by increased superoxide (O2~) and O2~ positive cells (Bravo et al.. 2007;
      November 2012                            5-313                  Draft - Do Not Cite or Quote

-------
 1                  Vargas-Robles et al.. 2007). elevated urinary malondialdehyde (MDA, a measure of lipid
 2                  peroxidation) (Bravo et al.. 2007). and increased 3-nitrotyrosine (Vargas-Robles et al..
 3                  2007). Inhibition of NAD(P)H oxidase, an enzyme that generates O2 and hydrogen
 4                  peroxide, was able to block Pb-induced (1 ppm) aortic contraction
 5                  to 5-hydroxytryptamine (5-HT) (Zhang et al.. 2005). Increases in systolic BP, intra-aortic
 6                  mean arterial pressure, and MDA after Pb exposure (100 ppm; mean blood Pb level:
 7                  23.7 (ig/dL) were also prevented by treatment with the immunosuppressant,
 8                  mycophenolate mofetil (MMF) (mean blood Pb level in MMF-treated animals: 27 (ig/dL)
 9                  (Bravo et al.. 2007). MMF has been shown to inhibit endothelial NAD(P)H oxidase,
10                  which could explain how it decreases Pb-induced increases in oxidative stress and BP.
11                  MMF was not found to alter blood Pb levels of animals. Red grape seed extract and
12                  ascorbic acid supplementation were also able to protect rats from Pb-induced (100 ppm)
13                  increased BP and heart rate, perhaps through the antioxidant properties of the extract
14                  (Badavi et al.. 2008) and vitamin C (Mohammad et al.. 2010). Red grape seed extract did
15                  not alter the accumulation of Pb in blood, indicating that its protective effect was not
16                  mediated through altered Pb toxicokinetics; however, internal doses of Pb were not
17                  measured in the vitamin C study to clarify the mechanism of action of vitamin C. Another
18                  study found that the antioxidant, anti-inflammatory chemical, curcumin, as well as
19                  physical exercise training reversed Pb-induced  increases in serum creatinine kinase-MB
20                  (CK-MB), low density lipoprotein (LDL), heart high-sensitivity C-reactive protein
21                  (hs-CRP), and MDA. Pb-induced decreases in serum total antioxidant capacity, high
22                  density lipoprotein (HDL), and heart glutathione peroxidase (GPx) were also reversed by
23                  curcumin and exercise. However, internal doses of Pb were not measured to clarify the
24                  mechanism of action in this study (Roshan et al.. 2011).

25                  Exposure to Pb can also affect the activity and levels of antioxidant enzymes. Male ($)
26                  and female (9) rats exposed to Pb for 18 weeks (100-1,000 ppm) had altered responses in
27                  antioxidant enzymes in heart tissue (Sobekova et al.. 2009; Alghazal et al.. 2008a). Pb
28                  exposure in female rats increased the activity of cardiac SOD, GST, GR, and GPx
29                  (>100 ppm) and increased cardiac thiobarbituric acid reactive substances (TEARS, a
30                  measure of lipid peroxidation) (1,000 ppm). Pb exposure in male rats did not affect the
31                  activity of SOD or production of TEARS, however decreased the activity of GST and GR
32                  (>100 ppm). Male and female rats also accumulated different amounts of Pb in the
33                  cardiac tissue after similar Pb  exposure (<$ 100 ppm: 205% of control, 1,000 ppm: 379%;
34                  9 100 ppm: 246%, 1,000 ppm: 775%), which could explain the sex differences observed
35                  in antioxidant enzyme responses.

36                  Oxidative stress can trigger a cascade of events that promote cellular stress, renal
37                  inflammation, and hypertension. As was shown previously (Rodriguez-Iturbe et al..
38                  2005). Pb exposure can increase renal NF-KB, which was associated with
      November 2012                            5-314                 Draft - Do Not Cite or Quote

-------
 1                  tubulointerstitial damage and infiltration of lymphocytes and macrophages (Bravo et al.,
 2                  2007). These events could also be ablated by MMF treatment, likely due to its anti-
 3                  inflammatory and antioxidant properties. Pb also was found to induce inflammation in
 4                  human endothelial cells as a model for vessel intima hyperplasia (Zeller et al.. 2010). The
 5                  pro-inflammatory cytokine, interleukin (IL)-8 protein and mRNA were increased,
 6                  concentration- and time-dependently, after in vitro Pb exposure (5-50 (JVI). Enhanced
 7                  IL-8 production was mediated through activation of the transcription factor Nrf2 (but not
 8                  NF-KB, hypoxia inducible factor-1, or aryl hydrocarbon receptor), as shown through
 9                  increased nuclear translocation and Nrf2 cellular knockdown experiments. Additionally,
10                  measures of endothelial stress, NQO1 and HO-1 protein, were induced by Pb exposure
11                  (Zeller et al.. 2010). Pb treatment (20 ppm, i.p., 3 days/week, 8 weeks) increased the
12                  inflammatory markers hs-CRP and CK-MB in rat hearts (Roshan etal.. 2011).

13                  Oxidative stress affects vascular reactivity and tone through inactivation and
14                  sequestration of NO, causing a reduction in biologically active NO. Recent studies
15                  affirmed past conclusions on the  interplay of ROS and NO metabolism in the
16                  cardiovascular effects of Pb. Elevated systolic BP and altered vasorelaxation after Pb
17                  exposure was accompanied by a decrease in total nitrates and nitrites (NOX) (Mohammad
18                  etal.. 2010: Zhang et al.. 2007a: Hevdari et al.. 2006). Serum NOX levels in Pb-treated
19                  rats remained depressed for 8 weeks and then reversed after 12 weeks, despite continued
20                  elevation in systolic BP (Hevdari et al.. 2006). This return of serum NOX levels to levels
21                  similar in controls could be a result of compensatory increases in endothelial NOS
22                  (eNOS) attempting to replenish an over-sequestered NO supply. With this in mind,
23                  studies showed increased eNOS protein expression after long-term Pb exposure in kidney
24                  (Zhang et al.. 2007a) and isolated cultured aorta (Vargas-Robles et al.. 2007). No change
25                  in inducible NOS was observed in isolated cultured aorta after 1 ppm Pb exposure (Zhang
26                  et al.. 2007a). In contrast to long-term exposure, Pb treatment over a short time period
27                  (daily injections resulting in mean [SEM] blood Pb levels of 9.98 [1.7] (ig/dL) was found
28                  to increase iNOS and phosphorylated eNOS protein (Fiorimet al.. 2011) which may
29                  cause an increase in NO production and a short-term increase in NO bioavailability. This
30                  increase in NO bioavailability early after Pb exposure could be the immediate
31                  compensatory mechanism against the elevation in BP.

32                  NO, also known as endothelium-derived relaxing factor, is a potent endogenous
33                  vasodilator. Toxicological studies continued to investigate the effects of Pb on
34                  NO-dependent vascular reactivity by using NO stimulating vasodilators, such as
35                  acetylcholine (ACh) and sodium  nitroprusside (SNP), and NO inhibiting
36                  vasoconstrictors, such as L-NAME. Studies provided mixed evidence; however, results
37                  suggested that Pb disrupts the vasorelaxant response to NO in the aorta due to damage to
38                  the endothelium. Pb exposure (1 ppm and 100 (iM, 1 hour) decreased ACh-induced
      November 2012                             5-315                  Draft - Do Not Cite or Quote

-------
 1                  vasorelaxation, which triggers the release of NO from the endothelial cell, in isolated rat
 2                  tail artery, suggesting damage to the endothelium (Silveira et al.. 2010; Zhang et al..
 3                  2007a). In aortic rings of perinatally exposed rats (1,000 ppm through pregnancy and
 4                  lactation, mean blood Pb level: 58.7 (ig/dL), blocking NOS with L-NAME abolished the
 5                  relaxant response evoked by ACh (Grizzo and Cordellini. 2008). However, there was no
 6                  change observed in the relaxation response to ACh by Pb alone (Fiorim et al.. 2011; Rizzi
 7                  et al., 2009; Grizzo and Cordellini, 2008). Conversely, Skoczynska and Stojek (2005)
 8                  found that Pb exposure (50 ppm; blood Pb level 11.2 (ig/dL) enhanced NO-mediated
 9                  vasodilation by ACh in rat mesenteric arteries, and NOS inhibition enhanced the ACh
10                  relaxant response. A number of studies found that Pb exposure did not affect smooth
11                  muscle integrity since SNP-induced vasorelaxation, which is endothelium independent,
12                  was unchanged (Fiorim et al., 2011; Silveira et al.. 2010; Rizzi et al.. 2009; Grizzo and
13                  Cordellini. 2008).

14                  NO also was found to play a role in the interaction between Pb and the vasoconstrictor
15                  response. Blocking NOS with L-NAME or inhibiting iNOS specifically, which decreases
16                  NO production, increased the contraction of aortic rings in response to the
17                  vasoconstrictor phenylephrine (PHE), and Pb exposure potentiated this response (Fiorim
18                  et al.. 2011). Also, L-NAME increased the Pb pressor response to PHE after perinatal Pb
19                  exposure (1,000 ppm through pregnancy and lactation, blood Pb level 58.7 (ig/dL)
20                  (Grizzo and Cordellini. 2008). Conversely, in rat renal interlobar arteries, Pb exposure
21                  blunted the increase in renal angiotensin II (Angll)-mediated contraction from NOS
22                  inhibition by L-NAME (Vargas-Robles et al.. 2007). Treatment with the SOD mimetic
23                  tempol, which would increase NO bioavailability, decreased, but did not eliminate, the Pb
24                  pressor response (Silveira et al.. 2010).

25                  In summary, recent studies continued to provide evidence for the role of ROS in
26                  Pb-induced hypertension and vascular  disease by indicating Pb-induced increases  in ROS
27                  and modulation of cardiovascular responses by antioxidant substances. Additionally,
28                  recent studies continued to show that Pb-induced hypertension and vascular responses  are
29                  mediated primarily via inactivation of NO not via inhibition of NO production.


                    Vascular Reactivity

30                  Alteration of the adrenergic system from Pb exposure, which can increase peripheral
31                  vascular resistance, and thereby arterial pressure, may be one mediator of Pb-induced
32                  hypertension. Pb exposure in animals can increase stimulation of the sympathetic nervous
33                  system (SNS), as  shown by increased plasma levels of norepinephrine (NE) and other
34                  catecholamines (Carmignani et al.. 2000; Chang et al.. 1997) and decreased (3 adrenergic
35                  receptor density and  (3 agonist-stimulated cAMP production in the aorta and heart  (Tsao
      November 2012                            5-316                  Draft - Do Not Cite or Quote

-------
 1                   et al.. 2000; Chang etal., 1997). These stimulatory effects on the SNS paralleled the
 2                   effects of Pb on BP, cardiac contractility, and carotid blood flow. Pb-induced elevations
 3                   in arterial pressure and heart rate were abrogated by ganglionic blockade (Simdes et al.,
 4                   2011; Lai et al.. 2002). Arterial pressure and heart rate gradually decreased 7 months
 5                   after Pb exposure cessation as did the Pb-induced SNS alterations (Chang et al., 2005).

 6                   Increases in BP can be caused by activation of the SNS, which can lead to vascular
 7                   narrowing, in turn, resulting in increased total peripheral resistance. In this neural
 8                   mechanism, activation of the SNS leads to vasoconstriction, whereas inhibition leads to
 9                   vasodilation. It has been suggested that Pb leads to increased vascular reactivity to
10                   catecholamines (i.e., epinephrine, NE, and dopamine), hormones of the  SNS. Indeed, the
11                   isolated mesenteric vessel bed from Pb-treated rats (50 ppm with blood  Pb level:
12                   11.2 (ig/dL, but not 100 ppm with blood Pb level: 17.3 (ig/dL) exhibited increased
13                   reactivity to NE (Skoczynska and Stojek. 2005). However, in another study, 100 ppm Pb
14                   did not affect the NE-induced contractile response after 10 months of exposure (Zhang et
15                   al., 2009a). suggesting a small  range of Pb doses affects pressor response to NE.
16                   Catecholamines act primarily through the adrenergic and dopaminergic  receptors.
17                   Antagonists of a 1-adrenergic, a2-adrenergic, (3-adrenergic,  and dopamine Dl receptors
18                   were found to abolish Pb-induced aortic  contraction (Fazli-Tabaei et al.. 2006; Heydari et
19                   al., 2006). However, the a 1-adrenergic receptor agonist, PHE, induced aortic contractions
20                   and these were enhanced by treatment with Pb (100 ppm; blood Pb level: 26.8 (ig/dL),
21                   indicating a specific role for the a 1-adrenergic receptor (Silveira et al., 2010; Grizzo and
22                   Cordellini. 2008; Heydari et al.. 2006). Removal of the endothelium blunted the PHE-
23                   induced contraction. Conversely, short-term Pb treatment (7 days, i.p.) decreased the
24                   contractile response induced by PHE in rat aortas resulting in a decreased vascular
25                   reactivity (Fiorim etal.. 2011). This decrease may be playing a compensatory role in
26                   attempting to correct the Pb-induced BP elevation. Additionally, Pb blunted the
27                   isoproterenol-induced relaxation, supporting a role for the (3-adrenoceptors (Vassallo et
28                   al.. 2008; Hevdari et al.. 2006).

29                   Recently, there was mixed evidence for Pb disrupting vascular reactivity to other pressor
30                   agents. Pb (1 ppm) treatment of isolated rat thoracic aorta increased 5-HT induced
31                   contraction, which was endothelium dependent, but not due to 5-HT2B  receptor
32                   expression (Zhang et al., 2005). Follow-up of this study in whole animals found, on the
33                   contrary, that Pb (100 ppm; blood Pb level: 28.4 (ig/dL) decreased the maximum
34                   contractile response to 5-HT, but did not affect 5-HT plasma levels or 5-HT2B receptor
35                   expression (Zhang et al.. 2009a). In addition, Pb exposure (100 ppm,  12 weeks) increased
36                   the renal vascular response to Angll in isolated perfused kidneys from Pb-exposed rats
37                   (Vargas-Robles et al.. 2007).
      November 2012                             5-317                  Draft - Do Not Cite or Quote

-------
 1                   Studies continued to investigate the effects of Pb on NO-dependent vascular reactivity by
 2                   using NO stimulating vasodilators, such as ACh and SNP, and NO inhibiting
 3                   vasoconstrictors, such as L-NAME. These studies were discussed in the preceding
 4                   subsection (Oxidative Stress Response).


                     Renin-Angiotensin-Aldosterone and Kininergic Systems

 5                   The adrenergic system also affects the renin-angiotensin-aldosterone system (RAAS),
 6                   which is responsible for fluid homeostasis and BP regulation, and has been shown to be
 7                   affected by Pb exposure. A meta-analysis found that Pb exposure (resulting in blood Pb
 8                   levels: 30-40 (ig/dL) increased plasma renin activity and renal tissue renin in young but
 9                   not old rats fVander. 1988). Exposure of experimental animals to Pb also induced
10                   increases in plasma, aorta, heart, and kidney angiotensin converting enzyme (ACE)
11                   activity; plasma kininase II, kininase I, and kallikrein activities; and renal Angll positive
12                   cells (Rodriguez-Iturbe et al.. 2005; Sharifi et al.. 2004; Carmignani etal.. 1999). ACE
13                   activity declined over time while arterial pressure stayed elevated, suggesting that the
14                   RAAS may be involved in the induction, but not the maintenance of Pb-induced
15                   hypertension in rats.

16                   Recent studies continued to implicate the RAAS in the development of Pb-induced
17                   hypertension, especially during early exposure in young animals. Angll, a main player in
18                   the RAAS, induces arteriolar vasoconstriction leading to increased BP. Pb exposure
19                   increased the vascular reactivity to Angll (Vargas-Robles et al.. 2007). Acute
20                   (60 minutes)  or short-term (7 days) treatment of rats to Pb increased the plasma ACE
21                   activity (Fiorim et al.. 2011; Simdes et al.. 2011). and Fiorim et al. (2011) additionally
22                   found this increase to be correlated with the  Pb-induced increase in systolic BP.
23                   However, at these  short time points there were no changes in the Angll receptors 1 or 2
24                   protein levels or expression. Treatment with the Angll receptor (ATiR) blocker,
25                   Losartan, or the ACE inhibitor, Enalapril, blocked the Pb-induced systolic BP increase
26                   (Simdes et al.. 2011) and decreased the PHE-induced vasoconstrictor response in
27                   Pb-treated aortas (Fiorim et al.. 2011). Similarly, treatment with Losartan resulted in a
28                   greater decrease in systolic BP in highly Pb-exposed rats (10,000 ppm Pb, 40 days; blood
29                   Pb level >240 (ig/dL after exposure, 12-13 (ig/dL after chelation after 1 year) compared
30                   to control rats that continued into later periods of follow-up (day 283) (Bagchi  and
31                   Preuss. 2005). Increased systolic BP after early exposure to Pb corresponded with
32                   increased water intake, urine output, potassium excretion, and decreased urinary sodium
33                   and urine osmolality. These functional changes in renal behavior are consistent with the
34                   actions of a stimulated RAAS.  Lower level Pb (100 ppm, 14 weeks; range of blood Pb
35                   levels: 23.7-27 (ig/dL) exposure increased renal cortical Angll content and the number of
36                   tubulointerstitial Angll-positive cells (Bravo et al.. 2007). This heightened intrarenal

      November 2012                            5-318                  Draft - Do Not Cite or Quote

-------
 1                  angiotensin corresponded with sodium retention and increased systolic BP and was
 2                  ablated by the anti-inflammatory antioxidant, MMF. Sodium reabsorption is important
 3                  for the maintenance of BP, and Na+ transporters play a key role in this process. In other
 4                  studies, Pb exposure increased activity and levels of the a-1 subunit protein of
 5                  Na+/K+ATPase, which plays a major role in Na+ reabsorption and is regulated by the
 6                  RAAS (Fiorim et al.. 2011; Simdes et al.. 2011). These studies point to the activation of
 7                  the RAAS in the course of Pb-induced hypertension, particularly in the early stages of
 8                  elevated BP.


                    Vasomodulators

 9                  The balance between production of vasodilators and vasoconstrictors is important in the
10                  regulation of BP and cardiovascular function. The 2006 Pb AQCD reported that Pb did
11                  not affect all vasomodulators in the same way. Urinary excretion of the vasoconstrictor,
12                  thromboxane (TXB2), and the vasodilatory prostaglandin, 6-keto-PGFla, was unchanged
13                  in rats with Pb-induced hypertension (Gonick et al.. 1998). However, in vitro Pb
14                  exposure promoted the release of the prostaglandin precursor, arachidonic acid, in
15                  vascular smooth muscle cells (VSMCs) via activation of phospholipase A2 (Dorman and
16                  Freeman. 2002). Plasma concentration and urinary excretion of the vasoconstrictive
17                  peptide, endothelin (ET) 3 was increased after low (100 ppm), but not high-level
18                  (5,000 ppm) Pb exposure  in rats (Gonick et al.. 1997; Khalil-Manesh et al.. 1994; Khalil-
19                  Manesh et al.. 1993b). Antagonism of the ET receptor A blunted the downregulation of
20                  sGC and cGMP production by Pb in isolated rat artery segments, suggesting that some of
21                  the hypertensive effects of Pb exposure may be mediated through ET (Courtois et al..
22                  2003). Additionally, Pb-exposed animals exhibited fluid retention and a
23                  concentration-dependent decline in the vasodilator, atrial natriuretic factor (ANF)
24                  (Giridhar and Isom. 1990). Results from these studies suggest that Pb may interfere with
25                  the balance between vasodilators and vasoconstrictors that contribute to the complex
26                  hormonal regulation of vascular contraction and BP.

27                  The imbalance in vasomodulators is one explanation for the concentration-dependent
28                  vasoconstriction observed in some animals after Pb exposure (Valencia et al.. 2001;
29                  Watts et al.. 1995; Piccinini et al.. 1977). However, vasoconstriction after Pb exposure
30                  was not reported in all studies (Shelkovnikov and Gonick, 2001) and is likely varied
31                  depending on the type of vessel used, the Pb concentration employed, and the animal
32                  species being studied. Studies have reported Pb-induced attenuation of ACh- and NO-
33                  mediated vasodilation (Marques et al.. 2001; Oishi et al.. 1996) in some, but not all
34                  vascular tissues and in some, but not all studies (Purdy et al.. 1997). These effects have
35                  been variably attributed to Pb-mediated activation of PKC and direct action on the
      November 2012                            5-319                 Draft - Do Not Cite or Quote

-------
 1                  VSMCs through the Ca2+ mimetic properties of Pb among other possibilities (Valencia et
 2                  al..2Q01: Watts etal. 1995: Piccinini et al.. 1977).

 3                  A recent study investigated the role of the endothelial-derived vasoconstrictor, ET-1, in
 4                  Pb-induced hypertension. ET-1 from the endothelium acts on the ETA-type receptors
 5                  located on the vascular smooth muscle layer and may be involved in vascular reactivity
 6                  by NO and COX derivatives. Pb exposure (1 ppm, 24 hours) to rat aortic segments
 7                  decreased expression of sGC-(31 subunit, an enzyme involved in NO-induced
 8                  vasodilation, and increased expression of COX-2 in an endothelium-dependent manner
 9                  (Molero et al.. 2006). Even though Pb treatment did not alter ET-1 or ETA-type receptor
10                  protein expression in this system, blocking the ETA-type receptors partially reversed
11                  Pb-induced changes in sGC and COX-2 in vascular tissue. These results suggest that the
12                  endothelium and ET-1 may contribute to Pb-induced hypertension through activation of
13                  ETA-type receptors that alter expression of COX-2 and sGC-(31 subunit, which affects
14                  NO signaling.

15                  COX-2 blockade has been shown to prevent Pb-induced downregulation of sGC
16                  expression (Courtois et al., 2003). Inhibition of COX-2 also decreased the Pb-induced
17                  pressor response to ACh (Grizzo and Cordellini. 2008) and PHE (Silveira et al.. 2010) in
18                  experimental animals. These results suggest that Pb-induced vascular reactivity may
19                  depend on the participation of a COX-derived vasoconstrictor, such as prostaglandins,
20                  prostacyclins, orthromboxanes.

21                  In summary, recent studies continued to show that Pb exposure affects vasomodulatory
22                  pathways that are important for the  maintenance of vascular tone; however, results
23                  indicated that not all vascular cell types are similarly affected by Pb exposure. Further,
24                  effects appeared to vary according to the concentration of Pb exposure. Pb exposure has
25                  been shown to interrupt baseline or endogenous NO-mediated vasodilation of vessels via
26                  alterations in PKC, sGC, VSMC, endothelial cells, NADPH oxidase, and Ca2+ levels.
27                  Recent studies indicated that Pb exposure may affect vascular reactivity by increasing
28                  COX-2 and COX-2-dependent vasoconstrictors. Also, the vasoconstrictor endothelin may
29                  contribute to Pb-induced vasomodulation via similar pathways as NO including effects
30                  on sGC and COX-2.
                    5.4.2.4      Summary of Blood Pressure and Hypertension

31                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported a clear association between higher
32                  blood Pb levels and higher BP. The effect was modest, but robust, as determined by a
33                  meta-analysis (TSfawrot et al.. 2002) of over 30 cross-sectional and prospective studies
34                  comprising over 58,000 adults (Figure 5-22). In the meta-analysis, each doubling of

      November 2012                             5-320                 Draft - Do Not Cite or Quote

-------
 1                  concurrent blood Pb was associated with a 1 mmHg increase in systolic BP and a 0.6
 2                  mmHg increase in diastolic BP. Recent epidemiologic studies supported this association
 3                  at lower concurrent blood Pb levels (in populations with mean blood Pb levels <2 (ig/dL)
 4                  and added to the evidence base regarding populations potentially at increased risk
 5                  (i.e., high stress, genetic variants) and regarding associations of bone Pb levels with BP
 6                  and hypertension in populations with mean bone Pb levels less than 20 (ig/g. As these
 7                  studies were mostly cross-sectional in design and were conducted in adults whose
 8                  concurrent blood Pb levels are influenced both by current Pb exposures and past Pb
 9                  exposures mobilized from bone, uncertainty exists over the Pb exposure conditions that
10                  contributed to the associations observed between concurrent blood Pb level with
11                  increased BP and hypertension (Sections 4.3 and 4.7.3). However, U.S. EPA (1990a)
12                  reviewed studies available prior to 1990, a period when Pb exposures from air were
13                  probably at the highest level, that examined Pb exposure and BP outcomes which
14                  included evaluation of several studies of the population represented in NHANES II
15                  (1976-80).  They noted that across a range of 7 to 34 (ig/dL, no evident threshold was
16                  found below which the blood Pb level was not significantly related to BP.  U.S. EPA
17                  (1990a) concluded that a small but positive association exists between blood Pb levels
18                  and increases in BP.

19                  A recent prospective study in Pb workers found independent associations of both baseline
20                  blood Pb level and subsequent changes in blood Pb over follow-up with changes in BP
21                  over follow-up and bone Pb level with hypertension (Glenn et al.. 2006). Although these
22                  Pb workers had higher current Pb exposure compared with nonoccupationally-exposed
23                  adults, the results indicated that different mechanisms may mediate shorter-term
24                  Pb-associated increases in BP and longer-term Pb-associated development of
25                  hypertension.

26                  Key evidence was further provided by a recent cross-sectional study in an  ethnically
27                  diverse community-based cohort of women and men aged 50-70 years of age that found
28                  associations of both blood and tibia Pb levels with BP  with extensive consideration of
29                  potential confounding factors (Martin et al.. 2006). Additionally, a recent epidemiologic
30                  study provided evidence for associations in an adult cohort between blood Pb level and
31                  BP and hypertension with relatively low blood Pb levels;  a positive relationship was
32                  found in the NHANES adult data (1999-2002) with a geometric mean blood Pb level of
33                  1.64 (ig/dL (Muntner et al..  2005). However, as noted above, in adults, uncertainty exists
34                  regarding the magnitude, timing, frequency, and duration of Pb exposure that contribute
35                  to the associations observed with concurrent blood Pb  levels.
      November 2012                            5-321                  Draft - Do Not Cite or Quote

-------
POCOCK81.
KRCMHOUT 85
ORSSAUD 85
WEISS 86
DE KORT 87
PARKINSON 87 K
RABINOWITZ87
ELWOOD (Cl 88 i-
ELWOOO [HP) 88 H
ELWOOO [HP) 88 *
GARTSIOE IW) 88
GART5IOE [W18B t-

NERI 88 i
GRANDJEAN89 1—
GRANDJEAN89 i —

APOSTOL1 90
MORRIS 90
MORRIS 90
SHARP [BI90

M01LER92
HEN5E 93
HENSE93 i-
MENDITTD9*.
PROCTOR 96 i—
STAESSEN \p) 96 \ »
STAESSEN (P) 96 i 	
SQf Ac |W) 97 i •


CHU99
CHU 99 I
ROTHENBERG INI] 99 i-
ROTHENBERG [I) 99
SCHWARTZ 00
DEN HOND (W) 01 i-
DEN HOND (B) 01
DEN HOND (W) 01 •
DEN HOND IB) 01
ALL
* o-




* o-
h*H $
H O*
H er
H 9
H^H 9

-• 	 < 9

• 9
*~ 9


• 4 C
I c^
-1-A f
+—* cr
g



J B
• O*
9
MH 9
-•H (f
%
i^^ 0
• o-9
f I 1 1 1 1 1 1 1
-15 -10 -50 5 10 15 20 25 30
SBP
7379
15?
89
105
116
3851
1136
665
B56
350
28Z7
288
W
58
2?r
US
106
in
132
398
133
(39
1703
809
1319
798
339
345
'86
2563
2763
K7]
mi
(.39
lisa
51.3
1,685
1761
5138
Z197
58518
     Source: Reprinted with permission of MacMillan Press, Nawrot et al. (2002)
     Study Key: C - Caerphilly Study; HP - Welsh Heart Program; W - Whites; B - Blacks; Nl - Non-immigrants; I - Immigrants;
     FW- Foundry Workers; CS - Civil Servants; P - PheeCad (Public Health and Environmental Exposure to Cadmium) Study.
     Note:  Individual study results are presented in each row. The rightmost columns indicate the sex of subjects and study sample size.
     Circles represent individual groups and squares represent the combined association sizes. Open circles denote a nonsignificant
     association size that was assumed to be zero.

     Figure 5-22      Meta-analysis of change in systolic BP (SBP), in mmHg with 95%
                        Cl, associated with a  doubling in the blood Pb concentration.
i
2
3
4
5
6
7
In concordance with epidemiologic evidence, collectively, the animal toxicological
studies providing blood Pb level and BP measurements reported higher BP with higher
blood Pb levels in adult rodents (Figure 5-21). While the contribution of low concurrent
blood Pb levels to the findings is difficult to ascertain in adult humans, animal
toxicological studies provide support for  low blood Pb level effects with increases in BP
observed in groups of animals with long-term dietary Pb exposure resulting in blood Pb
levels as low as 2 (ig/dL (Rizzi et al.. 2009: Tsao et al.. 2000: Nakhoul et al.. 1992).
However, the majority of animal toxicological studies showing Pb-induced hypertension
     November 2012
                            5-322
Draft - Do Not Cite or Quote

-------
 1                   were conducted at higher Pb exposure levels that result in blood Pb levels >10 (ig/dL. In
 2                   addition, recent animal evidence suggests the potential for increased BP following short-
 3                   term (4 weeks) Pb treatment that included injected bolus doses that may have uncertain
 4                   relevance to human routes of Pb exposure (Fiorim et al.. 2011; Simdes et al.. 2011;
 5                   Sharifi et al., 2004). A recent study also demonstrated partial reversibility (not to levels in
 6                   controls) of Pb-induced elevations in BP following Pb exposure cessation or chelation
 7                   (Chang et al.. 2005).

 8                   Epidemiologic studies continued to investigate the relationship between bone Pb and BP.
 9                   A recently published meta-analysis (Figure 5-23)  (Navas-Acien  et al.. 2008) included
10                   several studies (three prospective, five cross-sectional) that individually showed that bone
11                   Pb level was associated with systolic BP but not diastolic BP. In the cross-sectional
12                   studies, a pooled estimate indicated an increase in systolic BP of 0.26 mmHg (95% CI:
13                   0.02, 0.50) per 10  (ig/g tibia Pb. In the longitudinal studies, a 0.33 mmHg (95% CI: -0.44,
14                   1.11) increase was estimated per 10 (ig/g bone Pb. Most studies  also reported associations
15                   of bone Pb with hypertension. Pooled odds ratios for hypertension of 1.04 (95% CI: 1.01,
16                   1.07) per 10 (ig/g increase in tibia Pb and 1.04 (95% CI: 0.96, 1.12) per 10 (ig/g increase
17                   in patella Pb were reported.
      November 2012                             5-323                  Draft - Do Not Cite or Quote

-------
First author, year
Tibia lead
Prospective
Glenn 2006"
Cheng 2001 '3
Glenn 200312

Cross-sectional
Lee 2001'"
Hu 96'VCheng 01
Martin'8 2008
SchwartzK2000
KorrickJ51999

Patella lead
Cheng'3 2001
Hu 96's/Cheng 01
Korrick'5 1999


Mean Increase in SBP (95%CI) Increase in DBP (95%CI) Hypertension RR or OR (95%CI)
Lead
fuo/a)


3S-4 -0.02 (-0.03 to 0.004) 1
21-9 --
14.7 0.78 (0.24 to 1.31)
Overall: 0.33 (-0.44 to 1. 11)
-1
Increase i
37.2 0.20 (-0.05 to 0.45}
5 32.1 1.01 (0.01 to 2. 02)
18.8 Q.20(-Q.SQto 1.10)
14.4 0.74 (-0.73 to 2. 21)
13.3 -
Overall: 0.26 {0.02 to 0.50)

31 .4 --
5 32.1 0.29 (-0.36 to 0.95)
17.3 -

1
• 0.07 (-0.30 to 0.45) ^|



II 1 1 1 1 1 1 1 1
17. -1 fl 1 5 OSS 1 1 5 1 S
i SBP (mmHg / year) Increase in DBP {mrnHg / year) Hypertension RR
-R- -0.02 (-0.20 to 0.1 7) ^
• 0 "0 ' 0 nn to 0 ^0)
f Cj 05 f.f) 7K [Q -5 ^gl


0.02 (-0.1 5 to 0.19)


— • 	
__

^ 1.05(1.00101.11)
1.15 (0.97 to 1.35}
"• 1.13 (0.98 !o 1.29}
— • 	 0.90 (0.70 to 1 .17} ^ — • 	
1.03(1.00101.05)
1.04 (1.01 10 1.07)

1.14 (1.01 to 1.28}
1.09(0.98101.22}
1 .00 (0.98 to 1 .03} |
1.04(0.96101.12)
-•—
	 • 	

•


	 • 	
	 • 	
1

-1012 -1012 0.85 1 1.2 1.5
increase in SBP (mrnHg) Increase in DBP (mrnHg) Hypertension OR
      In the Normative Aging Study, Hu et al. (1996a) reported the cross-sectional association between bone Pb levels and the
      prevalence of hypertension and Cheng et al. (2001) reported the cross-sectional association between bone Pb levels and systolic
      BP in study participants free of hypertension at baseline.
      Note: The studies are ordered by increasing mean bone Pb levels. The area of each square is proportional to the inverse of the
      variance of the estimated change or log relative risk. Horizontal lines represent 95% confidence intervals. Diamonds represent
      summary estimates from inverse-variance weighted random effects models. Because of the small number of studies, summary
      estimates are presented primarily for descriptive purposes. RR indicates risk ratio.
      Source: Reprinted with permission of Elsevier Publishers, Navas-Acien et al. (2008)

      Figure 5-23     Meta-analysis  of an  increase in systolic BP (SBP) and diastolic BP
                         (DBP) and relative risk of hypertension per 10 ug/g  increase  in
                         bone Pb levels.
 i
 2
 o
 6
 4
 5
 6
 9
10
A few recent epidemiologic studies also emphasized the potential interaction between
measures of long-term Pb exposure, i.e., bone Pb levels, and factors such as chronic
stress and HFE genetic variants to moderate or modify the relationship of BP and
hypertension with Pb. For example, among NAS men, tibia Pb level was associated with
a larger risk of developing hypertension in an originally nonhypertensive group among
men with higher self-reported stress (Peters et al.. 2007).

In addition to stress, recent epidemiologic studies investigated effect modification by
race/ethnicity and genetic variants. In the NHANES 1988-1994 population of adults, the
association of concurrent blood Pb with systolic BP was higher among Mexican
Americans.  In the same NHANES population, the association between blood Pb level and
      November 2012
                                                    Draft - Do Not Cite or Quote

-------
 1                  hypertension was higher among non-Hispanic Blacks with the ALAD2 allele (see Figure
 2                  5-18 and Figure 5-19 for results) (Scinicariello et al.. 2010). Additionally, the association
 3                  between blood Pb and PP was larger among NAS men with the HFE H63D variant
 4                  (Figure 5-18) (Zhang etal.. 2010a). PP represents a good predictor of cardiovascular
 5                  morbidity and mortality and an indicator of arterial stiffness. The aforementioned genes
 6                  are related to iron metabolism and have been  linked with differences in Pb distribution in
 7                  blood and bone. Park et al. (2009b) provided  further evidence of variants in iron
 8                  metabolism genes impacting the association of bone Pb levels with QT interval changes
 9                  (see Table 5-21 for results).

10                  Animal toxicological evidence continued to build on the evidence characterizing  the
11                  mechanisms leading to these Pb-induced cardiovascular alterations. Biological
12                  plausibility for the consistent associations observed between blood and bone Pb and
13                  cardiovascular effects is provided by enhanced understanding of Pb-induced oxidative
14                  stress including NO inactivation, endothelial dysfunction leading to altered vascular
15                  reactivity, activation of the RAAS, and vasomodulator imbalance.
      5.4.3        Vascular Effects and Cardiotoxicity

16                  Not only has Pb been shown to increase BP and alter vascular reactivity, but Pb can alter
17                  cardiac function, initiate atherosclerosis, and increase cardiovascular mortality. Past
18                  toxicological studies have reported that Pb can increase atheromatous plaque formation in
19                  pigeons, increase arterial pressure, decrease heart rate and blood flow, and alter cardiac
20                  energy metabolism and conduction (Prentice and Kopp. 1985; Revis etal.. 1981). A
21                  limited number of available epidemiologic studies discussed in the 2006 Pb AQCD (U.S.
22                  EPA. 2006b) provided evidence of associations of blood Pb level with ischemic heart
23                  disease (IHD) and peripheral artery disease (PAD).
                    5.4.3.1      Effects on Vascular Cell Types

24                  The endothelial layer is an important constituent of the blood vessel wall, which regulates
25                  macromolecular permeability, VSMC tone, tissue perfusion, and blood fluidity. Damage
26                  to the endothelium is an initiating step in development of atherosclerosis, thrombosis, and
27                  tissue injury. Given that epidemiologic and toxicological evidence suggests that long-
28                  term Pb exposure is associated with a number of these conditions, numerous
29                  toxicological studies have investigated and found  an effect of Pb on endothelial
30                  dysfunction. A recent occupational study found that endothelial function assessed by
      November 2012                            5-325                  Draft - Do Not Cite or Quote

-------
 1                  flow-mediated dilatation was impaired in highly Pb-exposed workers (mean blood Pb
 2                  levels: 24.1 in workers versus 7.8 (ig/dL in unexposed controls) (Por^baet al.. 2010).

 3                  The endothelial layer makes up only a small part of the vascular anatomy; the majority of
 4                  the vessel wall is composed of VSMCs, which work in concert with the endothelial cells
 5                  (EC) in contraction and relaxation of the vessel, local BP regulation, and atherosclerotic
 6                  plaque development. Since Pb has been shown repeatedly to result in hypertension and
 7                  vascular disease in experimental animals, studies continued to investigate and find an
 8                  effect of Pb on VSMCs.

 9                  In in vitro assays, Pb (50 (iM, 2 weeks) stimulated VSMC invasiveness in isolated human
10                  arteries leading to the invasion of medial VSMC into the vessel intima and development
11                  of intimal hyperplasia, a key step in atherosclerotic progression (Zeller et al.. 2010). In
12                  addition, treatment with Pb (50 (iM, 12 hours) promoted VSMC elastin expression and
13                  increased arterial extracellular matrix in isolated human arteries. VSMC invasiveness was
14                  also increased in culture by treatment with  supernatant of Pb-treated human EC (50 (JVI),
15                  suggesting that Pb-exposed ECs secrete an activating compound. This compound was
16                  confirmed to be IL-8. Pb exposure (5-50 (iM) was able to, in a concentration-dependent
17                  manner, increase IL-8 synthesis and secretion in human umbilical vein EC cultures
18                  through activation of the transcription factor Nrf2. Neutralization of IL-8  could block
19                  VSMC invasion and arterial intima thickening (Zeller et al.. 2010). This study provides
20                  evidence that Pb exposure stimulates ECs to secrete IL-8 in an Nrf2-dependent manner
21                  which stimulates VSMC invasion from the vessel media to intima leading to a vascular
22                  thickening and possibly atherogenesis.

23                  A number of CVDs, including atherosclerosis, are characterized by increased
24                  inflammatory processes. Numerous studies have shown that Pb exposure is associated
25                  with an inflammatory environment in vascular tissues  of humans and animals as indicated
26                  by higher levels of inflammatory mediators like prostaglandin E2 (PGE2). Human aortic
27                  VSMCs treated with Pb (1 (JVI, 1-12 hours) exhibited increased secretion of PGE2 time-
28                  dependently through enhanced gene transcription (Chang et al.. 2011). This was preceded
29                  by a Pb-induced increase in the gene expression of cytosolic phospholipase A2 (cPLA2)
30                  and COX-2, two rate limiting enzymes in the regulation of prostaglandins. The induction
31                  of these enzymes was mediated by activation of ERK1/2, MEK1, and MEK2. Further
32                  investigation of the entrance of Pb into the  cell revealed that inhibition of the store-
33                  operated calcium channels (SOC) could only partially suppress cPLA2 and COX
34                  activation by Pb; however inhibition of epidermal growth factor receptor (EGFR)
35                  attenuated Pb-induced PGE2 secretion and activation of cPLA2 and COX. A follow-up  to
36                  this study found that Pb treatment (1(JVI) of a human epithelial cell line increased COX-2
37                  gene expression, promoter activity, and protein (Chou et al.. 2011). Inhibition of NF-KB
      November 2012                            5-326                  Draft - Do Not Cite or Quote

-------
 1                  decreased the Pb-induced COX-2 activation; whereas EGFR inhibition blocked COX-2
 2                  upregulation and NF-KB nuclear translocation. Overall these results suggest that Pb can
 3                  induce pro-inflammatory events in VSMC in the form of increased PGE2 secretion and
 4                  expression of cPLA2 and COX-2 through activation of EGFR via ERK1/2 and NF-KB
 5                  pathways.

 6                  Damage to the endothelium is a hallmark event in the development of atherosclerosis.
 7                  Past studies have shown that Pb exposure results in de-endothelialization, impaired
 8                  proliferation, and inhibition of endothelium repair processes after injury (Fujiwara et al.,
 9                  1997; Uedaetal..  1997; Kajietal.. 1995; Kishimoto et al. 1995). However, Pb exposure
10                  was not found to lead to nonspecific cytotoxicity at low exposure levels (2-25 (iM) as
11                  shown by the lack of release of lactate dehydrogenase (LDH) from Pb-treated bovine
12                  aortic EC (Shinkai etal. 2010). Instead, Pb induced specific apoptosis (caspase3/7
13                  activation) through endoplasmic reticulum (ER) stress that was protected against by the
14                  ER chaperones glucose-regulated protein 78 (GRP78) and glucose-regulated protein 94
15                  (GRP94). GRP78  and GRP94 play key roles in the adaptive unfolded protein response
16                  that serves as a marker of and acts to alleviate ER stress. Exposure of ECs to Pb induced
17                  GRP78 and GRP94 gene (2-25 \M) and protein (GRP78 [5-25 \M\ and GRP94
18                  [10-25 \\M\) expression through activation of the IREl-JNK-AP-1 pathways (Shinkai et
19                  al., 2010). This finding suggests that the functional damage in ECs caused by Pb
20                  exposure may be partly attributed to induction  of ER stress.
                    5.4.3.2      Cholesterol

21                  As blood cholesterol rises so does the risk of coronary heart disease. Previous
22                  occupational studies (Ademuviwa et al., 2005a; Beneretal.. 200Ib; Kristal-Boneh et al.,
23                  1999) examining higher than current adult blood Pb levels (>40 (ig/dL) reported higher
24                  total cholesterol levels related to Pb exposure, but mixed results for HDL, LDL, and
25                  triglycerides. More recently, Poreba et al. (2010). in an occupational study, reported no
26                  significant differences in parameters of lipid metabolism between Pb exposed workers
27                  (mean blood Pb level: 25 (ig/dL) and unexposed individuals. Conversely, Kamal et al.
28                  (2011) reported that occupational Pb exposure (mean blood Pb level: >40 (ig/dL) was
29                  associated with higher levels of triglycerides, total cholesterol, and LDL, and decreased
30                  HDL-C. Other Pb studies adjusted models for total cholesterol to control for this coronary
31                  heart disease risk factor. Higher mean total cholesterol with higher blood Pb levels has
32                  been reported in aNHANES study (Menke et al., 2006). In developing models to predict
33                  bone Pb levels, Park et al. (2009c) noted in a NAS study that total and HDL cholesterol
34                  were selected as 2 of 18 predictors for the bone  Pb level model. Their findings suggested
35                  that higher Pb exposure in nonoccupationally-exposed men may be associated with


      November 2012                            5-327                  Draft - Do Not Cite or Quote

-------
 1                   higher total and HDL cholesterols. In support of epidemiologic evidence, a recent
 2                   toxicological study reported increased LDL and decreased HDL in rats treated with Pb
 3                   (20 ppm, i.p., 3 days/week, 8 weeks) (Roshan et al., 2011). The major risk factor that
 4                   lipids represent for heart disease make relating  lipid levels to Pb exposures an interesting
 5                   but challenging hypothesis to test.
                     5.4.3.3      Atherosclerosis

 6                   A small number of toxicological and cross-sectional epidemiologic studies provide
 7                   evidence for increased atherosclerosis and intimal medial thickening (IMT) due to Pb
 8                   exposure. The association of stroke subtypes and severity of cerebral atherosclerosis was
 9                   examined in relation to a single concurrent blood Pb level and total 72-hour urinary Pb
10                   level (body Pb store-EDTA mobilization test) in a cross-sectional study of 153 patients
11                   (mean age 63.7 years) receiving digital subtraction angiography in Chang Gung
12                   Memorial Hospital in Taiwan from 2002 to 2005 (Lee et al., 2009). In an analysis
13                   adjusted for age, sex, hypertension, diabetes, triglyceride, uric acid, smoking, and alcohol
14                   consumption, a 1 (ig increase in urine Pb was associated with > 50% stenosis in the
15                   intracranial carotid system with an OR of 1.02 (95% CI: 1.00, 1.03). Urine Pb was not
16                   associated with greater stenosis in the extracranial or vertebrobasilar systems. Blood Pb
17                   level was not associated with greater stenosis in any region. As the development of
18                   atherosclerosis is a lifelong process, body Pb stores, analyzed by total  72-hour urine Pb
19                   amount, may more strongly be associated with atherosclerosis than are single blood Pb
20                   measurements.

21                   A recent study correlated greater carotid artery IMT with higher concurrent serum Pb
22                   levels (mean [SD] 0.41 [0.38] ng/dL) in hemodialysis patients (Ari et al.. 2011). A few
23                   available recent occupational studies also presented evidence for increased measures of
24                   atherosclerosis in highly Pb-exposed adult populations with mean blood Pb levels around
25                   25 (ig/dL. Por^ba et al. (2011 a) reported increased local arterial stiffness and more
26                   frequent left ventricular diastolic dysfunction in Pb-exposed workers with hypertension
27                   compared to nonexposed controls with hypertension. Occupational exposure to Pb (mean
28                   blood Pb levels: 24 (ig/dL in workers, 8.3 (ig/dL in nonexposed group) was also
29                   associated with greater IMT and atherosclerotic plaque presentation, analyzed by Doppler
30                   ultrasound (Porebaet al., 2011).

31                   Zeller et al. (2010) examined human radial and internal  mammary arteries exposed to Pb
32                   in culture and reported a concentration-dependent increase  in arterial intimal thickness
33                   (statistically nonsignificant at 5 (iM Pb, significant at 50 (iM Pb, 2 week treatment) and
34                   intimal extracellular matrix accumulation (50 (JVI). Also, Pb promoted EC proliferation
      November 2012                             5-328                 Draft - Do Not Cite or Quote

-------
 1                  (5 and 50 (JVI, 72 hours) and VSMC elastin expression (50 (JVI, 12 hours), as discussed
 2                  above (Section 5.4.3.1) (Zeller et al.. 2010). Another study showed that Pb exposure
 3                  (100 ppm in drinking water for 10 months; mean blood Pb level 28.4 (ig/dL) of rats also
 4                  increased the aortic media thickness, media-lumen ratio, and medial collagen content
 5                  (Zhang et al., 2009a). These morphological changes to the vessel due to Pb exposure
 6                  indicate initiation of arteriosclerosis and could be the cause of decreased contractile
 7                  response of the vessel due to altered visco-dynamic vessel properties. Alternatively, these
 8                  vascular changes could be an effect of Pb-induced hypertension.
                    5.4.3.4      Heart Rate Variability

 9                  HRV and BP are regulated, in part, by the sympathetic and parasympathetic nervous
10                  systems. Changes in either may increase the risk of cardiovascular events. HRV is
11                  defined as the oscillation in the interval between consecutive heart beats and between
12                  consecutive instantaneous heart rates. Decreases in HRV have been associated with
13                  cardiovascular mortality/morbidity in older adults and those with significant heart disease
14                  [(1996). Task Force of the European Society of Cardiology and the North American
15                  Society of Pacing and Electrophysiology]. In addition, decreased HRV may precede
16                  some clinically important arrhythmias, such as atrial fibrillation, as well as sudden
17                  cardiac death, in high risk populations (Chen and Tan. 2007; Sandercock and Brodie.
18                  2006).

19                  Pb has been shown not only to affect vascular contractility in animals, but also is
20                  associated with cardiac contractility. The 2006 Pb AQCD (U.S. EPA. 2006b) described
21                  one study that investigated Pb-induced alterations in HRV (Cheng etal.  1998). Cheng et
22                  al. (1998) found increasing duration of corrected QT interval (QTc) with increasing bone
23                  Pb levels in men <65 years, but not in men > 65 years. Bum et al. (2011)  and Park et al.
24                  (2009b) followed up this previous NAS cohort (Cheng etal., 1998) (details found in
25                  Table 5-21). Bum et al. (2011) prospectively examined the association between blood and
26                  bone Pb levels and the development of electrocardiographic (ECG) conduction
27                  abnormalities among 600 men who were free of ECG abnormalities at the baseline
28                  assessment. A second ECG was obtained for 496 men 8.1 (SD: 3.1) years later on
29                  average. Baseline Pb concentrations in blood (mean [SD]: 5.8 [3.6] (ig/dL), patella bone
30                  (mean [SD]: 30.3 [17.7] (ig/g), and tibia bone (mean [SD]: 21.6 [12.0] (ig/g) were similar
31                  to those found in other samples from the general U.S. adult population and much lower
32                  than those reported in occupationally exposed groups. Higher tibia Pb was associated
33                  with increases in QTc interval and QRSc duration. Compared with those  in the lowest
34                  tertile of baseline tibia Pb (<16 (ig/g), participants in the highest tertile  (>23  (ig/g) had a
35                  7.94 msec (95% CI: 1.42, 14.45) greater increase in QTc interval and a 5.94 msec (95%


      November 2012                            5-329                  Draft - Do Not Cite or Quote

-------
 1                  CI: 1.66, 10.22) greater increase in QRSc duration over 8 years after adjusting for
 2                  covariates: age, education, smoking, BMI, albumin-adjusted serum calcium, and diabetes
 3                  status at baseline, and years between ECG tests and QT-prolongation drugs at the time of
 4                  ECG measurement. There were no statistically significant associations with patella or
 5                  blood Pb levels. These associations with tibia bone Pb levels were observed in men with
 6                  relatively low blood and bone Pb concentrations who were free of cardiac conduction
 7                  abnormalities at baseline and were examined prospectively. Thus, they indicate that long-
 8                  term cumulative Pb exposure may increase the risk of developing cardiac abnormalities.
 9                  Uncertainty exists as to the specific Pb exposure level, timing, frequency, and duration
10                  contributing these associations observed for tibia Pb levels. A recent occupational study
11                  reported lower HRV and abnormal parameters of heart rate turbulence in Pb-exposed
12                  workers (mean blood Pb levels: -25 (ig/dL) compared to control subjects (Por^ba et al.,
13                  201 Ib).

14                  Park et al. (2009b) cross-sectionally examined whether polymorphisms in genes known
15                  to alter iron metabolism (HFE, transferrin [TF] C2, heme oxygenase-1 [HMOX-1])
16                  modify the association between Pb biomarker levels and the QT interval. Investigators
17                  examined associations in data stratified on polymorphisms in the three genes. They also
18                  analyzed interaction models with cross-product terms for genotype and the Pb biomarker.
19                  The distributions of all genotypes but the HFE variant, H63D, were in Hardy-Weinberg
20                  equilibrium. Subjects homozygous for the other HFE variant, C282Y, had higher bone Pb
21                  levels and those homozygous for H63D and heterozygous with both C282Y and H63D
22                  had lower bone Pb levels. The antioxidant HMOX-1 L variant (longer repeats of GT,
23                  associated with lower enzyme inducibility) alone, compared to the wild type, showed a
24                  statistically significant interaction with tibia Pb (11.35 msec longer QTc interval for each
25                  13 (ig/g increase in bone Pb in L-allele variants). No other gene variant alone showed
26                  different Pb-associated  QTc intervals from those in wild types, either for tibia and patella
27                  Pb or for (linear) concurrent blood Pb. Lengthening of QTc with higher tibia and blood
28                  Pb was more pronounced with an increase in the total number of gene variants, driven by
29                  a joint effect between HFE variant and HMOX-1 L allele. There was a trend observed
30                  with blood and tibia Pb-associated  QTc interval increasing with increasing number of
31                  gene variants from 0 to 3. This study provided further evidence of gene variants
32                  modifying associations of Pb biomarkers with cardiovascular effects.

33                  The interaction of key markers of the metabolic syndrome with bone Pb levels in
34                  affecting HRV was cross-sectionally investigated in a group of 413 older adults with
35                  patella Pb measurements in the NAS (Park et al.. 2006). Metabolic syndrome was defined
36                  to include three or more of the following: waist circumference >102 cm,
37                  hypertriglyceridemia (>150 mg/dL), low HDL cholesterol (<40 mg/dL in men), high BP
38                  >130/85 mmHg, and high fasting glucose (>110 mg/dL). Men using antihypertensive
      November 2012                            5-330                 Draft - Do Not Cite or Quote

-------
 1                  medication or diabetes medications were counted as high BP or high fasting glucose,
 2                  respectively. The strongest relationships between patella Pb levels and lower HRV were
 3                  observed among those with three or more metabolic abnormalities. A trend was observed
 4                  for larger patella Pb-associated decreases  in HRV with increasing number of metabolic
 5                  abnormalities. These results suggest multiplicative effects of cumulative Pb exposure and
 6                  metabolic abnormalities on key predictors of CVD. Park et al. (2006) also reported the
 7                  penalized spline fits to bone Pb in models assessing only main effects of bone Pb. The
 8                  optimal degree of smoothing determined by the generalized cross-validation criterion for
 9                  all HRV measures was 1, which indicated that the associations were nearly linear. The
10                  spline fits and associated statistics showed that the bone Pb main effects on HRV
11                  measures were linear. However, the relationship with LF/HF was linear with log(LFTHF).

12                  Increased incidence of arrhythmia and atrioventricular conduction block was found in
13                  rats after 12 weeks of Pb exposure (100 ppm; mean blood Pb level 26.8 (ig/dL) (Reza et
14                  al.. 2008). Also, Pb exposure for 8 weeks  increased heart rate and systolic BP. These
15                  increases corresponded with increased cardiac contractile force and prolonged ST
16                  interval, without alteration in QRS duration or coronary flow. In contrast, another study
17                  using rat right ventricular strips found that Pb (100 (iM) exposure, in a concentration-
18                  dependent manner, reduced myocardial contraction by reducing sarcolemmal Ca2+ influx
19                  and myosin ATPase activity (Vassallo et al.. 2008). This study also found that Pb
20                  exposure changed the response to inotropic agents and blunted the force produced during
21                  contraction. Conversely, past studies have found that Pb exposure increases intracellular
22                  Ca2+ content (Laletal..  1991: Favalli et al..  1977: Piccinini et al.. 1977). which could
23                  result in increased cardiac output and hypertension.
                     5.4.3.5      Peripheral Artery Disease

24                   Peripheral artery disease (PAD) is an indicator of atherosclerosis and measured by the
25                   ankle brachial index, which is the ratio of BP between the posterior tibia artery and the
26                   brachial artery. PAD is typically defined as an ankle brachial index of less than 0.9.
27                   Muntner et al. (2005). whose results describing the association of blood Pb and
                                  v	/ 7                      O
28                   hypertension in the NHANES 1999-2002 data set for adults were discussed previously,
29                   also examined the association of blood Pb with PAD (details found in Table 5-21). The
30                   authors observed an increasing trend in the odds of PAD with increasing concurrent
31                   blood Pb level. The OR for PAD comparing the fourth quartile of blood Pb (>2.47 (ig/dL)
32                   to the first quartile of blood Pb (<1.06 (ig/dL) was 1.92 (95% CI: 1.02, 3.61). Key
33                   potential confounding factors were adjusted for in the analysis. These results are
34                   consistent with those from a previous NHANES analysis  by Navas-Acien et al. (2004)
35                   reviewed in the 2006 Pb AQCD.
      November 2012                             5-331                  Draft - Do Not Cite or Quote

-------
 1                  Navas-Acien et al. (2004) reported a trend of increasing OR for PAD with increasing
 2                  quartile of concurrent blood Pb or Cd in adults who were 40 years of age in the
 3                  1999-2000 NHANES population. These authors tested both Pb and Cd in separate
 4                  models, tested the metals simultaneously, and tested the interaction between the metals.
 5                  The correlation coefficient between natural log Pb and natural log Cd was 0.32
 6                  (p <0.001). Although the interaction was not statistically significant, when blood Pb and
 7                  blood Cd were in the same model, the ORs were diminished slightly. Both showed
 8                  statistically significant trends of increasing OR with increasing quartile of the metal.
 9                  These results indicate that blood Cd levels did not confound the association between
10                  blood Pb level and PAD. In a subsequent analysis, Navas-Acien et al. (2005) used the
11                  same 1999-2000 NHANES dataset, but constructed PAD models using a suite of urine
12                  metal concentrations. Power was reduced in this study because only 659-736 subjects
13                  (compared to 2,125) had spot urine metal tests in the data set. Urinary Cd, but not urinary
14                  Pb, was consistently associated with PAD in all models. Associations also were observed
15                  with urinary antimony and tungsten. Spot urine Pb measurements are less reliable
16                  compared to blood Pb measurements. In Navas-Acien et al. (2005). the urinary Pb level
17                  association with PAD was sensitive to adjustment for urinary creatinine, indicating that
18                  spot urine Pb measurements are affected by differences in urine dilution.  This finding
19                  illustrates the limited reliability of spot urine Pb measurements compared to blood Pb
20                  measurements.
                    5.4.3.6     Ischemic Heart Disease

21                  A few cross-sectional studies discussed in the 2006 Pb AQCD (U.S. EPA. 2006b)
22                  indicated associations between Pb biomarker levels and increased risk of cardiovascular
23                  outcomes associated with IHD, including left ventricular hypertrophy (Schwartz. 1991)
24                  and myocardial infarction (Gustavsson et al.. 2001). Recently, Jain et al. (2007) provided
25                  prospective evidence for the incidence of IHD (physician confirmed MI, angina pectoris)
26                  among older adult males enrolled in the NAS that were followed during the period of
27                  September, 1991 to December, 2001 (details found in Table 5-21). All subjects had blood
28                  Pb and bone Pb measurements with no IHD at enrollment. Fatal and nonfatal cases were
29                  combined for analysis. Baseline blood, tibia, and patella Pb levels were log-transformed.
30                  Blood Pb level and patella Pb level were associated with increased risk of IHD over the
31                  10-year follow-up period. When blood Pb and patella Pb were included simultaneously in
32                  the model, each of their HRs was only moderately attenuated (HR: 1.24 [95% CI: 0.80,
33                  1.93] per SD increase in blood Pb and HR: 2.62 [95% CI: 0.99, 6.93]  per SD increase in
34                  patella Pb). When blood Pb and tibia Pb were included simultaneously in the model, their
35                  risk estimates were only moderately attenuated (HR:  1.38 [95% CI: 0.89, 2.13] per SD
      November 2012                            5-332                 Draft - Do Not Cite or Quote

-------
 1                  increase in blood Pb and HR: 1.55 [95% CI: 0.44, 5.53] per SD increase in tibia Pb).
 2                  These findings indicate that both blood and bone Pb levels are independently associated
 3                  with IHD incidence.

 4                  IHD, characterized by reduced blood supply to the heart, may result from increased
 5                  thrombosis. In support of the epidemiologic evidence, a recent animal study suggested
 6                  that Pb exposure promotes a procoagulant state that could contribute to thrombus
 7                  formation (Shin et al.. 2007). In a rat model of venous thrombosis, Pb treatment (i.v.
 8                  25 mg/kg) resulted in increased thrombus formation, although i.v. Pb treatment may have
 9                  uncertain relevance to human routes of Pb exposure. Additionally, Pb treatment to human
10                  erythrocytes (red blood cells, RBCs)  increased coagulation at a dose of 5 (iM and
11                  thrombin generation in a concentration-dependent manner at doses from 2-5(iM. This
12                  enhanced procoagulant activity in Pb-treated RBCs was the result of increased outer cell
13                  membrane phosphatidylserine (PS) surfacing (human RBCs: 2-5  (iM Pb; rat RBCs: 5 (iM
14                  Pb).  Similar to these in vitro results, PS externalization on erythrocytes was increased in
15                  Pb-treated rats (i.v. 50-100 mg/kg, not 25 mg/kg). Increased PS externalization was likely
16                  the result of increased intracellular calcium (5 (iM Pb), enhanced scramblase activity
17                  (5-10 (iM Pb), inhibited flippase activity (5-10 (iM Pb), and ATP depletion (1-5 (iM Pb)
18                  after Pb exposure (Shin et al.. 2007).
     November 2012                            5-333                 Draft - Do Not Cite or Quote

-------
Table 5-21   Characteristics and quantitative data for associations of blood and
            bone Pb with other CVD measures HRV, PAD, and IHD in recent
            epidemiologic studies.
Study



(Ordered as Study
they appear in Population/
the text)
Heart rate
Eum et al.
(2011)












Parket al.
(2009b)







Methodology
variability
Longitudinal
600 men free of
electrographic
abnormalities at
the time of
baseline ECG
from MAS in
Greater Boston,
MA area (496
with follow-up
ECG 8 years
later)











Cross-sectional
61 3 men from
MAS in Greater
Boston MA area
(8/1 991'-
12/1995)





Parameter Pb Data

ECG Baseline Blood Pb:
conduction" Mean (SD): 5.8
(QTc, QRSc, (3.6) ug/dL
JTc,
QT prolongation,
JT prolongation, Baseline Patella Pb:
IVCD° AVCD Mean (SD): 30.3
Arrhythmia) ' (17.7) ug/g
Baseline Tibia Pb:
Mean (SD): 21.6
(12.0) ug/g
Q1: <16 ug/g (n = 191)
Q2: 16.0-23 ug/g
(n = 208)
Q3: >23 ug/g (n = 195)







QTcb interval Blood Pb:
Median (IQR):
5 (4-7) ug/dL
Patella Pb:
Median (IQR):
26 (18-37) ug/g

Tibia Pb:
Median (IQR):
19 (14-27) ug/g
Statistical Analysis

Repeated measures
linear regression
adjusted forage,
education, smoking,
BMI, albumin-adjusted
serum Ca +, and
diabetes status at
baseline, and years
between ECG tests
and QT-prolongation
drugs at the time of
ECG measurement.











Linear regression
models adjusted for
age, BMI, smokina
status, serum Ca ,
and diabetes. No SES
indicator was
considered.






Effect Estimate (95%
r*na
Cl)

Tibia Pb:
Adjusted 8-year
change
(95% Cl):
QTc:
Q2 vs. Q1 (reference):
7.49(1.22, 13.75)
msec,
Q3vs. Q1:
7.94(1.42,14.45)
msec
p for trend = 0.03

QRSc:
Q2vs. Q1:
0.52 (-3.60, 4.65)
msec
Q3vs. Q1:
5.94(1.66,10.22)
msec
p for trend = 0.005
No associations with
patella or blood Pb
Per IQR (3 ug/dL)
increase in blood Pb:
1.3 (-0.76, 3.36) msec
after 8-year follow up
Per IQR (19 ug/g)
increase in patella Pb:
2.64(0.13,5.15) msec

Per IQR (13 ug/g)
increase in tibia Pb:
2.85 (0.29, 5.40) msec
November 2012
Draft - Do Not Cite or Quote

-------
Study
(Ordered as
they appear in
the text)
Study
Population/
Methodology
Parameter
Pb Data
Statistical Analysis
                                                                Effect Estimate (95%
                                                                Cl)a
Parket al.
(2006)
               Cross-sectional
               413 men from
               MAS in Greater
               Boston, MA area
               (11/14/2000-
               12/22/2004)
                 HRV

                 (SDNN, HF,
                 r~irnormi Lr,
                 LFnorm, LF/HF)
                 Patella Pb (measured
                 within 6 mo of HRV:
                 Median (IQR):
                 23.0 (15-34) ug/g
                 Estimated3: Median
                 (IQR):
                 16.3(10.4-25.8) ug/g


                 Tibia Pb:
                 Median (IQR):
                 19.0(11-28) ug/g
                         Log linear regression
                         models adjusted for
                         age, cigarette
                         smoking, alcohol
                         consumption, room
                         temperature, season
                         (model 2) BMI, fasting
                         blood glucose, HDL
                         cholesterol,
                         triglyceride, use of p-
                         blockers, Ca2+ channel
                         blockers, and/or ACE
                         inhibitors. No SES
                         indicator was
                         considered.
                      Tibia Pb: Model 2
                      Change (95%CI)
                      HF:
                      -0.9 (-3.8, 2.1)
                      normalized units (nu)
                      LF:
                      0.9 (-2.0, 3.9) nu
                      Log LF/HF:
                       3.3 (-10.7, 19.5) (%)
                      Per 17 ug/g tibia Pb


                      Patella Pb:
                      Model 2 Change
                      (95%CI)
                      HF:-0.6 (-3.1, 1.9)
                      nu
                      LF: 0.6 (-1.9, 3.1)nu
                      Log LF/HF: 3.0 (-8.7,
                      16.2) (%)
                      Per 15.4 ug/g patella
                      Pb
                                                                                                 Effect estimates were
                                                                                                 more pronounced
                                                                                                 among those with
                                                                                                 greater* metabolic
                                                                                                 abnormalities.
Peripheral artery disease
Muntner et al. Cross-sectional PAD
(2005) g 961 NHANES
(1999-2002)
participants



Navas-Acien Cross-sectional PAD
et al. (2QQ5) 7go partjcjpants,
age > 40 yr,
from NHANES
(1 999-2000)








Range Concurrent Blood
Pb:
Q1:<1.06ug/dL,
Q2: 1 .06-1 .63 ug/dL
Q3: 1 .63-2.47 ug/dL
Q4: >2.47 ug/dL



Concurrent urinary Pb:
Mean (1 Oth-90th
percentile)):
0.79 ug/L (0.2-2.3)









Logistic regression
models adjusted for
age, race/ethnicity,
sex, diabetes mellitus,
BMI, cigarette
smoking, alcohol
consumption, high
school education,
health insurance
status
Logistic regression
adjusted for the
following:
Model 1: age, sex,
race, and education
Model 2: covariates
above plus smoking
status
Model 3:covariates
above plus urinary
creatinine


OR (95% Cl):
Q1 : 1 .00 (Reference),
Q2: 1.00(0.45,2.22),
Q3: 1.21 (0.66,2.23),
Q4: 1.92(1.02,3.61)



Model 1:
OR: 1.17(0.81, 1.69)
Model 2:
OR: 1.1 7 (0.78, 1.76)
Model 3:
OR: 0.89 (0.45, 1.78)


Per IQR increase in
i irinsrx/ Ph
ui M icti y ~u

Array of metals in
urine also evaluated.
November 2012
                                   5-335
                                            Draft - Do Not Cite or Quote

-------
      Study
      (Ordered as
      they appear in
      the text)
Study
Population/
Methodology
Parameter
Pb Data
Statistical Analysis
                                                        Effect Estimate (95%
                                                        Cl)a
Ischemic Heart Disease
Jain et al. Longitudinal IHD
t^uU' ' 837 men from (Ml or anoina
NAS in Greater pectoris)
Boston, MA area
(1991-2001)







Baseline Blood Pb Mean
(SD):
Non-cases
6.2 (4.3) ug/dL;
Cases
7.0 (3.8) ug/dL

Baseline Patella Pb
Mean (SD):
Non-cases
30.6(19.7) ug/dL;
Cases
36.8 (20.8) ug/dL
Cox proportional
hazards models
adjusted forage, BMI,
education, race,
smoking status, pack-
years smoked, alcohol
intake, history of
diabetes mellitusand
hypertension, family
history of
hypertension, DBP,
SBP, serum
triglycerides, serum
HDL, and total serum
cholesterol
Blood Pb level
> 5 ug/dL
OR over 10-year
follow-up: 1.73(1.05,
2.87)

Ln [blood Pb] OR:
1.45(1.01,2.06)
Ln [patella Pb level]
OR'
2.64(1.09,6.37)
Ln [tibia Pb level ]OR:
1.84(0.57,5.90)
                                                 Baseline Tibia Pb Mean
                                                 (SD):
                                                 Non-Cases
                                                 21.4 (13.6) ug/g;
                                                 Cases
                                                 24.2(15.9) ug/g


                                                 Cases:
                                                 Blood Pb range:
                                                 1.0 to 20.0 ug/dL
                                                 Patella Pb range:
                                                 5.0 to 101 ug/g
                                                 Tibia Pb range:
                                                 -5 to 75 ug/g
                                                                       Per 1 SD increase in
                                                                       Pb biomarker
      "Estimated patella Pb accounts for declining trend in patella Pb levels between analysis of bone Pb and HRV.
      bHeart-rate-corrected QT interval calculated by Bazett's formula
      °IVCD, intraventricular conduction defect; AVCD, atrioventricular conduction defect
 1

 2

 o
 J

 4

 5

 6

 7
10

11

12
  5.4.3.7      Summary of Vascular Effects and Cardiotoxicity


  There are a limited number of studies in a limited number of populations that investigate

  the associations between Pb biomarkers and cardiovascular effects other than BP or

  hypertension (Table 5-21). As presented in Table 5-21. these studies demonstrated

  associations between various biomarkers of Pb exposure and clinical cardiovascular

  outcomes such as atherosclerosis, IHD, PAD, and HRV occurrence in adult populations

  after adjusting for potential confounding by variables such as age, sex, education, BMI,

  smoking, alcohol consumption, and diabetes. In a limited body of studies, mixed

  evidence of association between occupational exposure to Pb and altered cholesterol was

  reported.


  Few studies have evaluated markers of subclinical atherosclerosis such as PAD and IMT

  following Pb exposure in humans or animals. Concurrent blood Pb levels (population

  means >2.5 (ig/dL) were associated with greater odds of PAD in adults in NHANES
      November 2012
                              5-336
                                       Draft - Do Not Cite or Quote

-------
 1                   analyses (Muntner et al.. 2005; Navas-Acien et al., 2004). Since these effects are
 2                   observed in adults that may have had higher past exposure to Pb, there is uncertainty as to
 3                   the specific Pb exposure level, timing, frequency, and duration that contributed to the
 4                   observed associations. A recent study involving both human and toxicological studies
 5                   observed Pb-mediated arterial IMT, an early event in Pb-induced atherogenesis (Zeller et
 6                   al.. 2010). A second study in rats report increased aortic media thickness following Pb
 7                   exposure (Zhang et al., 2009a). Toxicological studies of Pb-induced endothelial
 8                   dysfunction, VMSC invasiveness, and inflammation in isolated vascular tissues and cells
 9                   provide mechanistic evidence to support the biological plausibility of these vascular
10                   effects and cardiotoxicity. Studies in isolated tissues and cells found that Pb stimulated
11                   the synthesis and secretion of IL-8 in ECs, which was responsible for stimulating VSMC
12                   invasion into the vessel intimal layer. Pb treatment also increased extracellular matrix and
13                   elastin, primary sites for lipid deposition in the vessel wall.

14                   Several studies report associations between biomarkers of Pb exposure and diseases
15                   associated with coronary heart disease (CHD), such as HRV, IHD, and MI. A prospective
16                   NAS study reported that higher baseline tibia Pb was associated with increases in QTc
17                   interval and QRSc duration over an 8-year follow-up period (Eumet al., 2011). In
18                   addition, in the NAS cohort of older adult men, blood Pb (> 5 (ig/dL) and patella Pb
19                   levels were associated with increased incidence of IHD (Jain et al., 2007). A recent study
20                   provided evidence for the interaction between biomarkers of Pb exposure and the HFE
21                   C282Y and HMOX-1 L variant on the prolonged QT interval in
22                   nonoccupationally-exposed older men (Park et al.. 2009b). Also, in the NAS population,
23                   bone Pb  levels were associated with larger decreases in HRV parameters among subjects
24                   identified as having metabolic abnormalities (Park et al.. 2006). These metabolic
25                   abnormalities, abdominal obesity, hypertriglyceridemia, low HDL cholesterol, high
26                   BP/medication use, or high fasting glucose, have been  shown to be associated with
27                   increased risk of cardiovascular events.

28                   Overall,  the relatively few available studies provide support for associations between Pb
29                   biomarkers and other cardiovascular conditions including subclinical atherosclerosis and
30                   CHD. A number of these are quality studies from two cohorts, NAS and NHANES with
31                   adequate sample size that account for potential confounding, with some being conducted
32                   prospectively.
      November 2012                             5-337                 Draft - Do Not Cite or Quote

-------
      5.4.4        Cardiovascular Function and Blood Pressure in Children
                     5.4.4.1      Introduction

 1                   The study of cardiovascular function effects in relation to blood Pb levels in children
 2                   potentially offers unique information on several topics. First, by examining endpoints
 3                   predictive of future cardiovascular pathology, these studies may offer information on the
 4                   potential cardiovascular effects of Pb exposure in an understudied population. Second,
 5                   examination of cardiovascular changes that are antecedent to increased BP and changes
 6                   in other CVD-related endpoints at later lifestages may inform uncertainties in regards to
 7                   the time course of cardiovascular changes associated with Pb exposure. Finally, these
 8                   studies address gaps  in knowledge regarding Pb exposure effects in populations of
 9                   children with mean blood Pb levels in the range of <10 (ig/dL.

10                   An important aspect  to the literature about the association between cardiovascular effects
11                   and blood Pb levels in children is that the blood Pb levels of children may better reflect
12                   relatively recent Pb exposure and its effect on CVD than blood Pb levels do in adults
13                   because of the much  longer exposure history of adults during which Pb exposures were
14                   commonly much higher than they are today. However, in older children there is still
15                   uncertainty regarding the frequency, duration, timing,  and magnitude of exposure
16                   contributing to the blood Pb levels measured. The much lower prevalence of
17                   cardiovascular effects in children, however, poses a challenge to investigations of
18                   potential relationships with Pb exposures. For example, the prevalence of hypertension in
19                   children (9 to  10 years old) ranges from to 2 to 5 percent (Daniels. 2011; Steinthorsdottir
20                   et al.. 2011). while more than half of people aged 60 to 69 years have hypertension
21                   (Chobanian et al., 2003). Accordingly, much larger study populations are required to
22                   provide similar statistical power for such studies in children as compared to adult studies.
23                   Further, in drawing interpretations from such studies with regard to potential effects of Pb
24                   exposures at later ages, it is additionally important to recognize that compensatory
25                   mechanisms in children may be more active than in adults, and the cardiovascular tissue
26                   of the young may be  less susceptible to damage than that  of adults.
      November 2012                             5-338                  Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
The limited numbers of cardiovascular studies published on children have examined
endpoints such as total peripheral resistance (TPR), BP, and autonomic nervous system
activation. These recent and earlier studies are presented in Table 5-22. Multiple single
pollutant studies in New York State evaluated two child cohorts born in the 1990s after
Pb was removed from gasoline in the U.S. with mean blood Pb levels of 4.62 and
1.01  (ig/dL (Gump etal.. 2011: Gump et al.. 2009: Gump et al. 2007: Gump et al.. 2005).
Zhang et al. (2011 a) examined children in Mexico City born from 1994 to 2003, when Pb
was being taken out of gasoline in Mexico as  indicated by Martinez et al. (2007). The
geometric means for cord and concurrent blood Pb levels of the children in the Mexico
City  cohort were 4.67 and 2.56 (ig/dL.
     Table 5-22   Studies of child cardiovascular endpoints and Pb biomarkers.
Study
(Ordered as
they appear in
the text)
Gump et al.
(2005)







Gump et al.
(2007)









Study
Population/
Methodology Parameters
Prospective SBP, TPR
122 children (total
age 9.5 yr in peripheral
Oswego, NY vascular
(born at a resistance)
single hospital
in New York
from 1991-94)




Prospective SBP, TPR
122 children
age 9.5 yr in
Oswego, NY









Blood Pb Data3
Cord blood Pb:
GM (GSD):
2.56ug/dL(1.16)

Childhood (mean
age of measurement:
2.6 yr) blood Pb: GM
/'OQnv
^ooUJ.
4.06ug/dL(1.14)


Childhood (mean
age of measurement:
2.6 yr) blood Pb:
GM (GSD):
4.06ug/dL(1.14)








Statistical Analysis
Multivariate linear
regression models
examined the
relationship of blood Pb
with change in z-score
for outcome (post- and
pre-stress). Potential
confounders
considered: HOME
score, SES, birth
weight, child BMI, child
sex.
Linear regression
models adjusting for
the same covariates as
in Gump et al. (2005).
Separate models
testing whether Pb is a
mediator of SES
associations, (Sobel
test) and whether Pb
moderates SES
associations (Pb-SES
interaction).



Effect Estimates/Results
Per 1 ug/dL increase in
childhood blood Pb level, 0.088
(95% Cl: 0.023, 0.153) dyne-
s/cm5 change in TPR
Per 1 ug/dL increase in cord
blood Pb level, 12.16 (95% Cl:
2.44, 21 .88) mmHg higher SBP




Blood Pb was a mediator of the
SES-TPR relationship
SES alone: -0.62 dyne-s/cm5
(p <0.05)
SES with Blood Pb: -0.40
dyne-s/cm5 (p >0.10), change
in R2 attributable to SES:
-55.3%

Blood Pb was a potential
moderator of the SES-TPR
relationship. Blood Pb x SES
interaction: p = 0.07 .
                                                                             Blood Pb was a moderator of
                                                                             SES-SBP relationship
                                                                             Pb x SES interaction:
                                                                             p = 0.007
                                                                             At blood Pb levels >4 ug/dL,
                                                                             SES not significantly
                                                                             associated with SBP
     November 2012
                          5-339
Draft - Do Not Cite or Quote

-------
Study
(Ordered as     Study
they appear in   Population/
the text)        Methodology   Parameters
                               Blood Pb Data3
                                     Statistical Analysis     Effect Estimates/Results
Gump et al.
(2009)







Prospective Salivary
122 children cortisol
age 9.5 yr in
Oswego, NY





Cord blood Pb:
GM (GSD):
2.56ug/dL(1.16)


Childhood (mean
age of measurement:
2.6 yr) blood Pb: GM
(GSD): 4.06 ug/dL
(1.14)
Linear regression to
examine whether blood
Pb level mediates or
moderates the
relationship between
SES and salivary
cortisol as in Gump et
al. (2007)

Blood Pb was a mediator of the
SES-cortisol association. SES
was no longer significantly
associated with cortisol after
adjusting for blood Pb level. R2
for SES decreased by 40, 33,
and 50% for cortisol measured
at 21, 40, and 60 min.
Blood Pb was not a significant
n-irtrlarotrt r r\f Q P Q_*^rt rtiort 1
                                                                                          association. Blood Pb x SES
                                                                                          interaction term was not
                                                                                          statistically significant
Gump et al.
(2011)
Cross-
sectional
140 children
ages 9-11 yr
Oswego, NY
SBP, TPR,
HRV (heart
rate
variability) in
response to
acute stress
(mirror
tracing task)
Concurrent blood Pb:
GM: 1.01 ug/dL
Quartiles:
Q1: 0.14-0.68 ug/dL
Q2: 0.69-0.93 ug/dL
Q3: 0.94-1.20 ug/dL
Q4: 1.21-3.76 ug/dL
Outcomes were
analyzed as continuous
variables for the pre-
stress values or the
change post- and pre-
stress. Regression
models were adjusted
for sex, SES, BMI, and
age.
Blood Pb levels associated
with autonomic and
cardiovascular dysregulation in
response to stress -greater
vascular resistance, reduced
stroke volume, and cardiac
output
Change in SBP (mmHg) across
quartiles: Q1: 5.30, Q2: 7.33,
Q3: 7.07, Q4: 7.23, p for
trend = 0.31
Change in TPR (%) across
quartiles: Q1: 2.91, Q2: 8.18,
Q3: 9.55, Q4: 9.51, p for
trend = 0.03
Change in Stroke Volume (%)
across quartiles: Q1: 2.23, Q2:
0.91, Q3:-3.47, Q4:-0.89, p
for trend = 0.04
Zhang et al.
(2011 a)









Prospective SBP
457 mother
child pairs in a
birth cohort,
born 1 994 to
2003 in
Mexico City.
Children were
evaluated
2008-201 0 at
ages 7-15 yr


Cord blood Pb:
GM (GSD):
4.67ug/dL(1.18)
(N=323)

Concurrent blood Pb:
GM (GSD):
2.56ug/dL(1.16)
(N=367)

Maternal post-
partum bone Pb:
Multiple regression
models and
generalized estimating
equations (log linear for
cord blood, linear for
concurrent blood and
maternal bone). The
base model considered
maternal education,
birth weight, BMI, sex,
and child concurrent
age as covariates.

Prenatal Pb exposure may be
associated with higher BP in
female offspring.
Among girls, an IQR (13 ug/g)
increase in maternal tibia Pb
was associated with a 2.11
(95% Cl: 0.69, 3.52) mmHg
increase in SBP
IQR (16 ug/g) increase in
maternal patella Pb was
associated with a 0.87 (95%
Cl: -0.75, 2.49) mmHg
increase in SBP
                                             Median (IQR):
                                             Tibia Pb: 9.3 (3.3,
                                             16.1) ug/g
                                             Patella Pb: 11.6(4.5,
                                             19.9) ug/g
                                                                            IQR (4 ug/dL) increase in cord
                                                                            blood Pb was associated with
                                                                            a 0.75 (95% Cl:-1.13, 2.63)
                                                                            mmHg increase in SBP
Factor-
Litvak et al.
(1999: 1996)
Cross-
sectional
260 children
ages 5.5 years
old in
K. Mitrovica
and Pristina,
Yugoslavia
                              SBP
               Concurrent blood Pb
               range:
               4.1 to 76.4 ug/dL
                      Linear regression
                      analysis. Potential
                      confounders
                      considered: sex,
                      maternal education,
                      birth weight, HOME
                      score, and BMI.
                        Per 1 ug/dL increase in
                        concurrent blood Pb level, 0.05
                        (95% Cl:-0.02, 0.13) mmHg
                        higher SBP
                        Blood Pb level at birth and
                        cumulative blood Pb level were
                        not as strongly associated with
                        SBP at age 5.5 yr.
November 2012
                                     5-340
                                                Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
Study
(Ordered as Study
they appear in Population/
the text) Methodology Parameters
Gerretal. Cross- BP
(2002) sectional
508 young
adults age
19-29 years,
born
1965-1975,
male and
female; half of
the subjects
had grown up
around an
active Pb
smelter in
Silver Valley,
Idaho







aBlood Pb data are estimates of geometric mean


Blood Pb Data3
While the concurrent
mean blood Pb level
was 3.15 ug/dL for
the highest bone Pb
category (>10 ug/g),
early childhood
mean blood Pb
levels in this group
were substantially
elevated for all bone
Pb level categories
and were highest
among participants
in the highest bone
Pb level category.
The mean blood Pb
level was 65 ug/dL
among participants
with bone Pb level
>10 ug/g. Bone Pb
was measured at the
time of entry into this
study.


Statistical Analysis
Multiple linear
regression models
always included age,
sex, height, BMI,
current smoking status,
frequency of alcohol
consumption, current
use of birth-control
medication,
hemoglobin level,
serum albumin, and
income, regardless of
significance levels.
Both blood Pb (as a
linear term) and bone
Pb (a four category
ordinal variable from
<1 ug/g to >10 ug/g)
were tested together.






Effect Estimates/Results
Group in highest quartile of
tibia Pb level (>10 ug/g) had
4.26 (95% Cl: 1.36, 7.16)
mmHg higher SBP and 2.80
(95% Cl: 0.35, 5.25) mmHg
higher DBP compared to the
lowest tibia Pb group (<1 ug/g).
















(GM) and geometric standard deviation (GSD) using the arithmetic mean and SD.
5.4.4.2      Cardiovascular Functioning in Children

The relationship between cardiovascular functioning (TPR, BP, stroke volume, and
cardiac output,) and blood Pb levels was examined prospectively by Gump et al. (2007;
2005) in a cohort born at a single New York hospital. Higher early childhood Pb levels
(average age 2.6 years) were associated with greater TPR response to acute stress induced
by mirror tracing on a computer at age 9.5 years as shown in Figure  5-24. Testing blood
Pb with linear, quadratic, and cubic terms did not produce significantly different Pb-TPR
associations, and the authors suggested that these effects were concentration-dependent
and notably, were not emergent at a specific exposure threshold. TPR increased with
increasing quartile of blood  Pb level. A mediational analysis indicated that Pb was a
significant mediator of the SES-TPR reactivity association; some evidence also suggested
moderation, whereby the inclusion of blood Pb into the model reduced the effect estimate
for SES. Observations that Pb exposure increases TPR in toxicological studies and
mechanistic evidence indicating that Pb-induced changes in SNS activity may mediate
such effects (Section 5.4.2.3) provides some biological plausibility for a  role of Pb in
affecting the TPR response to acute  stress in this child population. Additionally, higher
blood Pb level measured at age 2.6 years was associated with a smaller stroke volume
and cardiac output responses to acute stress at age 9.5 years (Gump et al.. 2007). In a
further analysis in this cohort, Gump et al. (2009) examined the possibility that Pb may
mediate an association between SES and cortical responses to acute  stress. Elevated
     November 2012
                           5-341
Draft - Do Not Cite or Quote

-------
                    cortisol has been associated with hypertension (Whitworth et al., 2000). Gump et al.
(2009) found that lower family income was associated with greater cortisol levels
following an acute stress task and that blood Pb was a mediator of this association.
30
25
e
8 H3 20
c »
SS
ff$
'1 | 15
1 If 10
•S.3
II 5
Ji »
-5
-10
(
r-
-
» »
* » »*
* * 4 * * ^
•.^H^T^-
•S'..-'-. •
:*•
*
i i i i i i i
) 2 4 6 8 10 12 14
                                               Blood lead level ({ig/dL)
     Source: Reprinted with permission of Elsevier (Gump et al.. 2005)
     Figure 5-24      Children's adjusted total peripheral resistance (dyn-s/cm5)
                        responses to acute stress tasks, as a function of childhood Pb
                        levels.
 4
 5
 6
 7
 8
 9
10
11
12
13
14

15
16
In a different cohort of 140 children 9 to 11 years of age recruited from local pediatrician
offices and from mailings to homes with children in this age group, Gump et al. (2011)
used a similar acute stress-producing paradigm as in previous studies to examine the
cross-sectional associations of concurrent blood Pb with cardiovascular responses. TPR
significantly increased in a concentration-dependent relationship with blood Pb, with
most of the increase occurring between the first quartile blood Pb (0.14-0.68 (ig/dL) and
the second quartile blood Pb (0.69-0.93 (ig/dL). This result is consistent with those of
Gump et al. (2005). Also, these newer findings provided evidence of associations with
concurrent blood Pb levels  and with lower blood Pb levels (Gump et al.. 2011) than were
previously examined by Gump et al. (2005) and in a large group of children without
higher Pb  exposures earlier in childhood.

Studies in adults and animals indicate Pb-associated decreases in HRV  (Section 5.4.3.4).
In Gump et al. (2011). cardiac autonomic regulation decreased in a
     November 2012
                           5-342
Draft - Do Not Cite or Quote

-------
 1                  concentration-dependent manner with increasing concurrent blood Pb quartile, with the
 2                  largest change relative to the first quartile (0.14-0.68 (ig/dL) measured in the highest
 3                  blood Pb quartile (1.21-3.76 (ig/dL). Also, high frequency HRV, decreased more with
 4                  acute stress in the highest Pb quartile group (1.21-3.76 (ig/dL). In the earlier cohort, early
 5                  childhood (mean age at collection: 2.6 years) blood Pb level was associated with reduced
 6                  stroke volume and cardiac output (Gump et al.. 2007; Gump et al.. 2005). In this recent
 7                  study, Gump et al. (2011) found the same but for concurrent blood Pb level and  at lower
 8                  blood Pb levels.
                    5.4.4.3     Blood Pressure in Children

 9                  Zhang et al. (2011 a) conducted a longitudinal study that examined changes in BP in 323
10                  girls and boys aged 7 to 15 years old in a Mexico City cohort and associations with
11                  maternal bone Pb measured one month post-partum (a measure of cumulative exposure
12                  that could expose fetuses to Pb through accelerated mobilization of bone Pb during
13                  pregnancy) and with cord blood Pb at delivery. This was the first study to examine the
14                  association of maternal bone Pb, as a marker of prenatal exposure, with offspring BP.
15                  The model including both girls and boys (without adjustment for concurrent blood Pb)
16                  showed no statistically significant association overall for any Pb biomarker with child
17                  BP. A significant interaction was found between maternal tibia Pb and sex, and in models
18                  stratified by sex, maternal tibia Pb was associated with adjusted systolic and diastolic BP
19                  in females, but not males. Maternal post-partum median tibia Pb was 9.3 (ig/g (IQR: 3.3,
20                  16.1 (ig/g) with no significant differences between mothers of male and female offspring.
21                  Suboptimal growth in utero is associated with accelerated weight gain in offspring during
22                  childhood and greater risk of later hypertension (Barker and Bagby. 2005; te Velde et al..
23                  2004; Barker et al.. 1989). The relationship between birth weight and Pb biomarkers is
24                  discussed in Section 5.8.3. These may represent biologically plausible mechanisms by
25                  which prenatal Pb exposure may result in increased BP later in  childhood as was
26                  demonstrated in female offspring.

27                  Gump et al. (2011; 2005) examined the relationship of blood Pb level with BP in their
28                  two cohorts of contemporary children around age 10 years in New York State. Gump et
29                  al. (2005) reported an  association of cord blood levels with systolic BP (12.16 mmHg
30                  [95% CI: 2.44, 21.88] increase per 1 (ig/dL increase in cord blood Pb level). Gump et al.
31                  (2011) found that with acute stress, children in higher quartiles  of concurrent blood Pb
32                  level (>0.69 (ig/dL) had larger increases in systolic BP. For example, children with blood
33                  Pb levels between 1.21 and  3.76 (ig/dL had a 7.23 mmHg change, and children with
34                  blood Pb levels between 0.14 and 0.68 (ig/dL had a 5.30 mmHg change. A linear trend
35                  was not observed across quartiles. An interaction between long-term perceived stress and


      November 2012                            5-343                  Draft - Do Not Cite or Quote

-------
 1                  bone Pb levels in association with BP and hypertension also was reported in a study of
 2                  adults (Peters et al.. 2007) (described in Section 5.4.2.1). An earlier study (Factor-Litvak
 3                  et al., 1999; Factor-Litvak et al.. 1996) of children with higher blood Pb levels ranging
 4                  from 4.1 to 76.4 (ig/dL found that a 1 (ig/dL increase in concurrent blood Pb was
 5                  associated with a 0.05 (95% CI: -0.02, 0.13) mmHg increase in systolic BP. An
 6                  additional study (Gerr et al.. 2002) reported that systolic BP for young adults (ages 19-29
 7                  years) with bone Pb levels greater than 10 (ig/g (mean concurrent blood Pb = 65 (ig/dL)
 8                  was 4.26 mmHg higher compared with young adults with bone Pb levels <1 (ig/dL
 9                  compared to young adults with bone Pb levels <1 (ig/dL.

10                  The pathogenesis of CVD has been hypothesized to begin in childhood (Kapuku et al..
11                  2006). Early markers observable in youth in association with Pb biomarkers include
12                  increased BP during stress, reduced HRV, increased IMT, and vascular endothelium
13                  dysfunction. Kapuku et al. (2006) state that endothelial dysfunction is the center of the
14                  CVD paradigm. The factors measured in childhood or as a cumulative burden since
15                  childhood are predictors of outcomes in young adults who are still too young to
16                  experience coronary events  (Li et al.. 2003). and early-life exposures may induce changes
17                  in arteries that contribute to the development of atherosclerosis (Raitakari et al.. 2003).
18                  Berenson et al.  (2002) observed that the effects of multiple risk factors on coronary
19                  atherosclerosis  support evaluation of cardiovascular risk in young people. Thus, evidence
20                  relating levels of biomarkers of Pb exposure in children to cardiovascular function in the
21                  groups of studies presented  in the preceding text when combined with the evidence for
22                  the potential pathogenesis of CVD starting in childhood that yield effects in adulthood,
23                  provides coherence with evidence in adults supporting the effects of long-term,
24                  cumulative Pb exposures in the development of cardiovascular effects.

25                  Few animal studies have examined the effect of Pb exposure during pregnancy and
26                  lactation on BP in offspring as adults and those that have used high levels of exposure.
27                  Recently, pups  of Pb-exposed dams (1,000 ppm through pregnancy and lactation)
28                  exhibited increased blood Pb level (mean blood Pb level: 58.7 (ig/dL) and increased
29                  arterial systolic BP after weaning (Grizzo and Cordellini. 2008) suggesting a role for
30                  childhood Pb exposure leading to adult disease.
                    5.4.4.4     Summary of Child Cardiovascular Studies

31                  The 2006 Pb AQCD (U.S. EPA. 2006b) described three studies on the effects of Pb on
32                  cardiovascular function in children; however, no conclusions were made as to the
33                  strength of the evidence. Studies have reported antecedent cardiovascular changes such as
34                  TPR responses to acute stress tasks as a function of childhood blood Pb levels. Also, a
      November 2012                            5-344                  Draft - Do Not Cite or Quote

-------
 1                   study reported associations with acute stress-induced autonomic and cardiovascular
 2                   dysregulation responses. Biomarkers of prenatal Pb exposure (maternal post-partum
 3                   patella and tibia Pb levels) were related to later higher BP. Other lines of evidence have
 4                   linked increased intrauterine growth restriction to later accelerated weight gain in
 5                   childhood, and this may indicate greater risk of hypertension later in life. The results are
 6                   not uniform with respect to the important lifestages of Pb exposure and can differ by sex
 7                   and other factors. Uncertainties in these studies may be related to sample size, single
 8                   measures of BP, variation in the age of onset of puberty, and cross-sectional design.
 9                   However,  some of these uncertainties may result in the attenuation of observed
10                   associations rather than the generation  of spurious associations. Overall, recent study
11                   findings indicate that in children with mean blood Pb levels in the range of <10 (ig/dL,
12                   increasing blood Pb level may be associated with small increases in BP and changes in
13                   the cardiovascular system that may be related to later development of CVD.

14                   Factors may limit the ability of studies to detect statistically  significant Pb-associated
15                   changes with BP. The relatively young age of the subjects may have limited the ability of
16                   these studies to detect significant BP effects (as opposed to early function effects) if
17                   longer duration Pb exposure is  necessary to produce the cardiovascular changes
18                   considering the lower prevalence and strength of compensatory mechanisms  in children.
19                   There is uncertainty in the shape of the concentration-response relationship to
20                   cardiovascular endpoints at lower blood Pb levels since most studies modeled a linear
21                   relationship. A nonlinear concentration-response relationship has been found for Pb with
22                   other outcomes in children,  most notably, decrements in cognitive function (See
23                   Section 5.3.2).

24                   Cardiovascular endpoints other than baseline BP may be more sensitive outcomes for
25                   measuring Pb-associated cardiovascular effects in very young children. The series of
26                   studies by Gump et al. (2011; 2009; 2007; 2005) evaluating much smaller samples than
27                   did the adult studies, was able to demonstrate statistically significant relationships of
28                   blood Pb levels with cardiovascular responses such as TPR,  related to acute stress. These
29                   results  suggest that the stress paradigm may be useful to detect associations of blood Pb
30                   levels with effects on the cardiovascular system of children.  Selection of the appropriate
31                   cardiovascular outcome in children is an important factor to  consider in the design of
32                   future studies.  Rather than using indicators of cardiovascular effects,  such as BP,
33                   evaluation of cardiovascular changes that are antecedent to increased BP and changes in
34                   other CVD-related endpoints that present at later lifestages be informative to
35                   understanding the time course of cardiovascular changes that may be associated with
36                   early Pb exposure.
      November 2012                             5-345                  Draft - Do Not Cite or Quote

-------
 1                   Overall this small body of evidence, based on different cohorts, locations, and study
 2                   designs, begins to form a literature base that suggests a relationship between biomarkers
 3                   of Pb exposure and cardiovascular effects in children. One longitudinal study ties in
 4                   maternal bone Pb level, and cord and concurrent blood Pb level for the children.
 5                   Limitations exist in the studies. While blood pressure increases are more prevalent in
 6                   older adults than in children, BP increases have been related to higher blood Pb level in
 7                   earlier studies of children  and young adults (Gerr et al., 2002; Factor-Litvak et al., 1999;
 8                   Factor-Litvak et al.. 1996). The recent Gump studies of children provide information in
 9                   populations with mean blood Pb levels in the range of <10 (ig/dL for BP and potential
10                   antecedents for CVD such as increases in TPR and changes in cardiac autonomic
11                   regulation.
      5.4.5        Mortality

12                   The 2006 Pb AQCD (U.S. EPA. 2006b) stated that available evidence suggested an effect
13                   of Pb on cardiovascular mortality in the general U.S. population but cautioned that these
14                   findings should be replicated before these estimates for Pb-induced cardiovascular
15                   mortality could be used for quantitative risk assessment purposes (U.S. EPA, 2006b).
16                   Previous results involved NHANES II and III analyses that examined prospectively the
17                   association of adult blood Pb measured at the time of the study with all cause and cause-
18                   specific mortality ascertained 8-16 years later (Schober et al.. 2006; Lustberg and
19                   Silbergeld. 2002). As blood Pb levels in adults reflect contributions from both recent Pb
20                   exposure and mobilization of historic Pb from bone, it is  unclear to what extent recent,
21                   past, or cumulative Pb exposures contributed to  the observed associations. Given the
22                   decline in ambient air Pb concentrations and population blood Pb levels, it is likely that
23                   study subjects had a much higher past Pb exposure compared to exposure during the
24                   study period. Using NHANES II (1976-1980) data, Lustberg and Silbergeld (2002) found
25                   significant increases in all-cause mortality, circulatory mortality, and cancer mortality,
26                   comparing adults with blood Pb levels of 20-29  (ig/dL to those with blood Pb levels  less
27                   than 10 (ig/dL (measured 12-16 years before ascertainment of vital status). Using
28                   NHANES III data, Schober et  al. (2006) found significant increased all-cause,
29                   cardiovascular, and cancer mortality comparing  adults with blood Pb levels
30                   from 5-9  (ig/dL and above 10 (ig/dL compared to those with blood Pb levels less than
31                   5 (ig/dL (measured a median of 8.8 years before ascertainment of vital status).

32                   Several recent studies substantially strengthen the evidence base for Pb-associated
33                   mortality. A further analysis of the NHANES III database by a different research group
34                   using different methods addressed uncertainties  from earlier analyses by considering a
35                   greater number of potential confounding factors and by characterizing concentration-


      November 2012                             5-346                 Draft - Do Not Cite or Quote

-------
 1                   response relationships. Additionally, two longitudinal prospective studies in different
 2                   U.S. cohorts conducted by different researchers with different methods demonstrate
 3                   consistency within the evidence base for blood Pb and add new evidence for mortality
 4                   associated with bone Pb levels.

 5                   Menke et al. (2006) examined all-cause and cause-specific mortality using NHANES III
 6                   data. Subjects at least 18 years of age were followed up to 12 years after their blood Pb
 7                   was measured, and 1,661 deaths were identified. Those with baseline blood Pb levels
 8                   from 3.63 to 10 (ig/dL had significantly higher risks of all-cause (HR: 1.25 [95% CI:
 9                   1.04, 1.51]), cardiovascular (HR: 1.55 [95% CI:  1.08, 2.24]), MI (HR: 1.89 [95% CI:
10                   1.04, 3.43]), and stroke (HR: 2.51 [95% CI: 1.20, 2.26]) mortality compared to those with
11                   baseline blood Pb levels less than 1.93 (ig/dL and increased risk of cancer mortality (HR:
12                   1.10 [95% CI: 0.82, 1.47]). Effect estimates adjusted for demographic characteristics
13                   were robust to the additional adjustment for factors  such as smoking, alcohol
14                   consumption, diabetes, BMI, hypertension, and level of kidney function. The consistency
15                   of HRs across models with a varying number of control variables indicated little residual
16                   confounding. Hazard ratios were not higher comparing adults with blood Pb levels from
17                   1.94 to 3.62 (ig/dL to those with blood Pb levels <1.93 (ig/dL.  However, tests for linear
18                   trend were statistically significant for all mortality outcomes except for cancer mortality.
19                   Menke et al. (2006) evaluated several of the model covariates (e.g., diabetes,
20                   hypertension, and glomerular filtration rate [GFR])  in a subgroup analysis. The
21                   comparisons for these are shown in Figure 5-25. The authors reported that there were no
22                   interactions between blood Pb and other adjusted variables.

23                   The results from Menke et al. (2006) generally were consistent with those from the
24                   previous NHANES III analysis of the association of blood Pb with mortality by Schober
25                   et al. (2006) that included participants greater than 40 years of age (N = 9686) and
26                   adjusted for covariates including age, sex, ethnicity, and smoking rather than the full suite
27                   of covariates evaluated by Menke et al. (2006). Schober et al. (2006). which was
28                   discussed in the 2006  Pb AQCD (U.S. EPA. 2006b). reported increased HRs comparing
29                   adults with blood Pb levels > 10 (ig/dL to those with blood Pb  levels <5 (ig/dL for all-
30                   cause (HR:  1.59 [95% CI: 1.28, 1.98]), CVD (HR: 1.55 [95% CI: 1.16, 2.07]), and cancer
31                   (HR:  1.69 [95% CI: 1.14, 2.52]) mortality. In general, HRs were higher but
32                   nonsignificant, comparing adults with blood Pb levels from 5-9 (ig/dL to those with
33                   blood Pb levels <5 (ig/dL. The median follow-up time between measurement of blood Pb
34                   and death ascertainment was 8.55 years.
      November 2012                             5-347                  Draft - Do Not Cite or Quote

-------
              Subgroup
Hazard ratio of all-cause mortality (95% C!)
Hazard ratio of cardiovascular mortality (95% Ci)
Age (years)
< 60
>=60
Race-ethnicity
Non-Hispanic white
Non-Hispanic black
Mexican- American
Sex and menopausai status
Male
Female
Pre-menopausai
Post-menopausai
Residence
Rural
Urban
Smoking
Never
Former
Current
Body mass index (kg/m2)
<25
>=25
Hypertension
No
Yes
Diabetes
No
Yes
Estimated glomerular filtration
rate (ml/mi n/1.73m2)
< 60
>=60
Overall


1 ,75 (1.25
1.31 (1.08

1.32(1.09
1 .23 (0,99
1.17(0.86

1.41 (1.11
1.24(1.00
1 .02 (0.54
1.24(1.00

1.28(1.05
1.42(1.18

1.21 (0.93
1.61 (1.33
1 .34 (0.96
1.51 (1.16
1.28(1.03

1.31 (1.08
1.32(1.09

1.37(1.19
1.12(0.73


1 .44 (1.01
1.32(1.12
1.34(1.16


2.44)
1,58)

1.60)
1.52)
1.60) —

1.78)
1.54)
1 .95)
1.54)

1.54)
1.72)

1 .58)
1.94)
1 .87)
1.96)
1.58)

1,58)
1.60)

1.58)
1.71) 	


2.06)
1,56)
1,54)

—4—

— ^ —
— •-! —

I
	 !• 	
— *r~
— »j—

— •; —
— !• —

— »-^ —
i~*

— 1_ • —
— * —

— * —
— ^ —

p
— * — ; 	


-V
+-
0,5 1 23

1,49(1.12
1,490.12


1.63(1.25









1.49(1.15
1.59(1.31

1.49(1.18
1.53(1.21

1.99)
1,99)


2.11)









1,94)
1.92)

1.89)
1.94)

	 m
	 •












_™_fl

	 •
•^

—


* 	









	
•—

—
*-
                                                                          0,5
      Note: Hazard ratios were calculated for a 3.4 ug/dL increase in blood Pb level with log-blood Pb as a continuous variable. This
      increase corresponds to the difference between the 80th and 20th percentiles of the blood Pb distribution (4.92 ug/dL versus
      1.46 ug/dL, respectively).
      Source: Reprinted with permission of Lippincott Williams & Wilkins, Menke et al. (2006)

      Figure 5-25      Multivariate adjusted relative hazards of all-cause and
                         cardiovascular mortality per 3.4 ug/dL increase in blood Pb.

 1                   Both Menke et al. (2006) and Schober et al. (2006) presented mortality curves that plot
 2                   the HRs against blood Pb level. Figure 5-26 shows the mortality hazard ratio curves (not
 3                   absolute cases of mortality) for both stroke and MI reported by Menke et al. (2006).
 4                   Nonlinear associations were modeled. The curves were fitted using predetermined
 5                   restricted  quadratic splines with knots at the  10th percentile (1.00 (ig/dL), the 50th
 6                   percentile (2.67 (ig/dL), and the 80th percentile (5.98 (ig/dL) blood Pb levels. The
 7                   authors did not explain the shape of the blood Pb-mortality curves in detail; however, the
 8                   knots corresponded with the inflection points in the curve. In the tails of the blood Pb
 9                   distribution, hazard ratios decreased with increasing blood Pb level. However, hazard
10                   ratios remained above 1 over most of the blood Pb distribution (blood Pb level greater
11                   than 2 (ig/dL and between 2 and 3 (ig/dL for stroke and myocardial infarction,
12                   respectively), and in the most heavily populated portion of the blood Pb distribution,
13                   hazard ratios increased with increasing blood Pb level. Using a referent group of persons
14                   with blood Pb level less than 1.94 (ig/dL, the hazard ratio for persons with blood Pb level
      November 2012
             5-348
   Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
greater than 3.63 (ig/dL was significant at the 5% level 1.51 (1.07-2.14), but not
significant for persons with blood Pb level in the range of 1.94 to 3.62 (ig/dL. Hazard
ratios peaked for all outcomes at a blood Pb level of approximately 6 (ig/dL. Lower
concentration-response functions at higher blood Pb levels also have been found for
blood Pb-cognitive function relationships in children (Section 5.3.2).
                      2.0 1
                   — 1.5
                    O
                    •**
                    TO
                    B
                   I
                      1.0-
              All cause
              Myocardial infarction
              Stroke
              Cancer
                      0.7
                           12                     5                10
                                               Blood Lead, pg/dL

      Source: Reprinted with permission of Lippincott Williams & Wilkins, Menke et al. (2006)
      Note: A histogram of blood Pb levels is superimposed in the background and displayed on the right axis.

      Figure 5-26      Multivariate-adjusted relative hazard (left axis) of mortality
                         associated with blood Pb levels between  1 ug/dL and 10 ug/dL.

 6                   Schober et al. (2006) examined proportional hazard assumptions, tested for a linear trend
 7                   across blood Pb tertiles, and evaluated log-transformed continuous blood Pb level as
 8                   a 5-knot cubic spline (position of knots not reported). A statistically significant increasing
 9                   linear trend for mortality was observed across blood Pb tertiles. The results of the spline
10                   fit of the continuous blood Pb level term to relative hazard of all cardiovascular diseases
11                   reported by Schober et al. (2006) are shown in Figure 5-27. Schober et al. (2006) shows
12                   the upper 95% confidence band (dashed lines) of the  relative risk for all cause mortality
13                   spline is greater than 1 for all blood Pb levels greater than 1.5 (ig/dL using the referent
14                   group of persons with blood Pb levels less than 1.5. The hazard ratio was fixed at 1.0 for
15                   the referent blood Pb level of 1.5 (ig/dL. Also, the lower 95% confidence band is greater
      November 2012
                            5-349
Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
than 1 when the blood Pb levels is greater than about 4.5 (ig/dL. Using a referent group of
persons with blood Pb levels less than 5 (ig/dL, they found statistically significant
relative risks of CVD for persons with blood Pb levels in the range of 5 to 9, and those
with blood Pb levels greater than 10. In contrast to the curve presented by Menke et al.
(2006). Schober et al. (2006) found the relative hazard axis and the blood Pb axis largely
to be linear (solid line). Both Menke et al. (2006) and Schober et al. (2006) agree that
persons with blood Pb levels greater than 4.5 (ig/dL are at increased risk for mortality;
however, these studies report different shapes for the concentration-response curves.
Despite differences in the age groups included, follow-up time, categorization of blood
Pb levels, and differences in hazard ratio across the blood Pb range, results reported by
Menke et al. (2006) and Schober et al. (2006) are find associations between higher blood
Pb and increased CVD mortality (see Figure 5-30).
                              2.0
                              1.8
                              1.6
                           •21.4
                           J 1.2
                            g 1.0
                           J 0.8
                           K 0.6
                              0.4
                              0.2
                              0.0
                                  1.0   2.0  3.0  4.0   5.0   6.0  7.0   8.0   9.0  10.0
                                               Blood lead (ug/dL)

      Source: Schober etal. (2006)
      Note: The solid line shows the fitted five-knot spline relationship; the dashed lines are the point-wise upper and lower 95% CIs.

      Figure 5-27     Relative risk of all cause mortality for different blood Pb levels
                        compared with referent level of 1.5 ug/dL (12.5th percentile).

13                  In addition to the NHANES analyses described above, studies of older adult, primarily
14                  white, males (Weisskopf etal.. 2009) and older adult females (Khalil. 2010; Khalil et al..
15                  2009b) were conducted recently. Weisskopf et al. (2009) used data from the NAS to
16                  determine the  associations of blood, tibia, and patella Pb with mortality. The authors
17                  identified 241  deaths over an average observation period of 8.9 years (7,673 person-
18                  years). The strongest associations were observed between mortality and baseline patella
      November 2012
                                                  Draft - Do Not Cite or Quote

-------
 1                   Pb concentration. Baseline tibia Pb levels were more weakly associated with CVD
 2                   mortality. Tibia bone Pb level is thought to reflect a longer cumulative exposure period
 3                   than is patella bone Pb level because the residence time of Pb in trabecular bone is shorter
 4                   than that in cortical bone. IHD contributed most to the relationship between patella Pb
 5                   and all CVD death with an individual HR of 2.69 (95% CI: 1.42, 5.08). Although there
 6                   was high correlation between tibia and patella Pb (Pearson r = 0.77), compared with
 7                   cortical bone Pb, trabecular bone Pb may have more influence on circulating blood Pb
 8                   level and thus, local organ concentration of Pb because of its shorter residence time in
 9                   bone. In contrast to the NHANES analyses, the NAS study found that baseline blood Pb
10                   was not significantly related to cardiovascular mortality. This discrepancy may be related
11                   to differences in sample size and resulting power, modeling strategies (e.g., linear versus
12                   log-linear blood Pb level terms), or age range of the study populations. The duration of
13                   follow-up was similar across studies. In the Weisskopf et al. (2009) study of NAS data,
14                   the youngest subjects at baseline were approximately  50-55 years old, compared to the
15                   youngest in the Menke et al. (2006) and Schober et al. (2006) NHANES studies, who
16                   were 18 and 40 years, respectively. Further, the blood Pb tertile analysis of Weisskopf
17                   et al. (2009) could have been affected if the majority of a hypothesized nonlinear
18                   association was contained largely in the lowest (reference) blood Pb tertile.

19                   Weisskopf et al. (2009) also conducted a concentration-response analysis. A linear trend
20                   was observed for increasing HR across tertiles of both tibia and patella Pb levels. The
21                   linear relationship using tertile patella Pb was confirmed in other models in which
22                   continuous patella Pb and nonlinear penalized spline terms (higher order terms) were not
23                   statistically significant. The number of knots and their placement within the Pb variable,
24                   which can influence these results, were determined by an iterative best fit procedure.
25                   Concentration-response relationships shown in Figure 5-28 were approximately linear for
26                   patella Pb on the log HR scale for all CVD, but appeared nonlinear for IHD (p <0.10).
27                   The peak HR is shown around 60 (ig/g, beyond which the HR tends to decrease. It is
28                   important to note the wide confidence limits, which increase uncertainty at the lower and
29                   upper bounds of patella Pb levels.
      November 2012                             5-351                   Draft - Do Not Cite or Quote

-------
           
-------
 1                   The association of adult blood Pb with mortality has also been examined among women
 2                   enrolled in the Study of Osteoporotic Fractures (SOF) (Khalil et al. 2009b). This
 3                   prospective cohort (N = 533) enrolled female volunteers (age 65-87 years) from two U.S.
 4                   locations, Baltimore, MD and Monongahela Valley, PA and followed women for an
 5                   average of 12 years after blood Pb measurement. All-cause mortality is significantly
 6                   higher comparing women with blood Pb levels >8 (ig/dL to those with blood Pb levels
 7                   <8 (ig/dL (HR: 1.59 [95% CI: 1.02, 2.49]). Hazard ratios for combined cardiovascular
 8                   disease mortality (HR: 1.78 [95% CI:  0.92,  3.45]), coronary heart disease mortality (HR:
 9                   3.08 [95% CI: 1.23, 7.70]), but not stroke mortality (HR: 1.13 [95% CI: 0.34, 3.81]) were
10                   higher among the women enrolled in this study with blood Pb levels >8 (ig/dL. In
11                   addition, analyses of blood Pb tertiles  and quintiles indicated that blood Pb-mortality HRs
12                   were consistently elevated in groups with blood Pb levels >7 (ig/dL (Khalil 2010). The
13                   findings for elevated mortality HRs with the highest blood Pb levels are reinforced by the
14                   results displayed in Figure 5-29.  The HR curve for all-cause mortality is relatively flat
15                   over most of population blood Pb distribution (represented by the blue dots) and
16                   increases only in the upper tail of the blood Pb distribution where there are relatively few
17                   subj ects (i. e.,  fewer dots).

18                   Other studies  also reported Pb-associated increased in mortality but have limited
19                   implications due to their weaker  analytic methods. Two studies reported standardized
20                   mortality ratios (SMR) to compare observed deaths in a Pb-exposed population versus
21                   expected deaths, calculated from a reference group (Neuberger et al.. 2009; Cocco et al..
22                   2007). Mortality studies that compare  populations by calculating SMRs based on an
23                   "exposed group" versus the population within which the exposed group resides have
24                   several drawbacks, including the ecologic nature  of the analysis and the absence of Pb
25                   exposure data or biological markers of Pb exposure. Neuberger et al. (2009)carried out a
26                   retrospective mortality study of a Superfund site that was highly contaminated with heavy
27                   metals, principally Zn, Pb, and Cd. Not knowing  the metal concentrations in the
28                   population obscures interpretation of the significantly elevated county-state SMRs and
29                   the insignificant or significantly  lowered SMRs in the county comparison.

30                   A retrospective study of causes of death among Pb smelter workers in Sardinia, Italy
31                   followed 933  male production and maintenance workers (Cocco et al.. 2007). SMRs for
32                   cardiovascular disease-related deaths were calculated based on age-specific and calendar-
33                   year specific mortality of the entire region. Significantly reduced mortality was reported
34                   in the worker groups. The authors attributed the results to the healthy worker effect based
35                   on health criteria applied at hiring and the small size of the cohort.  The usual caveats
36                   regarding population comparison mortality studies apply.
      November 2012                             5-353                  Draft - Do Not Cite or Quote

-------
              00 -
              o -
                                         5                      10
                                               LEAD VALUE
                                                                    15
                                   relative hazard
                                         Median spline
     Source: Khalil et al. (2010)
     Figure 5-29     Multivariate adjusted relative hazard (left axis) of mortality as a
                       function of blood Pb levels between 1  ug/dL and 15 ug/dL.
 i
 2
 o
 6
 4
 5
 6
 7
 8
 9
10
11
5.4.5.1     Summary of Mortality

The mortality results in this review supported and expanded upon findings from the
2006 Pb AQCD (U.S. EPA. 2006b). which included a few NHANES mortality studies
(Schober et al., 2006; Lustberg and Silbergeld. 2002). The recent NHANES mortality
study discussed above (Menke et al.. 2006) addressed many of the limitations of the
earlier studies, including control for a wider range of potential confounders, testing for
interactions with Pb, consideration of concentration-response relationships, extensive
model evaluations, and examination of mortality from specific CVDs. Further, an
association with increased mortality was observed at lower mean population blood Pb
levels. The mean blood Pb level of the NHANES III population was 2.58 (ig/dL. In the
recent analysis, the Pb risk of increased cardiovascular mortality increased with
increasing blood Pb level over the most heavily populated portion of the blood Pb
     November 2012
                                               Draft - Do Not Cite or Quote

-------
 1                  distribution, with maximum blood Pb levels between 6 and 7 (ig/dL. It is important to
 2                  note that the relative contributions of recent, past, and cumulative Pb exposure to
 3                  associations observed with the baseline blood Pb levels is uncertain. In addition, the first
 4                  evidence that bone Pb, a metric of cumulative Pb exposure, is associated with increased
 5                  mortality was reported recently among NAS men (Weisskopf et al.. 2009).

 6                  Quantitative differences in Pb-associated hazard for death between studies may be
 7                  influenced by age range of the study groups, follow up time to death, variation in model
 8                  adjustment, central tendency and range of the Pb biomarker levels, assumptions of
 9                  linearity in relationship with Pb biomarkers, and choice of Pb biomarker. Quantitative
10                  differences in Pb-associated mortality across NHANES II and NHANES III studies or
11                  between different NHANES III analyses may be explained by the use of continuous or
12                  ordered blood Pb terms and different data selection strategies. Further, studies using
13                  ordered categories of blood Pb level may obtain different results, as the range of blood Pb
14                  level represented in the reference category will affect the calculated coefficients of the
15                  remaining percentiles  or groups.

16                  Specifically, Menke et al. (2006) is the strongest study presently published for estimating
17                  the effects of Pb on cardiovascular disease-related mortality.  The study uses the
18                  nationally representative NHANES III (1988-1994) sample of men and women. The
19                  results corroborate of earlier published NHANES  studies but address some of the key
20                  weaknesses noted in those studies. For example, Menke  et al. (2006) examined potential
21                  confounding by a large number of factors, including hypertension and kidney function.
22                  Weisskopf et al. (2009) is the first published mortality study using bone Pb as an
23                  exposure index. The study is a prospective study with nearly  100% successful follow-up
24                  of deaths. This  rigorous study found increased cardiovascular disease mortality in
25                  association with patella bone Pb with weaker associations for tibia Pb level. The Khalil
26                  et al. (2010; 2009b) study of SOF subjects provides supporting  results for a cohort
27                  consisting of white females aged 65-87 years. Further, a number of prior studies found
28                  associations between accumulated Pb reflected in bone Pb measurements and higher
29                  CVD morbidity (Sections 5.4.2.1 and 5.4.3). This evidence base is augmented with new
30                  findings indicating that biomarkers of longer-term cumulative Pb exposure increases
31                  CVD mortality. The NAS and SOF examine only men and women, respectively.
32                  However, the consistency of findings between the two studies indicates that the results of
33                  either study may be applicable widely. Despite the differences in design and methods
34                  across studies, associations between higher levels of Pb biomarkers and higher risk of
35                  mortality were generally observed (Figure 5-30 and Table 5-23). One exception is that
36                  stroke  mortality was not significantly elevated in the SOF study although it was positive.
37                  Mortality from  specific CVD causes, MI and IHD mortality, which are related to higher
38                  BP and hypertension,  were elevated with higher Pb biomarker levels.
      November 2012                             5-355                  Draft - Do Not Cite or Quote

-------
  Reference
  Menkeetal. (2006)
   n = 13,946
   mn age=44
   mn blood Pb=2.58
  Schoberetal. (2006)
   n = 9757
   age > 40 y
  Lustberg & Silbergeld
  [2002]; n = 4,190
    mn age=54 y
    mn blood Pb=14,0

  Weisskopf etal. (2009)
    n = 868 men
    mn age-67.3 y
    mn blood Pb 5.7
  Khalil etal. (2009)
    n - 533 women
    rnn age=70 y
    mn blood Pb=5.3
Outcome


All Cause

CVD

Ml

Stroke


All Cause

CVD


All Cause

CVD


All Cause

CVD

IHD


All Cause
CVD
IHD

All Cause

CVD

IHD
All Cause
CVD
CHD
Stroke
Study       Pb Biomarker Comparison
                  Groups
           Blood Pb (ug/dL)
NHANESIII   23,63 vs. S1.93
           1.94-3.62 vs. S1.93
           >3.63 vs. S1.93
           1.94-3.62 vs. <1.93
           >3.63vs. S1.93
           1.94-3.62 vs. Sl.93
           £3.63 vs. £1.93
           1.94-3.62 vs. <1.93

           >10vs. <5
           5-9 vs. < 5
           HO vs. <5
           5-9 vs. < 5

NHANESII   20-29 vs. < 10
           10-19 vs. < 10
           20-29 vs. < 10
           10-19 vs. < 10
NAS
                                     SOF
           >6vs. < 4
           4-6 vs. < 4
           >6 vs. < 4
           4-6 vs. < 4
           >6 vs. < 4
           4-6 vs. < 4
           Tibia Pb (M«/g)
           Tertile 3 vs. 1 (NR)
           Tertile 3 vs. 1 (NR)
           Tertile 3 vs. 1 (NR)
           Patella Pb (u.g/g)
           > 35 vs. < 22
           22-35 vs. < 22
           > 35 vs. < 22
           22-35 vs. < 22
           > 35 vs. < 22
           22-35 vs. < 22
           Z 8 vs. < 8
           > 8 vs. < 8
1986-1988   a 8 vs. < 8
           2 8 vs. < 8
                                                                 0.0   1.0  2.0  3.0  4.0  5.0  6.0  7.0   8.0  9.0  10.0
                                                                                Hazard Ratio (95% Cl)
Note: Studies are presented in order of strength of study design and follow the order of discussion in the preceding text. Hazard
ratios represent the hazard in the higher blood or bone Pb group relative to that in the lowest blood or bone Pb group (reference
group).
Blood Pb (closed markers), or Bone Pb (open markers) associations with All-cause mortality (black diamonds) or
Cardiovascular mortality (blue circles).

Figure 5-30      Hazard ratios for associations of blood Pb or bone Pb with
                      all-cause mortality and cardiovascular mortality.
November 2012
                                                     Draft - Do Not Cite or Quote

-------
Table 5-23  Additional characteristics and quantitative data for associations of
            blood and bone Pb with CVD mortality for studies presented in
            Figure 5-30.

Study
Menke et al.
(2006)





Schober et
al. (2006)














Lustberg
and
Silbergeld
(2002)














Study
Population /
Methodology
Longitudinal
13,946 adult
participants of
NHANES III ,
> 17yr
(1988-1994)





Longitudinal
9,686 adult
participants of
NHANES III,
^ 40 yr












Longitudinal
4,190 adult
participants of
NHANES III,
yr (1976-1 980)
aged 30 to 74,
Studied
through
December 31,
1992












Parameter
All cause and cause-
specific mortality
Studied through
December 31, 2000
CVD:ICD-9 390-434;
ICD-10 IOO-I99), Ml
(ICD-9410-414and
429.2; ICD-10 I20-I25),
stroke (ICD-9 430-434
and 436-438; ICD-10
I60-I69).





All cause and cause-
specific mortality














All cause and cause-
specific mortality
















Pb Data
Baseline
Blood Pb
(measured an
average of 12
yr before
mortality):
Mean:
2.58 ug/dL
Tertiles:
<1.93ug/dL,
1 .94-3.62 ug/d
L,
> 3.63 ug/dL



Ordered
categorical
blood Pb level,
measured a
median of 8.55
yr prior to
death
<5 ug/dL
5-9 ug/dL
> 10 ug/dL








Categorical
blood Pb level
Mean: 14.0
(5.1)
Msdisrr
13 ug/dL
Isttertile:
<10 ug/dL
(Reference)
2nd tertile:
10-19ug/dL
3rd tertile:
20-29 ug/dL








Statistical Analysis
Survey-design adjusted Cox
proportional hazard
regression analysis (up to 12
yr follow-up) adjusted for
Model 1 : age, race/ethnicity,
sex, Model 2: urban
residence, cigarette smoking,
alcohol consumption,
education, physical activity,
household income,
menopausal status, BMI,
CRP, total cholesterol,
diabetes mellitus, Model 3:
hypertension, GFR category



Survey-design adjusted Cox
proportional hazard adjusted
for sex, age, race/ethnicity,
smoking, education level Did
not evaluate BMI nor
cormorbidities












Proportional Hazard model,
RRs adjusted for age, sex,
location, education, race,
income, smoking, BMI,
exercise















Hazard Ratio or
SMR (95% Cl)
All-cause (3rd vs.
Isttertile):
1.25(1.04, 1.51)
CVD (3rd vs. 1st):
1 .55 (1 .08, 2.24)
Ml (3rd vs. 1st):
1 .89 (1 .04, 3.43)
Stroke (3rd vs. 1st):
2.51 (1.20,5.26)

Cancer (3rd vs.
1st):
1.10(0.82, 1.47)
All-cause (2nd vs.
1st):
1 .24 (1 .05, 1 .48)
All-cause (3rd vs.
1st):
1.59(1.28, 1.98)

CVD (2nd vs. 1st):
1.20(0.93, 1.55)
CVD (3rd vs. 1st):
1.55(1.16,2.07)
Cancer (2nd vs.
1st):
1.44(1.12, 1.86)
Cancer (3rd vs.
1st):
1.69(1.14,2.52)
All-cause (2nd vs.
1st):
1.40(1.16-1.69)
All-cause (3rd vs.
1st)'
i 01;.
2.02 (1 .62-2.52)

Circulatory (2nd vs.
1st):
1 .27 (0.97-1 .57)
Circulatory (3rd vs.
1st):
1 .74 (1 .25-2.40)
Cancer (2nd vs.
1st):
1 .95 (1 .28-2.98)
Cancer (3rd vs.
1st):
2.89 (1 .79-4.64)
November 2012
Draft - Do Not Cite or Quote

-------
Study
Weisskopf et
al. (2009)



























Khalil et al.
(2QQ9b)








Study
Population /
Methodology Parameter
Longitudinal All cause and cause-
868 men, >55 specific mortality
yr, 95% white,
from MAS in
Greater
Boston
area, MA

























Longitudinal All cause and cause-
533 women, specific mortality
65-87 yr, from
Study of
Osteoporotic
Fractures
cohort in
Baltimore, MD
and
Monongahela
Valley, PA






Pb Data
Pb biomarkers
collected an
average of 8.9
years before
death
Blood Pb:
Mean (SD):
5.6 (3.4) ug/dL
Patella Pb:
Mean (SD):
31.2
(19.4) ug/g
Tertiles:
<22 ug/g,
22-35 ug/g,
>35 ug/g

Tibia Pb:
Mean (SD):
21.8
(13.6) ug/g












Blood Pb
measured an
average 12
(SD; 3) yr
before death:
Mean (SD;
range): 5.3
(2.3;
1-21)ug/dL







Statistical Analysis
Cox proportional hazard
regression analysis adjusted
forage, smoking, education.
Additional models adjusted
for alcohol intake, physical
activity, BMI, total cholesterol,
serum HDL, diabetes
mellitus, race, and
hypertension

























Cox proportional hazards
regression analysis adjusted
forage, clinic, BMI,
education, smoking, alcohol
intake, estrogen use,
hypertension, total hip bone
mineral density, walking for
exercise, and diabetes







Hazard Ratio or
SMR (95% Cl)
All-cause (3rd vs.
1st patella Pb
tertile):
1 .76 (0.95, 3.26)
All CVD (3rd vs. 1st
tertile):
2.45 (1 .07, 5.60)
IHD(3rdvs. 1st):
8.37 (1 .29, 54.4)

Cancer (3rd vs.
1st):
0.59(0.21, 1.67)


After excluding 154
subjects with CVD
and stroke at
baseline:
All-cause (3rd vs.
1st):
2.52(1.17-5.41)
All CVD (3rd vs.
1st):
5.63(1.73, 18.3)
All-cause (3rd vs.
1st blood Pb tertile):
0.93(0.59, 1.45)
All CVD (3rd vs.
1st):
0.99(0.55, 1.78)
IHD(3rdvs. 1st):
1.30(0.54,3.17)
> 8 ug/dL vs.
<8 ug/dL
All cause:
1 .59 (1 .02, 2.49)
CVD: 1 .78 (0.92,
3.45)
Coronary Heart
Disease:
3.08 (1 .23, 7.70)
Stroke: 1.13 (0.34,
3.81)
Cancer: 1.64(0.73,
3.71)
November 2012
Draft - Do Not Cite or Quote

-------
      Study
Study
Population /
Methodology   Parameter
                     Pb Data
              Statistical Analysis
                                                                                            Hazard Ratio or
                                                                                            SMR (95% Cl)
      "Neuberger
      et al. (2009)
Ecological
Residents at
or near Tar
Creek
Superfund
site, Ottawa
County, OK
(exposed pop.
5,852,
unexposed
pop. 16,210)
Cause-specific mortality
No biomarker
measurements
Standardized mortality ratio
(SMR) based on 2000 U.S.
Census data
Heart disease:
Both sexes:
114.1 (113.1, 115.2)
Men:
118(116.4, 119.6)
Women:
111 (109.5, 112.5)


Stroke:
Both sexes:
121.6(119.2,123.9)
Men:
146.7(107.4, 195.7)
Women:
106.5(80.2,138.6)
aCocco et al.
(2007)



Ecological
933 male Pb
smelter
workers from
Sardinia, Italy
(1973-2003)
All cause and cause-
specific mortality



No biomarker SMR
measurements



All cause: 56 (46,
68)

CVD: 37 (25, 55)

      "These references not included in Figure 5-30 because they reported standardized mortality ratios.
      5.4.6
 Air Pb-PM Studies
 i

 2

 3

 4

 5

 6

 7

 8

 9

10

11

12

13

14

15
   5.4.6.1      Cardiovascular Morbidity


   A relatively small number of studies used Pb measured in PMi0 and PM2 5 ambient air

   samples to represent Pb exposures. However, given that size distribution data for Pb-PM

   are fairly limited, it is difficult to assess the representativeness of these concentrations to

   population exposure (Section 3.5.3). Moreover, data illustrating the relationships of

   Pb-PMio and Pb-PM2 5 with blood Pb levels are lacking. A few available studies exposed

   rats, dogs, or humans to concentrated ambient air particles (CAPS) in which Pb and

   several other components were measured. Consistent with epidemiologic studies of blood

   and bone Pb and with studies of animals exposed to Pb, these studies show exposure to

   Pb-containing CAPS resulted in various changes related to increased vasoconstriction

   (Urch et al.. 2004: Wellenius etal.. 2003: Batalha et al.. 2002). While Pb-containing

   CAPS indicate cardiovascular effects with short-term exposure (2-6 hours over

   multiple days), they cannot be attributed specifically to the Pb component of the mixture.

   It is important to note that Urch et al. (2004) estimated the Pb effect on brachial artery

   diameter based on the ambient concentration of Pb, not direct exposure to Pb isolated

   from CAPs.
      November 2012
                                                       Draft - Do Not Cite or Quote

-------
 1                   A U.S. time-series study of almost 3 million pregnant women found that increases in
 2                   ambient Pb-TSP concentrations were associated with increased odds of pregnancy
 3                   induced hypertension (PIH) assessed at delivery (Chen et al., 2006c). In contrast,
 4                   epidemiologic studies provide weak evidence for an association between short-term
 5                   changes (daily average) in ambient air concentrations of Pb- PM25 and cardiovascular
 6                   morbidity in adults adjusting for weather and time trends. Some of these time-series
 7                   studies analyzed Pb individually, whereas others applied source apportionment
 8                   techniques to  analyze Pb as part of a group of correlated components.  In a time-series
 9                   study of 106 U.S. counties, Bell et al. (2009) found that an increase in lag 0 Pb- PM2 5
10                   was associated with an increased risk of cardiovascular hospital admissions among adults
11                   ages 65 years  and older. Quantitative results were not presented; however, the 95% CI:
12                   was wide and included the null value. In this study, statistically significant associations
13                   were observed for other PM metal components such as nickel, vanadium, and Zn. In the
14                   absence of detailed data on correlations among components or results  adjusted for
15                   copollutants, it is difficult to exclude confounding by ambient air exposures to these other
16                   components or copollutants. To address correlations among PM chemical components,
17                   some studies applied source apportionment techniques to group components into
18                   common source categories. In these source-factor studies, it is not possible to attribute the
19                   observed association (Sarnat et al.. 2008) or lack of association (Andersen et al.. 2007)
20                   specifically to Pb.
                     5.4.6.2      Mortality

21                   Time-series epidemiologic studies of ambient air Pb- PM25 reported positive associations
22                   with mortality. Although limited in number, these studies indicated associations in
23                   multiple cities across the U.S. In the Harvard Six Cities Study, Laden et al. (2000) found
24                   a 1.16% (95% CI: 0.20, 2.9%) increased risk in all-cause mortality per 461.4 ng/m3
25                   (5th-95th percentile) increase in Pb-PM2 5. In six California counties, Ostro et al. (2007)
26                   found that a 5 ng/m3 (interquartile range) increase in Pb-PM2 5 was associated with a
27                   1.89% (95% CI: -0.57, 4.40%) increased  risk of cardiovascular mortality and a 1.74%
28                   (95% CI: 0.24, 3.26%) increased risk of all-cause mortality during the cool season. The
29                   limitations of air-Pb studies were described in Section 5.4.6.1 above and also are relevant
30                   to the interpretation of these findings for mortality.
      November 2012                             5-360                  Draft - Do Not Cite or Quote

-------
      5.4.7        Summary and Causal Determination

 1                  Large bodies of epidemiologic and toxicological evidence indicate effects of Pb exposure
 2                  on a range of related cardiovascular effects.  For evaluation of causal relationships with
 3                  Pb exposure, evidence was grouped in categories using the U.S. Surgeon General's
 4                  Report on Smoking as a guideline (CDC. 2004).  The categories include hypertension,
 5                  subclinical atherosclerosis, coronary heart disease, and cerebrovascular disease. The
 6                  causal determination for hypertension and increased BP is not only informed by evidence
 7                  for hypertension and blood pressure, but also cardiovascular mortality. Coronary heart
 8                  disease is informed by evidence for HRV, MI, IHD, mortality from MI, IHD, and CHD,
 9                  and in animals, increased thrombosis, coagulation, and arrhythmia. The biological
10                  plausibility and mode of action for these cardiovascular effects is provided by evidence
11                  for oxidative stress, inflammation, vascular cell activation or dysfunction. The  sections
12                  that follow describe the evaluation of evidence for these four groups of outcomes,
13                  hypertension, subclinical atherosclerosis, coronary heart disease, and cerebrovascular
14                  disease, with respect to causal relationships with Pb exposure using the framework
15                  described in Table II of the Preamble. The key supporting evidence to the causal
16                  framework is summarized in Table 5-24.
                    5.4.7.1      Evidence for Hypertension and Increased Blood Pressure

17                  The 2006 Pb AQCD concluded that there was a relationship between higher blood Pb and
18                  bone Pb and cardiovascular effects in adults, in particular increased BP and increased
19                  incidence of hypertension (U.S. EPA, 2006b). and recent evidence strengthens this
20                  conclusion. This conclusion is informed by the coherence of effects observed between
21                  epidemiologic and toxicological findings and among related endpoints. Prospective
22                  evidence and animal toxicology studies demonstrate the temporal relationship of the
23                  exposure to effect, while meta-analyses provide indications of consistency and strength,
24                  and cross-sectional evidence support the consistently observed results. Consideration of
25                  numerous potential confounding factors in both the prospective and cross-sectional
26                  studies limit uncertainty from bias and other lines of evidence characterizing modes of
27                  action provide biological plausibility to the associations.

28                  Longitudinal prospective studies clearly support the relationship between biomarkers of
29                  Pb exposure and hypertension incidence and BP changes establishing the directionality of
30                  effects. High-quality studies are replicated by different investigators using different
31                  designs and in large cohorts in different locations (Peters et al.. 2007; Glenn et al.. 2006;
32                  Cheng etal.. 2001). Bone Pb coupled with high perceived stress was associated with an
33                  increased risk of developing hypertension in an originally nonhypertensive group of
      November 2012                             5-361                  Draft - Do Not Cite or Quote

-------
 1                   adults (Peters et al.. 2007). Cheng et al. (2001) examined subjects from the NAS cohort
 2                   without hypertension at baseline measurement and reported a significant increase for
 3                   hypertension with patella Pb analyzed by linear models. A recent prospective study in Pb
 4                   workers found independent associations of both baseline blood Pb level and subsequent
 5                   changes in blood Pb over follow-up with changes in BP over follow-up and bone Pb level
 6                   with hypertension (Glenn et al.. 2006). The results indicated that different mechanisms
 7                   may mediate short-term Pb-associated increases in BP and long-term Pb-associated
 8                   development of hypertension. Consideration for key potential confounding factors was
 9                   appropriate including baseline age, alcohol consumption, BMI, and use of BP lowering
10                   medications. Other factors such as smoking and education were evaluated but did not
11                   predict systolic BP. When subjects with hypertension were excluded from the model, the
12                   predicted change was not altered.  Thus, chance, bias, and confounding can be ruled out
13                   with reasonable confidence based on the consistent, positive, statistically significant
14                   results indicated in these studies. Figure 5-18 and Figure 5-19 and the meta-analysis
15                   indicate that results for effects of Pb exposure on BP and hypertension are positive and
16                   precise. This provides more confidence in this relationship and reduces the level of
17                   uncertainty.

18                   The prospective evidence is supported by meta-analyses that underscore  the consistency
19                   and reproducibility of Pb-associated increases in BP and hypertension across diverse
20                   populations and different study designs (Navas-Acien et al..  2008; Nawrot et al.. 2002).
21                   Nawrot et al. (2002) found that each doubling of concurrent  blood Pb level (between 1
22                   and >40 ug/dL) was associated with a 1 mmHg increase in systolic BP and a 0.6 mmHg
23                   increase in diastolic BP. Navas-Acien et al. (2008) found that all included studies showed
24                   a relationship between higher bone Pb levels and higher BP. Also, all but one that
25                   characterized hypertension showed higher relative risks or odds ratios associated with
26                   higher bone Pb levels.

27                   Further support for a causal relationship between blood and bone Pb levels and increased
28                   BP and hypertension is provided by many cross-sectional analyses conducted by
29                   numerous researchers using different study designs and analyses in large, diverse cohorts
30                   in different locations. A recent study in an ethnically diverse community-based cohort of
31                   women and men aged 50-70 years found hypertension risk to be  associated with blood
32                   and tibia Pb levels (Martin et al.. 2006). Recent epidemiologic studies in adults found
33                   associations with hypertension in populations with relatively low mean blood Pb levels.
34                   For example, a positive relationship was found in the nationally representative NFfANES
35                   III (1988-1994), in which the population geometric mean blood Pb level was 1.64 (ig/dL
36                   (Muntner et al., 2005). Despite the extensive evidence for associations at relatively low
37                   concurrent blood Pb levels, these cardiovascular outcomes were most often examined in
38                   adults that have been exposed to higher levels of Pb earlier in life, and uncertainty
      November 2012                             5-362                  Draft - Do Not Cite or Quote

-------
 1                   remains concerning the Pb exposure level, timing, frequency, and duration contributing to
 2                   the observed associations. However, evidence presented in the 1990 Pb Supplement to
 3                   the Addendum Q990a), indicated that in populations aged 20-74 years during 1976-1980
 4                   in NHANES II (Schwartz. 1991) across the range of 7-34 (ig/dL no evident threshold was
 5                   found below which the blood Pb level was not significantly related to BP. Further, as
 6                   described in Section 5.4.1. general population blood Pb levels across those aged 20 to 74
 7                   years as indicated by NHANES II and other studies probably peaked in the time frame of
 8                   1978-1988 achieving levels that were likely to persist over the long-term ranging from 10
 9                   to 30 (ig/dL.

10                   Further, recent cross-sectional epidemiologic studies also emphasized the interaction
11                   between Pb biomarker levels and factors, such as genetic variants, race/ethnicity, and
12                   metabolic syndrome, in modifying the association with BP or hypertension. Evidence
13                   was presented for a larger blood Pb-associated increase in BP in carriers of the ALAD2
14                   allele, which is associated with greater binding affinity for Pb in the bloodstream (see
15                   Figure 5-18 for results) (Scinicariello et al., 2010). Additionally, bone Pb level was
16                   associated with larger increases in PP, which represents a good predictor of
17                   cardiovascular morbidity and mortality and an indicator of arterial stiffness, among NAS
18                   adults with the HFE H63D and/or C282Y variant (Zhang etal.. 2010a) (Figure 5-18 for
19                   results). Park et al. (2009b) provided further evidence of HFE and transferrin gene
20                   variants, related to iron metabolism, impacting the associations of bone Pb levels with
21                   cardiovascular effects, evaluated by QT interval changes in the NAS cohort.

22                   Combined evidence  from prospective and cross-sectional studies helps limit the level of
23                   uncertainty  for bias from confounding with reasonable confidence. While the adjustment
24                   for specific  factors varied by study, the collective body of evidence adjusted for multiple
25                   potential known key confounding factors, including  age, diet, sex, BMI, blood pressure
26                   lowering medication use, SES, race/ethnicity, alcohol consumption, cholesterol, smoking,
27                   pre-existing disease  (i.e., diabetes), measures of renal function, and copollutant exposures
28                   (i.e., Cd).

29                   Cardiovascular effects of Pb exposure in children are discussed in Section 5.4.4. Overall
30                   this body of evidence, based on different cohorts, locations, and study designs provides a
31                   preliminary literature base examining the potential for a relationship between biomarkers
32                   for Pb exposure and cardiovascular effects in children. Recent studies provide
33                   information for BP and antecedents for cardiovascular disease such as increases in TPR
34                   and changes in cardiac autonomic regulation.

35                   A causal relationship is further supported by coherence between epidemiologic and
36                   toxicological evidence for the  effects of long-term exposure on BP. Collectively, all
37                   animal toxicological studies providing blood Pb level and BP measurements reported


      November 2012                             5-363                  Draft - Do Not Cite or Quote

-------
 1                   increases in BP with increasing blood Pb level in the range relevant to humans (Figure
 2                   5-21). Whereas the majority of studies examined long-term Pb exposures that resulted in
 3                   mean blood Pb levels >10 (ig/dL, one animal toxicological study found a continuous
 4                   monotonic increase in BP in animals with a mean blood Pb level from 0.05 to 29 (ig/dL
 5                   with no evidence of a threshold (Tsao et al. 2000). Thus, most evidence demonstrated
 6                   such effects in adult animals with blood Pb levels >10 (ig/dL. Also, recent studies
 7                   demonstrated only partial reversibility of Pb-induced increased BP following Pb exposure
 8                   cessation or chelation and the possibility for short-term Pb exposure-induced increases in
 9                   BP. The short-term effects were found with routes of Pb exposure that may have
10                   uncertain relevance to humans.

11                   Coherence for BP and hypertension evidence was also provided by epidemiologic
12                   evidence indicating associations with related CV conditions. Studies in the medical
13                   literature show that increasing BP,  even within the nonhypertensive range, is associated
14                   with increased rates of death and cardiovascular disease, including CHD, stroke, and
15                   cardiac failure (Ingelsson et al., 2008; Chobanian et al., 2003; Pastor-Barriuso et al.,
16                   2003;  Prospective Studies Collaboration. 2002; Kannel. 2000a. b; Neatonetal.. 1995).
17                   Evidence for Pb-induced hypertension and increased BP is supported by, consistently
18                   observed associations between Pb biomarkers and both cardiovascular and all-cause
19                   mortality in prospective studies with follow-up periods ranging between 8 and 12 years.
20                   A recent analysis of the NHANES  III sample reported associations of adult blood Pb
21                   level with cardiovascular mortality (Menke et al., 2006). These findings were supported
22                   by a community-based cohort of women age 65-87 years,  in which higher effect
23                   estimates were observed  for mortality from cardiovascular disease (Khalil et al., 2009b).
24                   Weisskopf et al. (2009) published the first mortality study using bone Pb as an exposure
25                   index. This prospective study found that patella bone Pb levels were associated with
26                   increased mortality from cardiovascular disease.

27                   Animal toxicology studies further indicate coherence and strengthen the evidence for
28                   causality by providing strong biological plausibility for Pb-associated increases in BP and
29                   hypertension. Hypertension results from dysfunction in the regulation of blood flow and
30                   vascular resistance. Many systems, including the central and sympathetic nervous
31                   systems, the contractile processes in the vasculature, and various hormonal regulators,
32                   contribute to the maintenance of BP and disruption of these systems will alter BP
33                   homeostasis. Studies demonstrate that oxidative stress produced following Pb exposure
34                   inactivates the vasodilator NO which may lead to increased vasoconstriction and
35                   increased BP, leading to hypertension. In addition, oxidative stress can damage the
36                   endothelium, further disrupting endothelium-dependent vascular relaxation and
37                   increasing the contractile response. Studies also  suggest Pb exposure disrupts normal
38                   contractile processes by altering the sympathetic nervous system, the renin-angiotensin-
      November 2012                             5-364                 Draft - Do Not Cite or Quote

-------
 1                   aldosterone system, and the balance between production of vasodilators and
 2                   vasoconstrictors (Section 5.4.2.3).

 3                   Associations between biomarkers of Pb exposure and increased BP and hypertension
 4                   have been observed in a number of populations, including the large nationally
 5                   representative NHANES cohort (Menke et al.. 2006; Muntner et al.. 2005). In addition,
 6                   associations are found in other cohorts that include both men and women (Martin et al..
 7                   2006). Further, the meta-analyses assess cohorts both within the U.S. and international,
 8                   further supporting the generalizability of the relationship between Pb exposure and
 9                   increased BP and hypertension.

10                   Changes in BP that have been associated with biomarkers  of Pb exposure indicate a
11                   modest change for an individual; however, these modest changes can have a substantial
12                   public health implication at the population level. The reported effects represent a central
13                   tendency of Pb-induced cardiovascular effects among individuals; some individuals may
14                   differ in risk and manifest effects that are greater in magnitude. For example, a small
15                   increase in BP may shift the population distribution and result in considerable increases
16                   in the percentages of individuals with BP values that are clinically significant, i.e., an
17                   indication of hypertension and medication use.

18                   Overall, evidence in epidemiologic and toxicological studies demonstrates consistent
19                   effects of long-term Pb exposure on increased BP and hypertension in adults; however,
20                   uncertainty remains concerning the Pb exposure level, timing, frequency, and duration
21                   contributing to the effects. The epidemiologic studies are of high-quality, have been
22                   replicated by different researchers in different cohorts, and have adjusted for numerous
23                   potential confounding factors. Thus, collectively, they help limit the level of uncertainty
24                   for bias from confounding with reasonable confidence. In  addition, a biologically
25                   plausible potential mode of action is described. Thus, the overall evidence is sufficient to
26                   conclude that there is a causal relationship between Pb exposure and hypertension.
                     5.4.7.2      Evidence for Subclinical Atherosclerosis

27                   Measures of subclinical atherosclerosis provide the opportunity to assess the pathogenesis
28                   of vascular disease at an earlier stage. Studies that discuss markers of subclinical
29                   atherosclerosis, such as PAD (i.e., ankle-brachial index) and generalized atherosclerosis
30                   (i.e., IMT), are included in this category. A limited number of studies have evaluated
31                   markers of subclinical atherosclerosis following Pb exposure in adult humans or animals.
32                   One study described in the 2006 Pb AQCD (U.S. EPA. 2006b) indicated that Pb was
33                   associated with PAD in the NHANES population and coexposure with Cd did not
34                   confound the association (Navas-Acien et al., 2004). Recent epidemiologic findings  are

      November 2012                             5-365                 Draft - Do Not Cite or Quote

-------
 1                   limited to cross-sectional analyses, so uncertainty exists as to the specific Pb exposure
 2                   level, timing, frequency, and duration that contributed to the observed effects. One study
 3                   reported an increasing trend in the odds of PAD and concurrent blood Pb level in adults
 4                   within the NHANES population (Muntner et al.. 2005). which is consistent with the
 5                   results from the previous Navas-Acien et al. (2004) analysis. An occupational study of
 6                   Pb-exposed adults with a mean blood Pb level around 25 (ig/dL presented evidence for
 7                   increased measures of atherosclerosis analyzed by Doppler ultrasound (i.e., greater IMT
 8                   and atherosclerotic plaque presentation) in the Pb-exposed population (Poreba et al..
 9                   2011). Similarly, toxicological studies have provided limited evidence to suggest long-
10                   term Pb exposure may initiate atherosclerotic vessel disease. Pb exposure to human radial
11                   and internal mammary arteries resulted in a concentration-dependent increase in arterial
12                   intimal thickness (Zeller et al.. 2010). Also, exposure to Pb in rats increased aortic medial
13                   thickness (Zhang et al.. 2009a).

14                   Toxicological studies also present evidence to clearly describe a plausible biological
15                   mechanism. Atherosclerosis is considered an inflammatory disease with a clear role for
16                   oxidative stress in the pathogenesis of the disease. There is consistent evidence that Pb
17                   exposure promotes oxidative stress and increased inflammation in animal and cell culture
18                   models (Section 5.4.2.3). In addition, there is evidence that Pb will stimulate vascular cell
19                   activation and lead to endothelial cell dysfunction. Both events are key to the
20                   development and progression of atherosclerosis. Also, epidemiologic and animal
21                   toxicology studies have related higher blood Pb levels with higher cholesterol; high
22                   cholesterol is one of the principal risk factors for atherosclerosis (Section 5.4.3.3).

23                   In summary, the evidence includes one high-quality epidemiologic study with control for
24                   numerous potential confounders (Muntner et al.. 2005) and biological plausibility for the
25                   effects observed in humans. Thus, the evidence  for subclinical atherosclerosis is
26                   suggestive of a causal relationship.
                     5.4.7.3      Evidence for Coronary Heart Disease

27                   Coronary heart disease (CHD) results from interruption of the blood supply to a part of
28                   the heart resulting from atherosclerosis of the coronary arteries, with acute injury and
29                   scarring leading to permanent damage to the heart muscles. A disrupted HRV has been
30                   associated with a higher mortality after MI and is used as a predictor of the physiological
31                   processes underlying CHD (Buccelletti et al.. 2009). Studies that discuss incidence of MI,
32                   IHD, HRV, and mortality from CHD, MI,  or IHD are included in this category.

33                   There were a small number of studies discussed in the 2006 Pb AQCD (U.S. EPA.
34                   2006b) that indicated associations between Pb biomarker levels and increased risk of
      November 2012                             5-366                  Draft - Do Not Cite or Quote

-------
 1                  cardiovascular outcomes associated with CHD. However, recent longitudinal studies in
 2                  cohorts in different locations with follow-up periods ranging between 8 and 12 years
 3                  report that biomarkers of Pb exposure are associated with risk of mortality from
 4                  cardiovascular disease, specifically MI, IHD, or CHD. A recent analysis of the NHANES
 5                  III sample reported associations of adult blood Pb level with cardiovascular mortality,
 6                  with stronger associations observed with MI mortality (Menke et al.. 2006). These
 7                  findings were supported by a community-based cohort of women age 65-87 years, in
 8                  which higher effect estimates were observed for mortality from CHD (Khalil et al..
 9                  2009b). Weisskopf et al. (2009) published the first mortality study using bone Pb as an
10                  exposure index. This prospective study found that patella bone Pb levels were associated
11                  with increased mortality from IHD. Despite the differences in design and methods across
12                  studies, with few exceptions associations between higher levels of Pb biomarkers and
13                  higher risk of mortality were consistently observed (Figure 5-30 and Table 5-23).

14                  The body of evidence demonstrating associations with mortality from CHD is
15                  substantiated by several findings indicating associations between biomarkers of Pb and
16                  incidence of CHD-related outcomes. A prospective analysis examined the incidence of
17                  IHD (physician confirmed MI, angina pectoris) in the NAS cohort and reported findings
18                  indicating that both blood and bone Pb levels contribute independently to IHD incidence
19                  (Jain et al.. 2007). Earlier studies reported associations of increased Pb biomarkers with
20                  increased risk of left ventricular hypertrophy  (Schwartz. 1991). Coherence for the
21                  associations in humans is provided by a recent animal  study that suggested that Pb
22                  exposure promotes a procoagulant state that could contribute to thrombus formation
23                  which could reduce the blood supply to the heart (Shin et al.. 2007).

24                  Further support for a relationship between Pb exposure and CHD is provided by evidence
25                  from the NAS cohort for effects on disrupted HRV (Bum et al.. 2011; Park et al.. 2009b:
26                  Park et al.. 2006). which has been associated with a higher mortality from MI and is used
27                  as a predictor of the physiological processes underlying CHD. A prospective analysis
28                  reported that higher tibia Pb, but not blood or patella Pb, was associated with increases in
29                  QTc interval and QRSc duration (Eumetal..  2011). Park et al. reported associations of
30                  bone Pb with HRV measures and effect modification by increasing number of iron
31                  metabolism gene variants from 0 to 3. Park et al. (2006) reported associations of bone Pb
32                  with HRV measures and effect modification by increasing number of iron metabolism
33                  gene variants from 0 to 3. Park et al. (2006) reported the strongest relationships between
34                  patella Pb levels and lower HRV among those with three or more metabolic
35                  abnormalities. Also, bone Pb level was associated with larger decreases in HRV among
36                  adults with metabolic syndrome, which like reduced HRV is associated with increased
37                  risk of cardiovascular events (Park et al.. 2006).
      November 2012                            5-367                  Draft - Do Not Cite or Quote

-------
 1                   As CHD is the result of vascular blockage, the suggestive evidence for subclinical
 2                   atherosclerosis supports the observations of increased CHD morbidity and mortality. In
 3                   addition, the strong and consistent evidence for Pb-induced hypertension serves as further
 4                   biological plausibility for CHD. Hypertension may contribute to CHD development in a
 5                   number of ways. Hypertension may lead to thickening of the vascular wall or
 6                   exacerbation of atherosclerotic plaque development and thus contribute to plaque
 7                   instability. In addition, hypertension may increase the myocardial oxygen demand
 8                   priming for potential myocardial ischemia (Olafiranye et al.. 2011). Both subclinical
 9                   atherosclerosis and hypertension are supported by consistent evidence describing the
10                   mode of action including Pb-induced oxidative stress, inflammation, cellular activation
11                   and dysfunction, altered vascular reactivity, RAAS dysfunction, and vasomodulator
12                   imbalance.

13                   Building on this strong body of evidence, recent epidemiologic and toxicological studies
14                   substantiated the evidence that long-term Pb exposure is associated with CHD in adults;
15                   however, uncertainty remains concerning the Pb exposure level, timing, frequency, and
16                   duration contributing to the effects. Overall, high-quality studies examining CHD
17                   morbidity and mortality and contributing cardiovascular effects have been replicated by
18                   different researchers in different cohorts and report consistent associations that increase
19                   the confidence that a relationship exists between Pb exposure and CHD. In addition, both
20                   animal and human studies describe a biologically plausible potential mode of action.
21                   Thus the overall evidence is sufficient to conclude that there is a causal relationship
22                   between Pb exposure and coronary heart disease.
                     5.4.7.4      Evidence for Cerebrovascular Disease

23                   Cerebrovascular disease describes a group of conditions involving the cerebral blood
24                   vessels that result in transient or permanent disruption of blood flow to the brain. These
25                   conditions include stroke, transient ischemic attack, and subarachnoid hemorrhage. Both
26                   hypertension and atherosclerosis are risk factors for Cerebrovascular disease and the
27                   mechanisms for these outcomes also apply to Cerebrovascular disease. Despite strong
28                   evidence for hypertension and CHD and long-term Pb exposure, very few studies have
29                   examined the effects of Pb exposure on Cerebrovascular disease. Lee et al.  (2009)
30                   examined 153 patients in Taiwan cross-sectionally while adjusting for key confounders
31                   and reported increased stenosis greater than 50% in the intracarotid system related to
32                   urine Pb but not blood Pb level. Two epidemiologic studies prospectively evaluated
33                   mortality from stroke. Menke et al. (2006) reported a positive relationship with wide
34                   confidence intervals compared to other outcomes considered for blood Pb levels with
35                   stroke mortality in the NHANES study. Khalil et al. (2009b) provides a non significant


      November 2012                             5-368                  Draft - Do Not Cite or Quote

-------
1

2

3

4
result with imprecise confidence intervals. These few studies provide insufficient

evidence to inform the causal relationship between cerebrovascular disease and long-term

Pb exposure. Thus, the evidence at this time is inadequate to determine that a causal

relationship exists between Pb exposure and cerebrovascular disease.
     Table 5-24   Summary of evidence supporting cardiovascular causal
                      determinations.
     Attribute in Causal
     Framework3
      Key Supporting Evidence
References
Pb Exposure or Blood
Levels Associated with
Effects0
      Hypertension - Causal
     Consistent associations
     with relevant blood Pb
     levels from multiple,
     high quality
     epidemiologic studies
      Longitudinal prospective evidence for
      associations with incidence of
      hypertension and increase in blood
      pressure in adults.
      Large body of supportive cross-sectional
      studies applying differing designs across
      multiple cohorts of adults in different
      locations.
      Meta-analyses provide further support
      Associations found while adjusting for
      numerous potential confounding factors
      Studies provide C-R information
Peters et al. (2007).
Glenn et al. (2006).
Cheng et al. (2001)
Martin et al. (2006).
Scinicariello et al. (2010).
Park et al. (2009c)
Navas-Acien et al. (2008).
Nawrot et al. (2002)


Section 5.4.2.1
Adult, prospective :
Blood Pb level >20 ug/dL;
Bone Pb level >20 ug/g
                                                                                           Adult, concurrent:
                                                                                           Blood Pb level >2 ug/dL;
                                                                                           Bone Pb level >19 ug/g
Consistent toxicological
results provide
coherence with
epidemiologic evidence
Consistent cross-sectional evidence for
increases in BP in adults are supported by
studies in adult rodents with relevant
dietary long-term Pb exposure
Rodents:
Rizzi et al.(2009).
Bravo et al.(2007),
Chang et al.(2005),
Tsao et al. (2000)
Rat, adult:
Blood Pb level >10 ug/dL
                                                                  Section 5.4.2.2
     Consistent associations
     with relevant Pb levels
     in blood and/or bone
     and cardiovascular
     mortality from multiple,
     high quality
     epidemiologic studies
      Longitudinal, prospective studies find
      consistent associations of blood and/or
      bone Pb levels in adults with risk of
      cardiovascular mortality applying differing
      designs across multiple cohorts in different
      locations while controlling for potential
      confounding.
Khalil et al. (2009b).
Weisskopfetal. (2009).
Menke et al. (2006)
Schoberet al. (2006)
Lustberg and Silbergeld
(2002)


Section 5.4.5
Adult, prospective:
Blood Pb level >4 ug/dL
     November 2012
                               5-369
             Draft - Do Not Cite or Quote

-------
Attribute in Causal
Framework3
Key Supporting Evidence
References
Pb Exposure or Blood
Levels Associated with
Effects0
Evidence clearly
describes mode of
action

  Oxidative Stress
  Alteration of vascular
  reactivity
  Renin-angiotensin-
  aldosterone system
  dysfunction
  Vasomodulator
  imbalance
Consistent evidence of increased
oxidative stress leading to inactivation of
'NO and downregulation of sGC in
animals with relevant dietary Pb
exposures and cultured vascular cells.

Toxicological evidence for activation of the
sympathetic nervous system, increased
reactivity to catecholamines, and
activation of the adrenergic and
dopaminergic receptors in rats,  isolated
vessels, and cultured cells.
Mixed evidence for reactivity to  other
pressor agents (e.g., 5-HT) in rats.

Toxicological evidence that activation of
the RAAS may be involved in
development of Pb-induced hypertension
Evidence for increased RAAS activity in
rats and decreased BP following RAAS
inhibition and Pb exposure.

Limited available toxicological evidence
reporting vasomodulator imbalance in
Pb-exposed rats and cells.
                                                                 Section 5.4.2.3
Subclinical Atherosclerosis - Suggestive
Limited evidence in
humans of an
association with
subclinical
atherosclerosis and
peripheral artery
disease
One NHANES analysis reported
associations with PAD at relevant adult
blood Pb levels with control for potential
confounding.

Limited evidence for increased IMT or
arterial stiffness in adult human
populations.

Occupational studies report increased IMT
and atherosclerotic plaque presentation in
highly exposed adult populations.
Muntner et al. (2005)
Ari et al. (2011
                                                                 Poreba et al. (2011: 2011 a)
                                                                 Sections 5.4.3.3 and
                                                                 5.4.3.5
Adult, concurrent:
Blood Pb level >2.5 ug/dL
Adult, concurrent:
Serum Pb level >0.4 ug/dL
                           Adult workers:
                           Blood Pb level >24 ug/dL
Limited evidence in
animals of initiation or
progression of
atherosclerosis after Pb
exposure
Limited studies reporting increased IMT,
vascular morphological changes, and
endothelial and SMC alterations in rats
and human tissue.
Zelleretal. (2010).
Zhang et al. (2009a)


Section 5.4.3.3
Rat: 28.4 ug/dL
Human Tissue: 50 uM
Evidence clearly
describes mode of
action

  Oxidative Stress
  Inflammation
  Vascular Cell
  Activation and
  Endothelial
  Dysfunction
Consistent evidence of increased
oxidative stress in animals with relevant
dietary Pb exposures and cultured
vascular cells.

Toxicological evidence of increased
inflammation as indicated by increased
production of TNF-a,  IL-6, IL-8, and PGE2
by macrophages and vascular cells.

Toxicological evidence of VSMC
stimulation and endothelial dysfunction
and damage in culture.
Limited available evidence of impaired
flow-mediated dilatation in Pb exposed
workers.
                                                                 Section 5.4.2.3
Section 5.4.3.1
November 2012
                            5-370
              Draft - Do Not Cite or Quote

-------
Attribute in Causal
Framework3
Key Supporting Evidence
References
Pb Exposure or Blood
Levels Associated with
Effects0
Coronary Heart Disease - Causal
Consistent associations
with relevant bone
and/or blood Pb levels
and mortality from Ml,
IHD, CHD, and
cardiovascular disease
from multiple, high
quality epidemiologic
studies
Longitudinal, prospective studies find
consistent associations of bone and/or
blood Pb levels in adults with risk of
cause-specific cardiovascular mortality
applying differing designs across multiple
cohorts in different locations with control
for potential confounding.
Khalil et al. (2009b),
Weisskopfetal. (2009).
Menke et al.  (2006)
Schoberetal. (2006)
Lustberg and Silbergeld
(2002)


Section 5.4.5
Adult, prospective:
Blood Pb level >4 ug/dL
Limited evidence in
humans of an
association with
ischemic heart disease,
myocardial  infarction,
or HRV
One prospective study demonstrates an
association of adult blood and bone Pb
levels with incidence of IHD in the MAS
cohort

Associations of Pb levels in adults and left
ventricular hypertrophy and Ml

Prospective evidence of association  of
HRV with tibia bone Pb level in adults.

Evidence for interaction of markers of
metabolic syndrome and  genetic
polymorphisms with Pb-induced HRV.
Jainetal. (2007)




Schwartz (1991)


Eum et al. (2011)


Park et al. (2009b: 2006)
                                                                 Sections 5.4.3.4 and
                                                                 5.4.3.6
Adult, prospective:
Blood Pb level >5 ug/dL
                                                                                            Adult, prospective:
                                                                                            Bone Pb level >23 ug/g
Limited evidence in
animals of increased
thrombosis, enhanced
coagulation, and
arrhythmia
One study reporting increased thrombosis
and enhanced coagulation in rats and
cells.

One study reporting increased incidence
of arrhythmia and atrioventricular
conduction block in rats.
                           Rat
                           Blood Pb level: 26.8 ug/dL
Reza et al. (2008)
                                                                 Sections 5.4.3.4 and
                                                                 5.4.3.6
Evidence clearly
describes mode of
action

  Oxidative Stress
  Inflammation
  Atherosclerosis
  Hypertension
Consistent evidence of increased
oxidative stress in animals with relevant
dietary Pb exposures and cultured
vascular cells.

Toxicological evidence of increased
inflammation as indicated by increased
production of TNF-a, IL-6, IL-8, and PGE2
by macrophages  and vascular cells.

Suggestive evidence of subclinical
atherosclerosis in humans and animals
with relevant Pb exposure resulting in
narrowing of the blood vessels to the
heart.

Consistent evidence of increased BP and
hypertension following Pb exposure in
humans and animals at relevant Pb levels
across numerous studies with control for
confounding.
Association of increased  blood pressure
with manifestation of CHD has been well
documented.
                                                                 Section 5.4.2.3
                                                                 Section 5.4.3.1
                                                                 Sections 5.4.3.3 and 5.4.3.5
                                                                 Section 5.4.2
Cerebrovascular Disease - Inadequate
Evidence for
cerebrovascular
disease in humans and
animals is of insufficient
quality and quantity
One study reported an association of
intracranial carotid stenosis with urinary
Pb level.
Lee et al. (2009)


Section 5.4.3.3
Adult, concurrent:
Blood Pb level >5 ug/dL
November 2012
                           5-371
              Draft - Do Not Cite or Quote

-------
      Attribute in Causal
      Framework3
         Key Supporting Evidence
References
Pb Exposure or Blood
Levels Associated with
Effects0
      Limited evidence for
      increased mortality
      from stroke
         Limited evidence for increased risk of
         mortality from stroke across two cohorts in
         different locations.
Menke et al. (2006),
Khalil et al. (2QQ9b)


Section 5.4.5
Adult, prospective:
Blood Pb level >4 ug/dL
      Evidence for possible
      mode of action

        Hypertension
        Atherosclerosis
         Consistent evidence of increased BP and
         hypertension following Pb exposure in
         humans and animals at relevant Pb levels
         across numerous studies with control for
         confounding.
         Association of increased blood pressure
         with manifestation of CHD has been well
         documented.

         Suggestive evidence of subclinical
         atherosclerosis in humans and animals
         with relevant Pb exposure resulting in
         narrowing of the blood vessels to the
         heart.
                                                                Section 5.4.2
                                                                Sections 5.4.3.3 and 5.4.3.5
      "Described in detail in Table II of the Preamble.
      bDescribes the key evidence and references contributing most heavily to causal determination. Also noted are the sections where
      full body of evidence is described.
      °Describes the blood Pb levels in humans with which the evidence is substantiated and blood Pb levels in animals most relevant to
      humans.
      dBecause blood Pb level in nonoccupationally-exposed adults reflects both recent and past Pb exposures, the magnitude, timing,
      frequency, and duration of Pb exposure contributing to the observed associations is uncertain.
      5.5
Renal Effects
      5.5.1
  Introduction
 i

 2

 o
 J

 4

 5

 6

 7

 8

 9

10

11
    This section summarizes key findings with regard to effects of Pb on the kidney in animal

    toxicology and epidemiologic studies. Findings summarized across epidemiologic and

    toxicological studies indicate that long-term Pb exposure is associated with pathological

    changes in the renal system such as proximal tubule (PT) cytomegaly, renal cell

    apoptosis, mitochondrial dysfunction, aminoaciduria, increased electrolyte excretion,

    ATPase dysfunction, oxidant redox imbalance, altered glomerular filtration rate (GFR),

    chronic kidney disease (CKD) development, and altered NO homeostasis with ensuing

    elevated BP. As several of these outcomes are  most often observed in adults with likely

    higher past Pb exposures, uncertainty exists as to the Pb exposure level, timing,

    frequency, and duration contributing to the associations observed with blood or bone Pb

    levels.
      November 2012
                                  5-372
             Draft - Do Not Cite or Quote

-------
 1                   The cardiovascular and renal systems are intimately linked. Homeostatic control at the
 2                   kidney level functions to regulate water and electrolyte balance via filtration,
 3                   re-absorption and excretion and is under tight hormonal control. Pb exposure has been
 4                   shown to damage the kidneys and its vasculature with ensuing effects on systemic
 5                   hypertension and effects on the cardiovascular (Section 5.4) and renal systems. Chronic
 6                   increases in vascular pressure can contribute to glomerular and renal vasculature injury,
 7                   which can lead to progressive renal dysfunction and kidney failure. In this manner,
 8                   Pb-induced hypertension has been regarded as one potential contributor of Pb-induced
 9                   renal disease. However, the relationship between BP and renal function is more
10                   complicated. Not only does hypertension contribute to renal dysfunction but damage to
11                   the kidneys can also cause increased BP. Long-term control of arterial pressure is
12                   affected by body fluid homeostasis which is regulated by the kidneys. In examination of
13                   the physiological definition of BP (i.e., mean BP equates to cardiac output multiplied by
14                   total peripheral resistance [TPR]) the role of the kidneys in BP regulation is highlighted.
15                   Cardiac  output is driven by left ventricular and circulating blood volume. TPR is driven
16                   by vasomodulation and electrolyte balance. Thus, it is possible to dissect the causes of
17                   hypertension from features of primary kidney disease. Increased extracellular fluid
18                   volume results in increased blood volume which enhances venous return of blood to the
19                   heart and increases cardiac output.  Increased cardiac output not only directly increases
20                   BP, but also increases TPR due to a compensatory autoregulation or vessel constriction.
21                   In addition, damage to the renal vasculature will alter the intra-renal vascular resistance
22                   thereby altering kidney function and affecting the balance between renal function and BP.
23                   The interactions between these systems can lead to further exacerbation of vascular and
24                   kidney dysfunction following Pb exposure. As kidney dysfunction can increase BP and
25                   increased BP can lead to further damage to the kidneys, Pb-induced damage to both
26                   systems  may result in a cycle of further increased severity of disease.

27                   In general, associations between bone Pb (particularly in the tibia) and health outcomes in
28                   adults indicate chronic effects of cumulative Pb exposure. In adults without current
29                   occupational Pb exposure, blood Pb level represents both recent and cumulative Pb
30                   exposure. In particular, blood Pb level may represent cumulative exposure in
31                   physiological circumstances of increased bone remodeling or loss (e.g., osteoporosis and
32                   pregnancy) when Pb from bone of adults contributes substantially to blood Pb
33                   concentrations. Blood Pb level in children is also influenced by Pb stored in bone due to
34                   rapid growth-related bone turnover in children relative to adults. Thus, blood Pb in
35                   children is also reflective of cumulative dose. Additional details on the  interpretation of
36                   Pb in blood and bone are provided in Section 4.3.5. The toxicokinetics of Pb in blood and
37                   bone are important considerations in making inferences about etiologically-relevant Pb
38                   exposures that contributed to associations observed between blood and  bone Pb levels
39                   and health outcomes.

      November 2012                             5-373                 Draft - Do Not Cite or Quote

-------
                    5.5.1.1      Kidney Outcome Measures

 1                  The primary function of the kidneys is to filter waste from the body while maintaining
 2                  appropriate levels of water and essential chemicals, such as electrolytes, in the body.
 3                  Therefore, the gold standard for kidney function assessment involves measurement of the
 4                  GFR through administration of an exogenous radionuclide or radiocontrast marker
 5                  (e-g-,  1251-iothalamate, iohexol) followed by timed sequential blood samples or, more
 6                  recently, kidney imaging, to assess clearance through the kidneys. This procedure is
 7                  invasive and time-consuming. Therefore, serum levels of endogenous compounds are
 8                  routinely used to estimate GFR in large epidemiologic studies and clinical settings.
 9                  Creatinine is the most commonly measured endogenous compound; blood urea nitrogen
10                  (BUN) has also been examined. Increased serum concentration or decreased kidney
11                  clearance of these markers  both indicate decreased kidney function. The main limitation
12                  of endogenous compounds identified to date is that non-kidney factors impact their serum
13                  levels. Specifically, since creatinine is metabolized from creatine in muscle,  muscle mass
14                  and diet affect serum levels resulting in variation in different population subgroups
15                  (e.g., women and children compared to men), that are unrelated to kidney function.
16                  Measured creatinine clearance, involving measurement and comparison of creatinine in
17                  both serum and urine, can address this problem. However, measured creatinine clearance
18                  utilizes timed urine collections, traditionally over a 24-hour period, and the challenge of
19                  complete urine collection over an extended time period makes compliance difficult.

20                  Therefore equations to estimate kidney filtration that utilize serum creatinine but also
21                  incorporate age, sex, race, and, in some, weight (in an attempt to adjust for differences in
22                  muscle mass), have been developed. Although these are imperfect surrogates for muscle
23                  mass, such equations are currently the preferred outcome assessment method.
24                  Traditionally, the Cockcroft-Gault equation (Cockcroft and Gault. 1976). which estimates
25                  creatinine clearance, a GFR surrogate, has been used. In the last decade, the  abbreviated
26                  Modification of Diet in Kidney Disease (MDRD) Study equation (Levey et al.. 2000;
27                  Levey etal.. 1999). which estimates GFR, has become the standard in the kidney
28                  epidemiologic and clinical  communities. With widespread use of the MDRD equation, it
29                  became clear that the equation underestimates GFR at levels in the normal range.
30                  Therefore, the CKD-Epidemiology Collaboration (CKD-EPI) equation was recently
31                  developed to be more accurate in this range (Levey et al.. 2009). This is a decided
32                  advantage in nephrotoxicant research since most participants in occupational and many
33                  even in general population  studies have GFRs in a range that is underestimated by the
34                  MDRD equation.

35                  Both the MDRD and CKD-EPI equations use serum creatinine. Due  to the inability to
36                  adjust serum creatinine levels for muscle mass, alternative serum biomarkers have been
      November 2012                            5-374                 Draft - Do Not Cite or Quote

-------
 1                  evaluated such as cystatin C, a cysteine protease inhibitor that is filtered, reabsorbed, and
 2                  catabolized in the kidney (Fried. 2009). It is produced and secreted by all nucleated cells
 3                  thus avoiding the muscle mass confounding that exists with serum creatinine (Fried.
 4                  2009). However, recent research indicates that serum cystatin C varies by age, sex, and
 5                  race (Kottgen et al.. 2008). Thus, a cystatin C-based eGFR equation was recently
 6                  developed that includes age, sex, and race (Stevens et al.. 2008).

 7                  Most of the kidney outcome measures discussed above were developed for use in the
 8                  clinical setting. Unfortunately, they are insensitive for detection of early kidney damage,
 9                  as evidenced by the fact that serum creatinine remains normal after kidney donation.
10                  Therefore, in the last two decades, the utility of early biological effect (EBE) markers as
11                  indicators of preclinical kidney damage has been of interest. These can be categorized as
12                  markers of function (i.e., low molecular weight proteins that should be reabsorbed in the
13                  PT such as p2-microglobulin and retinol-binding protein [RBP]); biochemical alteration
14                  (i.e., urinary eicosanoids such as prostaglandin E2, prostaglandin F2 alpha, 6-keto-
15                  prostaglandin FI alpha, and thromboxane B2); and cytotoxicity (e.g., N-acetyl-(3-D-
16                  glucosaminidase [NAG]) (Cardenas et al..  1993). Elevated levels may indicate an
17                  increased risk for subsequent kidney dysfunction. However, most of these markers are
18                  research tools only, and their prognostic value remains uncertain since prospective
19                  studies of most of these markers in nephrotoxicant-exposed populations are quite limited
20                  to date. Recently, microalbuminuria has been identified as a PT marker, not just
21                  glomerular as previously thought (Comper and Russo. 2009). Kidney EBE markers are a
22                  major recent focus  for research in patients with acute kidney injury (AKI) and markers
23                  such as neutrophil gelatinase-associated lipocalin (NGAL) and kidney injury molecule-1
24                  (Kim-1), developed in AKI research, may prove useful for chronic nephrotoxicant work
25                  as well (Ferguson et al..  2008; Devarajan. 2007).
      5.5.2        Nephrotoxicity and Renal Pathology
                    5.5.2.1      Epidemiology in Adults

26                  A number of advances in research on the impact of Pb on the kidney in the 20 years
27                  following the 1986 Pb AQCD (U.S. EPA.  1986a) were noted in the 2006 Pb AQCD (U.S.
28                  EPA. 2006b). These included research in general and CKD patient populations at much
29                  lower blood Pb levels (5-10 (ig/dL) at the time of evaluation than were previously
30                  studied. These advances contributed to the understanding of the effects of Pb exposure on
31                  kidney dysfunction overall in the population. Pb, at much lower doses than those causing
32                  chronic Pb nephropathy, may act as a cofactor with other more established kidney risks to
      November 2012                            5-375                 Draft - Do Not Cite or Quote

-------
 1                  increase the risk of CKD and disease progression in susceptible patients. Marie and Hall
 2                  (2011) note that data from basic and clinical studies suggest that obesity, hypertension,
 3                  hyperglycemia, hyperlipedemia, and other elements of the metabolic syndrome are highly
 4                  interrelated and contribute to the development and progression of diabetic nephropathy
 5                  and thus represent populations potentially at increased risk for kidney dysfunction.

 6                  In the 2006 Pb AQCD (U.S. EPA. 2006b). several key issues could not be completely
 7                  resolved based on the Pb-kidney literature published to date. These included
 8                  characterizing the lowest Pb dose at which altered kidney function effects occur, the
 9                  impact of higher past exposures on associations with concurrent Pb biomarker levels, the
10                  impacts of Pb on the kidney in children, the use of paradoxical Pb-kidney associations on
11                  risk assessment in the occupational setting, and the impact of co-exposure to other
12                  environmental nephrotoxicants, such as Cd. In the intervening five years, relevant data
13                  addressing several of these challenges have been published.


                    General Population Studies

14                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported studies that examined associations
15                  between indicators of Pb exposure and kidney function in general populations. This was a
16                  new approach to Pb-kidney research in the two decade time period covered by the
17                  2006 Pb AQCD. As illustrated in Figure 5-31 and Table 5-25. studies consistently
18                  demonstrate associations between higher blood Pb level and lower renal function in
19                  adults. These general population studies provided critical evidence that the effects of Pb
20                  on the kidney occur at much lower doses than previously appreciated based on
21                  occupational exposure data. However, because blood Pb level in nonoccupationally-
22                  exposed adults reflects both recent and past Pb exposures, the magnitude, timing,
23                  frequency, and duration of Pb exposure contributing to the  observed associations was
24                  uncertain. The evidence of Pb-associated renal effects in general population studies was
25                  substantiated by results that were adjusted for multiple potential confounding factors
26                  including age, race, sex, education, household income, smoking, alcohol use, and various
27                  health indicators such as diabetes, SBP, BMI, and history of cardiovascular disease. A
28                  few studies also adjusted for Cd exposure.

29                  The landmark Cadmibel Study was the first large environmental study of this type that
30                  adjusted for multiple kidney risk factors, including urinary  Cd (Staessen et al.. 1992). It
31                  included 965 men and 1,016 women recruited from Cd exposed and control areas in
32                  Belgium. Mean concurrent blood Pb was 11.4 ug/dL (range 2.3-72.5) and 7.5 ug/dL
33                  (range 1.7-60.3) in men and women, respectively. After adjustment for covariates (Table
34                  5-25). log transformed blood Pb was negatively associated  with measured creatinine
35                  clearance. A 10-fold increase in blood Pb was associated with a decrease in creatinine
      November 2012                            5-376                  Draft - Do Not Cite or Quote

-------
 1                   clearance of 10 and 13 mL/min in men and women, respectively. Blood Pb was also
 2                   negatively associated with estimated creatinine clearance.

 3                   Multiple analyses assessing the kidney impact of Pb exposure have been conducted in the
 4                   NAS population (Tsaih et al.. 2004; Wu et al.. 2003a: Kimetal.. 1996; Pavton et al..
 5                   1994). Participants in this study were originally recruited in the 1960s in the Greater
 6                   Boston area. The inclusion criteria, male sex, age 21 to 80 years, and absence of chronic
 7                   medical conditions, limit the generalizability of the results to the rest of the U.S.
 8                   population. Longitudinal data contained in NAS publications remain essential to address
 9                   the dearth of prospective data on the kidney effects of Pb. The first of these included 459
10                   men whose blood Pb levels from periodic examinations, conducted every 3 to 5 years
11                   during 1979-1994, were estimated based on measurements  in stored packed red blood
12                   cell samples adjusted for hematocrit level (Kim etal.. 1996).  Participants were randomly
13                   selected to be  representative of the entire NAS  population in terms of age and follow-up.
14                   Kidney function was assessed with serum creatinine. Data from four evaluations were
15                   available  for the majority of participants. At baseline, mean (SD) age, blood Pb level, and
16                   serum creatinine, at baseline, were 56.9 (8.3) years, 9.9 (6.1)  ug/dL, and 1.2 (0.2) mg/dL,
17                   respectively. In the longitudinal analysis with random-effects modeling of repeated
18                   measures, In-transformed blood Pb was associated with an increase in serum creatinine
19                   from the previous to current follow-up period in the 428 participants whose highest blood
20                   Pb level was < 25 ug/dL (|3= 0.027 mg/dL [95% CI: 0.0, 0.054] per unit increase in In
21                   blood Pb); effect estimates  in the entire group and subsets with different peak blood Pb
22                   levels (<  10 or 40 ug/dL) also were positive (and larger for blood Pb levels < 10 ug/dL).

23                   This  study made two other key  contributions. In order to address the question of whether
24                   nephrotoxicity observed at current blood Pb levels is due to higher blood Pb levels from
25                   past exposure, these authors performed a sensitivity analysis in participants whose peak
26                   blood Pb  levels, dating back to  1979, were < 10 ug/dL. A statistically significant positive
27                   association between blood Pb and concurrent serum creatinine remained in a cross-
28                   sectional  analysis. These authors evaluated reverse causality, which attributes increased
29                   blood Pb  levels to lack of kidney excretion rather than as a  causative factor for CKD, by
30                   showing in adjusted plots that the association between blood Pb and serum creatinine
31                   occurred over the entire serum creatinine range (0.7-2.1 mg/dL), including the normal
32                   range where reverse causality would not be expected.

33                   Cortical and trabecular bone Pb measurements  were obtained in addition to whole blood
34                   Pb in evaluations performed in the NAS between 1991 and 1995. Associations between
35                   baseline blood, tibia, and patella Pb and change in serum creatinine over an average of 6
36                   years in 448 men were reported in a subsequent NAS publication (Tsaih et al.. 2004). At
37                   baseline, eligible participants were similar to nonparticipants with regard to age, BMI,
      November 2012                             5-377                  Draft - Do Not Cite or Quote

-------
 1                   alcohol consumption, smoking status, diabetic status, hypertensive status, baseline SCr,
 2                   and blood and bone Pb levels, indicating lack of selective follow-up by blood/bone Pb
 3                   level or kidney function. At baseline 6 and 26% of subjects had diabetes and
 4                   hypertension, respectively. Mean blood Pb levels and serum creatinine decreased
 5                   significantly over the follow-up period in the group. In cross-sectional analyses, both
 6                   patella and tibia Pb, but not blood Pb level, were positively but nonsignificantly
 7                   associated with serum creatinine. Baseline blood Pb level was not significantly associated
 8                   with change in creatinine in all participants. However, diabetes was observed to be an
 9                   effect modifier of the relations of blood and tibia Pb with change in serum creatinine. Per
10                   unit increase in In blood Pb, the increase in serum creatinine between follow-up periods
11                   was substantially stronger in diabetics ((3 = 0.076 mg/dL [95% CI: 0.031, 0.121])
12                   compared to non-diabetics ((3 = 0.006 mg/dL [95% CI: -0.004, 0.016]). A similar
13                   relationship was observed for tibia Pb. An interaction was also observed between tibia Pb
14                   and hypertension, although it is possible that many of the 26 diabetics were also included
15                   in the hypertensive group and were influential there as well. A sensitivity analysis was
16                   conducted to evaluate the potential for reverse causality by examining participants whose
17                   serum creatinine was <1.5 mg/dL; the authors reported that longitudinal associations did
18                   not materially change.

19                   In modeling the association between blood Pb level and change in serum creatinine,
20                   Tsaih et al. (2004) adjusted for baseline serum creatinine. Glymour et al.  (Glymour et al..
21                   2005) discusses how such adjustment may introduce bias. If there is no interaction
22                   between Pb exposure and unmeasured causes of kidney disease, the model is linear, and
23                   the slope does not change direction prior to and during the study period, the bias should
24                   be to underestimate the effect. However, Glymour (2012). noted that the direction of the
25                   bias is difficult to predict when the model is  nonlinear or the data are restricted to a
26                   specific  stratum of outcome.
      November 2012                             5-378                  Draft - Do Not Cite or Quote

-------
       Study
                        Population
                                            Blood Pb Median Blood Pb 10th-90th Outcome
                                             (IQR) (ug/dL)1  percentile d
       POSITIVE EFFECT ESTIMATES INDICATE POORER f UNCTION


       LONGITUDINAL RESULTS

       Kim et al, (1996)      NAS men, blood Pb < 40 ng/dL 8.4(5.7,12.4)  4.0-17,5
                        NAS men, blood Pb < 25 ng/dL
                        NAS men, blood Pb < 10 [ig/dL
             Change in Scr between visits
                  (mg/dL) x 10
                                             3 9 (2 8 5 6)  21-76
       Tsiah et al. (2004)     NAS men with diabetes
                        NAS men without diabetes
                        NAS men with hypertension
                        NAS men without hypertension


       CROSS-SECTIONAL RESULTS
       Kim etal. (1996)      NAS men, blood Pb < 40 Mg/dL 8.4(5.7,12.4)  4.0-17.5   5cr(mg/dL)
                        NAS men, blood Pb < 25 ug/dL
                        NAS men, blood Pb s 1
                                                                 chs"Be '" Scr Per Vear (mg/dl)
       Tsaih et al. (2004)
                        NAS men with diabetes
                        NAS men without diabetes
                        NAS men with hypertension
                        NAS men without hypertension
                                             Baseline Blood Pb
                                             5.5(3.7,8.1)  2.6-11.5   Scr (mg/dL)
                                             Repeated measures Blood Pb
IIM:J men wan aidoeiei
NAS men without diabetes
NAS men with hypertension
NAS men without hypertension
DeBurbure et al. (2006) Ctech, French, Polish Children
NEGATIVE EFFECT ESTIMATES INDICATE POORER FUNCT
LONGITUDINAL RESULTS
Yu et al. (2004) CKD Patients
CROSS-SECTIONAL RESULTS
Akesson et al. (2005) Swedish Women
Staessen etal. (1992) Belgian Women
Paytonet ol(1996) NAS Men
Akesson et al. (2005) Swedish Women
Fadrowski et al. (2010) NHANES III Adolescents
3.3 (i.O, 3.OJ 4.J-/.D

3.9 (2.6, 5,7) 1.8-8.1
ION
3.2 (2.5, 4.1) 2.0-5.1
2.2 (1.7, 3.0) 1.3-3.8
7.5 (5.2, 10.9) 3.7-15 1
7.3 (5.4, 9.9) 4.1-12.9
2.2 (1.7, 3.0) 1.3-3.8
1.4 (0.7, 2.9) 0.4-5.4


Log Scr (rng/L)b •« — »

Change In GFR over A yr/100 >
(rnL/mln)
Creatinine C lea ra nee/ 1 DO
(mL/min)
Creatinine Clearance/ 100
(mL/min)
Log Creatinine Clearance
(mL/min)
GFR/100(mL/rnin) -•-
GFR/100(mL/min/1.73m2) — •—
^
-• 	








                                                                                  0.1
                                                                                            0.0
                                                                                                      0.1
                                                                                                                0.2
                                                                           Change in Kidney Function Per 1 ng/dL increase in
                                                                         blood Pb level within the 10th-90th percentile interval
"Blood Pb data are presented as median and (IQR) in u.g/dL for blood Pb. For uniform presentation, median and IQR were
estimated from the given distributional statistics by assuming normal distributions.
bThe cross product of logged blood Pb and ranked urine Hg was included in the regression to model the interaction between these
two variates. The significant hyperfiltrative effect to these children  could be due to a biphasic time course sometimes seen in early
exposure.
Note: Results are  presented first for kidney function tests where an increase is considered impaired function (black circles) then for
tests where a decrease is considered impaired function (blue circles, outlined in box). Within a category, results are presented first
for longitudinal analyses followed by cross-sectional analyses. To  compare results for linear and nonlinear modeling, effect
estimates were standardized to a 1 ug/dL increase in blood Pb level within the 10th-90th percentile interval. Magnitudes of the effect
should not be compared among different kidney metrics.


Figure 5-31       Concentration-response  relationships for associations  between

                       blood Pb level  or bone  Pb level and  kidney function outcomes.
November 2012
5-379
Draft - Do Not Cite or Quote

-------
 1                  The impact of Pb on the kidney has been examined in multiple NHANES datasets
 2                  obtained over the last few decades (Figure 5-32 and Table 5-25). NHANES data analyses
 3                  benefit from a number of strengths including large sample size, ability to adjust for
 4                  numerous potential confounding factors, and the fact that the study population is
 5                  representative of the U.S. non-institutionalized, civilian population. The results, covering
 6                  different time frames, have been consistent in providing support for Pb as a CKD risk
 7                  factor, including NHANES III, conducted from 1988-1994, in which adults with
 8                  hypertension and diabetes were observed to be potentially at-risk populations (Muntner et
 9                  al.. 2003) and NHANES 1999-2002 (Muntner et al.. 2005). However, because the various
10                  NHANES analyses were cross-sectional in design, examining associations between
11                  concurrent measures of kidney function and blood Pb levels, a common limitation is the
12                  uncertainty regarding the temporal sequence between Pb exposure and renal function and
13                  the magnitude, timing, frequency, and duration of Pb exposure that contributed to the
14                  observed associations.

15                  A recent publication examined NHANES data collected from 1999 through 2006 (Navas-
16                  Acien et al.. 2009). The geometric mean concurrent blood Pb level was  1.58 ug/dL in
17                  14,778 adults aged > 20 years. After adjustment for survey year, sociodemographic
18                  factors,  CKD risk factors, and blood Cd, the odds ratios for albuminuria (> 30 mg/g
19                  creatinine), reduced eGFR (<60 mL/min/1.73 m2), and both albuminuria and reduced
20                  eGFR were 1.19 (95% CI: 0.96,  1.47), 1.56 (95% CI:  1.17, 2.08), and 2.39 (95% CI:
21                  1.31,  4.37), respectively, comparing the highest (>2.4 ug/dL) to the lowest (< 1.1 ug/dL)
22                  blood Pb quartiles. Thus, in the subset of the population with the most severe kidney
23                  disease  (both reduced eGFR and albuminuria), the magnitude of association with
24                  concurrent blood Pb was greater. When blood Cd was included as  a covariate, blood Pb
25                  remained significantly associated with renal function. In fact, the most important
26                  contribution of this recent NHANES analysis was the evaluation of joint Pb  and Cd
27                  exposure (discussed in Section 5.5.4.1).

28                  An important contribution of all NHANES publications is that they provide evidence that
29                  blood Pb remains associated with reduced kidney function (<60 mL/min/1.73 m2 as
30                  estimated with the MDRD equation cross-sectionally) despite steadily declining blood Pb
31                  levels in the U.S. population during the time periods covered. Other studies of adults
32                  participating in NHANES have also reported worse kidney function related to blood Pb
33                  levels (Lai et al.. 2008a: Hernandez-Serrato et al.. 2006; Goswami et al.. 2005).
      November 2012                            5-380                 Draft - Do Not Cite or Quote

-------
  Study
Quartiles of Blood Pb Distribution Used
  Muntner et al. (2003)

        Elevated Serum Creatinine

            Hypertensive
            Normotensive
  Navas-Acien et al. (2009)

        Albuminuria >30 mg/g creatinine
        eGFR <60 mL/min/1.73m2
                                              AL
                                               J 1
                                              Al
                                               1
                                               1
                                             ug/dL Blood Pb
                          -40       -10    10           50

                             % Change per ug/dL Blood Pb
Note: These results depicted are from studies that reported ORs of kidney function measures by grouping the population into
quartiles of blood Pb and then comparing each group to the quartile with the lowest blood Pb (reference group). The blood Pb
distribution of the examined group is shaded black and the reference group is shaded gray. To express these odds ratios in terms of
blood Pb concentration, a log normal distribution was fit to the statistics presented and then the medians of each group were
determined. The adjusted OR was the exponentiated quantity (log(OR) divided by the difference in the medians of the groups
compared). The resulting odds ratio is presented in terms of percent change=100*(OR-1).


Figure  5-32      Percent change in kidney outcomes across quartiles of blood Pb

                     level in NHANES.
November 2012
 5-381
Draft - Do Not Cite or Quote

-------
Table 5-25   Additional  characteristics and quantitative data for associations of blood and bone Pb with kidney
                outcomes for results presented in Figure 5-31 and Figure 5-32.
  Reference
Population
   Study
 Location;
Time Period
N
Pb Biomarker
    Data
                                                                              Outcome
Statistical Analysis
                                                                                                                              Effect Estimate
                                                                                                                                 (95% Cl)
Results for Figure 5-31: Positive Effect Estimates Indicate Poorer Function
Longitudinal Results
                                                                                                     Per 1 jjg'dL increase in blood Pb
                                                                                                     within the 10th-90th percentile
                                                                                                     interval
                Adult males
                                 Boston, MA;
                                 Multiple
                                 examinations
                                 1979-1994
                             459     Median baseline
                                     blood = 8.6 ug/dL
                                     10th-90th percentile:
                                     4.0-17.5
                                          Change in serum
                                          creatinine
                                          between visits x
                                          10(mg/dL)
                                           Random-effects modeling
                                           adjusted for baseline age,
                                           time since initial visit, BMI,
                                           smoking status, alcohol
                                           ingestion, education level,
                                           hypertension, baseline serum
                                           creatinine, and time between
                                           visits
                                                              Peak blood Pb < 40 ug/dL: 0.012
                                                              (-0.0001, 0.025)
                                                              Peak blood Pb < 25 ug/dL: 0.015
                                                              (0.0002, 0.03)
                                                              Peak blood Pb < 10 ug/dL: 0.021
                                                              (-0.005, 0.048)
Tsaih et al.
(2004)
                Adult males
                                 Boston, MA;
                                 8/1991-1995
                                 with mean
                                 6 year
                                 follow-up
                             448     Mean (SD) Baseline
                                     Blood Pb = 6.5
                                     (4.2) ug/dL
                                     10th-90th percentile:
                                     2.1-7.6
                                     Tibia Pb = 21.5
                                     (13.5)ug/g
                                     Patella Pb = 32.4
                                     (20.5) ug/g
                                          Change in serum
                                          creatinine per
                                          year
                                          x 10(mg/dL)
                                           Log linear regression adjusted
                                           forage, age squared, BMI,
                                           hypertension, diabetes,
                                           smoking status, alcohol
                                           consumption, analgesic use,
                                           baseline serum creatinine,
                                           serum creatinine squared
                                                              With diabetes: 0.18 (0.07, 0.29)
                                                              Without diabetes: 0.014 (-0.009,
                                                              0.037)
                                                              With hypertension: 0.019 (-0.027,
                                                              0.065)
                                                              Without hypertension: 0.021
                                                              (-0.007, 0.049)


                                                              Per unit increase in In-transformed
                                                              tibia Pb
                                                              With diabetes: 0.082 (0.03, 0.14)
                                                              Without diabetes: 0.005 (-0.01,
                                                              0.02)
                                                              With hypertension: 0.023 (0.003,
                                                              0.04)
                                                              Without hypertension: 0.0004
                                                              (-0.01,0.01)
November 2012
                            5-382
                                      Draft - Do Not Cite or Quote

-------

Reference Population
Study
Location;
Time Period N
Pb Biomarker
Data

Outcome

Statistical Analysis
Effect Estimate
(95% Cl)
Cross-Sectional Results
Kim et al. (1 996) Adult males



Tsaih et al. Adult males
(2004)













De Burbure et al. Children, mean
(2006) age = 1 0 years,
age
range = 8.5-12.3
years


Boston, MA; 459
Multiple
examinations
1979-1994

Boston, MA; 448
8/1991-1995
with mean 6
yr follow-up













France, 804
Czech
Republic, and
Poland; dates
not provided


Median baseline
blood = 8.6 ug/dL
10th-90th percentile:
4.0-17.5

Mean (SD) baseline
Blood
Pb = 6.5 (4.2) ug/dL
10th-90th percentile:
2.6-11 .5
Repeated measures
10th-90th percentile:
2.1-7.6
Tibia Pb = 21 5
(13.5)ug/g
Patella Pb = 32.4
(20.5) ug/g







Concurrent Blood Pb
Median (IQR) = 3.9
(2.6, 5.7) ug/dL
10th-90th percentile:
1.8-8.1


Serum creatinine
(mg/dL)


Serum creatinine
(mg/dL)













Log-transformed
serum creatinine,
cystatin C, and
p2-microglobulin


Random-effects modeling
adjusted for baseline age,
time since initial visit, BMI,
smoking status, alcohol
ingestion, education level, and
hypertension.
Log linear regression adjusted
forage, age squared, BMI,
hypertension, diabetes,
smoking status, alcohol
consumption, analgesic use













Log linear regression adjusted
for Cd, urinary creatinine,
urinary Hg


Peak blood Pb < 40 ug/dL: 0.0017
(0.0005, 0.003)
Peak blood Pb < 25 ug/dL: 0.0021
(0.0007, 0.0035)
Peak blood Pb < 10 ug/dL: 0.0033
(0.0012,0.0053)
Baseline blood Pb
With diabetes: -0.009 (-0.038,
0.020)
Without diabetes: -0.004
(-0.010, 0.003)
With hypertension: 0 (-0.013,
0.013)
Without hypertension: -0.005
(-0.011,0.002)
Follow-up blood Pb
With diabetes: 0.053 (-0.032,
0.138)
Without diabetes: 0.034 (0.007,
0.061)
With hypertension: 0.083 (0.038,
0.128)
Without hypertension: 0.014
(-0.016, 0.044)
Log serum creatinine (mg/L):
-0.062 (-0.1 06, -0.01 7)a
Log Cystatin C: -1 .3 (-2.4,
-0.21 )a
Log p2-microglobulin: -2.2
(-4.0, -0.54)3
November 2012
5-383
Draft - Do Not Cite or Quote

-------
Reference
Results for
Longitudinal
Population
Study
Location;
Time Period N
Fiqure 5-31: Neqative Effect Estimates
Results:
Yu et al. (2004) Adult CKD
patients

Taipei, 121
Taiwan;
48 month
longitudinal
study period
Pb Biomarker
Data
Indicate Poorer

Mean (SD) Baseline
blood = 4.2
(2.2) ug/dL
10th-90th percentile
2.0-5.1
Outcome
Function

Change in MDRD
eGFRover4
yr/100
(mL/min/1.73m2
body surface area)
Statistical Analysis


Cox proportional hazard
model examined whether a
predictor was associated
with renal function including
age, sex, BMI,
hyperlipidemia,
hypertension, smoking, use
of ACE inhibitor, baseline
serum creatinine, daily
protein excretion, daily
protein intake, underlying
kidney disease
Effect Estimate
(95% Cl)

Per 1 ug/dL increase
in blood Pb within the
10th-90th percentile interval
-0.040 (-0.072, -0.008)3
Cross-Sectional Results:
Akesson et al.
(2005)
Staessen et al.
(1992)
WHILA,
adult women
Adults
Sweden; 820
6/1999-1/2000
Belgium; 1,981
1985-1989
Median (5-95%)
concurrent
blood = 2.2 (1.1,
4.6) ug/dL
10th-90th percentile:
1.3-3.8
Creatinine
clearance/100
(mL/min)
Cystatin C-based
eGFR (Larsson et
al.. 2004V100
(mL/min)
Concurrent Blood Pb Creatinine
Mean (SD) clearance/100
Males:11.4ug/dL (mL/min)
Linear regression adjusted
for age, BMI, diabetes,
hypertension, regular use of
status
Log linear regression
adjusted for age, age
squared, sex, BMI, BP,
-0.01 8 (-0.03, -0.006)
-0.02 (-0.03, 0.007)
Females: -0.067
(-0.108, -0.027)3
Males- -O rWI C-D DQ7 -D ClA7\B
Payton et al.
(1994)
                  Adult males
Boston, MA;
1988-1991
                                                    744
                                                             Females: 7.5 ug/dL
                                                             10th-90th percentile:
                                                             3.7-15.1
Mean (SD)
concurrent
blood = 8.1
(3.9) ug/dL
10th-90th percentile:
4.1-12.9
Log-transformed
creatinine
clearance (mL/min)
ferritin level, smoking status,
alcohol ingestion,
rural/urban  residence,
analgesic and diuretic use,
blood and urinary Cd,
diabetes, occupational
exposure to heavy metals,
and gamma glutamyl
transpeptidase

Log linear regression
adjusted forage, BMI,
analgesic and diuretic use,
alcohol consumption,
smoking status, SBP, DBP
-0.040 (-0.079, -0.0015)
November 2012
                                          Draft - Do Not Cite or Quote

-------
Study
Location;
Reference Population Time Period
Fadrowski et al. NHANES, U.S.;
(2010) adolescents 19881994





Results for Figure 5-32: Analysis of Blood
Muntner et al. NHANES III, U.S.;
(2003) adults 1988-1994












Navas-Acien et NHANES III, U.S.;
al. (2009) adults 1999-2006






Pb Biomarker
N Data
769 Median concurrent
blood = 1.5 ug/dL
10th-90th percentile:
0.4-5.4

Q1: <1 .0
Q2: 1.0 to 1.5
Q3: 1.6 to 2.9
Q4: >2.9
Pb Quartiles
4813 Mean(SD)
concurrent blood Pb
With Hypertension:
4.2(0.14)ug/dL
Q1: 0.7 to 2.4
Q2: 2.5 to 3.8
Q3: 3.9 to 5.9
Q4: 6.0 to 56.0

Without
Hypertension: 3.3
(0.10) ug/dL
Q1: 0.7 to 1.6
Q2: 1.7 to 2.8
Q3: 2.9 to 4.6
Q4: 4.7 to 52.9



14,778 Geometric
concurrent blood
mean = 1.58 ug/dL
Q1: £ 1 .1

Q2: 1.2 to 1.6
Q3: 1.7 to 2.4
Q4: >2.4



Outcome
Cystatin C-based
eGFR/100
(mL/min/1.73 m2;
calculated using
the Filler and
Lepage equation)



Elevated Serum
Creatinine
(99th percentile of
each race-sex
specific distribution
for healthy young
adults)


CKD










eGFR <60
mL/minute/1.73 m2



Albuminuria and
eGFR <60
mL/minute/1 .73 m



Statistical Analysis
Log linear regression
adjusted for age, sex,
race/ethnicity, urban/rural
residence, smoking, obesity,
household income,
education level of family
reference person, BP, lipid
levels, glucose levels


Logistic regression adjusted
for age, race, sex, diabetes,
SBP, smoking, history of
CVD, BMI, alcohol
consumption, household
income, education level,
marital status, health
insurance












Logistic regression adjusted
for survey year, age, sex,
race/ethnicity, BMI,
education, smoking,
cotinine, alcohol intake,
- hypertension, diabetes,
menopausal status



Effect Estimate
(95% Cl)
-0.022 (-0.038, -0.0054)
Q1 : Referent
Q2: -1.4 (-7.4, 4.5)
Q3: -2.6 (-7.3, 2.2)
Q4: -6.6 (-12.6, -0.07)


% change in kidney outcome
Q1 : Referent
With hypertension
Q2:47%(3, 110)
Q3: 80% (34, 1 42)
Q4: 1 41 % (46, 297)
Without hypertension
Q2:11%(-44, 121)
Q3: 19% (-38, 125)
Q4:9%(-47, 122)

With hypertension
Q2:44%(0, 109)
Q3:85%(32, 159)
Q4: 1 60% (52, 345)
Without hypertension
Q2:-10% (-63, 116)
Q3:0%(-55, 122)
Q4:9%(-59, 189)
Q1 : Referent
Q2: 10% (-20, 51)
Q3: 36% (-1 , 85)
Q4: 56% (17, 108)


Q2:53%(-15, 177)
Q3:57%(-17, 198)
Q4: 139% (31, 337)
a95% Cl: estimated from given p-value.
November 2012
5-385
Draft - Do Not Cite or Quote

-------
                    Patient Population Studies

 1                  CKD as defined by the National Kidney Foundation (NKF) - Kidney Disease Outcomes
 2                  Quality Initiative workgroup (NKF. 2002) is the presence of markers of kidney damage
 3                  or GFR <60 mL/min/1.73 m2 for > 3 months. The MDRD equation is the most common
 4                  one used in the eGFR determination for this definition. Notably, decreased GFR is not
 5                  required for the first criterion and markers of kidney damage are not required for the
 6                  second criterion.

 7                  Several key studies in CKD patients provide prospective data that indicate that higher
 8                  baseline blood Pb level is associated with greater CKD progression over time (kidney
 9                  function decline) in patient populations (Table 5-26). Yu et al. (2004). discussed in the
10                  2006 Pb AQCD, followed 121 patients over a four year period. Eligibility required well-
11                  controlled CKD with serum creatinine between 1.5 and 3.9 mg/dL.  Importantly, EDTA-
12                  chelatable Pb <600 ug/72 h, a level below that traditionally thought to indicate risk for
13                  Pb-related nephrotoxicity, was required at baseline. Patients with potentially unstable
14                  kidney disease were excluded (i.e., due to systemic diseases such as diabetes). Mean
15                  blood Pb and EDTA-chelatable Pb levels were 4.2 ug/dL and 99.1 ug/72 hours,
16                  respectively. Cox proportional hazard modeling indicated lack of significant association
17                  between serum creatinine changes and various potential confounding factors (Table
18                  5-25). examined one at a time. Only chelatable Pb (body Pb burden indicator) was
19                  significantly associated with overall risk for the primary endpoint (doubling of serum
20                  creatinine  over the 4-year study period or need for hemodialysis). When the group was
21                  dichotomized by EDTA chelatable Pb level, Kaplan-Meier analysis demonstrated that
22                  significantly more patients (15/63) in the high-normal group (EDTA chelatable Pb level >
23                  80 but <600 ug/72 hours) reached the primary end point than did those in the lower
24                  EDTA chelatable Pb levels (<80 ug Pb/72 hours) group (2/58). Associations between
25                  baseline chelatable or blood Pb level and change in serial measurements of eGFR
26                  (estimated by the MDRD equation (Levey etal. 1999)) were modeled separately using
27                  generalized estimating equations. Based on these models, a 10 ug higher chelatable Pb
28                  level or 1 ug/dL higher blood Pb level reduced the GFR by 1.3 and 4.0 mL/min/1.73 m2,
29                  respectively, during the 4-year study period. The use of estimated GFR provides a better
30                  estimate of progressive changes of renal function than creatinine  clearance used in the
31                  other related studies. Recent studies expanded the CKD patient populations in which this
32                  effect was observed to include those with diabetic nephropathy (Lin et al.. 2006b) and
33                  with the lowest blood Pb levels studied to date (Lin et al.. 2006a). Results of these
34                  observational studies have been summarized in Table 5-26 (Weaver and Jaar. 2010).
      November 2012                            5-386                 Draft - Do Not Cite or Quote

-------
Table
Study
Linet
al.
(2003)
Yuet
al.
(2004)
Linet
al.
(2006b)
Lin et
al.
(2006a)
5-26 Prospective patient population studies: kidney function decline.
Baseline
mean
(SD)
blood
Pb
n (ua/dL)
202 5.3 (2.9)
121 4.2(2.2)
87 6.5 (3.4)
108 2.9(1.4)a
"Notably, mean blood Pb level
in Baltimore, MD (Martin et al..
Baseline mean
(SD) chelatable
Pb(ug/72 hours)
104.5(106.3)
99.1 (83.4)
108.5(53.8)
40.2(21.2)
(all <80)
Baseline
mean(SD)
eGFR
(mL/min
/1.73 m2)
41.6(14.4)
36.0 (9.8)
35.1 (9.0)
47.6 (9.8)
in this study was below that observed in a
2006).
Decline in
eGFR perl SD
higher Pb dose
Years of at baseline
follow-up per year
2 0.16
4 2.7 (chelatable)
2.2 (blood Pb)
1 3.87
2 1.1
Comments
Largest study to date
Longest follow-up;
1 jxg/dL higher blood
Pb, at baseline,
associated with
4.0mL/min/1.73m2
reduction in eGFR over
4 years
Type II diabetics with
nephropathy
Lowest Pb exposed
CKD patients
recent large general population study of 50- to 70-year olds
      Source: Reprinted with permission of UpToDate.com, Weaver and Jaar (2010)
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
A recent population-based case-control study examined occupational Pb exposure as a
risk factor for severe CKD (Evans et al.. 2010). The study included 926 cases with first
time elevations of serum creatinine >3.4 mg/dL for men and >2.8 mg/dL for women and
998 population-based controls. Occupational Pb exposure was assessed using an expert
rating method based on job histories; no biomarkers of Pb exposure were measured. In
multivariable logistic regression modeling, the OR for CKD (adjusted for age, sex,
smoking, alcohol consumption, diabetes, education, and BMI) was 0.97 (95% CI: 0.68,
1.38) in Pb-exposed compared to non-exposed participants. In addition, the CKD patients
were followed prospectively for a mean of 2.5 years for the 70 Pb-exposed patients and
2.4 years for the 731 patients without past occupational Pb exposure. Mean eGFRs (using
the MDRD equation) were 16.0 and 16.6 mL/min/1.73 m2 in exposed and non-exposed
patients, respectively, indicating severe disease in both groups. The results  overall did not
provide strong evidence that Pb exposure was associated with renal effects. The expert
ratings used in this study may have lower validity and reliability as compared to other
exposure assessment methods (Teschke et al.. 2002) including blood and bone
measurements  used in the majority of well-conducted studies. Strengths included
virtually complete case ascertainment and minimal loss to follow-up. Exposure
assessment was listed as both a strength and a limitation. Expert rating methods are
commonly used when biological monitoring is not an option and in case-control studies
where many occupational exposures are considered. In Pb-kidney research, this approach
is uncommon except in the case-control setting. However, given the challenges of
      November 2012
                           5-387
Draft - Do Not Cite or Quote

-------
 1                  interpreting blood Pb in dialysis patients (discussed below), this approach may have
 2                  advantages in this study of such severe CKD. Other case-control studies examining
 3                  occupational risk factors for CKD found Pb exposure to be a risk factor (Nuvts et al..
 4                  1995: Steenland et al.. 1990). Nuyts et al. (1995) found adults with history of
 5                  occupational Pb exposure to have elevated odds of CKD (OR for ever- versus never-
 6                  exposed: 2.11 [95% CI: 1.23, 4.36]). The association was weaker in Steenland et al.
 7                  (1990) (OR for ever- versus never-exposed: 1.73 [95% CI: 0.82, 3.65]). Regular
 8                  moonshine consumption, also a potential source of Pb exposure, was a stronger risk
 9                  factor for CKD (OR: 2.42  [95% CI: 1.10, 5.36]).

10                  The prospective observational aspect of Evans et al. (2010) is similar in design to the
11                  work of Lin and colleagues but differs in several important respects. In Evans et al.
12                  (2010). only occupational Pb exposure was considered whereas the work in Taiwan
13                  excluded occupational exposure and used blood and chelatable Pb measures. In the past
14                  in developed countries, environmental exposures were substantial. For example, mean
15                  tibia Pb levels were 21.5 and 16.7 |o,g/g bone mineral,  in environmentally-exposed 50- to
16                  70-year-old African-Americans and whites, respectively, in Baltimore (Martin et al..
17                  2006). In Korean Pb workers, mean baseline tibia Pb level was only twofold higher (35.0
18                  Mg/g) (Weaver et al.. 2003a) which illustrates the substantial body burden in middle- and
19                  older-aged Americans from lifetime Pb exposure. Declines in blood Pb levels in Sweden
20                  have been reported and attributed to the leaded gasoline phase-out (Stromberg et al..
21                  1995: Blinder etal.. 1986). although blood Pb levels were lower than those noted during
22                  the U.S. phase-out. Finally, the severe degree of CKD among subjects in Evans et al.
23                  (2010) creates a survivor bias at enrollment and limits the eGFR decline possible during
24                  follow-up, thus limiting the ability to identify factors that influence that decline.


                    ESRD  Patient Studies

25                  End stage renal disease (ESRD) is a well-established public health concern, and is
26                  characterized by the use of dialysis to perform the normal functions of the kidney.
27                  Incidence and prevalence in the U.S. continue to increase resulting in rates that are the
28                  third highest among nations reporting such data (U.S.  Renal Data System. 2009). Studies
29                  in patients with CKD requiring chronic hemodialysis have also been published in the past
30                  five years. A study of 271  adult patients on regular thrice weekly dialysis reported much
31                  higher blood Pb levels than had been appreciated by the treating clinicians (Davenport et
32                  al.. 2009). Blood Pb levels ranged from 3 to 36.9 ug/dL; 25.5% had levels >20 ug/dL,
33                  59% had values of 10-20 ug/dL, and 15.5% were <10  ug/dL. Few details on the statistical
34                  analysis were provided which complicates interpretation of the findings. However, blood
35                  Pb was positively correlated with hemodialysis vintage (months on dialysis; Spearman
      November 2012                             5-388                  Draft - Do Not Cite or Quote

-------
 1                   r = 0.38, p <0.001); negatively correlated with urine output (r = -0.44, p <0.001) and
 2                   higher in patients using single carbon filter and reverse osmosis water purification
 3                   devices. Another recent publication reported higher Pb in dialysate than in the tap water
 4                   used in its preparation (Chen et al.. 2009a). A systematic review of a wide range of trace
 5                   elements in hemodialysis patients reported higher Pb levels in patients compared to
 6                   controls although the difference was not large (Tonelli et al.. 2009). These data suggest
 7                   that blood Pb monitoring in dialysis patients may be useful.

 8                   Interpretation of blood and bone Pb in patients on dialysis is challenging for several
 9                   reasons. First, renal osteodystrophy, the bone disease related to kidney disease, may
10                   result  in increased release of Pb from bone stores. Thus, interpretation of blood and even
11                   bone Pb levels may require adjustment with one or more of a range of osteoporosis
12                   variables. Secondly, as observed above (Davenport et al.. 2009). residual kidney function
13                   may have a substantial impact on blood Pb levels in populations with such minimal
14                   excretion.  Third, as illustrated in the studies cited above (Chen et al.. 2009a: Davenport et
15                   al.. 2009), water and concentrates used in dialysis may be variable sources of Pb. A
16                   recent study reported decreased blood Pb in post-dialysis compared to pre-dialysis
17                   samples (Kazi et al.. 2008). Thus, substantial fluctuations in blood Pb are possible while
18                   on dialysis. Finally, anemia is common in CKD and Pb is stored in red blood cells. Thus,
19                   measurement of blood Pb in anemia may require adjustment for hemoglobin; no
20                   standardized approach to this currently exists.

21                   Given these caveats, a small cross-sectional pilot study observed higher median blood Pb
22                   levels in 55 African-American dialysis patients compared to 53 age- and sex-matched
23                   controls (6 and 3 ug/dL respectively; p <0.001) (Muntner et al.. 2007). However, median
24                   tibia Pb was higher in ESRD patients although the difference did not reach statistical
25                   significance (17 and 13 ug/g bone mineral, respectively [p = 0.13]). Further, the authors
26                   note the limitation related to the sample size based on too few cases needed to achieve
27                   statistical significance from power calculations.

28                   In order to determine the potential impact of renal osteodystrophy, median blood and
29                   tibia Pb levels in the dialysis patients were compared by levels of serum parathyroid
30                   hormone, calcium, phosphorus, and albumin and were not found to be significantly
31                   different (Ghosh-Narang et al.. 2007). A study of 211 diabetic patients on hemodialysis
32                   (Lin et al.. 2008) found parathyroid hormone and serum creatinine to be associated with
33                   blood Pb level  in crude but not adjusted associations. In contrast, a study of 315 patients
34                   on chronic peritoneal dialysis observed parathyroid hormone to be positively correlated
35                   and residual renal function to be negatively correlated with logarithmic-transformed
36                   blood Pb levels after adjustment (Lin et al.. 2010). In the prospective portion of this
37                   study, blood Pb levels at baseline were categorized by tertile (range of 0.1 to 29.9 ug/dL
      November 2012                             5-389                  Draft - Do Not Cite or Quote

-------
 1                  with cut points of 5.62 and 8.66 ug/dL). Cox multivariate analysis, after adjustment for
 2                  parathyroid hormone level, residual renal function, and 20 other variables, showed
 3                  increased all-cause mortality in the middle (5.62-8.66 ug/dL) and highest (>8.66 ug/dL)
 4                  compared to the lowest (<5.62 ug/dL) tertiles after 18 months of follow-up (hazard ratio=
 5                  2.1 [95% CI: 2.0, 2.2] and 3.3 [95% CI: 1.3, 13.5], respectively). A recent publication of
 6                  an 18-month follow-up of 927 patients on maintenance hemodialysis also reported
 7                  increased hazard ratios for all-cause (4.7 [95% CI: 1.9,  11.5]), cardiovascular-cause (9.7
 8                  [95% CI: 2.1, 23.3]), and infection-cause (5.4 [95% CI: 1.4, 20.8]) 18-month mortality in
 9                  the highest (>12.64 ug/dL) compared to the lowest tertile (<8.51 ug/dL) of baseline blood
10                  Pb level, after adjustment for sex, urban residence, hemodialysis vintage, hemoglobin,
11                  serum albumin, and ferritin (Lin et al.. 2011). Given other recent publications in
12                  hemodialysis patients by this group, it would be valuable to examine these risks after
13                  adjustment for hemoglobin A1C (Lin-Tan et al.. 2007a). and blood Cd (Yen et al.. 2011;
14                  Hsu et al.. 2009a).


                    Clinical Trials in Chronic Kidney  Disease Patients

15                  Randomized chelation trials in CKD patients, uncommon in nephrotoxicant research,
16                  provide unique information on the impact of Pb on the kidney. These studies have been
17                  performed by Lin and colleagues at the Chang Gung Memorial Hospital in Taipei,
18                  Taiwan and involve similar study designs. Initially, patients were observed to compare
19                  CKD progression prior to chelation. Then, CKD patients whose diagnostic EDTA
20                  chelatable Pb levels were within certain ranges (generally 60-600 |og/72 hours and thus
21                  below the level commonly considered for chelation) were randomized. The treated group
22                  received weekly chelation with 1 g EDTA intravenously for up to 3 months. The control
23                  group received placebo infusions. In the follow-up period, chelation was repeated  for
24                  defined indications such as increased serum creatinine or chelatable Pb levels above
25                  specified cut-offs. Placebo infusions were repeated in the controls as well. The results of
26                  the most recent of these trials are summarized in Table  5-27 below.
      November 2012                            5-390                  Draft - Do Not Cite or Quote

-------
Table 5-27 Clinical randomized chelation trials in chronic kidney disease
patients.
Baseline
mean(SD)
blood Pb
Reference Group n (ug/dL)
Lin et al. Chelated 32 6.1 (2.5)
(2003)
Control 32 5.9 (3.0)
Linetal. Chelated 15 7.5(4.6)
(2006b)
Control 15 5.9(2.2)
Linetal. Chelated 16 2.6(1.0)a
(2006a)
Control 16 3.0(1.1)
Lin-Tan et Chelated 58 5.0 (2.2)
al. (2007b)
Control 58 5.1 (2.6)
Baseline mean(SD)
chelatable Pb (ng/72 hr)
150.9(62.4)
144.5(87.9)
148.0(88.6)
131.4(77.4)
43.1 (13.7)
47.1 (15.8)
164.1 (111.1)
151.5(92.6)
aNotably, mean blood Pb level in this study was below that observed in a
olds in Baltimore, MD (Martin et al.. 2006).
Baseline Months
mean(SD) of
eGFR treatment
(mL/min / follow-
/1.73m2) up
32.0(12.1) 27
31.5(9.0)
22.4(4.4) 15
26.3 (6.2)
41.2(11.2) 27
42.6 (9.7)
36.8(12.7) 51
36.0(11.2)
recent large general population
Change
in eGFR
peryr
(mL/min
/1.73m2)
+ 1.07
-2.7
-3.5
-10.6
+3.0
-2.0
-0.3
-2.9
Comments


Subjects
with Type II
diabetes
and
nephropathy

Lowest Pb
exposed
and treated
range
Body Pb
Burden (72
h urinary Pb
excretion)
> 20- <80
^9

Subjects
without
diabetes

study of 50- to 70-year
 1
 2
 3
 4
 5
 6
 7
10
11
12
13
14
15
The unique body of work in patient populations by Lin and co-workers, both
observational and experimental, has numerous strengths including prospective study
design, randomization, Pb assessment that includes estimates of the bioavailable dose,
longitudinal statistical analysis, and control for multiple kidney risk factors. However, the
generalizability of the results to broader populations is unknown. In addition, the
association observed between Pb dose and decline in GFR has been variable; the annual
decline in eGFR per standard deviation higher Pb dose at baseline was much lower in the
2003 study than in subsequent publications (Table 5-27 above). Small sample sizes and
differences in renal diagnoses between groups may be factors in this variability.

The studies presented in Table 5-26 and Table 5-27 have a number of potential
limitations. These include small sample  size and lack of blinding and placebo control
except for Lin et al. (2003). which attempted to address this potential limitation. Another
possible limitation may be the shorter follow-up time in some of the studies. The use of
creatinine clearance to assess changes in renal function may limit interpretation of results
as discussed in Section 5.5.1.1. Also, the effects observed following chelation therapy
      November 2012
                           5-391
Draft - Do Not Cite or Quote

-------
 1                  may result from removal of other ions such as Zn, Cu, and Fe. In addition, changes in
 2                  kidney function after treatment with Pb chelating agents may be by mechanisms other
 3                  than reduction in Pb body burden. Chelating agents have been shown to act as
 4                  antioxidants. DMSA abolished reactive oxygen species formation (i.e., MDA and
 5                  nitrotyrosine in interlobular arteries) and was protective against nonPb-induced
 6                  nephrosclerosis in rats (Gonick et al.. 1996). EDTA administration enhanced endothelial
 7                  NO production and reduced kidney damage in a rat model of ischemia-induced acute
 8                  renal failure (Foglieni et al.. 2006). Improved renal function following administration of
 9                  chelating agents have been reported in rodent models of Pb-induced nephrotoxicity
10                  (Sanchez-Fructuoso et al., 2002a; Sanchez-Fructuoso et al., 2002b; Khalil-Manesh et al.,
11                  1992a). Chelation did not appear to improve Pb-induced structural damage (Khalil-
12                  Manesh et al., 1992a); again suggesting that improved hemodynamics may be a result of
13                  reduction in reactive oxidant species, which could be due to reduced Pb level and/or
14                  directly to the chelating agent (Gonick et al.. 1996). Despite these uncertainties and
15                  limitations, the most prudent explanation for the combination of the observational and
16                  experimental chelation work of Lin and colleagues is that reduced Pb is the underlying
17                  reason for improved kidney function. This study design requires replication in larger
18                  populations at multiple clinical centers to confirm that the change in  renal function may
19                  be due to removal of Pb.


                    Occupational Studies

20                  The vast majority of studies in the literature  on the impact of Pb on the kidney have been
21                  conducted in the occupational setting. In general, study size and extent of statistical
22                  analysis are much more limited than those in general population studies. Publications in
23                  few populations have reported adjusted results in occupationally exposed workers in the
24                  five years since the 2006 Pb AQCD. In a two-year prospective  cohort study, generalized
25                  estimating equations were used to model change in kidney function between each
26                  evaluation in relation to tibia Pb and concurrent change in blood Pb in 537 current and
27                  former Pb workers (Weaver et al., 2009). Tibia Pb was evaluated at the beginning of each
28                  follow-up period (yearly on average)  and Pb biomarker levels were adjusted for baseline
29                  levels and other covariates. In males,  serum creatinine decreased and calculated
30                  creatinine clearance increased over the course of the study; these changes were largest in
31                  participants whose blood Pb declined concurrently or whose tibia Pb was lower at the
32                  beginning of the follow-up interval. In females, decreasing serum creatinine was
33                  associated with declining blood Pb (as in males); however, increasing blood Pb was
34                  associated with a  concurrent increase in serum creatinine. Women (25.9% of the  study
35                  population) were older and more likely to be former Pb workers than were men which
36                  may have been important factors in the effect modification observed by sex.
      November 2012                             5-392                  Draft - Do Not Cite or Quote

-------
 1                  Chia and colleagues observed a significant, positive association between concurrent
 2                  blood Pb and urine NAG in linear regression models after adjustment for age, sex, race,
 3                  exposure duration, ALAD G177C polymorphism and the interaction between ALAD
 4                  genotype and blood Pb (Chiaetal.. 2006). Similar positive associations were observed
 5                  between blood Pb and a wider range of EBE markers in models that adjusted for age, sex,
 6                  race, exposure duration, and the HpyCH4 ALAD polymorphism (Chia et al.. 2005).
 7                  Other studies published in the last 5 years also focused on ALAD polymorphisms but did
 8                  not find effect modification to be in a consistent direction (Gao etal. 2010a: Wang et al..
 9                  2009a: Weaver et al.. 2006; Weaver et al.. 2005b). In adults with the ALAD2 genotype,
10                  Pb has been associated with better and poorer renal function in separate cohorts of Pb
11                  workers.

12                  Two studies of occupationally-exposed adults have performed benchmark dose
13                  calculations for the effect of Pb on the kidney. Both used only EBE markers and found
14                  NAG to be the most sensitive outcome; reported lower confidence limits on the
15                  benchmark doses were 10.1 ug/dL (Sun et al., 2008b). and 25.3 ug/dL (Lin and Tai-yi.
16                  2007).

17                  A number of other publications in the five years since the 2006 Pb AQCD, have reported
18                  significantly worse kidney  outcomes in unadjusted analyses in occupationally-exposed
19                  adults compared to unexposed controls (Onuegbu et al.. 2011; Patil et al.. 2007) and/or
20                  significant correlations between higher levels of Pb biomarkers and worse kidney
21                  function (Alasiaet al.. 2010; Khan et al.. 2008; Garcon et al.. 2007; Lin and Tai-yi. 2007;
22                  Alinovi et al.. 2005). A study of 155 male workers reported significant, positive
23                  correlations between blood and urine Pb and urine NAG and albumin after controlling for
24                  age and job duration (Sun et al.. 2008b). One small study found no significant differences
25                  (Orisakwe et al.. 2007). In a study of 108 Pb workers with mean blood  Pb level of
26                  36.2 (ig/dL, no significant correlations were observed between blood Pb concentration
27                  and GFR, creatinine clearance, uric acid clearance or uric acid excretion fraction
28                  (Karimooy et al.. 2010). However, interpretation of this study is limited by the fact that
29                  "only 30 subjects had a correct 24 hours urine volume" and no methods are described for
30                  kidney outcome measurement or analysis.

31                  Overall, the occupational literature published in the last five years on the kidney impact
32                  of Pb exposure has been more consistent in reporting statistically significant associations
33                  than were data reviewed for the 2006 Pb AQCD. This may reflect increased reliance on
34                  EBE markers as more sensitive outcome measures, publication bias,  or multiple
35                  comparisons due to a greater number of outcomes assessed.
      November 2012                            5-393                  Draft - Do Not Cite or Quote

-------
 1
 2
 3
 4
 5

 6
 7
 8
 9
10
11
12
13
In a study of Korean Pb workers, Weaver et al., (2003a) reported inconsistent results with
higher Pb measures associated with worse renal function in some models and better renal
function in other models. In models of effect modification by age, a pattern emerged in
which higher Pb exposure and dose measures were associated with worse renal function
in older workers and better renal function in younger workers (Weaver et al., 2003a).
A small number of publications that include concentration-response information provides
evidence of Pb-related nephrotoxicity in the occupational setting across the blood Pb
ranges analyzed (Weaver et al., 2003a; Ehrlich et al., 1998). Data in 267 Korean Pb
workers in the oldest age tertile (mean age = 52 years) did not provide evidence of a
threshold for a Pb effect on serum creatinine levels (added variable plot shown in Figure
5-33) (Weaver et al., 2003a). It is important to note the uncertainty regarding whether the
concentration-response information provided in these studies applies to lower blood Pb
levels or to populations with lower current environmental Pb exposures.
                   O)   ...
                   E,   ^
                   CL>
                   •-£=   o
                   CO   ^
                   o
                   E
                        CO
                        O
                        CD
                        O
                             0
                       20
40
60
                                      Adjusted  blood  Pb level (|jg/dL)
      Note: Both the adjusted regression line (straight line) and the line estimated by the smoothing method of the S-PLUS statistical
      software function lowess (line with curves) are displayed. Both have been adjusted for covariates. For ease of interpretation, axes
      have been scaled, so that the plotted residuals are centered on the means, rather than zero.
      Source: Reprinted with permission of the BMJ Publishing Group, Weaver et al. (2003a)
      Figure 5-33      Added variable plot of association between serum creatinine and
                        blood Pb  in 267 Korean Pb workers in the oldest age tertile.
      November 2012
                                                  Draft - Do Not Cite or Quote

-------
 1                    A major challenge in interpretation of the occupational literature is the potential for
 2                    Pb-related hyperfiltration. Hyperfiltration involves an initial increase in glomerular
 3                    hypertension which results in increased GFR. If persistent, the risk for subsequent CKD
 4                    increases. This pattern has been observed in diabetes, hypertension, and obesity (Nenov
 5                    et al.. 2000). As discussed in the 2006 Pb AQCD (U.S. EPA. 2006b). findings consistent
 6                    with hyperfiltration have been observed in occupational populations (Weaver et al..
 7                    2003a; Hsiao et al., 2001; Roels et al.. 1994). a study of adults who were Pb poisoned as
 8                    children (Hu. 1991). and a study in European children (de Burbure et al.. 2006).
 9                    Longitudinal data in Pb-exposed rodents provide evidence of a hyperfiltration pattern of
10                    increased, followed by decreased GFR,  associated with Pb exposure and are critical in
11                    interpretation of the human Pb-kidney literature (Khalil-Manesh et al.. 1992b). Pb could
12                    induce glomerular hypertension resulting in hyperfiltration by  several mechanisms
13                    including increased ROS, changes in eicosanoid levels, and/or an impact on the renin-
14                    angiotensin system (Vaziri. 2008b; Roels et al.. 1994). Whether hyperfiltration
15                    contributes to pathology in humans is unclear; longitudinal studies are needed.

16                    The 2006 Pb AQCD provided several explanations for this inconsistency (U.S. EPA.
17                    2QQ6b) (Chapter 6, pp 99):
                         "Some are unique to the occupational literature, such as smaller sample sizes. In addition,
                         employed workers are typically healthier and younger than the general population—
                         resulting in the healthy worker bias. This is a particular problem as susceptible risk groups
                         are identified. Survivor bias in cross-sectional studies is also a concern, since workers
                         whose renal function has declined are generally removed from exposure, particularly if
                         they are followed in a medical surveillance program.  Few studies have included former
                         workers. Also, statistical analyses have been more limited in occupational studies.
                         Analyses for some outcomes were limited to comparisons between exposed workers and
                         controls whose Pb levels were in the range associated with adverse renal outcomes in
                         environmental work. Use of multiple linear regression has generally involved more limited
                         adjustment for covariates than in most of the environmental studies. Many of these
                         limitations result in bias towards the null, which increases the risk that true associations
                         may not be detected."
18                    Regardless, significant findings could be obscured if opposite direction associations are
19                    present in different segments of the study population and interaction models are not
20                    performed to address this. In the Korean Pb workers (Weaver et al.. 2003a; Weaver etal..
21                    2003b). significant associations in opposite directions were observed only when relevant
22                    effect modifiers such as age or genetic variants in ALAD, VDR, and NOS were included
23                    in the model. This is a valid concern for risk assessment, since the factors involved in
24                    these inverse associations in Pb-exposed workers are not well defined at present.
      November 2012                              5-395                   Draft - Do Not Cite or Quote

-------
                    5.5.2.2     Epidemiology in Children

                    Pb Nephrotoxicity in Children

 1                  Both the 2006 and 1986 Pb AQCDs noted that the degree of kidney pathology observed
 2                  in adult survivors of untreated childhood Pb poisoning in the Queensland, Australia
 3                  epidemic (Inglis et al., 1978) has not been observed in other studies of childhood Pb
 4                  poisoning. Recent publications remain consistent with that conclusion; a recent study
 5                  observed an impact of childhood Pb poisoning on IQ but not kidney outcomes (Coria et
 6                  al., 2009). Chelation was raised as a potential explanation for this discrepancy in the
 7                  2006 Pb AQCD.

 8                  With declining Pb exposure levels, recent work has focused on studies in children with
 9                  much lower blood Pb levels. However, insensitivity of the clinical kidney outcome
10                  (i.e., GFR) measures for early kidney damage is a particular problem in children who do
11                  not have many of the other kidney risk factors that  adults do, such as hypertension and
12                  diabetes. As a result, such  studies have utilized EBE markers. However, data to
13                  determine the predictive value of such biomarkers for subsequent kidney function decline
14                  in Pb exposed populations are extremely limited (Coratelli et al.. 1988) and may pose
15                  particular challenges in children due to puberty-related biomarker changes (Sarasua et al.,
16                  2003). The few studies included the 2006 Pb AQCD (U.S. EPA. 2006b) that analyzed
17                  clinical kidney outcomes in children found associations with indicators of Pb exposure
18                  that were inconsistent in direction. Pels et al. (1998) found no difference in mean serum
19                  creatinine between 62 children living  near Pb-producing factories and 50 control children
20                  living in communities without Pb emission sources. In a study of 200 Belgian adolescents
21                  aged 17 years, higher concurrent blood Pb level was associated with higher serum
22                  cystatin-C (de Burbure et al.. 2006); however, among 300-600 European children (n
23                  varied by outcome), higher concurrent blood Pb level was associated with lower serum
24                  creatinine and cystatin C (Staessen et al.. 2001).

25                  Recent studies of children  with elevated Pb exposure did not consistently indicate that Pb
26                  exposure was associated with reduced kidney function. A study in 123 children of
27                  workers in Pakistani Pb smelters and battery recycling plants and  123 control children,
28                  ages 1-6 years, reported elevated blood Pb levels, serum creatinine and urea in children of
29                  Pb-exposed workers compared to controls (medians: 8.1 versus 6.7 (ig/dL; 56 versus
30                  52 (iM; and  4,500 versus 4,300 (iM, respectively (p < 0.01 for all) in unadjusted analyses
31                  (Khan et al.. 2010a). Blood Pb levels were correlated with serum creatinine (Spearman
32                  r = 0.13; p = 0.05). However, a study  of 77 participants, ages 10-25 years, who were
33                  previously Pb poisoned through contaminated flour and chelated,  reported no difference
34                  in renal effects between children with blood Pb levels >48 (ig/dL and <43 (ig/dL although
      November 2012                            5-396                  Draft - Do Not Cite or Quote

-------
 1                  lower IQ was observed in the subset who were exposed before the age of six years (Coria
 2                  et al.. 2009).

 3                  One of the key gaps identified in the 2006 Pb AQCD (U.S. EPA. 2006b) was limited data
 4                  in children and adolescents particularly with respect to GFR measures and in populations
 5                  without the elevated Pb exposure associated with Pb poisoning, living near a Pb source,
 6                  or having parents with occupational Pb exposures. A recently published NHANES
 7                  analysis in adolescents begins to fill this gap (Fadrowski et al.. 2010). Associations
 8                  between concurrent blood Pb and kidney function were investigated in 769 adolescents
 9                  aged 12-20 years in the U.S. NHANES III, conducted 1988-1994. Kidney function was
10                  assessed with two eGFR equations. One utilized serum cystatin C and the other used the
11                  more traditional marker, serum creatinine. Median concurrent blood Pb and cystatin C-
12                  based eGFR levels were 1.5 ug/dL and 112.9 mL/min/1.73 m2, respectively. Cystatin C-
13                  based eGFR was lower (-6.6 mL/min/1.73 m2 [95% CI: -0.7, -12.6]) in participants with
14                  blood Pb levels in the highest quartile (> 3.0 ug/dL) compared with those in the lowest
15                  (<1 ug/dL). A doubling of blood Pb level was associated with a -2.9 mL/min/1.73 m2
16                  (95% CI: -0.7, -5.0) lower eGFR. In contrast, the association between blood Pb and
17                  creatinine-based eGFR, although in the same direction, was not statistically significant.
18                  As these children were born between 1968 and  1982, some likely had higher Pb
19                  exposures in earlier childhood, although notably, not as high or as long in duration as did
20                  older adults examined in aforementioned studies. Nonetheless, in this study of NHANES
21                  adolescents, there also is uncertainty regarding the magnitude, timing, frequency, and
22                  duration of Pb exposure that contributed to the observed associations. Additional research
23                  in children is warranted, in particular studies with longitudinal follow-up, multiple
24                  outcome assessment methods, and examination of children born after Pb was banned
25                  from gasoline.
                    5.5.2.3      Associations between Pb Dose and New Kidney Outcome
                                 Measures

26                  As noted above, in an effort to more accurately estimate kidney outcomes, new equations
27                  to estimate GFR based on serum creatinine have been developed, and the utility of other
28                  biomarkers, such as cystatin C, as well as equations based on them, are being studied.
29                  However, few publications have utilized these state-of-the-art techniques when
30                  evaluating associations between Pb or Cd dose and renal function. In addition to the
31                  study in NHANES adolescents discussed above (Fadrowski et al.. 2010). a cross-
32                  sectional study of Swedish women reported that higher concurrent blood Pb (median:
33                  2.2  (ig/dL) and Cd (median: 0.38 (ig/L) levels were associated with lower eGFR based on
34                  serum cystatin C alone (without age, sex, and race) after adjustment for socio-
      November 2012                            5-397                  Draft - Do Not Cite or Quote

-------
 1                   demographic and CKD risk factors (Akesson et al., 2005). Associations were comparable
 2                   to those using creatinine clearance as the kidney outcome for Pb; however associations of
 3                   Cd dose measures were stronger for the cystatin C based outcome. Staessen et al. (2001)
 4                   found a statistically significant association between concurrent blood Pb level and serum
 5                   cystatin C in a cross-sectional study of adolescents; creatinine-based measures were not
 6                   reported. However, in a cross-sectional study of 804 European children aged range 8.5 to
 7                   12.3 years, higher concurrent blood Pb levels were associated with lower serum cystatin
 8                   C and creatinine; these inverse associations were attributed to hyperfiltration (de Burbure
 9                   et al.. 2006). A recent publication compared associations of blood Pb and eGFR using the
10                   traditional MDRD equation to those with four new equations: CKD-EPI, and cystatin C
11                   single variable, multivariable, and combined creatinine/cystatin C, in 3,941 adults who
12                   participated in the 1999-2002 NHANES cystatin C subsample (Spector et al.. 2011).
13                   Similar to the  NHANES adolescent analysis, associations with the cystatin C outcomes
14                   were stronger. After multivariable adjustment, differences in  mean eGFR for a doubling
15                   blood Pb were -1.9 (95% CI:  -3.2, -0.7), -1.7 (95% CI: -3.0, -0.5), and -1.4 (95% CI: -2.3,
16                   -0.5) mL/min/1.73 m2, using the cystatin C single variable, multivariable and combined
17                   creatinine/cystatin C equations, respectively, reflecting lower eGFR with increased blood
18                   Pb. The corresponding differences were -0.9 (95% CI: -1.9, 0.02) and -0.9 (95% CI: -1.8,
19                   0.01) using the creatinine-based CKD-EPI and MDRD equations, respectively.
                     5.5.2.4      Reverse Causality

20                   The reverse causality hypothesis suggests that the associations between blood Pb and
21                   kidney function may be due to reduced excretion of Pb rather than a causal association
22                   between Pb exposure and this outcome. Cross-sectional studies of populations that
23                   include participants with CKD frequently note the potential for their findings to be
24                   explained by reverse causality as a limitation of the study (Muntner et al.. 2003).  There
25                   are several techniques that can be used to assess the potential for reverse causality to
26                   underlie associations between higher Pb dose and worse kidney function. Prospective
27                   studies in which associations between baseline measurements of Pb biomarkers and
28                   subsequent changes in renal function are demonstrated provide the strongest evidence to
29                   evaluate the possibility of reverse causality. In the NAS, baseline blood Pb levels were
30                   associated with subsequent declines in renal function over follow-up periods ranging
31                   from 3 to 6 years (Tsaih et al.. 2004; Kim et al..  1996). Prospective data in CKD patients
32                   also revealed an association between baseline Pb dose and decline in eGFR over  follow-
33                   up periods as long as four years (Yu et al.. 2004). Another approach involves sensitivity
34                   analyses in which associations are explored in participants with normal glomerular
35                   filtration. This approach has been used in the NAS with plots which revealed that the
      November 2012                             5-398                 Draft - Do Not Cite or Quote

-------
 1                  association between blood Pb level and serum creatinine was present across the entire
 2                  range of serum creatinine levels, including those in the normal range where excretion is
 3                  not impaired (Tsaih et al., 2004; Kim et al., 1996). Analyses restricted to the population
 4                  with serum creatinine below 1.5 mg/dL were conducted in a later publication; the authors
 5                  reported that associations were consistent (Tsaih et al., 2004). The use of a serum
 6                  creatinine rather than an eGFR cut-off is a limitation since there can be substantial
 7                  decrements in renal function with 'normal' serum creatinine. The associations observed
 8                  in both NAS studies were not limited to the segment of the population with potentially
 9                  clinically significant renal dysfunction in whom reduced Pb excretion would be more
10                  likely.

11                  Research in which EDTA chelation decreases body Pb burden and improves kidney
12                  function also provides evidence informing the possibility of reverse causality but is not
13                  without limitation because EDTA may independently improve kidney function (Lin et al..
14                  2006b; Lin et al.. 2006a; Lin et al..  2003). Additional evidence evaluating  reverse
15                  causality was provided by findings  among 153 adults with chronic kidney disease, in
16                  which renal failure was not associated with increases in blood or bone Pb levels or
17                  chelatable Pb levels (Van de Vvver et al.. 1988). Batuman et al.  (1983) found that
18                  chelatable Pb levels were similar in 27 adults with renal disease  of unknown and known
19                  non Pb-related causes where bone Pb levels (group means: 18 and 19 (ig/g) are in the
20                  range of those measured in recent epidemiologic studies. A pilot study of 55 ESRD cases
21                  at Tulane clinics (Muntner et al., 2007) reported the median blood Pb level was
22                  significantly higher among the ESRD cases compared to their control counterparts. For
23                  ESRD patients the distribution of blood Pb was shown by pre-defined levels: 18.5% less
24                  than 5 ng/dl; 66.7% of ESRD cases had blood Pb levels between 5 and 9.9 (ig/dL; and
25                  14.8% equal or greater 10 (ig/dL.

26                  The reverse causality hypothesis can be broadened to include a physiologic process in
27                  which Pb levels are influenced over the entire range of kidney filtration rates. If this
28                  occurs, even normal kidney function would impact blood Pb levels such that higher GFR
29                  would result in greater  Pb excretion and lower blood Pb levels. Serum creatinine levels
30                  are influenced in this way over the  entire range of kidney function; as a result, these
31                  levels are used to estimate kidney function. However, creatinine is produced and excreted
32                  at a steady state in the body which is one reason it was selected as a biomarker to assess
33                  kidney function. This expanded hypothesis implies that low blood and bone Pb levels
34                  may reflect kidney function in addition to exposure. If so, this would increase
35                  misclassification bias, with Pb biomarkers reflecting both exposure and kidney function.
36                  Given the longstanding use of blood Pb as a dose marker in research for many non-
37                  kidney outcomes this seems unlikely. Thus, published research has not directly addressed
38                  this. One such approach involves comparing associations of blood and urine Pb in models
      November 2012                             5-399                  Draft - Do Not Cite or Quote

-------
 1
 2

 3
 4
 5
 6
 7
 8
 9
10
11
of kidney function. If they are consistent, this hypothesis is not valid. However, urine Pb
is rarely used and may not be as reliable a biomarker as blood Pb (Gulson et al.. 1998c).

In summary, several lines of evidence support that reverse causality does not contribute
substantially to associations between higher blood Pb levels and worse kidney function.
These lines of evidence include prospective data observing that baseline Pb measures are
associated with subsequent declines in renal function, that associations in prospective
studies persist among adults with normal renal function, that renal failure does not
increase Pb biomarker levels and that reduction of Pb levels by chelation improves
kidney function. However, this bidirectional relationship is still possible and additional
research is needed to fully exclude the hypothesis. In particular prospective data are
required as is research  to determine if normal kidney function influences blood Pb levels.
12
13
14
15
16
17
5.5.2.5      Toxicology

In animals, Pb has been found to induce changes in a wide range of indicators of renal
function. Most studies examined Pb exposure concentrations that resulted in higher blood
Pb levels (>20 (ig/dL) than those in the current U.S. general population. While
toxicological information on renal dysfunction with blood Pb levels <10 (ig/dL generally
is not available, dysfunction in kidney function measures, including urinary flow, ALP,
microalbumin, and NAG, was observed at blood Pb levels above 20 (ig/dL.
      Table 5-28   Animal toxicological studies reporting the effects of Pb exposure
                    (as blood Pb level) on kidney function.



Reference
Berrahal et
al. (2011)




Masso-
Gonzalez
etal.
(2009)
Roncal et
al. (2007)



Species;
Lifestage;
Sex
Rat; Adult





Rat;
Weanling
pups

Rat; Adult;
Male



Pb Dose;
Exposure
Duration
50 ppm
Pb acetate in
drinking water;
lactation day 1
to PND40or
PND65

300 ppm
Pb acetate in
drinking water;
GD1 to PND21
150 ppm
Pb acetate in
drinking water;
16 weeks with
remnant kidney
surgery at week
4

Blood Pb
Level with
Response
(ug/dL)
PND40: 12.7
PND65: 7.5




23


26






Responses
Oxidative stress - Increased renal MDA (PND40 and PND65),
decreased renal SOD activity (PND40)
Morphology- Increased relative kidney weight (PND65)
Kidney function - Increased blood creatinine (PND40 and
PND65), increased BUN (PND40), decreased uric acid
(PND65), increased kidney proteins (PND65).
Oxidative stress - Elevated TEARS and catalase activity
Morphology- Elevated relative kidney weight at PND21

Inflammation - Increased the number of macrophages & renal
MCP-1 mRNA.
Morphology - Pb induced pre-glomerular vascular disease of
kidney (i.e., sclerosis, fibrosis, peritubular capillary loss)
Kidney function - Decreased creatinine clearance, increased
serum creatinine, increased BUN, and increased serum uric
acid.
     November 2012
                           5-400
Draft - Do Not Cite or Quote

-------

Reference
Khalil-
Manesh et
al. (|993a)


Khalil-
Manesh et
al. (1992a)





Vyskocil et
al. (1995)
Ademuyiwa
etal.
(2009)
Navarro-
Moreno et
al. (2009)

Khalil-
Manesh et
al. (1992b)
Wang et al.
(201 Od)




Species;
Lifestage;
Sex
Rat; Adult;
Male



Rat; Adult;
Male





Rat, Adult,
Female
Rat; Adult
Rat; Adult;
Male

Rat; Adult;
Male
Rat; Adult;
Female




Pb Dose;
Exposure
Duration
100 ppm
Pb acetate in
drinking water;
12 months


5,000 ppm
Pb acetate in
drinking water;
1, 6, or
9 months





1,000 ppm
Pb acetate in
drinking water; 2
or 4 months
200, 300, and
400 ppm
Pb acetate in
drinking water;
12 weeks
500 ppm
Pb acetate in
drinking water;
28 weeks

5,000 ppm
Pb acetate in
drinking water;
12 months
300 ppm
Pb acetate in
drinking water;
8 weeks



Blood Pb
Level with
Response
(ug/dL)
Mean at
3 months
29.4
Mean at
12 months
22
Range 9-34
At 1 month
7.9
At 6 months
-30
At 9 months
52




37.6
200 ppm: 41
300 ppm: 61
400 ppm: 39
43

Max 125
Mean 45
20 (serum)




Responses
Morphology - Mild tubular atrophy and interstitial fibrosis seen
at 12 months, otherwise normal.
Kidney function -Increased GFR at 1 and 3 months, increased
NAG, and no change in GST.


At 1 month
Kidney function - no functional or pathological changes
At 6 months
Morphology - Prominent tubulointerstitial fibrosis and
segmental sclerosis. Increased kidney weight
Kidney function - Decreased GFR, increased serum creatinine
and SUN, no change urinary NAG or GST
At 9 months
Morphology - Severe tubulointerstitial disease
Kidney function - Decreased GFR, increased serum creatinine
and SUN
Kidney function - No change in kidney function or
nephrotoxicity
Kidney function - Renal phospholipidosis and depletion of
renal cholesterol.
Oxidative stress - Increased kidney lipid peroxidation
(i.e., TEARS)
Morphology - Electron micrography showed lumen reduction,
microvilli loss, brush border loss, and mitochondrial damage
Kidney function - Elevated urinary pH and protein, and glucose
and blood in the urine.
Morphology - Tubular atrophy and interstitial fibrosis after
6 months. Increased urinary brush border antigens.
Kidney function - Hyperfiltration at 3 months and decreased
GFR at 12 months. After 3 months, elevated urinary NAG and
GST.
Biomarker- Aberrant NAG, GGT, p2-microgobulin expression
Oxidative Stress- Increased lipid peroxidation (i.e., MDA
production), elevated kidney antioxidant enzymes (SOD, GPx,
CAT), and depleted GSH
Morphology - Electron micrography showed Pb damages
mitochondria, basement membrane, and brush border in
kidney tissue. Some focal tubal necrosis observed.
Kidney function - Elevated urinary total protein, urinary
albumin, and serum urea nitrogen.
1
2
o
J
4
5
Renal Function and Interstitial Fibrosis

Past studies have shown that chronic continuous or repeated Pb exposure can result in
interstitial nephritis and focal or tubular atrophy. A series of studies on Pb exposure in
rats (longitudinal 12-month exposure study to either 100 ppm or 5,000 ppm Pb in
drinking water) report an initially elevated GFR, consistent with hyperfiltration, and renal
hypertrophy (Khalil-Manesh et al.. 1993a: Khalil-Manesh et al.. 1992b: Khalil-Manesh et
     November 2012
                           5-401
Draft - Do Not Cite or Quote

-------
 1                  al.. 1992a). After 6 months of exposure, GFR decreased, albuminuria was present, and
 2                  pathology ensued with focal tubular atrophy and interstitial fibrosis formation. This
 3                  pathology and functional decrement was persistent out to 12 months, and at 12 months
 4                  glomeruli developed focal and segmental sclerosis.

 5                  The toxicological evidence for differences in GFR according to duration of Pb exposure
 6                  (i.e., hyperfiltration with 3-month exposure versus decreased GFR with 6- or 12-month
 7                  exposure) provides biological plausibility for epidemiological studies that observed a
 8                  similar phenomenon by age in adults in association with Pb biomarker levels. However,
 9                  the toxicological studies report blood Pb levels at  some exposure durations much higher
10                  than relevant to blood Pb levels in current human populations. Still, these duration-
11                  dependent dichotomous  changes in  GFR are consistent between the toxicological and
12                  epidemiologic literature.

13                  At exposure concentrations resulting in blood Pb levels within one order of magnitude of
14                  the upper range of current human population blood Pb level, animal toxicological studies
15                  present inconsistent results for the effects of Pb on kidney function. There are studies that
16                  have corroborated the previously observed increase in serum creatinine following Pb
17                  exposure in rats. Berrahal et al. (2011) reported on the effects of age-dependent exposure
18                  to Pb on nephrotoxicity in male rats (Table 5-29).  Pups  were exposed to Pb lactationally
19                  (as a result of dams consuming water containing 50 ppm Pb acetate) until weaning.
20                  Thereafter, the offspring were exposed to the same solution from weaning (day 21) until
21                  sacrifice. Male pups were sacrificed at age 40 days (puberty; blood Pb level 12.7 ug/dL)
22                  and at age 65 days  (post-puberty; blood Pb level 7.5 ug/dL). Serum creatinine was
23                  elevated at both 40 days and 65 days (0.54 and 0.60 mg/dL compared to control values of
24                  0.45 mg/dL [p <0.001]). Various parameters of Pb-induced renal dysfunction are listed in
25                  Table 5-29 below. The elevated serum creatinine in the  Pb-exposed animals compared to
26                  controls suggests that animals exposed to low dose (i.e., 50 ppm) Pb from birth may
27                  develop renal abnormalities. However, the lack of measurements of GFR or renal
28                  pathology weakens the conclusions.
      November 2012                            5-402                  Draft - Do Not Cite or Quote

-------
      Table 5-29  Indicators of renal damage in male rats exposed to 50 ppm  Pb for 40
                    and 65 days, starting at parturition.
Biomarker
(Mean ± SD)
Blood Pb level (|jg/dL)
Plasma Creatinine (mg/L)
Plasma Urea (mg/L)
Plasma Uric Acid (mg/L)
PND40 Control
1.8 ±0.33
4.5 ±0.21
0.37 ±0.01 9
7.51 ± 0.44
PND40 Pb
12.7± 1.7
5.35 ± 0.25a
0.47 ± 0.021 a
7.65 ± 0.32
PND65 Control
2.1 ±n0.35
4.55n± 0.27
0.29n± 0.009
9.39n± 0.82
PND65 Pb
7.5n± 0.78
6.04 ± 0.29a
0.29n± 0.009
5.91n±0.53a
      ap <0.001
      Source: Modified with permission of John Wiley & Sons, Berrahal et al. (2011)

 1                  Roncal et al. (2007) found that Pb accelerated renal function decrements,
 2                  tubulointerstitial injury, and arteriolopathy in non-Pb-related CKD. Sprague-Dawley rats
 3                  were administered Pb acetate at 150 ppm for 4 weeks, then subjected to remnant kidney
 4                  surgery (left kidney mass reduced by 2/3 and right kidney removed), and subsequently
 5                  exposed to Pb for an additional 12 weeks resulting in a blood Pb level of 26 (ig/dL.
 6                  Pb-treated rats had higher systolic BP, increased serum creatinine, lower creatinine
 7                  clearance, and higher proteinuria than did controls. Most striking was development of
 8                  worse  arteriolar disease, peritubular capillary loss, tubulointerstitial damage, and
 9                  macrophage infiltration. Pb treatment was associated with significant worsening of pre-
10                  glomerular vascular disease, as characterized by an increase in the media-to-lumen ratio.
11                  There was also a higher percentage of segmental sclerosis within glomeruli and a
12                  tendency for a higher number of sclerotic glomeruli. Additionally, a loss of peritubular
13                  capillaries, as reflected by a reduction in thrombomodulin staining, was observed. This
14                  was associated with worse tubular injury (osteopontin staining) due to more interstitial
15                  fibrosis (type III collagen staining)  and a greater macrophage infiltration in the
16                  interstitium. The increase in macrophages was associated with higher renal MCP-1
17                  mRNA. As a whole, these findings indicate that Pb exposure concomitant with existing
18                  renal insufficiency due to surgical kidney resection accelerated vascular disease and
19                  glomerular pathology. These findings are consistent with the previous work of Bagchi
20                  and Preuss (2005) also showing that Pb-exposed animals with non-Pb-related CKD
21                  (remnant surgery) had kidney dysfunction including impairment of the renin-angiotensin
22                  system (Losartan challenge), elevated systolic BP, and alterations in renal excretion of
23                  Pb, K+, and Na+. Thus, this model shows that Pb exposure may exacerbate pre-existing
24                  underlying kidney disease.

25                  Other investigators have shown that chronic Pb exposure has detrimental effects on renal
26                  function at higher blood Pb levels. A number of studies report increased serum creatinine
27                  following high level Pb exposure (e.g., blood Pb levels >55.6 (ig/dL) (Abdel Moneim et
28                  al..2011b:  Ozsov et al.. 2010: Javakumaretal.. 2009: Kharoubi et al.. 2008a). In
      November 2012                            5-403                  Draft - Do Not Cite or Quote

-------
 1                  addition, studies reporting high blood Pb levels and high Pb exposure levels report
 2                  increased urine, serum, or blood urea nitrogen (Wang et al.. 2010d: El-Nekeetv et al..
 3                  2009; Javakumar etal..2009; Kharoubi et al., 2008a). Jayakumar et al. (2009) reported
 4                  alterations in other markers of kidney toxicity, lysosomal marker and brush border
 5                  enzymes (i.e., ALP, ACP, y-GT, NAG, (3-D-glucuronidase), following Pb exposure
 6                  (2,000 ppm for 6 weeks). Similarly, Wang et al. (2010d) reported time-related increases
 7                  in urinary alkaline phosphatase, urinary GGT, urinary NAG, urinary total protein, urinary
 8                  (3-2 microglobulin, and urinary microalbumin following Pb exposure (300 ppm in
 9                  drinking water, serum Pb level 20 (ig/dL). Pb-exposed male rats (500 ppm Pb acetate in
10                  drinking water for 7 months, blood Pb level 43 (ig/dL) had elevated urinary pH, urinary
11                  glucose, and proteinuria (Navarre-Moreno et al.. 2009).

12                  Qiao et al. (2006) measured the effect of Pb on the expression of the renal nuclear factor-
13                  kappa B (NF-KB), transforming growth factor (TGF-(3) and fibronectin in Sprague-
14                  Dawley rat kidney. These growth (TGF-|3) and transcription (NF-KB) factors modulate
15                  the progression of renal function decrements through promotion of extracellular matrix
16                  (fibronectin) synthesis and promotion of fibrosis. Pb was administered at a dose of
17                  5,000 ppm Pb acetate, continuously for either one, two, or three months. All factors
18                  increased by the end of three months of treatment, but only NF-KB increased
19                  progressively at each time period. These changes were hypothetically related to the
20                  development of Pb-induced renal fibrosis in rats, but no histology was performed.

21                  The renal effects of chronic Pb exposure as detailed above were partially rescued in rats
22                  following lowering of blood Pb level with chelation therapy (i.e., DMSA) (Khalil-
23                  Manesh et al..  1992a) and after treatment with antioxidants (Abdel Moneim et al.. 201 Ib;
24                  Ozsovetal. 2010; Wang etal.. 2010d; El-Nekeetv et al.. 2009; Javakumar et al.. 2009;
25                  Kharoubi et al.. 2008a). DMSA treatment improved renal function; however, Pb-induced
26                  pathology remained (Khalil-Manesh et al., 1992a). Improvements include increased GFR,
27                  decreased albuminuria, and decreased inclusion body numbers but little change in
28                  tubulointerstitial scarring. DMSA also acts as an antioxidant, so the protective effects
29                  may not be entirely attributed to the lowering of blood Pb level. Similarly, several studies
30                  found that treatment with antioxidant compounds could protect against Pb-induced
31                  kidney dysfunction. Administration of flaxseed oil, L-carnitine, NAC, and several
32                  medicinal plants including, Achyranthes aspera, Artemisia absinthium, and Aquilegia
33                  vulgaris, to Pb-exposed rodents protected against injury to the kidney or restored kidney
34                  function (Abdel Moneim et al., 201 Ib;  Ozsoy etal., 2010; Wang  etal., 2010d; E\-
35                  Nekeetv et al.. 2009;  Javakumar et al.. 2009; Kharoubi et al.. 2008a). These studies
36                  suggest that a reduction in reactive oxygen species may attenuate the effects of Pb on
37                  kidney function implicating oxidative stress as a predominant mechanism for Pb-induced
38                  reduced kidney function.
      November 2012                            5-404                 Draft - Do Not Cite or Quote

-------
                     Histological Changes

 1                   Earlier studies discussed in previous Pb AQCDs have identified Pb-related renal damage
 2                   by the presence of dense intranuclear inclusion bodies, which are capable of sequestering
 3                   Pb (Goveretal.. 1970b). Pb-induced formation of intranuclear inclusion bodies in the
 4                   proximal tubule (PT) is considered protective; Pb is sequestered such that it is not in its
 5                   bioavailable, free, lexicologically active form. Intranuclear inclusion bodies are found in
 6                   the kidney with short-term (i.e., <4 weeks) Pb exposure but present to a lesser degree
 7                   with chronic exposures (See Section 5.2.3 for further discussion). Chelators such as
 8                   CaNa2EDTA have removed these inclusion bodies from affected nuclei (Goyer et al..
 9                   1978).

10                   Multiple ultrastructural changes indicate dysfunction in the PT and nephropathy after Pb
11                   exposure, including changes to the PT epithelium, endoplasmic reticulum dilation,
12                   nuclear membrane blebbing, and autophagosome enlargement (Fowler et al.. 1980; Goyer
13                   et al.. 1970a). Indications similar to the PT transport-associated Fanconi syndrome appear
14                   with Pb exposure, albeit often at high doses of Pb, i.e., Pb poisoning. These indications,
15                   which include increased urinary electrolyte excretion (Zn), decreased Na+/K+ATPase
16                   activity, mitochondrial aberrations, and aminoaciduria, also have been associated with
17                   blood Pb levels in children.

18                   Recent studies since the 2006 Pb AQCD are consistent with the earlier findings and build
19                   upon the literature base by including the role of antioxidants. Jabeen et al. (2010) exposed
20                   pregnant albino BALB/c mice to a daily oral dose of Pb acetate (10 mg/kg body weight,
21                   daily throughout pregnancy) until GDI8, at which point the fetal kidneys were processed
22                   for histological examination. Histology revealed Pb exposure induced decreased kidney
23                   cortical thickness, decreased diameter of renal corpuscles, and increased renal tubular
24                   atrophy (with desquamated epithelium and degenerated nuclei in the distal and proximal
25                   tubules). Blood Pb levels were not reported in this study. Nonetheless, these data show
26                   that in utero Pb exposure had significant histological effects on  the fetal kidney, which
27                   could contribute to altered renal function including clearance of waste products,
28                   electrolyte balance, and vasoregulation.

29                   Massanyi et al. (2007) reported on Pb-induced alterations in male Wistar rat kidneys after
30                   single i.p. doses of Pb acetate (50, 25, and 12.5 mg/kg); kidneys were removed and
31                   analyzed 48 hours after Pb administration. Qualitative microscopic analysis detected
32                   dilated Bowman's capsules and dilated blood vessels in the interstitium with evident
33                   hemorrhagic alterations. Quantitative histomorphometric analysis revealed increased
34                   relative volume of interstitium and increased relative volume of tubules in the
35                   experimental groups. The diameter of renal corpuscles and the diameter of glomeruli and
36                   Bowman's capsule were significantly increased. Measurement of tubular diameter
      November 2012                             5-405                  Draft - Do Not Cite or Quote

-------
 1                   showed dilatation of the tubule with a significant decrease of the height of tubular
 2                   epithelium compatible with degenerative renal alterations. These findings extend the
 3                   observations of Fowler et al. (1980) and Khalil-Manesh et al. (1992b; 1992a); in
 4                   particular, the enlarged glomeruli are consistent with the early hyperfiltration caused by
 5                   Pb.

 6                   Abdel Moneim et al. (20 lib) reported histological evidence of inflammation after Pb
 7                   treatment in rats (i.p. 20 ppm, 5 days). This evidence included increased inflammatory
 8                   cellular infiltrations, cytoplasmic vacuolation, and dilatation of some kidney tubules.
 9                   Inflammation was accompanied by an increase in apoptotic cells and increased oxidative
10                   stress.

11                   A recent study has also reported  inclusion body formation in the nuclei, cytoplasm, and
12                   mitochondria of PT cells of Pb-treated rats (50 mg Pb/kg bw i.p., every 48 hours for
13                   14 days) (Navarre-Moreno et al.. 2009). These inclusion bodies were not observed in
14                   chronically Pb-exposed rats (500 ppm Pb in drinking water, 7 months). However, chronic
15                   Pb exposure resulted in morphological alterations including loss of PT apical membrane
16                   brush border, collapse and closure of the PT lumen, and formation of abnormal
17                   intercellularjunctions.

18                   Vogetseder et al. (2008) examined the proliferative capacity of the renal PT (particularly
19                   the S3 segment) following i.v. administration of Pb to juvenile and adult male Wistar
20                   rats. Proliferation induction was  examined by detection of Bromo-2'-deoxyuridine
21                   (BrdU), Ki-67 (labels S, G2, and M phase cells), and cyclin Dl (an essential cell cycle
22                   progression protein). The cycling marker Ki-67 revealed a much higher proliferation rate
23                   in the S3 segment in control juvenile rats (4.8 ± 0.3%) compared with control adult rats
24                   (0.4 ± 0.1%). Pb administration (3.8 mg/100 g bw) increased the proportion of Ki-67-
25                   positive cells to 26.1 ± 0.3% in juvenile rats and 31.9±0.3%in adult rats. Thus, the
26                   increased proliferation caused by Pb was age independent. The proliferation induction
27                   caused by Pb administration may be a result of reduced cell cycle inhibition by p27klp"1.
28                   Acute Pb treatment increased the incidence of cyclin D1  labeling in the BrdU-positive
29                   cells suggesting Pb was able to accelerate re-entry of cells into the cell cycle and cause
30                   proliferation  in the PT. Pb-induced cellular proliferation has also been reported in the
31                   retina with gestational and early  postnatal rodent Pb  exposure (Giddabasappa et al..
32                   2011).

33                   Ademuyiwa et al. (2009) examined Pb-induced  phospholipidosis and cholesterogenesis in
34                   rat tissues. Sprague-Dawley rats  were exposed to 200,  300 and 400 ppm Pb acetate for
35                   12 weeks. The Pb exposure resulted in induction of phospholipidosis in kidney tissue,
36                   accompanied by depletion of renal cholesterol. The authors suggested that induction of
37                   cholesterogenesis and phospholipidosis in kidney may be responsible for some of the
      November 2012                             5-406                  Draft - Do Not Cite or Quote

-------
 1                   subtle and insidious cellular effects found with Pb-mediated nephrotoxicity. Drug-
 2                   induced PT phospholipidosis is seen clinically with use of the potentially nephrotoxic
 3                   aminoglycoside drugs, including gentamicin (Baronas et al., 2007).

 4                   Various antioxidants have been shown to attenuate Pb-induced histopathological changes
 5                   to the kidney. Ozsoy et al. (2010) found L-carnitine to be protective in a model of
 6                   experimental Pb toxicity in female rats. Markers of histopathological change in the
 7                   kidney, including tubule dilatation, degeneration, necrosis, and interstitial inflammation
 8                   were rescued by L-carnitine treatment in females. Male rats exposed to Pb (2,000 ppm for
 9                   6 weeks) also displayed tubular damage, whereas concomitant treatment with Pb and an
10                   extract of Achyranthes aspera ameliorated the observed damage (Jayakumar et al.. 2009).
11                   El-Sokkary et al. (2005)  reported Pb-induced (100 ppm s.c. for 30 days) tubular
12                   degeneration with necrotic cells that could be prevented with melatonin treatment.
13                   Melatonin is known to be an efficacious free radical scavenger and indirect antioxidant.
14                   El-Nekeety et al. (2009)  found an extract of the folk medicine plant Aquilegia vulgaris to
15                   be protective against Pb acetate-induced tubular dilatation, vacuolar and cloudy epithelial
16                   cell lining, interstitial inflammatory cell infiltration,  hemorrhage, cellular debris, and
17                   glomerulus hypercellularity. Concomitant exposure to Pb and extract produced histology
18                   indiscernible from that in controls. Post treatment with extract partially rescued the
19                   Pb-induced histopathology. El-Neweshy and El-Sayed (2011) studied the influence of
20                   vitamin C supplementation (20 mg/kg pretreatment every other day) on histopathological
21                   alterations in Pb-exposed male rats (20 mg/kg by intragastric feeding once daily for
22                   60 days).  Control rats showed normal histology, while Pb-treated rats exhibited
23                   karyomegaly with eosinophilic intranuclear inclusion bodies in the epithelial cells of the
24                   proximal tubules. Glomerular damage and tubular necrosis with invading inflammatory
25                   cells were also found. Rats treated with Pb acetate plus  vitamin C exhibited relatively
26                   mild or no karyomegaly with eosinophilic intranuclear inclusion bodies in the proximal
27                   tubules. Normal glomeruli were noted in animals exposed to Pb and vitamin C. These
28                   findings consistently show that some antioxidants are capable of preventing or rescuing
29                   Pb-induced renal histopathological changes, suggesting a role for oxidative stress in the
30                   development of Pb-induced nephropathy.

31                   Table 5-30 presents the acute and chronic renal effects of Pb exposure observed in recent
32                   and past animal toxicology studies.
      November 2012                             5-407                  Draft - Do Not Cite or Quote

-------
      Table 5-30  Effects of Pb on the kidney/renal system related to exposure
                    duration- evidence from animal toxicology studies.
      Effects with less than 3 months of exposure
                                   Effects with 6 or 12 months of exposure
      Mitochondrial dysfunction
      Renal cell apoptosis
      Nuclear Inclusion Body Formation
      Proximal Tubule Cytomegaly
      Glomerular Hypertrophy
      Increased GFR
                                   Mitochondrial dysfunction
                                   Renal cell apoptosis
                                   Oxidant redox imbalance
                                   Altered NO homeostasis
                                   ATPase dysfunction
                                   Aminoaciduria
                                   Increased electrolyte excretion
                                   Elevated blood pressure
                                   Decreased GFR
      5.5.3
Modes of Action for Pb-lnduced Nephrotoxicity
 i
 2
 o
 5
 4
 5
 6
 9
10
11
12
13
14
15
16
17
18

19
20
  5.5.3.1      Oxidative Damage

  A role for ROS in the pathogenesis of experimental Pb-induced hypertension and renal
  disease has been well characterized (Vaziri. 2008a. b; Vaziri and Khan. 2007). The
  production of oxidative stress following Pb exposure is detailed in respect to modes of
  action of Pb (Section 5.2.4). Past studies have shown that Pb treatment (single or three
  daily i.p. injections) can elevate kidney GST levels, affecting glutathione metabolism
  (Daggett et al.. 1998: Moseretal.. 1995: Oberlev et al.. 1995).

  Animal studies continue to provide evidence for increased oxidative stress playing a role
  in the pathogenesis of Pb-induced renal toxicity. Increased ROS, serum NO, and renal
  NO were observed after Pb injections in rats (i.p. 20 mg/kg, 5 days) (Abdel Moneim et
  al., 20lib). Pb exposure to rat proximal tubular cells (0.5-1  (iM) also increased ROS
  production, in a concentration-dependent manner (Wang et al.. 20 lib). Increased lipid
  peroxidation (i.e., MDA) was demonstrated in serum and renal tissue after Pb exposure
  (Abdel Moneim et al.. 201 Ib: Lodi etal.. 2011: Wang etal.. 20lib). Berrahal et al.
  (2011) reported increased MDA in Pb-exposed (50 ppm Pb  acetate pre- and post-natally)
  rat kidney relative to controls at both 40 (puberty; blood Pb  12.7 ug/dL) and 65 (post-
  puberty; blood Pb 7.5 ug/dL)  days of age. In addition,  total sulfhydryl groups were
  significantly decreased at 65 days. These increases in oxidative stress were  accompanied
  by age-dependent Pb nephrotoxicity in male rats.

  Alterations in endogenous antioxidants and antioxidant enzymes that may lead to
  oxidative stress have also been reported after Pb exposure. Pb treatment decreased the
      November 2012
                             5-408
Draft - Do Not Cite or Quote

-------
 1                  activity of the renal antioxidant enzymes, CAT, SOD, GST, GPx, and GR (Abdel
 2                  Moneim et al. 20 lib) and protein levels of CAT and GSH (Lodietal.. 2011).
 3                  Additionally, proteomic analysis of high-level Pb treated (1,500 ppm, 5 weeks; resulting
 4                  in blood Pb level of 53.4 (ig/dL) rat kidney identified decreased abundance of a rate-
 5                  limiting enzyme in the synthesis of GSH (glutamate cysteine ligase) (Chen etal.. 20 lib).

 6                  Conterato et al. (2007) examined the effect of Pb acetate on the cytosolic thioredoxin
 7                  reductase activity and oxidative stress parameters in rat kidneys. A single injection of
 8                  Pb acetate consisted of a single i.p. injection of 25 or 50 mg/kg Pb acetate, while repeated
 9                  injections consisted of one daily i.p. injection of Pb acetate (5 or 25 mg/kg) for 30 days.
10                  Measured were thioredoxin reductase-1, a selenoprotein involved in many cellular redox
11                  processes, SOD, 5-ALAD, GST, GPx, non protein thiol groups (NPSH), CAT,  as well as
12                  plasma creatinine, uric acid, and inorganic phosphate levels. The single injection at the
13                  25 mg Pb dose level resulted in increased SOD and thioredoxin reductase-1 activity,
14                  while the 50 mg dose level increased CAT activity and inhibited 5-ALAD activity in the
15                  kidney. Repeated injections at the 5 mg dose level of Pb inhibited 5-ALAD and increased
16                  GST, NPSH, CAT, and thioredoxin reductase-1.  Repeated injections at the 25-mg dose
17                  level reduced 5-ALAD but increased GST, NPSH, and plasma uric acid levels.  No
18                  changes were observed in TEARS, GPx, creatinine or inorganic phosphate levels after
19                  either single or repeated injection dosing. As both dosing regimens increased thioredoxin
20                  reductase-1 activity, the authors suggest that this enzyme may be a sensitive indicator of
21                  renal changes with low dose Pb treatment.

22                  Jurczuk et al. (2006) published a study of the involvement of some low molecular weight
23                  thiols in the peroxidative mechanisms of action of Pb in the rat kidney. Wistar rats were
24                  fed a diet containing 500 ppm  Pb acetate for a period of 12 weeks and were compared to
25                  a control group receiving distilled water for the same time period. GSH, metallothionein
26                  (MT), total and nonprotein SH groups (TSH and NPSH) were measured, as were the
27                  blood activity and urinary concentration of 5-ALA. The concentrations of GSH and
28                  NPSH were decreased by Pb administration, while MT concentration was unchanged.
29                  5-ALAD in blood was decreased, whereas urinary 5-ALA was increased by Pb
30                  administration. Negative correlations were found between the kidney GSH concentrations
31                  and previously reported concentrations of Pb and MDA in kidneys of these rats. It is
32                  apparent from graphical presentation of the data that GSH was reduced by more than
33                  50% following Pb administration, while TSH was reduced by approximately 15%. No
34                  values for either blood or kidney Pb levels or kidney MDA were reported in this article.
35                  In 2007, the  same authors (Jurczuk et al.. 2007) reported on the renal concentrations of
36                  the  antioxidants, vitamins C and E, in the kidneys of the same Pb-treated and control rats.
37                  Exposure to  Pb significantly decreased vitamin E concentration by 13% and vitamin C
38                  concentration by 26%. The kidney concentration of vitamin C negatively correlated with
      November 2012                            5-409                  Draft - Do Not Cite or Quote

-------
 1                  MDA concentration. The authors concluded that vitamins E and C were involved in the
 2                  mechanism of peroxidative action of Pb in the kidney, and their protective effect may be
 3                  related to scavenging of free radicals.

 4                  Studies have used antioxidant compounds to investigate the role of oxidative stress in
 5                  Pb-induced nephrotoxicity. Abdel Moneim et al. (20lib) reported that flaxseed oil
 6                  treatment protected rats from Pb-induced (i.p. 20 mg/kg, 5 days) oxidative stress,
 7                  inflammation, and apoptosis. However, the flaxseed oil also decreased the accumulation
 8                  of Pb in renal tissue making it difficult to ascertain whether the protection was due to
 9                  decreased oxidative stress or to altered Pb uptake kinetics.

10                  El-Neweshy and El-Sayed (2011) studied the influence of vitamin C supplementation on
11                  Pb-induced histopathological alterations in male rats. Rats were given Pb acetate,
12                  20 mg/kg by intragastric feeding once daily for 60 days. Control rats were given 15 mg of
13                  sodium acetate per kg once daily, and an additional group was given Pb acetate plus
14                  vitamin C (20 mg/kg every other day) 30 minutes before Pb feeding. Control rats showed
15                  normal histology, while Pb-treated rats exhibited karyomegaly with eosinophilic
16                  intranuclear inclusion bodies in the epithelial cells of the proximal tubules. Glomerular
17                  damage and tubular necrosis with invading inflammatory  cells were also seen in
18                  Pb-treated animals. Among rats treated with Pb acetate plus vitamin C, five exhibited
19                  relatively mild karyomegaly and eosinophilic intranuclear inclusion bodies of proximal
20                  tubules and an additional five rats were normal. Normal glomeruli were noted in all.
21                  Thus, vitamin C was shown to ameliorate the renal histopathological effects of Pb
22                  intoxication, however no  measures of Pb accumulation were provided to clarify the
23                  mechanism of action of vitamin C.

24                  Masso-Gonzalez and Antonio-Garcia (2009) studied the protective effect of natural
25                  antioxidants (Zn, vitamin A, vitamin C, vitamin E, and vitamin B6) against Pb-induced
26                  damage during pregnancy and lactation in rat pups. At weaning, pups were sacrificed and
27                  kidneys were analyzed. Pb-exposed pups had decreased body weights. Blood Pb levels
28                  were 1.43 ug/dL in the control group, 22.8  ug/dL in the Pb group, 21.2 ug/dL in the Pb
29                  plus Zn plus vitamins group, and 0.98 ug/dL in the Zn plus vitamin group. The kidney
30                  TEARS were significantly elevated in Pb exposed pups, while treatment with vitamins
31                  and Zn returned TEARS to control levels. Kidney CAT activity was significantly
32                  increased above control with Pb treatment; however supplementation with Zn and
33                  vitamins reduced CAT activity toward normal. Pb exposure inhibited kidney Mn-
34                  dependent SOD but not Cu-Zn-dependent SOD activity. Thus, supplementation with Zn
35                  and vitamins during gestation and lactation was effective in attenuating the redox
36                  imbalance induced by developmental, chronic low-level Pb exposure, likely through the
37                  alteration of Pb accumulation.
      November 2012                            5-410                  Draft - Do Not Cite or Quote

-------
 1                  Bravo et al. (2007) reported further that mycophenolate mofetil (an immunosuppressive
 2                  agent used in renal transplantation which inhibits T and B cell proliferation)
 3                  administration reduces renal inflammation, oxidative stress and hypertension in
 4                  Pb-exposed rats. Thus, an inflammatory immune and oxidative stress component can be
 5                  seen as contributing to Pb-induced renal effects and hypertension.

 6                  Although the majority of studies of the effects of Pb exposure have been conducted in
 7                  male rats,  a couple of studies have compared the response of male rats with female rats
 8                  (Sobekova et al., 2009; Alghazal et al., 2008a). Sobekova et al. (2009) contrasted the
 9                  activity response to Pb on the antioxidant enzymes, GPx and GR, and on TEARS in both
10                  male and female Wistar rats of equal age. Males weighing 412 ± 47 g and females
11                  weighing 290 ± 19 g were fed diets containing either 100 ppm or 1,000 ppm Pb acetate
12                  for 18 weeks. In the male rats, kidney Pb content increased by 492% on the 100 ppm Pb
13                  diet and by 7,000% on the 1,000 ppm Pb diet. In the female rats, kidney Pb content
14                  increased by 410% on the 100 ppm Pb diet and by 23,000% on the 1,000 ppm Pb diet.
15                  There was virtually no change in GPx in the kidney of male rats given the 100 ppm Pb
16                  diet but there was a significant reduction in GPx in the female rats on both the 100 ppm
17                  diet and 1,000 ppm diet.  In male rats, GR was increased from 182 units/gram of protein
18                  in control  kidneys to 220 units on the 100 ppm Pb diet and 350 units on the 1,000 ppm
19                  diet. In female rats, kidney GR decreased from 242 units in control animals to 164 units
20                  in animals on the 100 ppm Pb diet and 190 units in animals on the 1,000 ppm diet. In
21                  male rats,  kidney TEARS content increased from 7.5 units/gram protein to 10.0 units
22                  (1,000 ppm Pb diet group). In female rats, there was a reduction in TEARS from 14.4
23                  units per gram protein to 10.0 units in rats on the 100 ppm Pb diet and to 11 units in rats
24                  on the 1,000 ppm Pb diet.

25                  Alghazal et al. (2008a) compared the activity responses of the antioxidant enzyme, SOD
26                  and the detoxifying enzyme, GST, of the same rats exposed to  100 ppm or 1,000 ppm
27                  Pb acetate for 18 weeks.  Similar to the previous study, kidney TEARS were increased
28                  only in male rats given the higher dose of Pb. Kidney SOD activity, on the other hand,
29                  was increased in both males and females at the higher dose of Pb, while GST activity was
30                  increased in kidney of males at the higher dose of Pb and decreased at the  lower dose, but
31                  was decreased at both doses of Pb in females. Thus there were significant  differences in
32                  the responses of male and female rats to Pb exposure. Differences may be  accounted for
33                  in part due to the greater deposition of Pb in female rat kidneys. Another explanation,
34                  offered by the authors, is that male rats are known to metabolize some foreign
35                  compounds faster than do females, so the biological half-life of xenobiotics in the
36                  females may be longer.
      November 2012                            5-411                 Draft - Do Not Cite or Quote

-------
                    5.5.3.2     Hypertension and Alteration of Renal Vasculature and
                                 Reactivity

 1                  As discussed in Section 5.5.1. changes in renal vasculature function or induction of
 2                  hypertension can contribute to further renal dysfunction. Pb exposure increases BP,
 3                  resulting in hypertension, through the promotion of oxidative stress and altered vascular
 4                  reactivity (Section 5.4). Antioxidants attenuated Pb-related oxidative/nitrosative stress in
 5                  the kidney and abrogated the Pb-induced increased BP (Vaziri etal.. 1999a). Chronic
 6                  increases in vascular pressure can contribute to glomerular and renal vasculature injury,
 7                  which can lead to progressive renal dysfunction and kidney failure. In this manner,
 8                  Pb-induced hypertension has been noted as one contributer to Pb-induced renal disease.

 9                  Also, Pb has been shown to act on known vasomodulating systems in the kidney. In the
10                  kidney, two vascular tone mediators, NO and ET-1, are found to be affected by Pb
11                  exposure. Administration of the vasoconstrictor endothelin-1 (ET-1) affected mean
12                  arterial pressure  (MAP) and decreased GFR (Novak and Banks.  1995). Acute high-dose
13                  Pb exposure (24 nmol/min for 15 or 30 minutes) completely blocked this ET-1-mediated
14                  GFR decrease but had no effect on MAP. Depletion of the endogenous antioxidant
15                  glutathione using the drug buthionine sulfoximine, a GSH synthase inhibitor, increased
16                  BP and increased kidney nitrotyrosine formation without Pb exposure, demonstrating the
17                  importance of GSH in maintenance of BP (Vaziri et al., 2000). Multiple studies have
18                  shown that Pb exposure depletes GSH stores. Catecholamines are vascular moderators
19                  that are also affected by Pb exposure (Carmignani et al., 2000). The effect on BP with Pb
20                  exposure is especially relevant to the kidney because it is both a  target of Pb deposition
21                  and a mitigator of BP. These earlier data detail the interaction of known modulators of
22                  vascular tone with Pb.

23                  The renin-angiotensin-aldosterone system plays an important role in kidney homeostasis
24                  and alteration of this pathway may affect renal function. Simoes  et al.  (2011) reported
25                  that acute Pb treatment (Pb acetate i.v. bolus dose of 320 (ig/kg bw, blood Pb of 37 (ig/dL
26                  at 120 minutes after Pb administration) in adult male Wistar rats increased serum
27                  angiotensin converting enzyme (ACE) activity. Systolic arterial pressure, but not diastolic
28                  arterial pressure  or heart rate, was also elevated 60 minutes after treatment. The
29                  Pb-induced altered systolic BP attenuated in animals co-treated with Losartan (Ang II
30                  receptor blocker) or Enalapril (ACE inhibitor), suggesting a regulatory role for the renin-
31                  angiotensin system (Simdes et al.. 2011). These data agree with earlier reports of
32                  Pb-related increases in ACE activity in young rats exposed to Pb for 2-8 weeks (Sharifi et
33                  al.. 2004) and adult rats exposed to  Pb for 10 months (Carmignani et al.. 1999).

34                  Recently, Vargas-Robles et al. (2007) examined the effect of Pb  exposure (100 ppm
35                  Pb acetate for 12 weeks) on BP and angiotensin II vasoconstriction in  isolated perfused

      November 2012                             5-412                  Draft - Do Not Cite or Quote

-------
 1                  kidney and interlobar arteries. Vascular reactivity was evaluated in the presence and
 2                  absence of the nitric oxide synthase inhibitor L-NAME in both Pb-treated and control
 3                  animals. Pb exposure significantly increased BP (134 ± 3 versus 100 ± 6 mmHg), eNOS
 4                  protein expression, oxidative stress, and vascular reactivity to angiotensin II. L-NAME
 5                  potentiated the vascular response to angiotensin II in the control group, but had no effect
 6                  on the Pb-treated group. Conversely, passive microvessel distensibility, measured after
 7                  deactivation of myogenic tone by papaverine, was significantly lower in the arteries of
 8                  Pb-exposed rats. Nitrites released from the kidney under the influence of angiotensin II in
 9                  the Pb group were lower as compared to the control group whereas 3-nitrotyrosine was
10                  higher in the Pb group. The authors concluded that Pb exposure increases vascular tone
11                  through nitric oxide-dependent and -independent mechanisms, increasing renal vascular
12                  sensitivity to vasoconstrictors.
                    5.5.3.3     Apoptosis and/or Ischemic Necrosis of Tubules and
                                 Glomeruli

13                  Apoptosis or programmed cell death in excess can cause cell atrophy while an
14                  insufficiency can lead to uncontrolled cell proliferation, such as cancer. Pb exposure has
15                  been shown to cause morphological changes to the kidney structure. Some of these
16                  Pb-induced changes are a result of cellular apoptosis or necrosis. Past studies have shown
17                  Pb-induced necrosis in proximal tubule cells (Fowler et al. 1980). Pb-induced apoptosis
18                  is known to act through the mitochondria (Rana. 2008). Pb-induced calcium overload
19                  may depolarize the mitochondria, resulting in cytochrome c release, caspase activation,
20                  and apoptosis. The apoptosis is mediated by  Bax translocation to the mitochondria and
21                  can be blocked by overexpression of Bcl-xl.  Also, Pb-induced ALA accumulation can
22                  generate ROS, which may damage DNA leading to apoptosis.

23                  Mitochondria are targets of Pb toxicity and often involved in apoptosis. Pb can induce
24                  uncoupling of oxidative phosphorylation, decreased substrate utilization, and
25                  modification of mitochondrial ion transport.  ATP energetics are affected when ATP-Pb
26                  chelates are formed and ATPase activity is decreased. ROS formation can contribute to
27                  these mitochondrial changes and to other changes within the kidney. Antioxidant
28                  supplementation after Pb exposure can remedy some changes. All of these outcomes, in
29                  conjunction with Pb-related depletion of antioxidants (e.g., GSH) and elevation of lipid
30                  peroxidation point to possible susceptibility of the kidney to apoptosis or necrosis.

31                  Rodriguez-Iturbe et al. (2005) reported that chronic exposure to low doses of Pb
32                  (100 ppm in drinking water for 14 weeks) results in renal infiltration of immune cells,
33                  apoptosis, NF-KB activation and overexpression of tubulointerstitial Ang(II). Similarly,
      November 2012                            5-413                 Draft - Do Not Cite or Quote

-------
 1                  higher level Pb treatment in rats (i.p. 20 mg/kg, 5 days) induced inflammatory cellular
 2                  infiltrations and an increase in apoptotic cells, accompanied by more pronounced BAX
 3                  staining in kidney tubule epithelial cells (Abdel Moneim et al., 201 Ib). Pb treatment
 4                  (0.5-1 (iM) of isolated rat proximal tubular cells increased cell death by apoptosis and
 5                  necrosis in a concentration- and time-dependent manner (Wang et al., 20 lib). This was
 6                  accompanied by increased morphological changes typical of apoptosis such as
 7                  fragmented chromatin, condensed chromatin, and shrunken nuclei. These cells also
 8                  exhibited decreased mitochondrial membrane potential, decreased intracellular pH,
 9                  inhibition of Na+/K+ATPase and Ca2+ATPase activity, and increased intracellular Ca2+
10                  following Pb treatment.

11                  Navarro-Moreno et al. (2009) examined the effect of 500 ppm Pb in drinking water over
12                  7 months on the structure (including intercellular junctions), function, and biochemical
13                  properties of PT cells of Wistar rats. Pb effects in epithelial cells consisted of an early
14                  loss of the apical microvilli, followed by a decrement of the luminal space and the
15                  respective apposition and proximity of apical membranes,  resulting in the formation of
16                  atypical intercellular contacts and adhesion structures. Inclusion bodies were  found in
17                  nuclei, cytoplasm, and mitochondria. Lipid peroxidation (TEARS measurement) was
18                  increased in the Pb-treated animals as compared to controls. Calcium uptake was
19                  diminished and neither proline nor serine incorporation that was present in controls was
20                  noted in the PT of Pb-exposed animals. The authors speculated that Pb may compete with
21                  calcium in the establishment and maintenance of intercellular junctions.

22                  Tubular necrosis was also observed in rats treated with Pb acetate  (100 ppm s.c.) for
23                  30 days (El-Sokkary et al.. 2005). Histological sections of kidneys from Pb-treated rats
24                  showed tubular degeneration with some necrotic cells. Similarly, El-Neweshy and
25                  El-Sayed (2011) reported glomerular damage and tubular necrosis with invading
26                  inflammatory cells after Pb treatment (20 mg/kg by intragastric feeding  once  daily for
27                  60 days) to male rats. The incidence of necrosis was decreased in both of these studies by
28                  pretreatment with either melatonin or vitamin C. Pretreatment with melatonin
29                  (10 mg/kg), an efficacious free radical scavenger and indirect antioxidant, resulted in a
30                  near normal tubular structure. The authors concluded that melatonin protected the liver
31                  and kidneys from the damaging effects of exposure to Pb through  inhibition of lipid
32                  peroxidation and stimulation of endogenous antioxidative defense systems (El-Sokkary et
33                  al.. 2005). Vitamin C supplementation (20 mg/kg pretreatment every other day) protected
34                  the renal architecture and histology (El-Neweshy and El-Saved. 2011).

35                  Wang et al. (2009c) examined the effect of Pb acetate (0.25, 0.5 and 1 (iM) on cell death
36                  in cultured rat primary PT cells. A progressive loss in cell viability, due to both apoptosis
37                  and necrosis, was observed in cells exposed to Pb. Apoptosis predominated and could be
      November 2012                             5-414                  Draft - Do Not Cite or Quote

-------
 1                  ameliorated with concomitant N-acetylcysteine exposure, whereas necrosis was
 2                  unaffected. Elevation of ROS levels and intercellular calcium, depletion of mitochondrial
 3                  membrane potential, and intracellular glutathione levels was observed during Pb
 4                  exposure. Pb-induced apoptosis was demonstrated morphologically (Hoechst 33258
 5                  staining) with condensed/fragmented chromatin and apoptotic body formation. CAT and
 6                  SOD activities were significantly elevated, reflecting the response to  accumulation of
 7                  ROS.
                    5.5.3.4      Renal Gangliosides

 8                  Gangliosides are constituents of the plasma membrane that are important for control of
 9                  renal GFR because they can act as receptors for various molecules and have been shown
10                  to take part in cell-cell interactions, cell adhesion, cellular recognition, and signal
11                  transduction. Aguilar et al. (2008) studied changes in renal gangliosides following Pb
12                  exposure (600 ppm Pb acetate in drinking water for 4 months) in adult male Wistar rats.
13                  Pb exposure caused an increase in blood Pb from 2.1 to 35.9 (ig/dL. There was no change
14                  in serum creatinine or in hemoglobin, but there was an increase in urinary 5-ALA. The
15                  following renal gangliosides were measured by immunohistochemistry and by thin layer
16                  chromatography: GM1, GM2, GM4, and 9-O-acetylated modified form of the
17                  GD3 ganglioside (9-O-Ac-GD3). The ganglioside pattern was mainly characterized by a
18                  decrease  in the GM1 ganglioside as well as by a mild increase in GM4 and GM2
19                  gangliosides, while the strongest alteration was observed in the 9-O-Ac-GD3, which was
20                  overexpressed. The latter was observed only in the glomerular zone. This was associated
21                  with a decrease in apoptotic glomerular cells, as assessed by the TUNEL assay. The
22                  authors hypothesized that the increase in GD3-O-acetylation could represent a strategy to
23                  attenuate the normal renal apoptotic process and therefore contribute to cell survival
24                  during Pb exposure.
                    5.5.3.5      Altered Uric Acid

25                  Higher occupational Pb exposure or blood Pb levels have been linked to increased risk
26                  for both gout and kidney disease (Shadick et al., 2000; Batuman. 1993). Pb is thought to
27                  increase serum uric acid by decreasing its kidney excretion (Emmerson and Ravenscroft.
28                  1975; Ball and Sorensen, 1969; Emmerson. 1965). Research during the past decade
29                  indicates that uric acid is nephrotoxic at lower levels than previously recognized (Johnson
30                  et al.. 2003). Therefore, the 2006 Pb AQCD (U.S. EPA. 2006b) reviewed literature
31                  implicating increased uric acid as a mechanism for Pb-related nephrotoxicity (Weaver et
32                  al., 2005a; Shadick et al., 2000). However, this does not appear to be the only

      November 2012                            5-415                  Draft - Do Not Cite or Quote

-------
 1                  mechanism, since associations between blood Pb and serum creatinine have remained
 2                  significant even after adjustment for uric acid (Weaver et al.. 2005a).

 3                  Alterations in serum uric acid have been studied in animal models exposed to Pb. In male
 4                  rats exposed to Pb in drinking water from lactation to puberty (40 days) or post-puberty
 5                  (65 days), Berrahal et al. (2011) found that plasma urea levels increased after 40 days of
 6                  exposure (puberty blood Pb level of 12.7 ug/dL) but decreased after 65 days of Pb
 7                  exposure (post-puberty blood Pb level of 7.5  ug/dL) (Table 5-29). Serum uric acid was
 8                  increased in male rats following long-term  Pb exposure (2,000 ppm for 6 weeks)
 9                  (Javakumar et al.. 2009). Conterato et al. (2007) followed various parameters of kidney
10                  function after single or multiple Pb injections in rats. The single dosing regimen consisted
11                  of a single  i.p. injection of 25 or 50 mg/kg Pb acetate, while the multiple injections
12                  involved once daily i.p. injection of either vehicle or Pb acetate (5 or 25 mg/kg) for
13                  30 days. Single and multiple injections at both dose levels increased plasma uric acid
14                  levels. Similarly, Abdel Moneim et al. (20lib) reported increased serum uric acid and
15                  urea levels after 5 days of Pb acetate treatment (i.p. 20 mg/kg).
                     5.5.3.6      Role of Metallothionein

16                   The metal-binding protein, metallothionein, may play a role in inclusion body formation
17                   and thus block potential interaction of Pb with cellular targets. Yu et al. (2009) described
18                   dichotomous effects of Pb acetate on the expression of MT in the liver and kidney of
19                   mice. Male mice were i.p. injected with Pb acetate in doses of 100, 200, and 300 (imol/kg
20                   and sacrificed 4, 8, and 24 hours after Pb treatment. Administration of Pb increased the
21                   levels of MT-1 mRNA in the liver and kidneys but increased MT protein only in the
22                   liver. Treatment of mouse PT cells in vitro with Pb also resulted in an increase in MT
23                   mRNA but little increase in MT protein. Thus, Pb appears to exert a dual effect on MT
24                   expression in the kidney: enhancement of MT gene transcription but suppression of MT
25                   mRNA translation.

26                   Zuo et al. (2009) examined the potential role of a-Synuclein (Sena) and MT in
27                   Pb-induced inclusion body formation. Unlike the parental wild type (WT)  cells, MT-I/II
28                   double knockout (MT-null) cells did not form inclusion bodies after Pb treatment;
29                   however, transfection of MT-1 into MT-null cells allowed inclusion body formation after
30                   Pb treatment. As inclusion bodies formed during Pb treatment, soluble MT protein in WT
31                   cells was lost. As Sena is a protein with a natural tendency to aggregate into oligomers,
32                   Sena was measured in WT cells and MT-null cells after Pb treatment. In both cell lines
33                   Pb-induced Sena expression rapidly increased and then decreased over 48  hours as
34                   Pb-induced inclusion bodies were formed. Pb exposure caused increased colocalization
      November 2012                             5-416                  Draft - Do Not Cite or Quote

-------
 1                  of MT and Sena proteins and MT was localized to the surface of inclusion bodies in WT
 2                  renal cortex samples following Pb treatment. Thus, Sena may be a component of
 3                  Pb-induced inclusion bodies and, with MT, may play a role in inclusion body formation.
      5.5.4       Effects of Exposure to Pb Mixtures

 4                  The effect of Pb on other cations, specifically calcium, is well established in the kidney
 5                  literature. Calcium-mediated processes involving receptors, transport proteins, and
 6                  second messenger signaling among other endpoints have been shown to be significantly
 7                  affected by Pb exposure. The disposition of Pb in the soft tissues (kidney and spleen) can
 8                  change with  exposure to Pb and other compounds. Pb plus  Cd exposure changed Pb
 9                  disposition with increased blood Pb (versus Pb alone group) and decreased metal
10                  concentration in the kidney and liver (versus Pb alone). An iron deficient diet
11                  significantly increased Pb deposition in adult animals (Hashmi et al.. 1989). pregnant
12                  dams, and maternally-exposed fetuses (Singh et al.. 1991).  Dietary thiamine plus Zn
13                  slightly reduced blood and kidney Pb in exposed animals (Flora etal.. 1989). Selenium
14                  (Se), a cofactor for GPx, attenuated Pb-induced lipid peroxidation and abrogated the
15                  Pb-induced attenuation of GR and SOD. Concomitant exposure to the cations aluminum
16                  and Pb protected animals from ensuing nephropathy (Shakoor et al.. 2000). In summary,
17                  Pb has been  shown to affect processes mediated by endogenous divalent cations. In
18                  addition, exposure to other metals or divalent cations can modulate Pb disposition and its
19                  effects in the body.
                    5.5.4.1      Lead(Pb) and Cadmium(Cd)

20                  Cd shares many similarities with Pb; it has been shown to be a ubiquitous PT
21                  nephrotoxicant and accumulates in the body. Despite the similarities, few studies have
22                  evaluated associations between Cd exposure and CKD or the impact of joint exposure of
23                  Pb and Cd or other metals on CKD. As discussed in the 2006 Pb AQCD (U.S. EPA.
24                  2006b). environmental exposure to Cd, at levels common in the U.S. and other developed
25                  countries, has been shown to impact substantially associations between indicators of Pb
26                  exposure and the kidney EBE marker, NAG, even in the presence of occupational level
27                  Pb exposure. In an occupational study, mean NAG, although higher in the Pb-exposed
28                  worker group compared to controls, was correlated with urine Cd but not blood or tibia
29                  Pb (Roels et al.. 1994). In another occupational population where both metals were
30                  significantly associated with NAG, a 0.5 ug/g creatinine increase in Cd had the same
31                  effect on NAG as did a 66.9 ug/g bone mineral increase in tibia Pb (Weaver et al..
32                  2003a).

      November 2012                             5-417                  Draft - Do Not Cite  or Quote

-------
 1                  The 2006 Pb AQCD noted that data examining the concentration-response relation
 2                  between environmental Cd and the kidney were too scarce to determine the impact of Cd
 3                  exposure on relations between Pb exposure and other kidney outcomes. A recent
 4                  publication in NHANES data collected from 1999 through 2006 addresses this need;
 5                  (results pertaining solely to Pb were discussed in  Section 5.5.2.1) (Navas-Acien et al..
 6                  2009). Geometric mean concurrent blood Cd level was 0.41 ug/L in 14,778 adults aged
 7                  > 20 years. After adjustment for survey year, sociodemographic factors, CKD risk
 8                  factors, and blood Pb, the ORs for albuminuria (> 30 mg/g creatinine), reduced eGFR
 9                  (<60 mL/min/1.73 m2), and both albuminuria and reduced eGFR were 1.92 (95% CI:
10                  1.53, 2.43), 1.32 (95% CI: 1.04, 1.68), and 2.91 (95% CI: 1.76, 4.81), respectively,
11                  comparing the highest with the lowest blood Cd quartiles. Both Pb and Cd remained
12                  significantly associated after adjustment for the other. Effect modification was not
13                  observed; however, ORs were higher for adults in the highest quartiles of both metals
14                  compared with the ORs for the highest quartiles of concurrent blood Cd or Pb alone
15                  (Table 5-25). Compared with adults with blood Cd levels < 0.2 ug/L and blood Pb levels
16                  < 1.1 ug/dL, adults with blood Cd levels >0.6 ug/L and blood Pb levels >2.4 ug/dL had
17                  ORs (95% CIs) of 2.34 (95% CI:  1.72, 3.18) for albuminuria, 1.98 (95%  CI: 1.27, 3.10)
18                  for reduced eGFR, and 4.10 (95% CI: 1.58,  10.65) for albuminuria and reduced eGFR
19                  together. These findings are consistent with other recent publications (Akesson et al.
20                  2005; Hellstrom et al.. 2001). support consideration of both metals as  independent CKD
21                  risk factors in the general population, and provide novel evidence of increased risk in
22                  those with higher environmental exposure to both metals.

23                  However, a very recent study suggests that interpretation of Cd associations with GFR
24                  measures may be much more complex. Conducted in Pb workers to address the fact that
25                  few studies have examined the impact of environmental Cd exposure in workers who are
26                  occupationally exposed to other nephrotoxicants such as Pb, the study assessed Cd dose
27                  with urine Cd, which is widely considered the optimal dose metric of cumulative Cd
28                  exposure. In 712 Pb workers, mean (SD) blood and tibia Pb, urine Cd, and eGFR using
29                  the MDRD equation were 23.1 (14.1) ug/dL, 26.6 (28.9)  ug/g, 1-15 (0.66) ug/g
30                  creatinine, and 97.4 (19.2) mL/min/1.73m2,  respectively (Weaver et al.. 2011). After
31                  adjustment for age, sex, BMI, urine creatinine,  smoking, alcohol use, education, annual
32                  income, diastolic BP, current or former Pb worker job status, new or returning study
33                  participant, and blood and tibia Pb, higher urine Cd was associated with higher calculated
34                  creatinine clearance, eGFR (P = 8.7 mL/min/1.73 m2 [95% CI: 5.4, 12.1] per unit
35                  increase in In-transformed urine Cd) and In-NAG, but lower serum creatinine. These
36                  unexpected paradoxical associations have been reported in a few other publications (de
37                  Burbure et al.. 2006; Hotz et al.. 1999) and have been observed in other populations.
38                  Potential explanations for these paradoxical results included a normal physiologic
      November 2012                            5-418                 Draft - Do Not Cite or Quote

-------
 1                  response in which urine Cd levels reflect renal filtration; the impact of adjustment for
 2                  urine dilution with creatinine in models of kidney outcomes; and Cd-related
 3                  hyperfiltration.

 4                  Wang et al. (2009c) studied the effects of Pb and/or Cd on oxidative damage to rat kidney
 5                  cortex mitochondria. In this study young female Sprague Dawley rats were fed for
 6                  8 weeks with either Pb acetate (300 ppm), Cd chloride (50 ppm), or Pb and Cd together
 7                  in the same dosage. Lipid peroxidation was assessed as MDA content. Renal cortex
 8                  pieces were also processed for ultrastructural analysis and for quantitative rtPCR to
 9                  identify the mitochondrial damage and to quantify the relative expression levels of
10                  cytochrome oxidase subunits (COX-I/II/III). Cytochrome oxidase is the marker enzyme
11                  of mitochondrial function, and COX-I, II, and III are the three largest mitochondrially-
12                  encoded subunits which constitute the catalytic functional core of the  COX holoenzyme.
13                  Mitochondria were altered by either Pb or Cd administration, but more strikingly by Pb
14                  plus Cd administration, as indicated by disruption and loss of mitochondrion cristae.
15                  Kidney cortex MDA levels were increased significantly by either Pb or Cd, given
16                  individually, but more so by Pb plus Cd. COX-I/II/III were all reduced by either Pb or Cd
17                  administration, but more prominently by Pb plus Cd administration. This study adds to
18                  knowledge of the synergistic effects of Pb and Cd on kidney mitochondria.
                    5.5.4.2      Lead(Pb), Cadmium(Cd), and Arsenic(As)

19                  Wang and Fowler (2008) present a general review of the roles of biomarkers in
20                  evaluating interactions among mixtures of Pb, Cd, and As. Past studies have found that
21                  addition of Cd to treatment of rats with Pb or Pb and As significantly reduced the
22                  histological signs of renal toxicity from each element alone, including swelling of the
23                  proximal tubule cells and intranuclear inclusion body formation. On the other hand,
24                  animals exposed to Cd in addition to Pb or Pb and As  showed an additive increase in the
25                  urinary excretion of porphyrins, indicating that, although measured tissue burdens of Pb
26                  were reduced, the biologically available fraction of Pb is actually increased (Mahaffey et
27                  al.. 1981: Mahaffev and Fowler.  1977).
      November 2012                            5-419                 Draft - Do Not Cite or Quote

-------
 1                  Stress proteins were examined after exposure to mixtures of Pb and other metals.
 2                  Induction of MT was strongest in groups with Cd treatment. However, co-exposure to Pb
 3                  and As induced higher levels of MT protein than did either Pb or As exposure alone in
 4                  kidney tubule cells. Heat shock proteins (Hsps) are commonly altered with exposure to
 5                  metal mixtures. A study found in vitro (low dose) and in vivo that Pb induced Hsps in a
 6                  metal/metalloid-, concentration-  and time-specific manner (Wang et al.. 2005a). Additive
 7                  or more than additive interactions occurred among Pb, Cd and As under combined
 8                  exposure conditions.
                    5.5.4.3      Lead (Pb) and Zinc (Zn)

 9                  Zinc has been investigated as a protective compound against the effects of Pb. Pb
10                  treatment (35 mg/kg i.p. for 3 days) caused a significant fall in hemoglobin content,
11                  significant increases in lipid peroxidation and decreased level of reduced glutathione in
12                  liver, together with diminished total protein content in liver and kidney. Co-treatment of
13                  Pb with Zn (10 mg/kg i.p.) or ascorbic acid (10, 20 and 30 mg/kg i.p.) showed a moderate
14                  therapeutic effect when administered individually, but more pronounced protective
15                  effects after combined therapy (Upadhyay et al., 2009).

16                  Jamieson et al. (2008) studied the effect of dietary Zn content on renal Pb deposition.
17                  Weanling Sprague Dawley rats were assigned to marginal zinc (MZ, 8 mg Zn/kg diet),
18                  zinc adequate control (CT, 30 mg Zn/kg), zinc-adequate diet-restricted (30 mg Zn/kg), or
19                  supplemental zinc (SZn, 300 mg Zn/kg) groups, with or without Pb acetate (200 ppm for
20                  3 weeks). Pb exposure did not result in nephromegaly or histological alterations. The MZ
21                  rats had higher renal Pb (35%) and lower renal  Zn (16%) concentrations than did CT rats.
22                  On the other hand, SZn was more protective than the CT diet was against renal Pb
23                  accumulation (33% lower). Standard procedures for indirect immunoperoxidase staining
24                  were used to determine MT localization in the kidney. Pb had no effect on MT staining
25                  intensity, distribution, or relative protein amounts. Western blot analysis confirmed that
26                  MT levels were responsive to dietary Zn but not to Pb exposure.
                    5.5.4.4     Lead(Pb) and Mercury(Hg)

27                  Stacchiotti et al. (2009) studied stress proteins and oxidative damage in a renal-derived
28                  cell line exposed to inorganic Hg and Pb. The time course of the expression of several
29                  Hsps, glucose-regulating proteins and MTs in a rat proximal tubular cell line (NRK-52E)
30                  exposed to subcytotoxic doses of inorganic mercury (HgCl2, 1-40 uM) and Pb (PbCl2,
31                  2-500 uM) were analyzed. ROS and reactive nitrogen species  (RNS) were detected by
      November 2012                            5-420                 Draft - Do Not Cite or Quote

-------
 1                  flow cytometric analysis. Endogenous total GSH content and the enzymatic activity of
 2                  GST were determined in cell homogenates. Western blot analysis and
 3                  immunohistochemistry were used for quantification of hsps and MTs. Reverse
 4                  transcription PCR was used for quantification of metallothionein. The higher doses of Hg
 5                  (20 uM and 40 uM) were shown to markedly inhibit growth of the cell line while the
 6                  higher doses of Pb (60 uM to 500 uM) inhibited cell growth to a lesser degree. After 24
 7                  hours of exposure of 20 (iM Hg, the cells presented abnormal size and pyknotic nuclei,
 8                  swollen mitochondria and both apoptosis and overt necrosis. In the presence of 60 or
 9                  300 (iM Pb, the cells lost cell-cell and cell-matrix contacts, showed a round size, irregular
10                  nuclear contour and often mitotic arrest, but no apoptosis or overt necrosis at 24 hours.
11                  Mercury (Hg) induced a significant increase in both ROS and RNS, maximal RNS at
12                  24 hours, and maximal ROS at 48 hours. Pb (60 or 300 (iM) did not cause an increase in
13                  ROS or RNS beyond the levels measured in control cells. Total GSH significantly
14                  increased in cells grown in the presence of Pb; the effect was concentration-dependent
15                  and GSH reached its maximal value at a dose of 300 (iM Pb. The effect of Hg was
16                  biphasic:  10 uM significantly enhanced GSH by 600%, while the amount of GSH
17                  detected after 20  (iM Hg only increased by 50% compared to control levels. GST activity
18                  was enhanced by both Pb and Hg. Hsp25 and Hsp72 were up-regulated by Hg but there
19                  was no effect on Grp78 as compared to control. On the contrary, Pb treatment only
20                  upregulated Grp78. Mercury (Hg) induced a time-dependent effect on MT mRNA
21                  expression, which reached its maximal value 3 hours after beginning treatment and
22                  reverted to control values at 24 hours. With Pb, on the other hand, mRNA transcription
23                  was concentration- and time-dependent. The transcripts remained overexpressed
24                  compared to controls up to 72 hours. The results of this study with regard to the Pb effect
25                  on MT synthesis  clearly differ from those of Jamieson et al. (2008). which found no
26                  increase in MT following Pb exposure. This discrepancy remains to be clarified.
      November 2012                            5-421                 Draft - Do Not Cite or Quote

-------
      5.5.5        Summary and Causal Determination

 1                  A large body of epidemiologic evidence and limited toxicological evidence indicates Pb
 2                  exposure leads to reduced kidney function. The causal determination for reduced kidney
 3                  function is informed by evidence for reduced GFR, reduced creatinine clearance, and
 4                  increased serum creatinine. Biological plausibility and mode of action for these effects is
 5                  provided by evidence for hypertension, oxidative stress, inflammation, vascular reactivity
 6                  and injury, increased uric acid, morphological changes, and apoptosis or necrosis. The
 7                  section that follows describes the evaluation of evidence for reduced kidney function,
 8                  with respect to causal relationships with Pb exposure using the framework described in
 9                  Table II of the Preamble. The key supporting evidence to the causal framework is
10                  summarized in Table 5-31.
                    5.5.5.1      Evidence for Reduced Kidney Function

11                  The 2006 Pb AQCD (U.S. EPA. 2006b) concluded that "in the general population, both
12                  circulating and cumulative Pb was found to be associated with a longitudinal decline in
13                  renal function," evidenced by increased serum creatinine and decreased creatinine
14                  clearance or eGFR over follow-up of 4 to 15 years in association with higher baseline
15                  blood and bone Pb levels (U.S. EPA. 2006b). Data in general and patient populations of
16                  adults provided consistent evidence of Pb-associated lower renal function in populations
17                  with mean concurrent or baseline blood Pb levels of 2-10 ug/dL (Akesson et al., 2005;
18                  Tsaih et al.. 2004; Yu et al.. 2004; Kimetal..  1996); associations with lower eGFR were
19                  observed in adults with hypertension with a mean concurrent blood Pb level of 4.2 ug/dL
20                  (Muntner et al.. 2003). The conclusion from the 2006 Pb AQCD was substantiated by the
21                  coherence of effects observed across epidemiologic and toxicological studies. However, a
22                  number of the animal toxicological studies were conducted at Pb exposure concentrations
23                  that resulted in blood Pb levels higher than what is relevant to the general U.S. adult
24                  human population. Both human and animal studies observed Pb-associated
25                  hyperfiltration. In animals during the first 3 months after Pb exposure, effects were
26                  characterized by increased GFR and increased kidney weight due to glomerular
27                  hypertrophy. However, exposure for 6 or 12 months resulted in decreased GFR,
28                  interstitial fibrosis, and kidney dysfunction. Additionally, toxicological  studies found that
29                  early effects of Pb on tubular cells were generally reversible, but continued exposure
30                  resulted in chronic irreversible damage. Toxicological studies provided mechanistic
31                  evidence to support the biological plausibility of Pb-induced renal effects, including
32                  oxidative stress leading to NO inactivation. Despite the strong body of evidence
33                  presented in the 2006 Pb AQCD, uncertainty remained on the contribution of past Pb
      November 2012                            5-422                 Draft - Do Not Cite or Quote

-------
 1                   exposures to associations observed in adults, the impact in children, the implication of
 2                   hyperfiltration, and reverse causality.

 3                   Consideration of results from recent epidemiologic studies does not alter the conclusions
 4                   of the previous AQCD. Prospective studies in adult men in the general population (Tsaih
 5                   et al.. 2004; Kimetal.. 1996): an adult patient population study (Yu et al. 2004): CKD
 6                   patient studies (Lin et al., 2006b: Lin et al., 2006a: Lin et al., 2003). and a Pb worker
 7                   cohort (Weaver et al.. 2009) provide evidence of a relationship between blood Pb level
 8                   and subsequent decreases in kidney function and better establish the temporal sequence
 9                   between Pb exposure and kidney function. The study designs and analysis are well
10                   designed with careful measurement of exposure, outcomes, and covariates, and
11                   adjustment for numerous potential confounding factors including age, race, sex,
12                   education, household income, smoking, alcohol use, Cd exposure, and various health
13                   indicators such as  diabetes, systolic BP, BMI, and history of cardiovascular disease.
14                   Large sample sizes provide strength to the general population studies, whereas some
15                   caution may be appropriate for the CKD patient studies where the smallest study has 87
16                   subjects.  Confidence in the relationship between Pb exposure and renal effects is
17                   provided by the combined results of a body of studies from different research groups
18                   using different designs in different cohorts.

19                   Limitations of these studies produce some uncertainty in the relationship between Pb
20                   exposure and reduced kidney function. Since the prospective studies are reported in
21                   patients, workers,  and primarily white men with a mean age of 60, they may not be
22                   generalizable to the entire U.S. population. Also, as these studies report effects most
23                   often observed in adults with likely  higher past Pb exposures, uncertainty exists as to the
24                   Pb exposure level, timing, frequency, and duration contributing to the associations
25                   observed with blood or bone Pb levels. In addition, it is possible that the CKD patient
26                   studies (Lin et al., 2006b: Lin et al., 2006a: Lin et al., 2003). in which blood Pb level was
27                   lowered compared to control groups by chelation, demonstrate an improvement in renal
28                   function by reducing reactive oxygen species, blood Pb level or both. The uncertainty
29                   related to this may reflect an involvement of both lowering of blood Pb levels and a
30                   reduction of reactive  oxygen species following chelation as both are possible to a varying
31                   extent. The potential  for a bidirectional relationship because of reverse causality is
32                   possible in observational epidemiologic studies and must be weighed in this discussion
33                   (see Section 5.5.2.4).

34                   Cross-sectional studies add support to the associations observed in prospective
35                   epidemiologic studies (Section 5.5.2.1). The majority of cross-sectional studies report
36                   associations between higher measures of Pb exposure and worse renal function. These
37                   studies include analyses from the NHANES cohort which provides a representative U.S.
      November 2012                             5-423                  Draft - Do Not Cite or Quote

-------
 1                   population sample that may be generalizable to the total U.S. population. Re-examination
 2                   of a study from the 2006 Pb AQCD (U.S. EPA. 2006b) provided data to conclude that in
 3                   a population with likely higher past exposures to Pb, a 10-fold increase in concurrent
 4                   blood Pb was associated with an 18 mL/min decrease in estimated creatinine clearance or
 5                   a 25% decrease from the mean, and that an increase in blood Pb from the 5th to the 95th
 6                   percentile (3.5 ug/dL) had the same negative impact on eGFR as did an increase of 4.7
 7                   years in age or 7 kg/m2 in body mass index (Akesson et al.. 2005). However, a small
 8                   number of studies report findings that are less consistent with the body of evidence
 9                   (Section 5.5.2.1). Overall, a relationship between higher Pb exposure and various
10                   indicators of lower kidney function is indicated by a set of high-quality studies that
11                   control for important potential confounders such as age, sex, BMI, comorbid
12                   cardiovascular conditions, smoking and alcohol use, and that are conducted with different
13                   designs in different cohorts by different researchers.

14                   At current blood Pb levels in the U.S. adult population, a downward shift in kidney
15                   function of the entire population due to Pb may not result in CKD in identifiable
16                   individuals; however, that segment of the population with the lowest kidney reserve may
17                   be at increased risk for CKD when Pb is combined with other kidney risk factors. For
18                   example, in adults with mean (concurrent or baseline measured 4-6 years before kidney
19                   function tests) blood Pb levels that are comparable to that of the general U.S. population
20                   (1.6 to 4.2 ug/dL), higher blood Pb level was found to be associated with clinically-
21                   relevant effects (e.g., eGFR <60 mL/min/1.73 m2, doubling of serum creatinine)
22                   (Fadrowski et al.. 2010; Yu et al.. 2004) and larger magnitudes of effect in potentially at-
23                   risk populations with cormorbidities for CKD such as diabetes mellitus (Tsaih et al.,
24                   2004) and hypertension (Tsaih et al.. 2004; Muntner et al.. 2003) or higher co-exposure to
25                   other environmental nephrotoxicants such as Cd (Navas-Acien  et al.. 2009).

26                   Research in the occupational setting has traditionally been far less consistent than that in
27                   environmentally exposed populations (Section 5.5.2.1). Limitations of the occupational
28                   evidence, which are discussed earlier in this section have been used to explain this
29                   inconsistency. The actual cause of paradoxical or inverse associations (higher Pb dose
30                   with lower serum creatinine, and/or higher eGFR or calculated  or measured creatinine
31                   clearance) in several of these studies may not be known. If associations are in opposite
32                   directions in different subgroups of the population and the relevant effect modifier is not
33                   considered, null associations will be observed. For these reasons, nonsignificant
34                   associations or paradoxical associations in the occupational setting cannot be used as a
35                   rationale for discounting Pb-related nephrotoxicity at lower environmental levels.

36                   CKD is an important risk factor for cardiac disease. As kidney dysfunction can increase
37                   BP and increased BP can lead to further damage to the kidneys, Pb-induced damage to
      November 2012                             5-424                  Draft - Do Not Cite or Quote

-------
 1                   either or both renal or cardiovascular systems may result in a cycle of further increased
 2                   severity of disease. Pb exposure has been causally linked to both increased BP and other
 3                   cardiovascular effects (Section 5.4). Interestingly, animal studies have shown Pb-induced
 4                   vascular injury in the kidney was associated with increased glomerular sclerosis,
 5                   tubulointerstitial injury, increased collagen staining, and an increase in macrophages
 6                   associated with higher levels of MCP-1 mRNA. It is possible that the cardiovascular and
 7                   renal effects of Pb observed are mechanistically linked and thus, Pb-induced hypertension
 8                   has been noted as one cause of Pb-induced renal disease.

 9                   Recently available animal toxicological studies strengthen the evidence regarding the
10                   modes of action for Pb exposure leading to renal alterations, including the influence of
11                   Pb-induced oxidative stress. The mode of action of Pb in the kidneys has been extended
12                   to the field of immunology, evidenced by observations that Pb exposure resulted in
13                   infiltration of lymphocytes and macrophages associated with increased expression of
14                   NF-KB in proximal tubules and infiltrating cells (Roncal et al.. 2007). Additionally,
15                   recent findings expand on the evidence of acute effects of Pb, including mitochondrial
16                   dysfunction, renal cell apoptosis, and glomerular hypertrophy. These mechanisms are
17                   useful in understanding the occurrence of acute hyperfiltration followed by chronic
18                   kidney dysfunction. Lower concentration Pb exposures and lower blood Pb levels in
19                   animals have not been widely examined. As indicated in Table 5-28. studies  found
20                   dysfunction in various kidney function measures, including urinary flow, ALP,
21                   microalbumin, and NAG at blood Pb levels greater than 20 (ig/dL.

22                   Changes in renal function that have been associated with biomarkers of Pb exposure may
23                   indicate a modest change for an individual; however, these modest changes can have a
24                   substantial public health implication at the population level. The reported effects
25                   represent a central tendency of Pb-induced renal function effects among individuals;
26                   some individuals may differ in risk and manifest effects that are greater in magnitude. For
27                   example, a small worsening of renal function may shift the population distribution and
28                   result in considerable increases in the percentages of individuals with worse renal
29                   function that are clinically significant.

30                   Overall, recent studies evaluated in the current review support and expand upon the body
31                   of evidence presented in the 2006 Pb AQCD indicating that Pb exposure is associated
32                   with reduced kidney function.  In addition, animal studies provide biological plausibility
33                   for the associations observed in epidemiologic studies between blood Pb levels and
34                   reduced kidney function with evidence for Pb-induced hypertension, renal oxidative
35                   stress, inflammation, apoptosis, and glomerular hypertrophy. However, a number of
36                   limitations remain including the representativeness of worker studies and older all male
37                   cohorts to the U.S. population, the potential for reverse causality to play a role in the
      November 2012                             5-425                  Draft - Do Not Cite or Quote

-------
1                   findings of cross-sectional studies, and inconsistent findings in occupational studies.
2                   Collectively, the evidence integrated across epidemiologic and toxicological studies is
3                   sufficient to conclude that a causal relationship is likely to exist between Pb exposures
4                   and reduced kidney function.
     November 2012                              5-426                  Draft - Do Not Cite or Quote

-------
Table 5-31    Summary  of evidence supporting renal causal determinations.
Attribute in Causal
Framework3
Key Supporting Evidence
References
Pb Exposure or Blood
Levels Associated
with Effects0
Reduced Kidney Function - Likely Causal
Consistent associations    Multiple prospective epidemiologic
in multiple high quality
epidemiologic studies
with relevant blood Pb
levels
studies in nonoccupationally exposed
adults for associations with change in
serum creatinine and GFR in MAS,
Korean workers, and CKD patients
Studies provide concentration-response
information
Associations found with adjustment for
potential confounding factors including
age, pre-existing cardiovascular disease,
baseline kidney function, and SES.
Supporting cross-sectional evidence for
associations between concurrent blood
Pb level and serum creatinine, creatinine
clearance, and GFR
Studies had population based
recruitment (NHANES) with high follow-
up participation

Uncertainty related to reverse causality;
the bidirectional association is possible
Uncertainty due to baseline serum
creatinine adjustment
                                                                 Section 5.5.2.1

                                                                 Muntner (2003).
                                                                 Navas-Acien (2009)
                                                                 Section 5.5.2.1
                                                                 Akesson (2005).
                                                                 Staessen (1992)
                                                                 Payton (1994)


                                                                 Section 5.5.2.4
                               Blood Pb level:
                               Adults,
                               <10ug/dLd
                               Concurrent Blood Pb
                               level:
                               Adults, means
                               1.58-4.2 ug/dL
                               Concurrent Blood Pb
                               level:
                               Adults, medians
                               2.2-11.4 ug/dL
Limited toxicological
evidence to support
epidemiologic evidence
Mixed evidence in animals report
decreased creatinine clearance,
increased serum creatinine, and
decreased GFR at both relevant and high
level long-term exposures.
Berrahal et al. (2011).
Roncal et al. (2007).
Khalil-Manesh et al.
(1993a: 1992b: 1992a)
Blood Pb level:
Rodents: >7.5 ug/dL
<65 days from birth,
26 ug/dL
12 weeks as adults,
29-125 ug/dL
Evidence clearly
describes mode of action
     Hypertension
     Oxidative Stress



     Increased Uric Acid


     Inflammation

     Morphology
Consistent evidence of increased BP and
hypertension following Pb exposure in
humans and animals at relevant Pb
levels across numerous studies with
control for confounding.
Association of increased blood pressure
with manifestation of CHD has been well
documented.
Consistent evidence for increased ROS,
enhanced lipid peroxidation, and
antioxidant enzyme disruption in Pb
exposed animals.
Evidence of increased plasma uric acid
and urea in animals.
Mixed results in  humans
Lymphocyte and macrophage infiltration
Increased MCP-1 expression
Glomerular hypertrophy
Cellular apoptosis and necrosis leading
to PT damage
                                                                 Sections 5.4 and 5.5.3.2
Section 5.5.3.1
Section 5.5.3.5
Sections 5.5.2.5 and 5.2.5
                                                                 Sections 5.5.2.5 and 5.5.3.3
"Described in detail in Table II of the Preamble.
bDescribes the key evidence and references contributing most heavily to causal determination. Also noted are the sections where full body of
evidence is described.
°Describes the blood Pb levels in humans with which the evidence is substantiated and blood Pb levels in animals most relevant to humans.
dBecause blood Pb level in nonoccupationally-exposed adults reflects both recent and past Pb exposures, the magnitude, timing, frequency, and
duration of Pb exposure contributing to the observed associations is uncertain.
November 2012
                          5-427
              Draft - Do Not Cite or Quote

-------
      5.6        Immune System Effects
      5.6.1       Introduction

 1                  With respect to studies conducted in laboratory animal and in vitro models, the immune
 2                  effects of Pb exposure have been extensively examined over several decades. Animal
 3                  studies of the effects of Pb exposure on host resistance date back to the 1960s while those
 4                  focusing on Pb-induced immune functional alterations, including developmental
 5                  immunotoxicity, were first conducted during the 1970s. Despite this long history of
 6                  research, Pb-associated immune effects in animals with blood Pb levels in the range of
 7                  current U.S. population levels (i.e., <10 (ig/dL), particularly early in life, have been
 8                  observed only within the last 10-15 years (Dietert and McCabe. 2007). Recent findings of
 9                  Pb-associated changes in immunological parameters in humans have increased
10                  understanding of the immune effects of environmental exposure to Pb.

11                  The pathways by which Pb exposure may alter immune cell function and consequently
12                  increase the risk of immune-related diseases are presented in (Figure 5-34). Rather than
13                  producing overt cytotoxicity, Pb exposure has  been associated with functional alterations
14                  in cellular and humoral immunity. In the 2006 Pb AQCD (U.S. EPA. 2006b). the
15                  hallmarks reported for Pb-induced changes in immune functional pathways were:
16                  (1) suppression of T cell-derived helper (Th)l-mediated immunity (i.e., suppressed Thl
17                  cytokine production and delayed type hypersensitivity [DTH] response); (2) stimulation
18                  of Th2 immunity (i.e., increased production of Th2 cytokines and immunoglobulin (Ig)E
19                  antibody); and (3) altered macrophage function (U.S. EPA. 2006b). The latter was
20                  characterized by increased production of reactive oxygen species (ROS), prostaglandin E2
21                  (PGE2), and inflammatory cytokines such as tumor necrosis factor-alpha (TNF-a) and
22                  interleukin (IL)-6 and decreased production of nitric oxide (NO). Changes in immune
23                  cells can alter cell-to-cell interactions, multiple signaling pathways, and inflammation,
24                  that in turn, can influence the risk of developing  infectious, allergic, and autoimmune
25                  diseases and exacerbate inflammatory responses  in other organ systems. Studies
26                  conducted in animal and in vitro models provided consistent evidence for Pb exposure
27                  inducing effects on the range of immune effects presented in this continuum. In the much
28                  smaller epidemiologic evidence base, most studies examined Pb-exposed male  workers
29                  and a  limited range of immune-related endpoints.
      November 2012                            5-428                  Draft - Do Not Cite or Quote

-------
            Elevated IL-4 and IL-5
            Suppressed INF-y
                                                          Macrophages and
                                                          Other Innate Immune Cells
                   I
               Skewed
               Th2-biased
               responses
            Elevated IL-10
            Suppressed IL-12

       Increased lipid
       and DNA oxidation
       in tissues
Elevated TNF-a
Overproduction of ROS
Depleted antioxidant
    defenses
Suppressed
Thl- mediated
anti-tumor
hostdefense     /  B Cells

       Increased tumor   \.
       cell formation   Increased IgE
                     production
                                 Increased tissue
                                 inflammation
                                 (e.g. lung, gut, skin)
                                                                           Reduced:
                                                                           Phagocytosis
                                                                           Nitric oxide production
                                                                           Peroxynitrite production
                                                                           Lysosomal activity
                                                              Removal of normal
                                                              myelomonocytic
                                                              suppression
                                       Damaged epithelia
                                       andmucosal barriers
                            \
                            /productio

                                 \
                                  Tissue damage
                                  and de novo
                                  antigen
                                  appearance
                                                                              Inappropriate
                                                                              Tcell proliferation
                                                                              activation
Increased risk of
later-life cancer

Increased risk of atopy
and allergic disease

Increased risk of tissue
inflammatory diseases

Increased risk of
autoimmunity

Reduced host resistance
to bacterial infection
      Figure 5-34      Immunological pathways by which Pb exposure may increase risk
                         of immune-related diseases.
 i
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
Reflecting suppressed Thl activity, toxicological evidence presented in the
2006 Pb AQCD linked Pb exposure of animals to impaired host resistance to viruses and
bacteria (U.S. EPA. 2006b). Indicating a hyperinflammatory state and local tissue
damage, a few available toxicological studies found Pb exposure-induced generation of
auto-antibodies, suggesting an elevated risk of autoimmune reactions. Additionally, the
shift toward Th2 responses  suggested that Pb could elevate the risk of atopic and
inflammatory responses. While the biological plausibility of such effects was supported
by toxicological evidence for Pb-induced increases in Th2 cytokines, IgE, and
inflammation, epidemiologic  evidence was too sparse to draw conclusions about the
effects of Pb exposure on these broader indicators of immune dysfunction in humans.
However, in concordance with toxicological evidence, a shift to a Th2 phenotype was
indicated in the few available studies by associations observed between higher concurrent
blood Pb level and higher serum IgE  levels in children. Because of lack of examination,
the immune effects of Pb exposure in adults without occupational exposures were not
well characterized.
      November 2012
                            5-429
                     Draft - Do Not Cite or Quote

-------
 1                  Changes in the spectrum of immune endpoints were found in association with a wide
 2                  range of blood Pb levels. Juvenile and adult animals with Pb levels in the range of
 3                  7-100 (ig/dL were found to have suppressed DTH, elevated IgE, and changes in cytokine
 4                  levels. Most epidemiologic studies examined and found lower T cell abundance and
 5                  higher serum IgE levels in association with population mean (or group) concurrent blood
 6                  Pb levels >10  (ig/dL.

 7                  With respect to important lifestages of Pb exposure, animal studies provided strong
 8                  evidence for immune effects in juvenile animals induced by prenatal Pb exposures and in
 9                  adult animals by postnatal exposures. There was uncertainty regarding important
10                  lifestages of Pb exposure in humans as epidemiologic studies of children primarily were
11                  cross-sectional and examined concurrent blood Pb levels. Several other limitations of
12                  epidemiologic studies were noted, including small sample sizes; little consideration for
13                  potential confounding factors such as age, sex, smoking, SES indicators, and allergen
14                  exposures; and comparisons of immune endpoints among groups with different blood Pb
15                  levels that provided little information on the concentration-response relationship.

16                  Collectively, the small numbers of toxicological and epidemiologic studies published
17                  since the 2006 Pb AQCD,  supported the previous findings of Pb-associated immune
18                  effects. Epidemiologic studies supported previous findings in children and provided new
19                  evidence for effects in nonoccupationally-exposed adults. Recent studies also expanded
20                  on the array of immunological parameters affected by Pb exposure as presented in Figure
21                  5-34. For example, a recent toxicological study indicated that Pb may modulate the
22                  function of dendritic cells. Results from recent toxicological and epidemiologic studies
23                  supported the  link between Pb-associated effects on immune cells  and immune- and
24                  inflammatory-related diseases by providing evidence for changes in intermediary
25                  signaling and inflammatory pathways (Figure 5-34). Several recent epidemiologic studies
26                  examined signaling molecules such as pro-inflammatory cytokines to provide coherence
27                  with toxicological findings. Recent toxicological studies further supported the broader
28                  role of Pb-associated immune modulation in mediating Pb effects in nonlymphoid tissues
29                  (e.g., nervous, reproductive, respiratory systems). Recent epidemiologic studies improved
30                  on the design of previous studies through greater examination of children and adults with
31                  blood Pb levels more comparable to current levels in the U.S. population and greater
32                  consideration for confounding by age, sex, smoking, SES indicators, and allergen
33                  exposures. This epidemiologic evidence particularly that from prospective studies,
34                  combined with the extensive toxicological evidence formed the basis of conclusions
35                  about the immune effects of Pb exposure (Section 5.6.8).
      November 2012                             5-430                  Draft - Do Not Cite or Quote

-------
      5.6.2        Cell-Mediated Immunity
                    5.6.2.1      T Cells

 1                  T cells have a central role in cell-mediated immunity including maturation of B cells,
 2                  activation of cytotoxic T cells and macrophages, and interactions with antigen presenting
 3                  cells (APCs). A majority of the evidence for the effects of Pb exposure on T cells was
 4                  provided by toxicological and epidemiologic studies reviewed in the 2006 Pb AQCD
 5                  (U.S. EPA. 2006b). Toxicological evidence consistently links Pb exposure with
 6                  alterations in T cells with observations of Pb-induced shifts in the partitioning of CD4+
 7                  (T helper) cell populations to favor Th2 cells in vitro (25-50 (iM Pb chloride, 3-5 days)
 8                  (Heoetal.. 1998; 1996). proliferation of Th2 cells over Thl cells in vitro (10, 100 (iM
 9                  Pb chloride, 7 days) (McCabe and Lawrence. 1991). and the production of Th2 cytokines
10                  over Thl cytokines in vitro and in vivo (wide range of Pb concentrations,
11                  Section 5.6.6.1). Epidemiologic findings are limited largely to associations observed
12                  between higher concurrent blood Pb level and lower T cell abundance in children.

13                  In vitro results indicated various mechanisms by which Pb exposure may induce a shift to
14                  Th2 responses including activation of transcription factor NF-KB (regulates T cell
15                  activation) in cultures of human CD4+ T cells (1 (iM Pb acetate, 30 minutes) (Pvatt et al..
16                  1996) and a concentration-dependent (10, 50 (iM Pb chloride, 24 hours) increased
17                  expression of MHC class II surface antigens (e.g., HLA-DR), which mediate the CD4+
18                  response to exogenous antigens (Guo etal.. 1996b). The few available recent
19                  toxicological studies described T cell-dependent and -independent pathways. Kasten-
20                  Jolly  et al. (2010) provided evidence in vivo and with relevant dietary Pb exposures.
21                  While results were based on a microarray analysis of hundreds of genes, which is subject
22                  to higher probability of chance findings, they were supported by the extant evidence base.
23                  In this study, gestational-lactational Pb acetate exposure of BALB/c mice (100 (iM in
24                  drinking water of dams GD8-PND21, resultant spleen homogenate level <3 (ig/dL)
25                  altered splenic cell gene expression of cytokines well documented in the literature to be
26                  affected by Pb, including the Th2 cytokine IL-4 and the Thl cytokine interferon-gamma
27                  (IFN-y). These changes occurred with increases in adenylate cyclase 8 and
28                  phosphatidylinositol 3-kinase in the absence of signaling molecules STAT4 or STAT6,
29                  which comprise the preferential signaling pathway for T cells. Similarly, in cultures of
30                  stimulated mouse T cells, Heo et al. (2007) showed that Pb chloride (25 (JVI, 12-24
31                  hours) decreased the IFN-y to IL-4 ratio (indicating a shift to Th2) in the absence of
32                  STAT6. Additionally, Pb blocked production of IFN-y not by affecting gene expression
33                  but by suppressing translation of the protein. This blockage was rescued with the addition
34                  of IL-12, which is a T cell stimulating factor. The STAT results indicated a T cell-


      November 2012                            5-431                  Draft - Do Not Cite or Quote

-------
 1                  independent pathway to skewing toward Th2 responses whereas the IL-12 results pointed
 2                  to a T cell-dependent pathway.

 3                  While a few available recent epidemiologic studies found associations of blood Pb levels
 4                  with Thl and Th2 cytokines in humans (Section 5.6.6.1). the extant evidence for effects
 5                  on T cells in humans is derived largely from previous cross-sectional studies describing
 6                  differences in the abundance of several T cell subtypes that mediate acquired immunity
 7                  responses to antigens. Most studies of children found that higher blood Pb levels were
 8                  associated with lower T cell abundance in serum, primarily CD3+ cells. These
 9                  associations were observed in studies that adjusted for some potential confounding
10                  factors (as described below) (Karmaus et al.. 2005; Sarasua et al.. 2000) and studies
11                  without consideration for potential confounding (Zhao et al., 2004; Lutz et al., 1999). In
12                  children, blood Pb level was less consistently associated with lower abundance of other T
13                  cell subtypes such as CD4+ (helper T) or CD8+ (cytotoxic T). Some studies did not
14                  provide evidence of blood Pb-associated decreases in T cell abundance but did not
15                  consider potential confounding (Hegazy et al., 2011; Belles-Isles et al., 2002).

16                  Associations between blood Pb level and T cell abundance were found in studies that
17                  generally had population-based recruitment. Most studies did not provide sufficient
18                  information to assess the potential for biased participation by Pb exposure and immune
19                  conditions. Most studies had multiple comparisons; however, associations were not
20                  isolated to T cell abundance. Most studies found lower T cell abundance in groups of
21                  U.S. and non-U.S. children (ages 6 months-10 years, n = 73-331) with concurrent blood
22                  Pb levels >10 ug/dL (Zhao et al.. 2004; Sarasua et al.. 2000;  Lutz et al.. 1999).
23                  Associations were inconsistent in comparisons of children with lower blood Pb levels.
24                  Among 331 children in Germany living near (15 km) and distant from industrial
25                  facilities, Karmaus et al. (2005) found that children (ages 7-10 years) with concurrent
26                  blood Pb levels 2.2-2.8 (ig/dL (2nd quartile) had a 9 to 11% lower abundance of several
27                  T cell subtypes (for some subtypes, p <0.05, t-test) compared with children with blood Pb
28                  levels <2.2 (ig/dL (lowest quartile). This study recruited children from schools and
29                  examined multiple exposures, reducing the likelihood of biased participation by Pb
30                  exposure. Monotonic decreases were not found across blood Pb groups. Compared with
31                  other studies of T cells, Karmaus et al. (2005) had greater consideration for potential
32                  confounding, adjusting for sex, age, number of infections in the previous  12 months,
33                  number of cigarettes/day smoked in the home in the previous 12 months, serum lipids,
34                  and blood organochlorine levels. SES was not examined. Cord blood levels of
35                  organochlorine and Hg but not Pb were associated with T cell abundance in 96 newborns
36                  from a subsistence fishing community and an urban center in Quebec, Canada with a
37                  population mean cord blood Pb level <2 (ig/dL (Belles-Isles et al.. 2002).
      November 2012                            5-432                  Draft - Do Not Cite or Quote

-------
 1                  Another study of children from multiple unspecified U.S. locations with and without
 2                  mining and smelting operations that considered confounding also found an association
 3                  between higher concurrent blood Pb level and lower T cell abundance, albeit limited to
 4                  the youngest subjects (Sarasua et al. 2000). Among 241 children ages 6-35 months, a
 5                  1 (ig/dL higher blood Pb level was associated with a 0.18% (95% CI: -0.34, -0.02) lower
 6                  CD3+ cell count, a 0.10% (95% CI: -0.24, 0.04) lower CD4+ cell count, and a 0.04%
 7                  (95% CI:  -0.15, 0.07) lower CD8+ cell count, with adjustment for location of residence,
 8                  age, and sex. In older age groups (36-71 months, 6-15 years), many effect estimates were
 9                  positive. Analysis of blood Pb level categories indicated that associations were influenced
10                  by lower T cell abundance (3-6%) among children ages 6-35 months with blood Pb levels
11                  >15 (ig/dL. Notably, 76% of subjects lived near a Pb  smelting operation, were likely to
12                  have higher blood Pb levels, and may have influenced the observed associations. Neither
13                  Karmaus et al. (2005) nor Sarasua et al. (2000) found a monotonic decrease in T cell
14                  abundance across blood Pb level groups.  Neither of these studies adjusted for SES, which
15                  has been associated with blood Pb levels  and immune-related conditions such as asthma,
16                  allergy, and respiratory infections.

17                  In the few studies with nonoccupationally-exposed adults (U.S, Italy), higher concurrent
18                  blood Pb levels were associated with higher T cell abundance (Boscolo et al.. 2000;
19                  Sarasua et al.. 2000; Boscolo etal.. 1999). the functional relevance of which is unclear.
20                  These  studies included healthy subjects and those with allergies, a wide  range of samples
21                  sizes (17-433), ages (16-75 years), and mean blood Pb levels (4.3-11.4 (ig/dL).
22                  Pb-exposed workers in the U.S. and Asia did not consistently have lower or higher
23                  abundance of various T cell subtypes than unexposed controls (Mishra et al., 2010;
24                  Pinkerton et al..  1998; Yiicesoy et al.. 1997b;  Undeger et al.. 1996; Fischbein et al..
25                  1993). The inconsistency among studies was not related to differences in sample size
26                  (20-145),  age (means: 22-58 years), or blood Pb levels (14.6-132 (ig/dL) among Pb
27                  workers. None of the studies of adults considered potential confounding factors,
28                  including other workplace exposures in occupational  studies.

29                  In summary, toxicological studies provided clear evidence  for the effects of Pb exposure
30                  on T cells by demonstrating Pb-induced expansion of Th2 cells and increased Th2
31                  cytokine production. Providing mechanistic evidence, a few recent toxicological studies
32                  found that Pb-induced Th2 skewing may  occur via T  cell-dependent (Heo et al.. 2007)
33                  and -independent pathways (Kasten-Jolly et al.. 2010; Heo et al.. 2007). The most
34                  consistent epidemiologic findings were associations between higher concurrent blood Pb
35                  level (>10 (ig/dL) and lower T cell abundance observed in  children ages 6 months to 10
36                  years (Karmaus etal.. 2005; Zhao et al.. 2004; Sarasua et al.. 2000; Lutzetal. 1999). An
37                  association was found with lower blood Pb levels, i.e., <3 (ig/dL (Karmaus et al.. 2005).
38                  albeit in children ages 7-10 years who may have had higher Pb exposures in earlier
      November 2012                            5-433                  Draft - Do Not Cite or Quote

-------
 1                  childhood. The uncertainties regarding blood Pb levels and the timing and duration of Pb
 2                  exposure contributing to the associations with T cell abundance apply to the evidence as a
 3                  whole since concurrent blood Pb levels in children also reflect past Pb exposures. There
 4                  are several other uncertainties in this evidence, including the temporal sequence between
 5                  Pb exposure and T cell changes, potential selection bias, the concentration-response
 6                  relationship, and potential confounding by factors such as SES and other environmental
 7                  exposures. The implications of Pb-associated changes in T cell abundance are limited
 8                  further by the uncertain functional relevance of small magnitudes of change in T cell
 9                  abundance (1-9% lower CD3+ abundance in groups of children with higher blood Pb
10                  levels) to downstream immune responses. Because toxicological studies examined effects
11                  related specifically to Thl or Th2 responses, toxicological evidence is the major
12                  consideration in drawing  conclusions about the  effects of Pb exposure on T cells.
                    5.6.2.2      Lymphocyte Activation

13                  Lymphocytes (T, B, and natural killer [NK] cells) are activated by reversing the normal
14                  suppression mediated by macrophage-like cells. Their activation is an indicator of
15                  response to antigens. A majority of data on the effects of Pb exposure on lymphocyte
16                  activation is provided by toxicological studies reviewed in the 2006 Pb AQCD that
17                  showed both mitogen-induced expansion and suppression of alloreactive B and T
18                  lymphocytes proliferation with Pb exposures in vivo and in vitro (U.S. EPA. 2006b).
19                  Adding to the mixed nature of evidence, a recent study found that 4-week oral exposure
20                  of 7 week-old Wistar rats to 200 ppm Pb acetate induced proliferation of lymphocytes
21                  within the thymus and submaxillary lymph nodes, primarily by affecting B cells (Teijon
22                  et al.. 2010). Overall T cell  proliferation did not change or was suppressed. Specific T
23                  cell subtypes, CD4+, CD8+ (decreased), CD4-CD8- (elevated) were affected only with
24                  i.p. Pb dosing (p <0.05) and not oral exposure. Using the local lymph node assay, Carey
25                  et al. (2006) found that Pb chloride increased antigen-induced (ovalbumin, OVA) T cell
26                  proliferation in adult female BALB/c mice but administered Pb via injection (25-50 (ig).
27                  The mechanistic basis for Pb effects on lymphocyte activation is not well characterized.
28                  As  discussed in Section 5.6.6.2. changes in NO production appear to be involved (Farrer
29                  et al.. 2008). Gao et al. (2007) described a potential role for dendritic cells. Dendritic
30                  cells that matured in the presence of 25 (iM Pb chloride enhanced alloreactive T cell
31                  proliferation in vitro compared to control dendritic cells.

32                  A few available cross-sectional epidemiologic studies in children and nonoccupationally-
33                  exposed adults provided indirect evidence for Pb-associated lymphocyte activation.
34                  Instead of directly measuring  lymphocyte proliferation, these studies measured the
35                  abundance of cells that expressed HLA-DR, a cell surface marker that indicates both


      November 2012                             5-434                  Draft - Do Not Cite or Quote

-------
 1                  activated lymphocytes and monocytes. These studies had limited consideration for
 2                  potential confounding, which also limits the implications of findings. In the study of
 3                  children (ages 9 months-6 years, Missouri), the mean percentage of HLA-DR+ cells was
 4                  about 2-fold higher (p >0.05, Kruskal-Wallis) in the 19 children with concurrent blood Pb
 5                  levels 15-19 (ig/dL than in children with blood Pb levels 10-14 (ig/dL (n = 61) or
 6                  <10 (ig/dL (n = 178) with adjustment for age (Lutz et al.. 1999). However, activated cells
 7                  were not elevated in 16 children with blood Pb levels 20-44 (ig/dL. Small studies of
 8                  adults without occupational Pb exposure in Italy found that concurrent blood Pb level
 9                  was correlated positively with the percentage of HLA-DR expressing cells in men ages
10                  19-52 years with and without allergies (Spearman r = 0.51, p <0.002, n=17 each, overall
11                  median blood Pb level:  11  (ig/dL) (Boscolo etal.. 1999) but only in women ages 19-49
12                  years without allergies (Spearman r = 0.44, p <0.05, n=25, median blood Pb level:
13                  5.5 (ig/dL) (Boscolo et al.. 2000). Associations also were found with other metals.

14                  Comparisons of Pb-exposed workers and unexposed controls indicated similar levels of
15                  lymphocyte proliferation (< 1% difference) in Pb-exposed workers (n = 10-33, mean age:
16                  32-40 years, blood Pb level range: 12-80 (ig/dL) (Queiroz et al.. 1994b: Cohen et al..
17                  1989) or lower lymphocyte proliferation (8-25%) among Pb-exposed workers (n = 15-39,
18                  mean age: 30-49 years, mean blood Pb level: 14.6-129 (ig/dL) (Mishraet al.. 2003;
19                  Fischbein et al.. 1993; Alomran and Shleamoon. 1988; Kimber et al.. 1986). In the
20                  combined epidemiologic and toxicological evidence, Pb was associated with both
21                  activation and suppression of lymphocyte activation. None of the  studies considered
22                  potential confounding factors, including other workplace exposures, and inconsistency
23                  among studies could not be explained by differences in sample size, age of subjects, or
24                  blood Pb level either. None of the studies provided concentration-response information.
25                  Toxicological studies have demonstrated the selective expansion of Th2 cells and
26                  suppression of Thl cells (Section 5.6.2.1). Therefore, the differential activation of
27                  specific subtypes may not be discernible in studies that measure overall lymphocyte
28                  proliferation.
                    5.6.2.3     Delayed-type Hypersensitivity

29                  Although not widely examined recently, several toxicological studies reviewed in the
30                  2006 Pb AQCD (U.S. EPA. 2006b) and recent reviews (Mishra. 2009; Dietert and
31                  McCabe. 2007) identified a suppressed DTH response as one of the most consistently
32                  observed immune effects of Pb exposure in animal models. A recent study indicated that
33                  this effect may be mediated by dendritic cells. The DTH assay commonly is used to
34                  assess the T cell-mediated response to antigens, i.e., induration and erythema resulting
35                  from T cell activation and recruitment of monocytes to the site of antigen deposition. The


      November 2012                            5-435                 Draft - Do Not Cite or Quote

-------
 1                  DTH response is largely Thl-dependent in that Thl cytokines induce the production of T
 2                  cells specifically directed against the antigen (sensitization) and recruitment of antigen-
 3                  specific T cells and monocytes to the site of antigen deposition (elicitation phase).

 4                  Previous studies demonstrated suppressed DTH in animals after gestational (Chen et al..
 5                  2004: Bunnetal. 200la: 2001b. c; Lee et al.. 200Ib: Chenetal.. 1999: Miller et al.
 6                  1998: Faith etal.. 1979) and postnatal (McCabe et al.. 1999: Laschi-Loquerie et al.. 1984:
 7                  Muller et al.. 1977) Pb acetate exposures. Such observations were made in F344 and CD
 8                  rats, BALB/c and Swiss mice, and chickens. Most studies exposed animals to Pb in
 9                  drinking water and found suppressed DTH in animals with blood Pb levels relevant to
10                  humans (means: 6.75, 25 (ig/dL) (Chen et al.. 2004: Bunnetal.. 200 la) and higher (51 to
11                  >100 (ig/dL) (Bunnetal.. 200 Ib: Chenetal..  1999: McCabe et al..  1999). The
12                  associations of DTH with lower blood Pb levels occurred with gestational Pb exposure.

13                  In some studies that examined Pb exposures at multiple stages of gestation, exposures
14                  later in gestation suppressed DTH in animals (Bunnetal.. 2001c: Lee etal.. 200 Ib).
15                  These latter findings may reflect the status of thymus and T cell development. A recent
16                  study contributed to the robust evidence by indicating a role for dendritic cells in the
17                  Pb-induced suppression of the DTH response. Gao et al. (2007) exposed bone marrow-
18                  derived dendritic cells in vitro to Pb chloride (25 (JVI, 10 days) then the antigen OVA and
19                  injected the cells into naive adult mice. Mice treated with Pb-exposed dendritic cells had
20                  a diminished OVA-specific DTH footpad response compared with mice treated with
21                  dendritic cells not exposed to  Pb.

22                  Evidence indicates Pb-induced suppression of DTH in animals with blood Pb levels
23                  relevant to humans (6.75-25 (ig/dL) produced by gestational Pb exposure via drinking
24                  water of dams. The mode of action  is strongly supported by observations that Pb
25                  suppresses production of the Thl cytokine IFN-y (Section 5.6.6.1).  IFN-y is the primary
26                  cytokine that stimulates recruitment of macrophages, a key component of the DTH
27                  response. In animal studies that also examined IFN-y, the suppressed DTH response was
28                  accompanied by a decreased production of IFN-y (Lee etal.. 2001b: Chen et al.. 1999).
29                  Observations of a concomitant decrease in IFN-y strengthen the link between Pb-induced
30                  inhibition of Thl functional activities and suppression of the DTH response.
                    5.6.2.4      Macrophages and Monocytes

31                  As reported in the 2006 Pb AQCD, based on a large body of toxicological evidence and
32                  some supporting epidemiologic evidence, Pb-induced alteration of macrophage function
33                  was considered to be a hallmark of Pb-associated immune effects (U.S. EPA. 2006b).
34                  Macrophages, which are produced by the differentiation of blood monocytes in tissues,

      November 2012                             5-436                 Draft - Do Not Cite or Quote

-------
 1                  mediate host defense through their role in phagocytosing pathogens and stimulating other
 2                  immune cells. Pb exposure was found to induce macrophages into a hyperinflammatory
 3                  phenotype as indicated by enhanced production of TNF-a, IL-6, and ROS and increased
 4                  metabolism of arachidonic acid into PGE2. Observations in macrophages of Pb-induced
 5                  enhanced production of ROS, suppressed production of NO, impaired growth and
 6                  differentiation, and potentially altered receptor expression [e.g., toll-like receptors])
 7                  provided coherence with the effects of Pb observed on tissue damage and diminished host
 8                  resistance in animals. Several of these findings are described in detail  in Sections 5.6.6.2
 9                  and 5.6.6.3. Because macrophages are major resident populations in most tissues and
10                  organs and also are highly mobile in response to microbial signals and tissue alterations,
11                  their functional impairment in response to Pb exposure may serve as a link between
12                  Pb-induced immune effects and impaired host defense, tissue integrity, and organ
13                  homeostasis in numerous physiological systems.

14                  Some rodent studies indicated reduced macrophage generation or phagocytosis with
15                  gestational or postnatal dietary Pb acetate exposure that produced blood Pb levels (upon
16                  cessation of exposure) relevant to humans, i.e., 8.2 (ig/dL in F344 rats (Bunn et al..
17                  200Ic) and 18 (ig/dL in CBA/J mice (Kowolenko et al., 1991).  Similarly, Knowles and
18                  Donaldson (1997) found that Pb acetate trihydrate (PND1-PND21) induced a decrease in
19                  macrophage phagocytosis but in turkeys with higher blood Pb levels, 42 (ig/dL. In one set
20                  of experiments, CBA/J mice exposed to Pb acetate for 2 weeks  with blood Pb levels of
21                  18  (ig/dL had reduced macrophage generation (Kowolenko et al.. 1991) in response to
22                  Listeria infection but no change in macrophage phagocytosis (Kowolenko et al.. 1988).
23                  Other animal studies administered Pb through routes that may not be directly relevant to
24                  those in humans. Effects such as decreased macrophage yield, viability, phagocytosis,
25                  chemotaxis, and killing ability were reported in Swiss mice following bacterial infection
26                  and Pb treatment by oral gavage (40 mg/kg Pb nitrate,  oral gavage, 40 days) (Lodi et al..
27                  2011) or injection 10 mg/kg, i.p., 15 days) (Bishayi and Sengupta. 2006).  Lee et al.
28                  (2002) found no change in monocyte abundance in 5-6 week-old chickens treated with
29                  200 (ig Pb acetate via the air sac in ovo at embryonic day 5 or 12. Some (Bussolaro et al..
30                  2008; Zhouetal.. 1985) but not all in vitro studies (De Guise et al.. 2000) also found
31                  Pb-induced (0.2-1,000 (iM Pb chloride or Pb nitrate) reduced phagocytosis. In particular,
32                  Bussolaro et al. (2008) found such effects with a relatively low  concentration of Pb
33                  exposure (0.2 (iM Pb nitrate, 72 hours).

34                  The effects of Pb exposure on macrophages in humans have not been widely examined.
35                  Pineda-Zavaleta et al. (2004) was unique in examining the hyperinflammatory state
36                  specifically in macrophages, and  consistent with the large body of toxicological studies,
37                  found associations of higher concurrent blood Pb level with lower NO release and higher
38                  superoxide anion release from macrophages isolated from child sera (Sections 5.6.6.2 and
      November 2012                            5-437                  Draft - Do Not Cite or Quote

-------
 1                  5.6.6.3). Other studies in humans examined macrophage abundance in Pb-exposed
 2                  workers, and evidence overall did not clearly indicate an association with concurrent
 3                  blood Pb level. Pinkerton et al. (1998) considered potential confounding and found a
 4                  
-------
 1                  neurodegeneration in brain tissue (Struzvnska et al.. 2007). Bone osteoblasts have been
 2                  shown to be affected by Pb exposure (Section 5.9.4). which, given the interactions
 3                  between osteoblasts and osteoclasts (Chang et al.. 2008a). could have implications for
 4                  development of arthritis [reviewed in Zoeger et al. (2006)1. In vitro, 1 (iM Pb acetate
 5                  elevated TGF-(3 production and cartilage formation in limb bud mesenchymal cells
 6                  (Zuscik et al.. 2007). Kaczynska et al. (2011) reported effects on alveolar macrophages
 7                  after Pb acetate treatment (i.p. 25 mg/kg, 3 days, resulting in blood Pb levels of
 8                  2.1 (ig/dL) in Wistar rats. Macrophages infiltrated airways, limiting air space available to
 9                  gas exchange and contained parts of phagocytized surfactant and alveolar lining. Resident
10                  immune cells in reproductive organs have been shown to be affected by high
11                  concentration Pb exposure. In male BALB/c mice, Pace et al. (2005) found that 0.1 ppm
12                  Pb acetate exposure in drinking water PND1-PND42 (mean peak blood Pb level:
13                  59.5 (ig/dL) resulted in sterility concomitantly with a decrease in the testicular
14                  macrophage population and an increase in apoptotic testicular cells.

15                  In summary, an extensive toxicological evidence base demonstrates that Pb exposure
16                  decreases functionality of macrophages and promotes a hyperinflammatory phenotype.
17                  Animals with dietary Pb exposure resulting in blood Pb levels (upon cessation of
18                  exposure) relevant to humans, 8.2 and  18 (ig/dL, had reduced macrophage generation and
19                  phagocytosis (Bunn et al., 200Ic; Kowolenko et al., 1991). Some in vitro studies
20                  (Bussolaro et al.. 2008; Zhouetal.. 1985) provided supporting evidence. Several
21                  observations link Pb exposure to impaired function and/or structure of specialized
22                  macrophages in nonlymphoid tissue, including liver Kupffer cells and alveolar
23                  macrophages. The results suggest that immune dysfunction may contribute to the effects
24                  of Pb on dysfunction in nonlymphoid tissues  and provides a link between immune
25                  dysfunction and disease in other organ systems. However, the implications of findings to
26                  effects in humans are uncertain because in several studies, animals were treated with Pb
27                  by injection and/or had high blood Pb levels.  Evidence  for Pb-induced decreases in
28                  macrophage functionality provides mode of action support for observations of
29                  Pb-induced decreased host resistance in animals (Section 5.6.5.1). The sparse
30                  epidemiologic evidence is not conclusive but was a lesser consideration than the
31                  toxicological evidence in drawing conclusions about the effects of Pb exposure on
32                  macrophages because  most epidemiologic studies did not examine the functional state of
33                  macrophages. The study that examined the functional state of macrophages,
34                  i.e., mediators of host defense or inflammation, found an association with blood Pb level
35                  in children (Pineda-Zavaleta et al., 2004) consistent with toxicological evidence.
36                  Occupational studies examined abundance of monocytes or markers of activated
37                  macrophages plus other antigen presenting cells, and evidence did not clearly indicate a
38                  difference in Pb-exposed workers. Inconsistency among studies  was not related to
39                  differences in sample size, age, or blood Pb levels of Pb-exposed workers. None of the

      November 2012                            5-439                  Draft - Do Not Cite or Quote

-------
 1                   studies considered the potential influence of other occupational exposures or provided
 2                   concentration-response information.
                     5.6.2.5      Neutrophils

 3                   In the 2006 Pb AQCD, Pb exposure was not judged to have a strong effect on
 4                   neutrophils, which comprise the majority of polymorphonuclear cells (PMNs) (U.S. EPA.
 5                   2006b). This conclusion was based on the limited available toxicological evidence as
 6                   compared with that for effects on other immune cells. However, the modulation of
 7                   neutrophil activity may have important consequences on the dysregulation of
 8                   inflammation and ability of organisms to respond to infectious agents. Studies of cultured
 9                   human PMNs (Governaet al.. 1987) and occupationally-exposed adults (Queiroz et al..
10                   1994a; Queiroz et al., 1993; Valentino et al.. 1991; Bergeret et al.. 1990) found
11                   Pb-associated reductions in PMN functionality, as indicated by reduced chemotactic
12                   response, phagocytic activity, respiratory oxidative burst activity, or reduced ability to
13                   kill ingested antigen. These observations were made in groups of Pb-exposed workers
14                   (n = 10-60) with range of mean age 34-41 years and blood Pb levels 14.8-91.4 (ig/dL.
15                   These studies were focused on neutrophil function, and while the evidence did not appear
16                   to be influenced by multiple comparisons, it could have been influenced by publication
17                   bias. In these cross-sectional studies, the temporal sequence between Pb exposure and
18                   neutrophil function cannot be determined. Other limitations across all studies include the
19                   lack of consideration for potential confounding factors, including other workplace
20                   exposures, and high blood Pb levels of Pb-exposed workers.

21                   Instead of examining neutrophil functional activities, the few available recent studies of
22                   animals and occupationally-exposed adults examined neutrophil counts, an increase in
23                   which has been interpreted by some investigators to be a compensatory response to
24                   Pb-induced impairment in neutrophil chemotactic activity and a hyperinflammatory
25                   response. A study in male Wistar rats found that 12 mg Pb spheres implanted in brains
26                   (compared with control glass spheres) resulted in greater neutrophil filtration from day
27                   7-28 with inflammatory-related damage that included apoptosis and indications of
28                   neurodegeneration (Nakao et al., 2010). However, the route of Pb administration has
29                   uncertain relevance to the typical routes of Pb exposure in humans.

30                   Occupational studies produced contrasting results that were not related to the blood  Pb
31                   levels of workers. DiLorenzo et al. (2006) found a Pb-associated higher absolute
32                   neutrophil count (ANC) in analyses that adjusted for potential confounding factors and
33                   showed a concentration-dependent relationship. In an analysis combining 68 ceramic, Pb
34                   recycling, or bullet manufacturing workers and 50 control food plant workers in Italy, a
      November 2012                             5-440                  Draft - Do Not Cite or Quote

-------
 1                   1 (ig/dL higher concurrent blood Pb level was associated with a 21.8 cells/(iL (95% CI:
 2                   11.2, 32.4 cells/(iL) higher ANC. While these results were adjusted for age, current BMI,
 3                   and current smoking status, other workplace exposures were not examined. Pb-exposed
 4                   workers had a mean age 44  years and a geometric mean concurrent blood Pb level of
 5                   20.5 (ig/dL. Controls had a mean age 46.8 years and mean blood Pb level of 3.5 (ig/dL.
 6                   Neutrophilia (n >7,500 cells/mm3) was found in 8 workers described to have medium to
 7                   high Pb exposures (exact blood Pb levels not reported) but no controls had suggesting
 8                   that long-term, higher-level Pb exposures can lead to a biologically meaningful excess of
 9                   circulating neutrophils. Additionally, a blood Pb concentration-dependent relationship
10                   was indicated by observations of a monotonic increase in ANC across increasing blood
11                   Pb level groups: controls, workers with blood Pb levels < 30 (ig/dL, and workers with
12                   blood Pb levels >30 (ig/dL.  Results further indicated an interaction between concurrent
13                   blood Pb level and current smoking. ANC increased across the three blood Pb groups
14                   among current smokers but not nonsmokers. In contrast,  in a study that did not consider
15                   any potential confounding factors, Conterato et al. (In Press) found lower neutrophil
16                   concentrations among 23 battery workers and 50 painters in Brazil with mean concurrent
17                   blood Pb levels of 50.0 and 5.4 (ig/dL, respectively, than among 36 controls with a mean
18                   blood Pb level of 1.5 (ig/dL. Pb-exposed workers did not consistently have higher levels
19                   of eosinophils, basophils, monocytes, or total lymphocytes either.

20                   Support for the decreased neutrophil function found in Pb-exposed workers is provided
21                   by findings of Pb-associated increases in TNF-a (Section 5.6.6.1) and complement,
22                   which are mediators of neutrophil proliferation, survival, maturation, and functional
23                   activation. The complement system is a component of the innate immune system that is
24                   involved in chemotaxis of macrophages and neutrophils and phagocytosis of antigens.
25                   The few available occupational studies found lower complement in Pb-exposed workers
26                   with mean blood Pb levels >60 (ig/dL (Undeger et al.. 1996; Ewers etal.. 1982). higher
27                   than those relevant to the U.S. general population. Neither study considered potential
28                   confounding factors, including other workplace exposures.  The evidence has limited
29                   implications also because the cross-sectional nature  of studies cannot establish the
30                   temporal sequence between Pb exposure and complement.

31                   In summary, previous occupational studies provided evidence for the effects of Pb
32                   exposure on neutrophils by  finding that compared with controls, Pb-exposed workers had
33                   lower neutrophil functionality (Queiroz et al.. 1994a: Queiroz etal.. 1993; Valentino et
34                   al.. 1991; Bergeret et al.. 1990) and lower complement (Undeger et al.. 1996; Ewers et
35                   al.. 1982). which is a mediator of phagocyte functionality. The limited number of recent
36                   epidemiologic studies examined only neutrophil abundance and conclusively did not find
37                   Pb-exposed workers to have higher or lower neutrophil abundance. While there is
38                   evidence for Pb-associated reduced neutrophil functionality, firm  conclusions are not
      November 2012                            5-441                  Draft - Do Not Cite or Quote

-------
 1                  warranted because the results are based on cross-sectional examination of male workers
 2                  with relatively high blood Pb levels (range: 18.6-100 (ig/dL), and they lack consideration
 3                  for potential confounding factors including other occupational exposures, concentration-
 4                  response information, and analogous toxicological evidence.
                     5.6.2.6      Dendritic Cells

 5                   Whereas as research reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b) focused on
 6                   examining T cells (Section 5.6.2.1). recent ex vivo and in vitro results suggest that the
 7                   effects of Pb on suppressing Thl activity and promoting Th2 activity may be a
 8                   consequence of the direct action of Pb on the function of dendritic cells, a major APC.
 9                   Gao et al. (2007) found that 25 (iM Pb chloride exposure for 10 days stimulated dendritic
10                   cell maturation in bone marrow cultures by changing the ratio of cell surface markers
11                   (e-g-, CD86/CD80) that promote Th2 cell development. Additionally, upon activation
12                   with LPS, Pb-matured dendritic cells produced less IL-6, TNF-a, and IL-12 (stimulates
13                   growth and differentiation of T cells) than did control cells but the same amount of IL-10
14                   (inhibits production of Thl cytokines). The effect of Pb in altering the cytokine
15                   expression profile of dendritic cells, in particular, the lower IL-12/IL-10 ratio, may serve
16                   as an important signal to shift naive T cell populations toward a Th2 phenotype.
17                   Supporting a role for dendritic cells in skewing to a Th2 phenotype, ex vivo results from
18                   the same study showed that Pb-naive adult BALB/c mice implanted with Pb-exposed
19                   dendritic cells had suppressed DTH (Section 5.6.2.3) and IgG2a antibody  (Section 5.6.3)
20                   responses (Gao et al.. 2007).
                     5.6.2.7      Natural Killer Cells

21                   Based mostly on studies reviewed in the 2006 Pb AQCD, evidence does not clearly
22                   indicate that Pb exposure affects the innate immune NK cells, which mediate host
23                   defense by killing infected cells. Some epidemiologic studies adjusted for factors such as
24                   age, sex, and smoking but did not find differences in NK cell abundance or level of
25                   functional activity by blood Pb level in children or adults with or without occupational
26                   exposures (n = 145-675) (Karmaus et al.. 2005; Sarasuaet al.. 2000; Pinkerton et al..
27                   1998). Studies in children did not examine potential confounding by SES, and the study
28                   in Pb-exposed workers did not examine other workplace exposures. Other smaller
29                   (n = 30-108) studies that did not consider potential confounding factors found a positive
30                   correlation between blood Pb level and NK cell abundance in adults in Italy (Boscolo et
31                   al.. 2000; 1999) or reported no significant association in children (quantitative results not
32                   reported) (Belles-Isles et al.. 2002). Pb-exposed workers in the U.S., Europe, and Asia

      November 2012                             5-442                  Draft - Do Not Cite or Quote

-------
 1                  (n = 25-141, mean ages: 26-49 years) with higher concurrent blood Pb levels (means:
 2                  6.5-128 (ig/dL) had similar means of NK cell abundance or functional activity as did
 3                  unexposed controls  (n = 10-84, mean blood Pb levels: <2-16.7 (ig/dL, mean ages: 28-47
 4                  years) (Garcia-Leston et al.. 2011; Mishra et al.. 2003; Pinkerton et al.. 1998; Yiicesov et
 5                  al.. 1997b: Undeger et al..  1996; Fischbein et al.. 1993; Kimberetal.. 1986).

 6                  The epidemiologic evidence is not sufficiently informative for drawing conclusions about
 7                  the effects of Pb exposure on NK cells because of its many limitations including cross-
 8                  sectional nature, limited consideration for potential confounding, and lack of
 9                  concentration-response information. However, toxicological evidence equally does not
10                  clearly indicate an effect of Pb on NK cells. A decrease in NK cell activity was found in
11                  6-8 week-old BALB/c mice but with higher Pb exposure than that relevant to humans
12                  (1,300 ppm Pb acetate in drinking water, 10 days, blood Pb level -100 (ig/dL) (Queiroz
13                  etal.. 2011). In an in vitro  study, Fortier et al. (2008) found that Pb chloride
14                  (7.5-20.7  (ig/dL) did not affect NK cytotoxicity compared with the control. However,
15                  Pb chloride was not found  to affect monocytes or lymphocytes either.
      5.6.3        Humoral Immunity

16                  The 2006 Pb AQCD (U.S. EPA. 2006b) described another hallmark effect of Pb on the
17                  immune system to be an enhanced humoral immune response as characterized by
18                  increased production of IgE antibodies (U.S. EPA. 2006b). Several toxicological and
19                  epidemiologic studies (Table 5-32) demonstrated Pb-associated increases in IgE, which
20                  mediates inflammation in allergic and allergic asthma responses by binding to mast cells
21                  and releasing histamines, leukotrienes, and interleukins upon exposure to an allergen.
22                  Neither toxicological nor epidemiologic evidence (Table 5-32) consistently indicated that
23                  Pb exposure was associated with changes in other classes of Igs including IgG, IgM, and
24                  IgA, which function in complement activation and host resistance or activation of
25                  immune cells.

26                  In toxicological evidence, there was a lack of coherence between results for IgE and
27                  activation of B cells, which regulate IgE production through differentiation into antibody-
28                  producing cells. Pb chloride exposure in vitro (10 (iM up to 5 days) was found to increase
29                  markers of B cell activation, including cell surface markers and levels of plaque forming
30                  cells (PFC), which are a measure of antibody-forming cells (McCabe and Lawrence.
31                  1990; Lawrence. 198 la). However, several studies in animals (Swiss mice and rabbits)
32                  found a wide range of Pb concentrations (0.5 to 250 ppm Pb acetate or tetraethyl Pb for
33                  3-10 weeks, postnatal via drinking water) to decrease levels of PFC (Blakley et al.. 1980;
34                  Koller and Kovacic. 1974; Koller. 1973). Among many mice strains tested, Mudzinski et
      November 2012                            5-443                  Draft - Do Not Cite or Quote

-------
 1                  al. (1986) found an increase in PFC only in BALB/c mice with 8-week postnatal dietary
 2                  Pb acetate exposure that produced high blood Pb levels, 70 (ig/dL. Epidemiologic studies
 3                  of children (Table 5-32) and adults (Table 5-32) with group comparisons and (Boscolo et
 4                  al.. 2000; Boscolo et al.. 1999) with correlation analyses) did not find a consistent
 5                  association between blood Pb level and the abundance of B cells, which may not reflect
 6                  activation. Inconsistencies among studies did not appear to be related to differences in
 7                  age, group blood Pb levels, or the extent of consideration for potential confounding
 8                  (Table 5-32).

 9                  Most animal studies found Pb-induced increases in IgE, with key evidence provided by
10                  studies that examined Pb acetate exposure through drinking water during the gestation
11                  and/or lactation period and IgE (Snyder et al.. 2000; Miller et al.. 1998). In particular,
12                  Snyder et al. (2000) found elevated IgE in juvenile BALB/c mice with relevant blood Pb
13                  levels, means 5-20 (ig/dL measured 0-1 week after gestational and/or lactational Pb
14                  exposure. In Miller et al. (1998). elevated IgE was found in adult mice exposed
15                  gestationally to Pb via drinking water of dams who had blood Pb levels of 30-39 (ig/dL.
16                  Chen et al. (2004) did not find gestational dietary  Pb acetate exposure to result in an
17                  increase in IgE in adult F344 rats who had blood Pb levels of 6.75 and 8 (ig/dL, measured
18                  one week postweaning. In BALB/c and OVA-transgenic (produce OVA-specific T cells)
19                  mice, Heo et al. (1997; 1996)  found concomitant Pb-induced increases  in IgE and IL-4,
20                  consistent with the mode of action of IL-4 to induce class switching of B cells to IgE
21                  producing cells. However, these effects were observed with Pb administered via s.c.
22                  injection (50 (ig/100 (iL, 3 times per week for 3 weeks) and with higher blood Pb levels,
23                  38 (ig/dL, than those relevant to humans. Some of these studies examined multiple
24                  immune endpoints; however, chance findings due to multiple comparisons likely are not
25                  responsible for the findings because the pattern of results consistently pointed to a shift
26                  from Thl to Th2 responses (Miller etal.. 1998; Heoetal.. 1997; 1996).
      November 2012                            5-444                  Draft - Do Not Cite or Quote

-------
Table 5-32   Comparison of serum immunoglobulin levels and B cell abundance
               among various blood Pb groups.

Study

Study Population and
Methodological Details
Blood Pb
Level
(ug/dL)

|gEa

igGb

igMb

igAb

B cells0
Children
Karmaus et
al. (2005)
331 children, ages 7-1 0 yr, Hesse,
Germany
Cross-sectional. School-based
recruitment. No information of
participation rate. Most subjects live
<2.2
2.21 -2.83
2.84-3.41
>3.41
46 (1 .0)
30 (0.65)
59 (1 .28)
59 (1 .28)d
1210
1214
1241
1201
150
143
153
148
123
121
133
136
418e(1.0)
353 (0.84)
389 (0.93)
393 (0.94)
           near industrial facilities. Multiple
           exposures examined. Results adjusted
           for age, sex, number cigarettes/day
           smoked in home in previous 12 months,
           number of infections in the previous
           12 months, serum lipid level, and blood
           organochlorine level. No consideration
           for potential confounding by SES,
           allergens. Monotonic C-R for IgE except
           in highest blood Pb group.
Sarasua et
al. (2000)







Sarasua et
al. (2000)


Sarasua et
al. (2000)


Lutz et al.
(1999)






Hegazy et
al. (2011)



Zhao et al.
(2004)
382 children, ages 6-30 mo, Multiple
U.S. locations
Cross-sectional. No information on
participation rate. Large proportion with
residence near Pb sources.
Comparison group age- and
demographically-matched. Results
adjusted for age, sex, and study
location. No consideration for potential
confounding by SES, allergens.
Inconsistent C-R.
562 children, ages 36-71 mo, Multiple
U.S. locations
Same methodology as above.

675 children ages 5-16 yr, Multiple U.S.
locations
Same methodology as above.

279 children, ages 9 mo-6 yr,
Springfield, MO
Cross-sectional. Recruitment from
public assistance and Pb poisoning
prevention program. No information on
participation rate. Results adjusted for
age. Lack of rigorous statistical
methods. No consideration for potential
confounding by SES, allergens.
Monotonic C-R for IgE except in highest
blood Pb group.
31 8 children, ages 6 mo-7 yr, Egypt
Cross-sectional. Clinic-based
recruitment. No information on
participation rate. Lack of rigorous
statistical methods. Potential
confounding not considered. No
monotonic C-R.
73-75 children, ages 3-6 yr, Zhejiang
Province, China
0.6-4.9
5-9.9
10- 14.9
> 15






0.6-4.9
5-9.9
10- 14.9
> 15
0.6-4.9
5-9.9
10- 14.9
> 15
<10
10-14
15-19
20-69





<5
5-9
10-14
15-19
20-44
45-69
<10
> 10
609
666d
680d
630






817
813
856
835
1,031
1 ,094d
1,048
1,221
51.8(1.0)
74.0 (1 .43)
210.7(4.07)
63.7 (1 .23)d





13.0(1.0)
12.0(0.92)
20.8 (1 .60)
14.9(1.15)
20.4 (1 .57)
10.2(0.78)d


103 50.1
108 55.0
105 58.2
124d 61 .4d






120 88.6
116 90.9
125 96.3
121 94.1
128 140
131 143
136 140
106 108















19.1 (1.0)
20 (1 .05)
20.4 (1 .07)
22.2(1.16)






18.4(1.0)
17.6(0.96)
19.2(1.04)
18.6(1.01)
16.1 (1.0)
15.8(0.98)
15.3(0.95)
20.1 (1.25)
13.4(1.0)
12.6(0.94)
16.9(1.26)
11.1 (0.83)










16.6(1.0)
16.8(1.01)
November 2012
5-445
Draft - Do Not Cite or Quote

-------
Study
Sun et al.
(2003)
Study Population and
Methodological Details
Cross-sectional. School-based
recruitment. No information on
participation rate. Lack of rigorous
statistical methods. Potential
confounding not considered.
Blood Pb
Level
(ug/dL) |gEa lgGb
<10 32.50 40.53
>10 41.33f 34.76f
lgMb lgAb B cells0
41.53
31 .74f
Adults without Occupational Pb Exposures
Sarasua et
al. (2000)
Adults with
Pinkerton et
al. (1998)
Fischbein et
al. (1993)
Kimber et
al. (1986)
Heoetal.
(2004)
Anetorand
Adeniyi
(1998)
Ewers et al.
(1982)
433 children and adults, ages 16-75 yr,
Multiple U.S. locations
Same methodology as that in children.
Occupational Pb Exposures
84 hardware factory controls, mean age
30 yr
145 male Pb smelter workers, mean
age 33 yr, U.S., exact location NR
Cross-sectional. Results adjusted for
age, race, current smoking status, and
workshift. No consideration for potential
confounding by other workplace
exposures, SES
36 industrial worker controls, mean age
47 yr
36 firearms instructors, mean age 49 yr
15 firearms instructors, mean age 48 yr
New York metropolitan area
Cross-sectional. Lack of rigorous
statistical methods. Potential
confounding not considered.
21 unexposed male controls, ages
20-60 yr
39 male tetraethyl Pb plant workers,
ages 25-61 yr, U.K.
Cross-sectional. Lack of rigorous
statistical methods. Potential
confounding not considered.
606 Pb battery plant workers, Korea
Cross-sectional. Monotonic C-R found;
lack of rigorous statistical methods.
Potential confounding not considered.
50 male controls, ages 22-58 yr
80 male Pb-exposed workers, ages
21-66yr, Nigeria
Cross-sectional. Lack of rigorous
statistical methods. Potential
confounding not considered.
53 male various occupation controls,
ages 21-54yr
72 male Pb battery/smelter workers,
ages 16-58 yr, Germany
Cross-sectional. Lack of rigorous
statistical methods. Potential
confounding not considered.
0.6-4.9 1 ,099
5-9.9 1 ,085
10-14.9 1,231
>15 1,169

<2 1 ,090
Mean: 39 1,110
NR
Mean: 14.6
Mean: 31.4
Mean: 11. 8 1062
Mean: 38.4 1018
<10 112.5(1.0)
10-29 223.3(1.99)
> 30 535.8 (4.76)d
Mean: 30.4 1,997
Mean: 56.3 1,1 87d
Mean: 11. 7 1939
Mean: 59.0 171
175 252 13.9(1.0)
175 242 13.0(0.94)
262d 283 12.4(0.89)
139 193 14.8(1.06)

94.5 180 14.6(1.0)
106.2 202 13.2(0.90)
8.6(1.0)
10.5(1.22)
11.2(1.3)"
1294 2235
1040 2425

215 188
191 144d
1619 1409
127 128
November 2012
5-446
Draft - Do Not Cite or Quote

-------
                  Study Population and
        Study    Methodological Details
                                 Blood Pb
                                 Level
                                 (ug/dL)
            igEa
lgGb
lgMb   lgAb   B cells0
      Undeger et   25 male university worker controls,
      al. (1996)     ages 22-56 yr
                  25 male Pb battery plant workers, ages
                  22-55 yr, Turkey
                  Cross-sectional. Lack of rigorous
                  statistical methods. Potential
                  confounding not considered.
                                                   Mean: 16.7
                                 Mean: 74.8
                                                          1202.1    140.4   210.3  545.5e(1.0)
                        854.6d    93.3d   168.1   635.9e (1.2)
      Alomran
      and
      Shleamoon
      (1988)
18 management personnel controls
39 Pb battery workers, mean age 35 yr
Iraq
Cross-sectional. Controls age matched.
Lack of rigorous statistical methods.
Potential confounding not considered.
NR
NR
1713
1610
       183
       170
      Note: Results are presented in order of quality of study design and methodology.
      algE data are presented as lU/mL unless otherwise specified. (In parentheses are ratios of IgE in the higher blood Pb group to IgE in
      the lowest blood Pb group.)
      bOther Ig data are presented as mg/dL unless otherwise specified.
      °B cell data are  presented as the percentage of B cells among all lymphocytes unless otherwise specified. (In parentheses are the
      ratio of B cells in the higher blood Pb group to B cells in the lowest blood Pb group.)
      dp <0.05 for group differences.
      eData represent the number of cells/uL serum.
      'Data represent the mean rank for Mann-Whitney U test, p = 0.07 for IgE.
      9Data are presented as lU/mL.
 1
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11

12
13
14
15
16
17
18
19
20
    A small number of available recent toxicological studies examined IgG subtypes, and as
    in previous studies, found inconsistent effects of Pb exposure. Kasten-Jolly et al. (2010)
    examined 100 (iM Pb acetate in drinking water of BALB/c dams GD8-PND21 because it
    produced blood Pb levels relevant to humans in another study, i.e., 10-30 (ig/dL (Snvder
    et al.. 2000). Pb-exposed pups had increases in the expression of genes encoding Ig
    antibodies and those involved in B lymphocyte function and activation. These genes
    included those for the heavy chain of IgM, IL-4, IL-7 and IL-7 receptor, IL-21, RAG-2,
    CD antigen 27, B-cell leukemia/lymphoma 6, RNA binding motif protein 24,
    Histocompatibility class II antigen A (beta 1), Notch gene homolog 2, and histone
    deacetylase 7A. These results were produced by amicroarray analysis of hundreds of
    genes, which is subject to a higher probability  of finding an effect by chance.

    Other recent studies examined specific IgG subtypes, did not find Pb-induced changes in
    a consistent direction, and thus did not clearly  indicate a shift to Th2 responses. A
    limitation of this evidence is the higher blood Pb levels of animals than those relevant to
    humans. Fernandez-Cabezudo et al. (2007) reported evidence for a shift to Th2 responses
    following Salmonella infection in C3H/HeN mice exposed postnatally to IxlO4 (iM
    Pb acetate in drinking water for 16 weeks (resultant mean blood Pb level: 106 (ig/dL).
    Relative to control mice, production of the Th2 cytokine IL-4 increased in spleen cells of
    Pb-exposed mice after infection as did serum levels of Salmonella-specific IgGl.
    Infection increased Thl-mediated IgG2a levels in control but not Pb-exposed mice.
      November 2012
                                 5-447
                       Draft - Do Not Cite or Quote

-------
 1                  In contrast, Gao et al. (2006) found a Pb-induced increased IgG2a/IgGl ratio, albeit via
 2                  i.p. injection (50 (ig Pb chloride, 3 times per week for 3 weeks), high blood Pb levels
 3                  (65 (ig/dL),  and in a highly-specialized strain of adult knockout mice lacking the ability
 4                  to produce IFN-y. This result was surprising given evidence that IFN-y usually directs
 5                  secretion of IgG2a; however, the authors suggested that in these knockout mice, Pb may
 6                  initiate a Thl response via an IFN-y independent pathway to enhance IgG2a production.
 7                  Carey et al.  (2006) found concentration-dependent increases in both IgG2a- and IgGl-
 8                  producing cells (after 7 days) in adult BALB/c mice treated with subsensitizing doses of a
 9                  T cell-independent (Trinitrophenyl-Ficoll [TNP-Ficoll]) or T cell-dependent (TNP-
10                  ovalbumin [TNP-OVA]) hapten-protein conjugate and 25-50 (ig Pb chloride by bolus
11                  injection. These results indicated stimulation of both Thl- and Th2-mediated
12                  mechanisms. Pb treatment also increased the numbers of T and B cells and IgM-
13                  producing cells in the lymph node against both TNP-Ficoll and TNP-OVA. The increase
14                  in IgM-producing cells against TNP-Ficoll indicated a T-cell independent mechanism.
15                  Despite finding increases in both IgGl- and IgG2a-producing cells, the authors concluded
16                  that Pb skewed the response toward Th2 based on observations of Pb-induced increases
17                  in T and B cells and suppression of DTH. Thus, the results  indicated the potential for Pb
18                  to promote allergic sensitization against T-dependent antigens.

19                  Observations of Pb-induced increases in IgE in animals provide biological plausibility for
20                  associations observed between higher Pb biomarkers levels and higher serum IgE levels
21                  in various populations of children, although a monotonic concentration-dependent
22                  increase was not consistently observed (Hegazy et al.. 2011; Hon et al.. 2010; Hon et al..
23                  2009; Karmaus et al., 2005; Annesi-Maesano et al.. 2003; Sun et al., 2003; Lutz et al..
24                  1999) (Table 5-32). The evidence was based on cross-sectional analyses which preclude
25                  establishing the temporal sequence between Pb exposure and IgE. Associations between
26                  blood Pb level and IgE were found in studies that generally had population-based
27                  recruitment. Most studies did not provide sufficient information to assess the potential for
28                  biased participation by Pb exposure and immune conditions. Most studies examined
29                  multiple immune endpoints; however, associations were not isolated to IgE.

30                  Karmaus et al. (2005) had greater adjustment for potential confounding factors and
31                  examined associations with lower blood Pb levels. Compared with 82 children (ages 7-10
32                  years) in Germany with concurrent blood Pb level <2.2 (ig/dL, children with blood Pb
33                  levels 2.84-3.41 (ig/dL (n = 86) and >3.4 (ig/dL (n = 82) had 28% higher serum IgE
34                  levels (p = 0.03, F-test). These differences were found with the adjustment for age, blood
35                  organochlorine levels, serum lipid levels, number of infections in the previous 12 months,
36                  and number of cigarettes/day smoked in the home in the previous  12 months. SES was
37                  not examined. A monotonic increase in IgE was found across blood Pb quartiles, except
38                  for the highest group (Table 5-32). Although IgE was elevated in children with relatively
      November 2012                            5-448                  Draft - Do Not Cite or Quote

-------
 1                  low blood Pb levels (>2.84 (ig/dL), in these children ages 7-10 years, the contribution of
 2                  higher past Pb exposures cannot be excluded. Similar differences in IgE count on
 3                  basophils were not found among the blood Pb quartiles. Although serum IgE and
 4                  basophil-bound IgE have been correlated in adults (Malveaux et al.. 1978; Conrov et al..
 5                  1977). few data are available in children (Dehlink et al., 2010). A study in children found
 6                  that serum IgE levels were not correlated with basophil-bound IgE (Spearman r = -0.003)
 7                  but were correlated with other IgE receptor-expressing cells such as dendritic cells and
 8                  monocytes (Spearman r = 0.43 to 0.65, p <0.05) (Dehlink etal. 2010). The number of
 9                  IgE-bound basophils also has been highly variable across individuals, particularly
10                  children (Hausmann et al., 2011; Dehlink et al., 2010). Thus, it is not unexpected that
11                  higher blood Pb level was associated with higher serum IgE but not basophil-bound IgE
12                  counts in Karmaus et al. (2005). In this study, blood Pb level was not associated with
13                  serum levels of IgG, IgA, IgM or B cell abundance. Lutz et al. (1999) found higher serum
14                  IgE in low  SES children (9 months-6 years) on public assistance in Springfield, MO after
15                  only adjusting for age, albeit with concurrent blood Pb levels >10 (ig/dL (n = 105/279).

16                  Recent studies in children (non-U.S.) also reported associations between concurrent
17                  blood Pb level and elevated serum IgE but did not adjust for potential confounding
18                  factors (Hegazy et al.. 2011: Honetal.. 2010: Hon et al.. 2009). Ron et al. (2010: 2009)
19                  demonstrated associations  in 110 children (mean age 9.9 years) with atopic dermatitis in
20                  Hong Kong with low blood Pb levels (range: 1.4-6.0 (ig/dL) and found that blood Pb
21                  level  also was correlated with severity of atopic dermatitis, a condition commonly
22                  characterized by elevated IgE levels. Among 318 children, ages 6 months to 7 years, in
23                  Egypt, Hegazy et al. (2011) did not find a monotonic increase across the blood Pb groups.

24                  Sarasua et al. (2000) did not examine IgE but found associations of higher concurrent
25                  blood Pb level with higher IgA, IgG, and IgM in 372 U.S. children ages 6-35 months but
26                  not older (36-71 months, 6-15 years, 16-75 years, n = 433-673). In the youngest age
27                  group, a 1 (ig/dL higher blood Pb level was associated with a 0.8 [95% CI: 0.2, 1.4], 4.8
28                  [95% CI: 1.2, 8.4], and 1.0 [95% CI: 0.1, 1.9] mg/dL higher IgA, IgG, and IgM,
29                  respectively, adjusted for age, sex, and location. In the youngest children, serum levels of
30                  all three examined Igs were elevated among the 24 children with concurrent blood Pb
31                  levels > 15 (ig/dL than among the 165 children with blood Pb levels <5 (ig/dL.

32                  While most epidemiologic studies examined concurrent blood Pb levels, some studies
33                  indicated that prenatal Pb exposure may impact Ig levels in newborns (Annesi-Maesano
34                  et al.. 2003: Belles-Isles et al.. 2002). While these studies better indicated the temporal
35                  sequence between Pb exposure and Ig levels, an important limitation was the lack of
36                  extensive consideration for potential confounding factors. Belles-Isles et al. (2002)
37                  examined 97 newborns in Quebec, Canada (from subsistence fishing communities and
      November 2012                            5-449                  Draft - Do Not Cite or Quote

-------
 1                  two other towns, geometric mean blood Pb levels:  1.64 and 1.32 (ig/dL, respectively) and
 2                  found an association between higher cord blood Pb level and higher cord serum IgG
 3                  adjusted for prenatal maternal smoking status. However, IgG also was associated with
 4                  cord plasma organochlorines, which also are elevated with high fish diets. Annesi-
 5                  Maesano et al. (2003) found that Pb level in infant hair (mean: 1.38 ppm, n = 234) but not
 6                  cord blood (mean: 6.7 (ig/dL, n = 326) or placenta (mean: 9.6 (ig/dL, n = 332) was
 7                  associated with cord serum IgE in newborns in Paris. Potential confounding was not
 8                  considered. The authors inferred a stronger effect of Pb exposure integrated over the
 9                  entire gestational period compared to exposures closer to birth. However, an empirical
10                  basis for interpreting Pb levels in hair has not been established (Section 4.3.4.2). Cotinine
11                  was not associated significantly with Pb biomarker levels or IgE. The correlation
12                  (Spearman r = 0.21, p <0.01)  was larger among the 67% infants with mothers without
13                  allergies than infants with maternal allergies (Spearman r = 0.12), pointing to the possible
14                  masking of a blood Pb-IgE association by the stronger association of family history of
15                  allergy.

16                  Blood Pb  level also was associated with IgE in adults without (Pizent et al..  2008) and
17                  with occupational Pb exposure (Heo et al.. 2004). Pizent et al. (2008)  adjusted for
18                  potential confounding by age, pack years  smoking, and alcohol consumption and found
19                  that higher concurrent blood Pb level was associated with higher IgE in 166 women in
20                  Zagreb, Croatia of similar SES (i.e., white-collar office workers) (Pizent et al.. 2008).
21                  Among women not on hormone replacement therapy or oral contraceptives, a 1 (ig/dL
22                  higher blood Pb level was associated with a 0.60 (95% CI: 0.58, 1.18) higher log of IgE.
23                  Concurrent blood Pb levels were low in these women who were aged  19-67 years (mean:
24                  2.16 (ig/dL, range: 0.56-7.35  (ig/dL); however, the cross-sectional study design makes it
25                  difficult to characterize the temporal sequence between exposure and  outcome or the
26                  timing, level, frequency, and duration of Pb exposure that contributed to the observed
27                  association. Authors did not report an effect estimate in men because it did not attain
28                  statistical  significance. Without quantitative results, it is difficult to ascertain whether
29                  there was  suggestion of association in men but insufficient power to indicate statistical
30                  significance due to the smaller number of men examined (50 men versus 166 females).
31                  Another study of 34 men with and without allergy in Italy also did not report quantitative
32                  results and only indicated a lack of statistically significant correlation between concurrent
33                  blood Pb level (median: 11 (ig/dL) and IgE without considering potential confounding
34                  (Boscolo et al.. 1999V

35                  Limitations of the collective epidemiologic evidence for IgE include the cross-sectional
36                  analyses with limited adjustment for potential confounding. Karmaus  et al. (2005) found
37                  a blood Pb-IgE association with adjustment for age, blood organochlorine levels, serum
38                  lipid levels, number of infections in the previous 12 months, and number of cigarettes
      November 2012                             5-450                  Draft - Do Not Cite or Quote

-------
 1                  a day smoked in the home in the previous 12 months. Lutz et al. (1999) comprised a low
 2                  SES population on public assistance; however, none of the studies adjusted for SES or
 3                  allergen exposure. Lower SES has been associated with poorer housing conditions,
 4                  higher exposures to Pb, allergens, and other factors associated with allergy and asthma.
 5                  Allergen exposure and lower SES are associated with higher IgE and related conditions
 6                  such as allergy and asthma (Bryant-Stephens. 2009; Dowd and Aiello. 2009; Aligne et
 7                  al., 2000). Most  studies did not provide detailed demographic or residential information.
 8                  Thus, there is uncertainty as to the extent to which the blood Pb-IgE associations
 9                  observed in children may be confounded by unmeasured SES and/or allergen exposure.

10                  The cross-sectional nature of evidence raises the possibility that associations are due to
11                  reverse causality, i.e., children with higher IgE may have increased lung permeability,
12                  and consequently, greater uptake of inhaled Pb into the blood. Studies have not directly
13                  compared Pb uptake in groups with different IgE levels; however, animals sensitized with
14                  allergens to produce higher IgE have not had greater uptake of radiolabeled particles than
15                  controls (Turi et al., 2011; Erjefalt and Persson, 1991). Histamine was shown only
16                  transiently to increase uptake of particles in a baboon (n =  1) (Yeates and Hameister.
17                  1992) and in humans, in both those with and without asthma (Rees etal.. 1985; Braude et
18                  al.. 1984; O'Byrne etal.. 1982). Histamine is released by IgE-bound mast cells and
19                  basophils upon exposure to sensitized antigens and leads to inflammation. Compared
20                  with healthy controls (n = 6-9), subjects with asthma (n = 9-13) did not consistently have
21                  greater particle uptake into blood (Del Donno et al., 1997; Rees etal.. 1985; O'Bvrne et
22                  al.. 1984; Elwoodetal..  1983). Thus, evidence does not strongly link higher
23                  inflammation with increased uptake of particles into the blood which reduces the
24                  likelihood that blood Pb-IgE associations observed in children or adults are attributable to
25                  reverse causality.

26                  Most of the epidemiologic evidence about the effects of Pb on IgA, IgG, and IgM levels
27                  is provided by previous studies of Pb-exposed workers (mostly males) from various
28                  industries with mean ages 32-36 years and mean blood Pb levels 38-74.8 (ig/dL (Anetor
29                  andAdeniyi. 1998; Pinkerton et al.. 1998; Undeger et al.. 1996; Queiroz et al.. 1994b:
30                  Alomran and Shleamoon. 1988; Kimber et al., 1986; Ewers et al..  1982). As are
31                  toxicological findings for these other Ig classes, epidemiologic evidence is mixed, with
32                  studies reporting higher, lower, and similar Ig levels in Pb-exposed workers (n = 25-145)
33                  compared with unexposed controls (n = 18-84). Some studies reporting lower Ig levels in
34                  Pb-exposed workers included workers with the highest mean blood Pb levels (>50 (ig/dL)
35                  (Anetor and Adenivi.  1998: Undeger etal.. 1996: Ewers etal.. 1982). The lack of
36                  analysis of potential confounding factors, including other workplace exposures, precludes
37                  characterization  of others factors that may contribute to inconsistent associations in
38                  occupational  studies.
      November 2012                            5-451                  Draft - Do Not Cite or Quote

-------
 1                  In summary, evidence for the effects of Pb exposure on humoral immunity largely
 2                  comprises consistent toxicological and epidemiologic observations of Pb-associated
 3                  increases in IgE. The combined toxicological and epidemiologic results do not clearly
 4                  indicate whether Pb exposure affects IgG, IgM, or IgA. Collectively, epidemiologic
 5                  studies conducted in the U.S., Europe, and Asia, some with large study populations
 6                  (n = 279-331) indicate higher serum IgE in children with concurrent blood Pb levels
 7                  >10 (ig/dL. There was some evidence of associations in non-U.S. children with lower
 8                  blood Pb level; however, the contribution  of higher Pb exposures earlier in childhood
 9                  cannot be excluded (HonetaL 2010; Karmaus et al.. 2005). Pb-associated increases in
10                  IgE were found in children with atopic dermatitis (Hon et al.. 2010; 2009) and children
11                  without immune conditions (Sun et al.. 2003). Other studies did not report the health
12                  status of subjects. Thus, sufficient information was not provided to assess the potential
13                  for selection bias. Among studies that provided concentration-response information,
14                  some found serum IgE to increase across blood Pb level groups, except for the highest
15                  blood Pb groups (Karmaus etal.. 2005: Lutzetal..  1999). In Hegazy et al. (2011). IgE
16                  did not increase monotonically across blood Pb groups. All evidence in humans is based
17                  on cross-sectional analyses, making it difficult to establish the temporal sequence
18                  between Pb exposure and increase in IgE.  Findings of similar particle uptake in subjects
19                  with and without acute inflammation of inflammatory conditions increase confidence that
20                  blood Pb-IgE associations are not due to reverse causality. Most studies did not consider
21                  potential confounding, and none adjusted for  SES. A study in children in Germany and a
22                  study in adults in Croatia adjusted for age  and smoking, with additional adjustment for
23                  blood organochlorine levels in children (Karmaus et al.. 2005) and alcohol consumption
24                  in adults (Pizent et al., 2008). Blood Pb level  was associated with IgE in a low SES
25                  population of children in Michigan (Lutz etal.. 1999) and in a population of female
26                  office workers of similar SES in Croatia (Pizent et al.,  2008). However, uncertainty
27                  remains regarding confounding in the associations observed between blood Pb level and
28                  IgE in humans. Biological plausibility for the epidemiologic evidence is provided by the
29                  Pb-induced increases in  IgE observed in most of the animal studies, with some evidence
30                  at blood  Pb levels relevant to humans (Snvder etal.. 2000; Miller etal.. 1998).
31                  Toxicological evidence indicates increases in IgE with gestational and/or postnatal
32                  juvenile Pb exposure, whereas epidemiologic evidence points to associations with
33                  concurrent blood Pb levels. Because concurrent blood  Pb levels in children reflect both
34                  recent and past Pb exposures, the combined evidence indicates that cumulative Pb
35                  exposures during childhood may affect IgE levels. While evidence for B cell activation is
36                  inconsistent, the mode of action for Pb-induced IgE production is well supported by
37                  extensive toxicological evidence for Pb-induced increases in the Th2 cytokine, IL-4
38                  (Section 5.6.6.1). The coherence between  epidemiologic and toxicological findings for
      November 2012                            5-452                  Draft - Do Not Cite or Quote

-------
 1                  IgE and evidence describing modes of action for increases in IgE supports a relationship
 2                  between Pb exposure and increases in IgE.
      5.6.4        Inflammation

 3                  The 2006 Pb AQCD (U.S. EPA. 2006b) identified misregulated inflammation a major
 4                  immune-related effect of Pb based primarily on consistent toxicological evidence for
 5                  Pb-induced increases in pro-inflammatory cytokines (Section 5.6.6.1). PGE2, and ROS
 6                  (Section 5.6.6.3). Inflammation has been characterized as a major mode of action for Pb
 7                  effects in multiple organ systems such as the liver, kidney, and vasculature given that
 8                  immune cells make up permanent residents and infiltrating cell populations of these other
 9                  organ systems (Section 5.2.5). Inflammation also provides a link between the evidence
10                  for the effects of Pb on modulating immune cell function and production of cytokines and
11                  IgE and the evidence for the effects of Pb on immune-based conditions such as infections
12                  and asthma and allergy. For example, IL-4-induces increases in IgE, which primes
13                  basophils and mast cells to secrete histamine, leukotrienes, and cytokines, which in turn,
14                  produce the inflammation associated with asthma and allergy, i.e., airway responsiveness,
15                  mucus secretion, respiratory symptoms. Pb-induced inflammation also has been
16                  associated with diminished host resistance by inducing local tissue damage. As described
17                  in Section 5.6.6. the few available recent toxicological studies support the effects of Pb
18                  exposure on inflammation with findings of Pb-induced increases in pro-inflammatory
19                  cytokines, ROS and PGE2.

20                  The few available epidemiologic studies have found Pb-associated changes in ROS
21                  release  from macrophages (Section 5.6.6.3) and cytokine levels (Section 5.6.6.1) in
22                  children and adults. Adding to this evidence, recent cross-sectional studies found
23                  associations between blood Pb  level and indicators of inflammation that may be related to
24                  multisystemic effects. As discussed in Section 5.6.3. evidence has not provided strong
25                  evidence for increased particle  uptake in subjects with acute inflammation or
26                  inflammatory conditions, reducing the likelihood of reverse causality. However, because
27                  of the cross-sectional design of studies, the temporal sequence between Pb exposure and
28                  inflammation cannot be established. The most compelling epidemiologic evidence was
29                  provided by studies in adults, which were larger had greater consideration for potential
30                  confounding. The consistent pattern of association observed across endpoints reduces the
31                  likelihood of chance findings due to multiple comparisons.

32                  Strengths of the study of adults (age > 40 years, n = 4,663- 7,342) participating in
33                  1999-2004 NHANES included the examination of several potential confounding factors,
34                  multiple exposures and outcomes in predominately healthy adults and statistical analyses
      November 2012                            5-453                  Draft - Do Not Cite or Quote

-------
 1                  to provide nationally-representative results. Higher concurrent blood Pb level was
 2                  associated with higher serum inflammation markers, C-reactive protein (CRP),
 3                  fibrinogen, and white blood cell (WBC) count, particularly among men (Songdej et al.,
 4                  2010). Results were adjusted for age; sex; race/ethnicity; education; current income;
 5                  physical activity; and several factors related to inflammation including, BMI, smoking
 6                  status, and history of diabetes, inflammatory disease, or cardiovascular disease. For
 7                  women, most ORs for associations between quintiles of blood Pb and tertiles of CRP,
 8                  fibrinogen, and WBC count were <1.0 whereas corresponding ORs in men mostly were
 9                  >1.0. For example, compared with men with concurrent blood Pb level <1.16 (ig/dL, men
10                  with blood Pb levels of 1.16-< 1.63 ng/dL, 1.63-<2.17, 2.17-<3.09 (ig/dL, and
11                  > 3.09 (ig/dL had elevated odds of higher CRP (OR [95% CI]: 2.22 [1.14, 4.32],  1.67
12                  [0.85, 3.28], 2.12 [1.07, 4.21], and 2.85 [1.49, 5.45], respectively). For all inflammation
13                  markers, although the highest OR was found in the highest quintile of blood Pb level
14                  (> 3.09 (ig/dL), monotonic concentration-dependent increases were not observed.
15                  Consistent with NHANES findings, higher concurrent blood Pb level was associated with
16                  higher levels of WBCs and IL-6 with adjustment for age, BMI, and current smoking
17                  status among 300 university  students  in Incheon, Korea (Kim et al.. 2007). Adults with
18                  allergic conditions or using anti-inflammatory medication were excluded; however,
19                  sufficient information was not provided to assess potential selection bias. Larger effects
20                  were  estimated for the 147 men in the upper two quartiles of blood Pb levels,
21                  2.51-10.47 (ig/dL than for the full range of blood Pb levels (n = 150).

22                  Low blood Pb levels also were associated with inflammation in a small genome-wide
23                  association study that included 37 children with autism and  15 children without autism
24                  (ages 2-5 years; blood Pb level range: 0.37 to 5.2 (ig/dL) in  California who were unlikely
25                  to have had higher past Pb exposures. In models that included age, sex, and autism
26                  diagnosis, concurrent blood Pb level was associated with the expression of several genes
27                  related to immune function and inflammation, including HLA-DRB and MHC Class II-
28                  associated invariant chain CD74 (involved in antigen presentation) (Tian et al.. 2011).
29                  Although blood Pb levels were similar between children with and without autism and
30                  correlations with gene expression were observed in both groups, they were in opposite
31                  directions (positive among children with autism and negative among children without
32                  autism). With gene expression arrays, there is a higher probability of chance. Further,
33                  there was limited consideration for potential confounding in this study, and the
34                  representativeness of findings in children with autism may be limited. However, the
35                  results are supported by observations  that Pb chloride (10-100 (iM) increases MHC
36                  molecule surface expression  in mouse and human HLA antigen presenting cells (Guo et
37                  al.. 1996a: McCabe and Lawrence. 1991).
      November 2012                             5-454                  Draft - Do Not Cite or Quote

-------
 1                  In summary, Pb-associated increases in indicators of inflammation such as CRP, WBCs,
 2                  and IL-6 were found in populations mostly comprising healthy adults with concurrent
 3                  blood Pb levels 1.16-10 (ig/dL with adjustment for several potential confounding factors
 4                  (Songdei etal.. 2010; Kim et al. 2007). The analysis of adults participating in NHANES
 5                  was particularly noteworthy for its representative population and adjustment for age, sex,
 6                  race/ethnicity, education, current income, physical activity, and several inflammatory
 7                  conditions (Songdei et al., 2010). Because all evidence was based on cross-sectional
 8                  analyses, the relative contributions of recent and past Pb exposures to the observed
 9                  associations are uncertain. Other lines of evidence do not strongly support reverse
10                  causality; however, the temporal sequence between Pb exposure and inflammation is
11                  difficult to establish. Despite the limited extent and cross-sectional nature of
12                  epidemiologic evidence, the biological plausibility is provided by findings of Pb-induced
13                  increases in Th2 cell partitioning (Section 5.6.2.1) and IL-6 (Section 5.6.6.1) in
14                  toxicological  studies. Th2 cells produce IL-6 which is  the primary stimulus for
15                  expression of CRP and fibrinogen (Hage and Szalai. 2007; Fuller and Zhang. 2001).
      5.6.5       Immune-based Diseases
                    5.6.5.1      Host Resistance

16                  The capability of Pb to reduce host resistance of animals to bacteria has been recognized
17                  for almost 40 years and was supported by several animal studies described in the
18                  2006 Pb AQCD. Several studies demonstrated increased mortality following Pb exposure
19                  through drinking water and infection with Listeria monocytogenes. Multiple
20                  investigations in the same laboratory indicated increases in body burdens of viable
21                  bacteria, mortality, and sickness behavior induced by Listeria exposure in juvenile or
22                  adult BALB/c or CBA/J mice exposed postnatally to 500 to 2,000 (iM Pb acetate in
23                  drinking water for 3 to 8 weeks (Dyatlov and Lawrence. 2002; Kim and Lawrence. 2000;
24                  Kishikawaetal.. 1997; Lawrence. 198 Ib). Decreased bacterial resistance was observed
25                  in mice with blood Pb levels (upon cessation of Pb exposure) relevant to humans,
26                  i.e., 25 (ig/dL in BALB/c mice exposed PND1-PND22 (Dyatlov and Lawrence. 2002)
27                  and 20 (ig/dL in adult C3H/HeN mice with 16-week exposure (Fernandez-Cabezudo et
28                  al., 2007). Other studies found mortality from Salmonella or E. coli or reduced clearance
29                  of Staphylococcus in mice or rats administered Pb acetate or nitrate via injection, a route
30                  of Pb exposure less relevant to humans (Bishavi and Sengupta. 2006; Cook etal., 1975;
31                  Hemphill etal..  1971; Serve etal.. 1966). Although not examined as  much, postnatal
32                  dietary Pb (mostly Pb acetate) exposure for 4-10 weeks increased mortality of mice and
33                  chickens from viral infection (Gupta et al., 2002; Youssef et al.,  1996; Exon etal.. 1979;

      November 2012                            5-455                  Draft - Do Not Cite or Quote

-------
 1                  Thind and Khan. 1978; Gainer. 1977). These effects were observed in animals with high
 2                  blood (71-313 (ig/dL) (Gupta et al. 2002: Thind and Khan. 1978) or tissue Pb levels
 3                  (0.12-0.71 ppm) (Exonetal.. 1979).

 4                  The mode of action for Pb-induced decreased host resistance is well characterized by
 5                  observations that Pb suppresses Thl-driven acquired immune responses and increases
 6                  inflammatory responses in target tissue, which may compromise host protective barriers.
 7                  Host resistance to bacteria such as Listeria requires effective Thl-driven responses
 8                  including the production of IL-12 and IFN-y (Lara-Tejero and Pamer. 2004) and these
 9                  have been found to be inhibited by Pb exposure (Section 5.6.6.1).  The lack of IFN-y can
10                  inhibit appropriate and timely macrophage activation. Strengthening the evidence for
11                  Pb-induced decreased host resistance, Fernandez-Cabezudo et al. (2007) found both
12                  increased mortality and decreased production of IL-12 and IFN-y (ex vivo in spleen cells)
13                  in Salmonella-exposed CH3/HeN mice with blood Pb levels of 20 (ig/dL. Nitric oxide is
14                  produced by activated macrophages and has been found to be suppressed by Pb exposure
15                  (Section 5.6.6.2). Pb-induced decreases in bacterial clearance have been found in
16                  conjunction with reduced NO and macrophage functionality (Bishayi and Sengupta.
17                  2006). Further, Pb-induced inflammation has been demonstrated as increases in ROS and
18                  PGE2 (Section 5.6.6.3). Additional mode  of action evidence was provided recently with
19                  observations that developmental Pb exposure of BALB/c mice (100 (iM Pb acetate in
20                  drinking water of dams from GD8 to PND21) upregulated splenic gene expression of
21                  caspase-12 (Kasten-Jolly et al.. 2010). Caspase-12 is a cysteine protease that functions in
22                  apoptosis and activation of pro-inflammatory cytokines  and has been linked with a role in
23                  the inhibition of bacterial clearance both systemically and in the gut mucosa (Saleh et al..
24                  2006).

25                  In the few available epidemiologic studies, a range of Pb exposure indicators (i.e., cord or
26                  concurrent blood Pb, Pb content in total deposition samples or lichen) was associated
27                  with viral and bacterial infections in children. An increase in infections was associated
28                  with cord blood Pb levels > 10 (ig/dL in children in Boston, MA (n = 283)  (Rabinowitz et
29                  al.. 1990) and a mean concurrent blood Pb level of 3.34  (ig/dL in children in Germany
30                  (n = 311) (Karmaus et al.. 2005).  Similarly, a study found higher frequency of self-
31                  reported colds or influenza among 66 Pb battery or smelter plant workers with blood Pb
32                  levels 21.3-85.2 (ig/dL than among 53 controls with blood Pb levels 6.6-20.8 (ig/dL
33                  (Ewers et al.. 1982). Because of the many limitations, the lack of consideration for
34                  potential confounding (Karmaus et al.. 2005; Rabinowitz et al.. 1990; Ewers etal.. 1982).
35                  lack statistical rigor in comparisons of mean blood Pb levels by number of infections
36                  (Karmaus et al.. 2005; Ewers et al.. 1982). and ecological design (Carreras et al.. 2009).
37                  conclusions about the effects of Pb exposure on viral or bacterial infections cannot be
38                  drawn based on epidemiologic evidence alone. And, the weak epidemiologic data do not
      November 2012                            5-456                 Draft - Do Not Cite or Quote

-------
 1                  detract from the consistent findings in animals for Pb-induced decreased host resistance,
 2                  including those in animals with relevant blood Pb levels, and evidence for modes of
 3                  action including decreased macrophage function and Thl cytokine production.

 4                  With few studies  available, the effect of Pb on resistance to eukaryotic parasites is not
 5                  clear. High concentration Pb acetate (> 30 (iM) diminished the ability of macrophages to
 6                  kill Leishmania enrietti protozoa in vitro (Mauel et al.. 1989). Survival of malaria-
 7                  infected mice was enhanced with 100 (iM Pb nitrate exposure via drinking water (Koka
 8                  et al.. 2007). which was attributed to Pb inducing eryptosis and removal of infected
 9                  erythrocytes and not to Pb-induced alterations in immune function. Nriagu et al. (2008)
10                  found that higher blood Pb level was associated with lower malaria prevalence among
11                  653 children (ages 2-9 years) from three Nigerian cities with a mean blood Pb level of
12                  8.9 (ig/dL. Results were adjusted for age, sex, number of siblings, and other
13                  comorbidities such as depressed mood, headaches, and irritability. Given the well-
14                  characterized effect of Pb in promoting Th2 activity, it is plausible for Pb to enhance host
15                  resistance to parasites that require robust Th2 responses such as helminths. However, this
16                  relationship is not well characterized.
                     5.6.5.2     Asthma and Allergy

17                   lexicological evidence and to a lesser extent, epidemiologic evidence, have supported
18                   the effects of Pb exposure on stimulating Th2 activity, including increasing production of
19                   Th2 cytokines such as IL-4 (Section 5.6.6.1). IgE antibody (Section 5.6.3). and
20                   inflammation (Section 5.6.4). These endpoints comprise a well-recognized mode of
21                   action for the development and exacerbation of asthma and allergy, which are atopic and
22                   inflammatory conditions. Thus, this mechanistic evidence provides support for the small
23                   body of epidemiologic evidence indicating associations of blood Pb levels with asthma or
24                   allergy in children (Figure 5-35 and Table 5-33). Whereas such evidence reviewed in the
25                   2006 Pb AQCD was too sparse to permit conclusions, findings from recent studies add
26                   supporting evidence. Children examined in studies of asthma and allergy encompassed a
27                   wide age range (i.e., <1-12 years) and across studies, blood Pb was measured at different
28                   lifestages. Studies ascertained outcomes with parent report of doctor diagnosis but also
29                   more objectively using a surveillance database or and clinical testing.  This variability in
30                   could contribute to between-study heterogeneity in results; however, the objective
31                   assessment of outcomes in some  studies is not likely to produce a spurious association.
32                   Further, some of the evidence was provided by large studies that prospectively
33                   ascertained outcome incidence after the measurement of blood Pb levels, did not indicate
34                   selection bias, and considered potential confounding by SES and other environmental
35                   exposures. These strengths reduce the likelihood of reverse causality and the influence of


      November 2012                            5-457                  Draft - Do Not Cite or Quote

-------
                 other risk factors and increase confidence that the observed associations reflect a
                 relationship with Pb  exposure.
 Study                 Population         Outcome

 Jedrychowskietal. (2011)  Children


 Rabinowitzet al. (1990)    Children           Eczema
                   Blood Pb Level Mean
                     or Group (ug/dL)
Positive skin prick test   prenatal: 1.16
                   concurrent: 2.02

                   >10vs. <10       —
 i    L  i  i /^nnr\     r*\--i i    i~>     •     IncidentAsthma        ,_    _
 Josephetal. (2005)     Children,Caucasian  _   ....,.,_    >5vs. <5
                                       Requiring Medical Care
                      Children, African American
                      Children, African American
 Rabinowitzet al. (1990)   Children

 Pugh Smith and Nriagu
       (2011)
                     Children
                                       Asthma
                                       Asthma
                   >5vs. <5
                   >10vs. <5
                                                          >10vs.
                                                          >10vs.
                                                                            12345
                                                                                Odds ratio (95% Cl)a
aFor analyses with blood Pb level as a continuous variable, odds ratios are standardized to a 1 ug/dL increase in blood Pb level.
Note: Results are presented first for allergy-related outcomes then for asthma. All results are from prospective analyses, except for
Pugh Smith and Nriagu (2011). Black diamond represents associations with concurrent blood Pb levels, green triangles represent
associations with prenatal (cord) blood Pb levels, and blue circles represent associations with blood Pb levels measured in
childhood up  to 12 months prior to outcome assessment.
Figure 5-35       Associations of blood Pb levels with  asthma and allergy in
                      children.
November 2012
          5-458
Draft - Do Not Cite or Quote

-------
Table 5-33    Additional characteristics and quantitative results for studies
                  presented in Figure  5-35.
                Study Population and
Study	Methodological Details
 Blood Pb Level Data (ug/dL)
        Outcome
Odds Ratio or
Relative Risk
(95% Cl)
Jedrychowski     224 children followed prenatally to
et al. (2011)      age 5 yr, Krakow, Poland
                Prospective. Study of multiple
                exposures and outcomes. Clinical
                assessment of atopy. No information
                on follow-up participation but no
                selective attrition. Logistic regression
                adjusted for sex, parity, maternal
                age, education, and atopy, cord
                blood cotinine, smoker in home
                during follow-up. Also considered
                potential confounding by
                breastfeeding and allergen levels in
                house dust.
 Prenatal (cord): Geometric mean:
 1.16(95% Cl: 1.12, 1.22)
 Concurrent: Geometric means:
 2.02 (95% Cl: 1.95,2.12)
        Positive Skin
        Prick Test
2.3(1.1,4.6)a
                       1.1 (0.7, 1.6)a
Rabinowitz et    159 children followed from birth to
al. (1990)        unspecified age, Boston area, MA
                Prospective. Low participation
                among eligibles. No information on
                differences with nonparticipants.
                Logistic regression with parental
                report of eczema. No consideration
                of potential confounding factors.
 Prenatal (cord) > 10 vs. <10
                                                                                    Eczema
                       1.0 (0.6, 1.6)b
Hon et al.        110 children with atopic dermatitis,
(2010: 2009)°    mean (SD) age: 9.9 (4.6) yr, Hong
                Kong, China,
                Not clear whether subjects were free
                of atopic dermatitis at time of blood
                Pb measurement. Recruitment from
                dermatology clinic. Clinical
                assessment of atopic dermatitis. No
                information on participation rate.
                Examination of multiple metals. Lack
                of rigorous statistical methods. No
                consideration of potential
                confounding factors.
 Serum Pb mean (SD): 1.86 (0.83)
        Atopic
        dermatitis
        severity
r = 0.329,
p O.001
Joseph et al.     4,634 children, ages 1 -3 yr followed
(2005)           for 12 months, Southeastern Ml
                Prospective. Large sample size.
                Indirect assessment of asthma
                diagnosis but ascertained from
                managed care organization claims
                database. Logistic regression
                adjusted for sex, birth weight, and
                average annual income available
                only at census block level. Lack of
                information on other environmental
                exposures.
 Caucasian a 5 vs. Caucasian <5

 African American a 5 vs.
 African American <5

 African American > 10 vs.
 African American <5
 Measured up to 12 mo before
 outcome
        Incident
        asthma
        requiring
        medical care
2.7(0.9,8.1)°

1.1 (0.8, 1.7)d

1.3(0.6, 2.6)d
Rabinowitz et    204 children followed from birth to
al. (1990)        unspecified age, Boston area, MA
                Same methodology as above.
 Prenatal (cord blood)
 > 10 vs. <10
        Prevalent
        asthma
1.3(0.8,2.0)"
November 2012
5-459
Draft - Do Not Cite or Quote

-------
                     Study Population and
      Study	Methodological Details
                              Blood Pb Level Data (ug/dL)
       Outcome
Odds Ratio or
Relative Risk
(95% Cl)
      Pugh Smith      356 children, ages 0-12 yr, Saginaw,
      and Nriagu      Ml
      (2011)          Cross-sectional. Recruitment from
                     blood Pb database. Moderate
                     participation rate. Parental report of
                     asthma diagnosis. Logistic
                     regression adjusted for age, sex,
                     family income, number of stories in
                     unit, cat in home, dog in home,
                     cockroach problem, number of
                     persons in home, smoker in home,
                     clutter, candles/incense, type of
                     cooking stove, main heating
                     source, months of residency, housing
                     tenure, type of air conditioning,
                     peeling paint, ceiling/wall damage,
                     age of housing, water
                     dampness/mold/mildew.
                              Highest blood Pb level at address
                              > 10 vs. <10
                              Levels ascertained from statewide
                              database, specific timing
                              unreported but varied among
                              subjects
       Prevalent
       asthma
       diagnosed
       within previous
       12 months
7.5(1.3,42.9)"
      aOdds ratio presented per 1 ug/dL increase in blood Pb level.
      bOdds ratio in children with blood Pb level a 10 ug/dL with children with blood Pb level <10 ug/dL serving as the reference group.
      °Results are not included in Figure 5-35 because only correlations are presented.
      dRelative risk in each specified subgroup with children with blood  Pb level <5 ug/dL serving as the reference group.
 1

 2
 o
 6

 4

 5

 6

 7

 8

 9

10

11

12

13

14

15

16

17

18

19

20

21

22
Key evidence for Pb-associated effects on allergy-related outcomes was provided by a

prospective study in 224 children in Poland at age 5 years that compared associations of

prenatal (cord and maternal) and concurrent blood Pb levels with incidence of allergic

sensitization, as ascertained by investigators (Jedrychowski et al., 2011). Subjects

assessed at five years did not differ from the full cohort, indicating lack of selective

attrition of subjects by blood Pb level or health status. The potential for selection bias

also is reduced because multiple exposures and outcomes were examined in this cohort.

A 1  (ig/dL increase in prenatal cord blood level was associated with greater risk of

positive skin prick test (SPT, rash/inflammatory  reaction) to dust mite, dog, or cat

allergen with an RR of 2.3 (95% CI: 1.1, 4.6). Concurrent blood Pb level was more

weakly associated with risk of positive SPT (Figure 5-35 and Table 5-33). For prenatal

Pb biomarkers, similar effect estimates were obtained before and after adjustment for sex,

parity, maternal age, education, and atopy, and prenatal (cord blood cotinine) and

postnatal (smoker in the home) smoking exposure. Results were not altered by the

addition of house dust allergen levels.  Cord and concurrent blood Pb  levels were weakly

correlated  (r = 0.29), providing support for an independent association for prenatal Pb

biomarkers. A relationship with Pb was substantiated by  observations that indicators of

other exposures,  including blood levels of Hg, poly cyclic aromatic hydrocarbon DNA

adducts, and residential levels of dust mite or pet allergen were associated with lower

risks of SPT than was blood Pb level. These associations were observed with relatively

low cord blood Pb levels (geometric mean:  1.16  (ig/dL [95% CI: 0.12, 1.22]). However,

cord blood Pb levels reflect the pregnancy blood Pb levels of mothers. Evidence indicates
      November 2012
                             5-460
Draft - Do Not Cite or Quote

-------
 1                   increased mobilization of Pb from bone to blood in pregnant women (Sections 4.2.2.4
 2                   and 4.3.5.2). Thus, there is uncertainty regarding the Pb exposure scenarios that
 3                   contributed to associations between cord blood Pb level and allergic sensitization in
 4                   children examined in Jedrychowski et al. (2011).

 5                   Contrasting results were produced by prospective studies of eczema or atopic dermatitis,
 6                   which are reactions of the skin to sensitized allergens; however, neither study considered
 7                   potential confounding factors. Rabinowitz et al. (1990) found no elevated risk of parental
 8                   reported eczema in children (n = 159) in the Boston, MA area with cord blood Pb levels
 9                   > 10 (ig/dL. Hon et al. (2010; 2009) found a correlation between concurrent serum Pb
10                   levels (mean: 1.86 (ig/dL) and clinical diagnosis of atopic dermatitis severity (e.g., skin
11                   area affected, intensity of rash and inflammation, symptoms) in 110 children
12                   approximately age 10 years in Hong Kong (Spearman r = 0.33, p O.005). The various
13                   other metals were  examined were negatively correlated with atopic dermatitis. Although
14                   Hon et al. (2010; 2009) examined incidence of atopic dermatitis, subjects were selected
15                   from patients referred to a dermatology clinic. The representativeness of the children to
16                   the source population is uncertain, and allergies may already have developed by the time
17                   serum Pb levels were measured.

18                   Prospective and cross-sectional evidence indicated associations with blood Pb levels with
19                   asthma in children. Among prospective studies, Joseph et al. (2005) accounted for
20                   potential confounding factors. Asthma-free children, ages 1-3 years, (n = 4,634) all
21                   members of the same managed care organization in southeastern Michigan were selected
22                   based on availability of blood Pb level data in the database then tracked for the following
23                   12 months for incidence of asthma. Incident asthma requiring a doctor visit or medication
24                   was defined from the medical claims database as four or more asthma medication
25                   dispensing events  and one or  more  asthma emergency department visit, hospitalizations,
26                   or outpatient visits with at least two asthma medication dispensing events in the previous
27                   12 months. While  this definition is not a direct diagnosis, it is used to define persistent
28                   asthma by the Healthcare Effectiveness Data and Information Set, which most U.S.
29                   health plans use to measure health care performance. The records-based analysis
30                   precluded bias due to selective participation of subjects by blood Pb level and health
31                   status. However, because a blood Pb measurement was required, it is uncertain whether
32                   the study population is representative of the managed care organization population. In
33                   analyses that adjusted for average annual income at the census block level, birth weight,
34                   and sex, an elevated risk of incident asthma requiring a doctor visit or medication was
35                   associated with blood Pb levels > 5 (ig/dL in Caucasian children (RR: 2.7 [95% CI: 0.9,
36                   8.1] compared with Caucasian children with blood Pb levels <5 (ig/dL) (Figure 5-35 and
37                   Table 5-33). In analyses restricted to African Americans, children with blood Pb levels
38                   > 10 (ig/dL had an elevated risk of asthma requiring medical care (RR: 1.3 [95% CI: 0.6,
      November 2012                             5-461                  Draft - Do Not Cite or Quote

-------
 1                  2.6] compared with children with blood Pb level <5 (ig/dL). There were small numbers of
 2                  children with asthma requiring medical care in the higher blood Pb level categories,
 3                  which could have accounted for the wide 95% CIs (5 Caucasian children with blood Pb
 4                  > 5 (ig/dL and 9 African American children with blood Pb level > 10 (ig/dL). In analyses
 5                  that used Caucasian children with blood Pb level <5 (ig/dL as the reference group, blood
 6                  Pb level was associated with increased risk of asthma requiring medical care among
 7                  African American children in all blood Pb level categories, which indicated a stronger
 8                  association with race. Nonetheless, results within race groups pointed to an association
 9                  with blood Pb level.

10                  Similarly, the prospective study of 204 children in the Boston, MA area found a
11                  Pb-associated increased risk of parental-reported asthma (age of assessment not
12                  reported), specifically in children with cord blood Pb levels >10 (ig/dL relative to cord
13                  blood Pb levels < 10 (ig/dL (Rabinowitz et al.. 1990) (Figure 5-35 and Table 5-33).
14                  However, potential confounding factors were not examined.

15                  Supporting the prospective evidence, a cross-sectional study  conducted in Saginaw, MI
16                  found a higher prevalence  of parental report of doctor-diagnosed asthma in children (ages
17                  < 12 years) with blood Pb levels > 10 (ig/dL (Pugh Smith and Nriagu. 2011V Similar to
18                  Joseph et al. (2005). the study population was predominately African American (78% of
19                  356). Children were randomly selected from a statewide database of initial blood Pb
20                  measurements collected at unspecified ages, and a positive bias is possible if parents of
21                  children with higher blood Pb levels and asthma were more likely to participate or recall
22                  an asthma diagnosis. Data were collected on asthma diagnosis in the previous 12 months;
23                  thus, for some subjects, blood Pb measurement likely preceded asthma diagnosis. A
24                  strength of this study was the adjustment for a large number of potential confounding
25                  factors such as age, sex, household pets, housing characteristics, and household smoking
26                  family income. Compared with children with initial blood Pb levels <10 (ig/dL, children
27                  with initial blood Pb levels > 10 (ig/dL had a higher odds of having a doctor diagnosis of
28                  asthma within the past 12 months (OR: 7.5 [95%  CI: 1.3, 42.9]). The results were
29                  imprecise, and while this study had more children with blood Pb levels > 10 (ig/dL
30                  (18.6%) than did Joseph et al. (2005). the analyses considered a much large number of
31                  covariates.

32                  As was discussed in Section 5.6.3. one may speculate that cross-sectional associations
33                  could be attributed to reverse causality. Individuals with asthma and animal models of
34                  asthma have been shown to have epithelial cell damage and exudation of cells and fluids
35                  into airways, which are indicators of increased lung permeability. With increased lung
36                  permeability, one may speculate the  potential for greater uptake of Pb from airways into
37                  blood. Most evidence does not demonstrate greater uptake of particles into the blood in
      November 2012                             5-462                  Draft - Do Not Cite or Quote

-------
 1                  subjects with asthma than in healthy controls (Del Donno et al.. 1997; Rees et al.. 1985;
 2                  O'Byrne etal.. 1984; Elwoodetal.. 1983). Histamine was found to increase particle
 3                  uptake transiently. In Rees et al. (1985). histamine equally increased particle uptake in
 4                  subjects with and without asthma whose mean lung function decreased by 33% and 55%,
 5                  respectively. This evidence combined with that from prospective epidemiologic studies
 6                  and that characterizing modes of action (i.e., increases in IgE, Th2 cytokines, and
 7                  inflammation) increases confidence that the associations observed between blood Pb
 8                  level and asthma and allergy in children are not due to reverse causality.

 9                  Among the studies in children that found associations of blood Pb level with asthma and
10                  allergy, several adjusted for potential confounding by indicators of SES and other
11                  environmental exposures. Joseph et al. (2005) adjusted for census block annual income,
12                  which may measure individual-level income with error. However, Pugh Smith and
13                  Nriagu (2011). which examined a primarily low SES  population of children in Michigan,
14                  adjusted for family annual income. In this study,  a Pb-associated higher asthma
15                  prevalence was found with adjustment for various additional factors associated with SES
16                  and allergen exposure, including multiple indices of housing condition and presence of
17                  pets and cockroaches in the home. Jedrychowski et al. (2011) adjusted for maternal
18                  education, and found similar magnitudes of association between cord blood Pb level and
19                  positive SPT as those in the unadjusted analysis.  Further, residential levels of dust mite or
20                  pet allergen were associated with lower risks of SPT than was blood Pb level. Blood Pb
21                  level also was associated with asthma or allergy after adjusting for concurrent exposure
22                  to smoking in the home (Jedrychowski et al.. 2011; Pugh Smith and Nriagu. 2011). with
23                  Jedrychowski et al. (2011) additionally adjusting for prenatal smoke exposure assessed as
24                  cord blood cotinine levels. The studies varied in the specific confounding factors
25                  considered, the method of measurement, and the  method of control. Some studies
26                  examined several potential confounding factors (Jedrychowski et al.. 2011; Pugh Smith
27                  and Nriagu. 2011). which increases confidence that the observed associations with
28                  asthma and allergy reflect a relationship with Pb. However, in the small evidence base,
29                  uncertainty remains regarding residual confounding, particularly by SES.  While there is
30                  no single complete measure of SES, these studies adjusted for different indicators of SES
31                  that may vary in the adequacy of control for confounding. Residual confounding also is
32                  possible by factors not examined.

33                  Cross-sectional evidence did not strongly indicate associations of biomarkers of Pb
34                  exposure with asthma or allergy in nonoccupationally-exposed adults (Mendy et al..
35                  2012; Pizent et al.. 2008). However, study limitations make the evidence less informative
36                  for drawing conclusions about the effects of Pb on asthma compared with evidence in
37                  children. Higher Pb  level in a spot urine sample was not associated with an increase in
38                  asthma prevalence (OR: 0.72 [95% CI: 0.46, 1.12] per 1 (ig Pb/g creatinine increase in
      November 2012                            5-463                  Draft - Do Not Cite or Quote

-------
 1                  urine) in the large U.S. NHANES 2007-2008 analysis of 1,857 adults ages 20 years and
 2                  older (geometric mean urinary Pb level: 0.59 (ig/g creatinine (Mendv et al.. 2012).
 3                  Associations were not found with other respiratory conditions such as emphysema or
 4                  chronic bronchitis either. This study examined several potential confounding factors,
 5                  adjusting for sex, race/ethnicity, education, income to poverty ratio, number of alcoholic
 6                  drinks/day, and current smoking status. However, spot urinary Pb level has an uncertain
 7                  relationship with long-term Pb exposure (Section 4.3.3). In a study of adult (19-67 years)
 8                  office workers in Zagreb, Croatia, Pizent et al. (2008) found a lower concurrent blood
 9                  Pb-associated odds of positive SPT to common inhaled allergens among 50 men (median
10                  blood Pb level: 3.2 (ig/dL) and lack of statistically significant association among 166
11                  women (median blood Pb level: 2.2 (ig/dL), adjusting for age, smoking  (current and
12                  history), and number of alcoholic drinks/day. The findings in women appeared to be
13                  discordant because there was an association between concurrent blood Pb level and
14                  serum IgE, which commonly mediates the acute inflammatory response to allergens.
15                  However, the interpretation of the findings is difficult because only statistically
16                  significant effect estimates were reported; thus it is not known whether odds ratios were
17                  in the same direction for SPT and IgE in women. Bener et al. (200la) found higher
18                  prevalence of asthma and allergy-related conditions in 110 Pb industrial workers (mean
19                  age: 35.5 years) than in 110 age-matched controls. However, the  implications are limited
20                  because blood Pb levels in both Pb-exposed workers and controls (geometric means: 77.5
21                  and 19.8 (ig/dL, respectively) were higher than those in the current U.S. adult general
22                  population, and potential confounding factors, including other occupational exposures,
23                  were not evaluated.

24                  In summary, evidence supports associations of higher Pb biomarker levels with asthma
25                  and allergy (Jedrychowski et al.. 2011; Pugh Smith and Nriagu. 2011; Joseph et al.. 2005)
26                  in children. Because of study limitations, the evidence in adults does not largely inform
27                  the conclusion (Mendv etal., 2012; Pizent et al., 2008; Bener etal., 200la). In children,
28                  evidence was limited to a few populations. Because of the heterogeneity in the relatively
29                  small body of evidence, it was difficult to identify whether the strength of association
30                  with asthma and allergy differed by age of children, lifestage of blood Pb measurement
31                  (prenatal, sometime in childhood prior to outcome assessment, concurrent), or blood Pb
32                  level. The prospective analysis in Jedrychowski et al. (2011) and Joseph et al. (2005)
33                  increase confidence that the observed associations are not due to  reverse causality. Lack
34                  of reverse causality also is indicated by observations that particle uptake generally does
35                  not differ between subjects with and without asthma and between animals sensitized with
36                  allergens and unsensitized controls. In these studies, the lack of selective participation of
37                  subjects and objective assessment of outcomes indicates lack biased reporting of asthma
38                  and allergy in children with higher blood Pb levels. In some studies, the method of
39                  recruitment of subjects from blood Pb surveillance databases may limit generalizability

      November 2012                             5-464                  Draft - Do Not Cite or Quote

-------
 1                  of findings. The adjustment for maternal education and exposure to smoking or allergens
 2                  in Jedrychowski et al. (2011) and family income, smoking, housing conditions, pets, or
 3                  pests in Pugh Smith and Nriagu (2011) increase confidence that the observed associations
 4                  in these studies are not due to confounding by SES, smoking, or allergen exposure.
 5                  Further, biological plausibility is well supported by evidence describing modes of action
 6                  for asthma and allergy, including Pb-associated increases in IgE (Section 5.6.3).
 7                  Th2 cytokines (Section 5.6.6.1). and inflammation (Section 5.6.4). However, because
 8                  there are few studies, uncertainty remains regarding residual confounding, particularly by
 9                  SES, which was examined with different indicators across studies.
                    5.6.5.3     Other Respiratory Effects

10                  Other respiratory effects associated with blood or air Pb levels were not well
11                  characterized in the 2006 Pb AQCD (U.S. EPA. 2006b) but have been examined recently
12                  in a small number of studies. As with asthma, associations between blood Pb level and
13                  respiratory effects in adults is inconsistent. Studies were cross-sectional, included
14                  similarly aged subjects, and considered similar confounding factors. Increased bronchial
15                  responsiveness (BR) is a characteristic feature of asthma and other respiratory diseases
16                  and can result from the activation of innate immune responses and increased airway
17                  inflammation. In the larger study of 523 adults (ages 19-58 years) in Seoul, Korea, Min et
18                  al. (2008a) found an association between concurrent blood Pb level and BR. A 1 (ig/dL
19                  higher concurrent blood Pb level was associated with a higher BR index (log [% decline
20                  in forced expiratory volume in 1 second (FEVi)/log of final methacholine concentration
21                  in mg/dL]) of 0.018 (95% CI: 0.004, 0.03), with adjustment for age, sex, height, smoking
22                  status, baseline FEVi, and presence of asthma (Min et al.. 2008a). The concurrent blood
23                  Pb levels in these adults were low (mean [SD]: 2.90 [1.59] (ig/dL); however, it is
24                  uncertain what timing, level, frequency, and duration of Pb exposures contributed to the
25                  observed association. In contrast to Min et al. (2008a). Pizent et al. (2008) found that
26                  higher concurrent blood Pb level was associated with lower BR in 47 men (2.4%
27                  decrease [95% CI:  -4.2, -0.52%] in percent change FEVi post-histamine challenge per
28                  1 (ig/dL increase in blood Pb level adjusted for age and serum Se). Smoking intensity and
29                  alcohol consumption were excluded as covariates by stepwise regression. Similarly,
30                  among these men, higher blood Pb level was associated with lower odds of positive SPT.

31                  Pb-associated respiratory effects were not clearly indicated in adults with occupational Pb
32                  exposures either. However, the lack of direct analysis of blood Pb levels and
33                  consideration for potential confounding limit the utility of this evidence in drawing
34                  conclusion about the respiratory effects of Pb related to airway responses. In bus drivers
35                  (mean age: 46 years) in Hong Kong (Jones et al.. 2008; Jones et al.. 2006).  129 drivers of


      November 2012                             5-465                  Draft - Do Not Cite or Quote

-------
 1                  non-air conditioned buses had lower exposures to PM10, lower blood Pb levels (mean
 2                  3.7 (ig/dL versus 5.0  (ig/dL in air conditioned buses) but lower indices of lung function
 3                  than did 358 drivers of air conditioned buses (Jones et al., 2006). The authors attributed
 4                  the slightly higher blood Pb levels of air conditioned bus drivers to the poor efficiency in
 5                  the filters resulting in higher PM10 levels on those buses. Blood Pb levels and various
 6                  lung function parameters were similar between 33 roadside vendors and 31 adjacent
 7                  shopkeepers (mean ages: 45.1 and 42.8 years, respectively and mean blood Pb levels:
 8                  5.61 and 5.14 (ig/dL) (Jones et al.. 2008). Pb industrial workers (n = 100, mean age: 34.6
 9                  years) in the United Arab Emirates had higher prevalence of respiratory symptoms such
10                  as cough, phlegm, shortness of breath, and wheeze than did  100 age- and sex-matched
11                  unexposed controls (Bener etal. 200la). Blood Pb levels in both the Pb industrial
12                  workers and the control group (geometric means: 77.5 and 19.8 (ig/dL,  respectively) were
13                  higher than those  in most of the current U.S. adult general population.

14                  An effect of Pb specifically on the lung was demonstrated in a recent study of Wistar rats
15                  with low blood Pb levels (2.1 (ig/dL) but produced by Pb acetate given  by injection
16                  (25 mg/kg, 3 consecutive days) (Kaczynska et al.. 2011). The lungs of Pb-treated rats
17                  exhibited pulmonary  fibrosis, epithelial cell damage, an increase in mast cells, an
18                  increased recruitment of monocytes and thrombocytes into capillaries, and increased
19                  macrophage accumulation in the alveolar space. While these pulmonary changes have
20                  been linked with functional pulmonary decrements and inflammation in other studies
21                  (unrelated to Pb exposure), the implications are uncertain because the results were
22                  obtained with a route of Pb exposure less relevant to those in humans.


                    Air-Pb Studies

23                  The 2006 Pb AQCD (U.S. EPA. 2006b) did not review studies that represent Pb exposure
24                  by Pb measured in PMi0 and PM2 5 air samples. However, recent studies have examined
25                  the respiratory effects of PM-Pb by analyzing the Pb component individually or as part of
26                  a group of correlated components using source apportionment or principal component
27                  analysis. Daily ambient air Pb-PM concentrations were associated with daily respiratory
28                  morbidity in children (Gent et al.. 2009; Hong et al.. 2007b). Gent et al. (2009) found that
29                  increases in lag 0  and 0-2 average Pb-PM2 5 were associated with increases in respiratory
30                  symptoms and asthma medication use in 149 children with asthma in Southern New
31                  England (ages 4-12 years), adjusting for season, day of the week, and date. Hong et al.
32                  (2007b) found that an increase in lag 1 Pb-PMi0 was associated with a decrease in lung
33                  function in 43  mostly healthy children in Korea, adjusting for age, sex, height, weight,
34                  household smoking, and weather. In  support of these results in children, toxicological
35                  studies found Pb-containing CAPs to induce pulmonary inflammation. Uzu et al. (2011)
36                  found that Pb-rich PM from a Pb recycling  plant increased the release of the cytokine

      November 2012                            5-466                 Draft - Do Not Cite or Quote

-------
 1                   granulocyte-macrophage colony-stimulating factor from human epithelial cells.
 2                   Pulmonary inflammation was found in animals exposed to CAPs in which Pb was one of
 3                   numerous components (Wei et al., 2011; Duvall et al., 2008; Godleski et al., 2002;
 4                   Saldiva et al.. 2002).

 5                   As with blood Pb, daily ambient air Pb-PM2 5 concentrations were not consistently
 6                   associated with daily respiratory effects in older adults with adjustment for weather and
 7                   temporal trends. Among adults ages 65 years and older in 6 California counties, a 4
 8                   ng/m3 increase in lag 3 Pb-PM25 was associated with an relative risk of respiratory
 9                   mortality of 1.01 (95% CI: 0.99, 1.03) in the all-year analysis and in a summer-only
10                   analysis (RR not reported) (Ostro et al.. 2007). However, among adults ages 65 years and
11                   older in 106 U.S. counties, Bell et al. (2009) found that an increase in lag 0 Pb-PM2s was
12                   associated with a decrease in respiratory hospital admissions.

13                   Despite evidence that indicates a relationship between respiratory effects in children and
14                   short-term (over a few days) changes in ambient air Pb-PM concentrations, uncertainties
15                   limit the utility of these findings in evaluating Pb-associated respiratory effects. Few data
16                   on size distribution of Pb-PM are available, so it is difficult to assess the
17                   representativeness of these concentrations to population exposure (Section 3.5.3).
18                   Moreover, few data are  available on the relationship between blood Pb and air Pb for the
19                   varying Pb-PM size distributions (see Section 4.5.1). In several air-Pb studies, other PM
20                   components such as elemental carbon (EC), copper  (Cu), and zinc (Zn) also were
21                   associated with respiratory effects. In the absence of data on correlations among PM
22                   components, measurements on co-occurring ambient pollutants, or results adjusted for
23                   copollutants, it is difficult to exclude confounding by ambient air exposures to other PM
24                   components or ambient pollutants. In several studies that analyzed PM component
25                   mixtures, of which Pb-PM comprised one component, it is not possible to attribute the
26                   observed associations or lack of associations specifically to Pb (Sarnat et al.. 2008;
27                   Andersen et al.. 2007; Veranth et al.. 2006; Macieiczyk and Chen. 2005).

28                   In summary, while air Pb-PM has  been associated with respiratory effects in children,
29                   main limitations of this  recent evidence include the confounding by other PM
30                   components and the uncertain representativeness of Pb-PM to population exposures. In
31                   adults, neither blood Pb nor Pb-PM was consistently associated with respiratory effects.
32                   Blood Pb studies of nonoccupationally-exposed adults were similar in cross-sectional
33                   design, age of subjects,  potential confounding factors examined, and respiratory
34                   endpoints which exhibit short-term changes. Studies of Pb-exposed workers were
35                   similarly limited by lack of rigorous statistical analysis with blood Pb levels and lack of
36                   consideration for potential confounding factors, including other occupational exposures.
      November 2012                             5-467                  Draft - Do Not Cite or Quote

-------
                    5.6.5.4     Autoimmunity

 1                  Autoimmunity is an immune response against self (e.g., generation of antibodies against
 2                  self antigens) and is linked with diseases such as lupus and rheumatoid arthritis. Evidence
 3                  for the effects of Pb on increasing the risk of autoimmunity is provided primarily by a
 4                  small number of toxicological studies reviewed in the 2006 Pb AQCD in which pre- and
 5                  post-natal Pb acetate exposure of animals, several by injection, was associated with the
 6                  generation of autoantibodies (Hudson et al., 2003; Bunn et al., 2000; El-Fawal et al.,
 7                  1999; Waterman et al.. 1994).  El-Fawal et al. (1999) found elevated auto-antibodies in
 8                  F344 rats with blood Pb levels 11-50 (ig/dL. Evidence was mixed in indicating a shift
 9                  toward Th2 or Thl responses as the underlying mechanism. While recent studies did not
10                  examine Pb-induced production of auto-antibodies,  some provided indirect evidence by
11                  indicating that Pb had the potential to induce formation of neo-antigens which in turn
12                  could induce the formation of auto-antibodies. For example, Kasten-Jolly et al. (2010)
13                  found that developmental Pb acetate exposure of BALB/c mice (100 (iM in drinking
14                  water, GD8-PND21) upregulated genes for digestive and catabolizing enzymes, which
15                  could lead to the generation of self-peptides, which  combined with other Pb-induced
16                  immune effects, had the potential to induce the generation of auto-antibodies. The
17                  potential for auto-antibody generation also was indicated by the activation of neo-
18                  antigen-specific T cells in adult BALB/c mice injected once with 25-50 (ig Pb chloride
19                  (Carey et al., 2006). Evidence  of Pb-associated autoimmune responses in humans is
20                  limited to findings of higher levels of IgM and IgG auto-antibodies to neural proteins in
21                  male battery-plant workers (n = 20, 56) with blood Pb level range 10-40 (ig/dL compared
22                  with controls (n = 7, 15, blood Pb levels not reported) (El-Fawal et al.. 1999). Pb workers
23                  and controls were matched by demographic and SES characteristics, but potential
24                  confounding by other workplace exposures was not examined. Similar to findings in
25                  Pb-exposed workers, modified neural proteins were found in CBA/J rats injected with
26                  native protein altered by Pb acetate in vitro (Waterman et al..  1994).
                    5.6.5.5      Tumors

27                  Toxicological evidence indicates that high concentration Pb exposures directly promote
28                  tumor formation or induce mutagenesis and genotoxicity (Section 5.10). and a study
29                  provided evidence for involvement of the immune system. Kerkvliet and Baecher-
30                  Steppan (1982) found that postnatal exposure of 6-8 week old male C57BL/6 mice to 130
31                  and 1,300 ppm Pb acetate in drinking water for 10-12 weeks transiently enhanced
32                  moloney sarcoma virus-induced tumor growth compared with control animals but did not
33                  prevent subsequent tumor regression. The Pb-induced tumor growth was accompanied by
34                  impaired macrophage phagocytosis  (indicating suppressed Thl responses) but not

      November 2012                            5-468                  Draft - Do Not Cite or Quote

-------
 1                  cytotoxicity. Cancer promotion is a relatively common outcome in chemical-induced
 2                  immunotoxicology, particularly when early life exposures are involved (Dietert. 2011).
      5.6.6        Modes of Action for Pb Immune Effects
                    5.6.6.1      Cytokine Production

 3                  As referenced in preceding sections, cytokines are signaling molecules that affect
 4                  immune cell function. For example, IL-4 induces B cells into IgE-producing cells, and
 5                  IFN-y induces macrophage recruitment and antigen presenting activity. The
 6                  2006 Pb AQCD (U.S. EPA. 2006b) presented a large body of evidence that clearly
 7                  demonstrated that pre- and postnatal Pb exposure of animals such as rodents and chickens
 8                  suppressed the production of Thl cytokine IFN-y and/or increased production of Th2
 9                  cytokines such as IL-4 [Table 5-7 of the 2006 Pb AQCD (U.S. EPA. 2006g)1. The
10                  combined evidence for Pb-induced cytokine changes in multiple cell types, including T
11                  cells and macrophages, indicates a  shift of acquired immunity responses away from Thl
12                  responses and toward Th2 responses. In turn, the Thl to Th2 shift provides mode of
13                  action support for downstream effects (Figure 5-34) such as inflammation, ROS
14                  production, impaired macrophage function, decreased host resistance observed primarily
15                  in toxicological studies, increased IgE production observed in both epidemiologic and
16                  toxicological studies, and asthma and allergy observed in epidemiologic studies. Previous
17                  toxicological studies found Pb to affect cytokine production via action on T cells and
18                  macrophages, and a recent study provided new evidence that Th2 skewing may be
19                  mediated via effects on dendritic cells.

20                  Many studies found a shift to Th2 cytokine production in animals with long-term
21                  (>4 weeks) dietary Pb exposure, and in some studies, the effects of prenatal exposure on
22                  cytokine production persisted to the adult lifestage (Chen et al.. 2004; Miller et al.. 1998).
23                  In the few studies that measured blood Pb levels shortly after cessation of Pb exposure
24                  (gestational plus postnatal or postnatal only), higher IL-4 and/or lower IFN- y were found
25                  in rodents with relevant blood Pb levels, means 6.75 and 17 (ig/dL (Chen et al.. 2004;
26                  Dyatlov and Lawrence. 2002). Some studies found an increase in IL-4 or decrease in
27                  IFN- y concomitantly with an increase in IgE (Heo et al.. 1997; 1996) or decrease in host
28                  resistance (Fernandez-Cabezudo et al.. 2007) further supporting changes in cytokine
29                  production as a mode of action for Pb-induced effects on downstream immune endpoints.
30                  A recent study found a shift to Th2 cytokine production in mice over wide range of Pb
31                  exposures, and provided evidence of effects at lower Pb exposures and a concentration-
32                  dependent relationship. In this study, lifetime (gestation through adulthood) exposure of
      November 2012                            5-469                  Draft - Do Not Cite or Quote

-------
 1                   Swiss mice to 0.06-400 ppm dietary Pb acetate produced blood Pb levels (upon
 2                   termination of exposure) of 1.23 to 61.48 (ig/dL (lavicoli et al.. 2006b). For IL-2 and
 3                   IL-4, nonlinear concentration-response relationships were found, with the largest
 4                   decrease and increase, respectively, found between animals with blood Pb levels of
 5                   0.8 (ig/dL (0.02 ppm Pb acetate, controls) and 1.23 (ig/dL. A linear concentration-
 6                   dependent decrease in IFN-y was observed in animals with blood Pb levels of
 7                   0.8-61.48 (ig/dL. Although examined in few in vivo studies, increases in IL-6 and IL-10,
 8                   also Th2 cytokines, were reported in juvenile and adult rodents exposed to Pb
 9                   gestationally or postnatally in drinking water [Table 5-7 of the 2006 Pb AQCD (U.S.
10                   EPA. 2006g)1.

11                   In vitro studies also reported a Pb-induced shift to production of Th2 cytokines. In
12                   concordance with other indicators of Th2 skewing (i.e., suppressed DTH) in BALB/c
13                   mice treated with Pb-exposed dendritic cells, Gao et al. (2007) observed that 10-day
14                   25 (iM Pb chloride exposure lowered the ratio of IL-12:IL-10 production by dendritic
15                   cells in vitro. Pb did not affect dendritic cell production of IL-6, IL-10, or TNF-a;
16                   however, in co-cultures of Pb-treated dendritic cells and T cells, most results indicated
17                   that dendritic cells stimulated T cells to produce Th2 cytokines. For example, although T
18                   cell  production of the Thl cytokine IL-2 increased, production of Th2 cytokines, IL-6
19                   and  IL-10 increased. Further, Pb-treated dendritic cells increased IL-4 production in
20                   OVA-specific T cells, indicating that Pb affected the antigen presenting cell function of
21                   dendritic cells. In another in vitro study, 24-hour Pb acetate exposures of 0.15 (ig/dL and
22                   higher suppressed expression of Thl cytokines, IFN-y, IL-1(3, and TNF-a, and increased
23                   secretion of Th2 cytokines, IL-5, IL-6, and IL-10 in cultures of human PMNs activated
24                   with Salmonella enteritidis or with monoclonal antibodies of CD3, CD28, and CD40,
25                   (Hemdan et al.. 2005).

26                   Several toxicological studies found Pb-induced increases in the cytokine TNF-a, in some
27                   cases, specifically from macrophages (Khan et al.. 2011; Cheng et al.. 2006; Flohe et al..
28                   2002; Zelikoff et al.. 1993). This provides mode of action support for toxicological
29                   evidence indicating Pb-induced decreases in resistance to bacterial infection since TNF-a
30                   is produced primarily by activated macrophages, is increased in response to infection, and
31                   induces inflammation. Among the in vivo studies, increases  in TNF-a were  found with
32                   prenatal dietary Pb acetate exposure (250 ppm) of F344 rats  (Chen et al.. 1999; Miller et
33                   al.. 1998). postnatal Pb oxide air exposure (31 (ig/m3, 3 hours/day, 4 days) of rabbits
34                   (Zelikoff et al.. 1993). and postnatal i.p. Pb acetate treatment (5.0 mg) of Swiss mice
35                   (Dentener et al.. 1989). The effects of prenatal dietary Pb exposure were found to persist
36                   to adulthood. In animals, the Pb-induced increases in TNF-a were accompanied by
37                   functional changes in host responses such as decreased macrophage phagocytosis
38                   (Zelikoff etal.. 1993). suppressed DTH (Miller etal.. 1998). and increased mortality to
      November 2012                             5-470                  Draft - Do Not Cite or Quote

-------
 1                  E.coli endotoxin (Dentener et al.. 1989). Blood Pb levels of animals were infrequently
 2                  reported; however, Chen et al. (1999) found increased TNF-a in rats with embryonic
 3                  blood Pb levels of 149 (ig/dL. In addition to finding Pb-induced increases in TNF-a
 4                  (Khan etal.. 2011: Gao et al.. 2007: Cheng et al.. 2006: Flohe et al.. 2002: Krocova et al..
 5                  2000: Guo etal.. 1996a). some in vitro studies provided mechanistic explanation by
 6                  finding that Pb acetate or chloride (10-50 (iM, 1.5 hours-10 days) induced
 7                  phosphorylation of mitogen-activated protein kinase (MAPK) signaling molecules (Khan
 8                  etal.. 2011: Gao et al.. 2007: Cheng et al.. 2006). Further, Cheng et al. (2006) found that
 9                  blocking protein kinase C or MAPK reduced TNF-a production by macrophages in vitro,
10                  which in turn, protected against Pb acetate + LPS-induced liver injury in A/J mice.

11                  The few available epidemiologic studies that examined cytokines found higher
12                  concurrent blood Pb levels in children and adults to be associated with higher Th2
13                  cytokine and lower Thl cytokine levels in serum. The epidemiologic evidence overall
14                  was based on cross-sectional analyses, which precludes identifying the temporal
15                  sequence between Pb exposure and cytokine changes. Other limitations include the lack
16                  of rigorous statistical analysis and limited consideration of potential confounding.
17                  Because of these limitations, the epidemiologic evidence is not a primary consideration in
18                  drawing conclusions about Pb-associated cytokine changes. However, it does not mitigate
19                  the consistent toxicological evidence. Among children ages 9 months to 6 years in
20                  Missouri recruited from a public assistance or Pb poisoning prevention program, Lutz et
21                  al. (1999) found that 8 children with concurrent blood Pb levels 15-19 (ig/dL had 4-5 fold
22                  higher serum levels of IL-4 (p = 0.08, Kruskal Wallis) and 3-fold higher IgE
23                  (Section 5.6.3) than did 90 children with lower blood Pb levels. IL-4 levels in 9 children
24                  with blood Pb levels 20-44 (ig/dL were lower than those  in 90 children with blood Pb
25                  levels <15 (ig/dL. The elevated IL-4 and IgE in children with blood Pb levels
26                  15-19 (ig/dL were consistent with the action of IL-4 to activate B cells to induce class
27                  switching to IgE. In another study of 214 children in grades 5 and 6 in Taiwan,
28                  investigators compared cytokine levels not by blood Pb level groups but by potential for
29                  Pb exposures due to age of home and location of residence (Hsiao etal.. 2011). Elevated
30                  concurrent blood Pb levels were found only among 64 children living near an oil refinery,
31                  in particular, among 34 children with known respiratory allergies (mean: 8.8 (ig/dL
32                  versus 3.2-3.8 (ig/dL in urban and rural groups). Children with allergies near the oil
33                  refinery also had the lowest serum levels of IFN-y (45-fold) and highest levels of IL-4 (6-
34                  fold) (lower p <0.05 for comparisons with any subgroup). While the results suggested
35                  that residence near the oil refinery contributed to differences in cytokine levels between
36                  healthy and allergic children, they do not specify a contribution of Pb, other exposures or
37                  co-occurring factors, or a combination of factors.
      November 2012                             5-471                  Draft - Do Not Cite or Quote

-------
 1                  Evidence of association between blood Pb levels and cytokine levels in is unclear.
 2                  However, a larger study that considered potential confounding found an association.
 3                  Sufficient information was not provided to assess potential selection bias. Among 300
 4                  (93% male, mean age 24 years) healthy university students in Incheon, Korea, higher
 5                  concurrent blood Pb level was associated with higher serum levels of TNF-a and IL-6,
 6                  adjusting for age, BMI, and smoking status (Kim et al.. 2007). Associations were larger
 7                  in magnitude among the  147 males in the upper two quartiles of blood Pb levels,
 8                  2.51-10.47 (ig/dL. Associations with these blood Pb levels may reflect contributions of
 9                  higher past Pb exposures. A 1 (ig/dL higher blood Pb level was associated with a 0.75
10                  (95% CI: 0.14, 1.36) pg/mL higher TNF-a and a 0.18 (95% CI: -0.02,  0.38) pg/mL higher
11                  IL-6. The association between levels of blood Pb and plasma TNF-a was greater among
12                  men who were GSTM1 null (n = 77) than men who were GSTM1 positive and men who
13                  had the TNF-a GG genotype (n = 131) than men who had the GA or AA genotype. For
14                  the association between blood Pb level and plasma IL-6, the effect estimate  was slightly
15                  elevated in TNF-a GG genotype but similar between GSTM1 genotypes. In this study,
16                  there were multiple comparisons, but a consistent pattern of association was observed
17                  across the immune endpoints examined. Subgroup analyses had fairly  large  sample sizes,
18                  and some results had biological plausibility, but they are prone to higher probability of
19                  findings by chance. Pb has been shown to increase ROS (Section 5.2.4). and cytokine
20                  expression has been shown to be modulated by ROS-sensitive transcription  factors.  Thus,
21                  it is biologically plausible that the null variant of GSTM1, which is associated with
22                  reduced elimination of ROS, may be associated with increased cytokine levels. The
23                  results for the  TNF-a variant are difficult to interpret. The GG genotype is associated
24                  with lower expression of TNF-a, but the literature is mixed with respect to which variant
25                  increases risk of inflammation-related conditions.

26                  A much smaller study of adults that did not consider potential confounding, did not report
27                  quantitative results but only reported lack of statistically significant correlations between
28                  concurrent blood Pb level and serum Th2 and Thl cytokine levels in men (n = 17 with
29                  and 17 without allergy, ages 19-52 years, median blood Pb levels: -11 (ig/dL) (Boscolo
30                  et al.. 1999) and women in Italy (n = 23 with and 25 without allergy, ages 19-49 years,
31                  median blood Pb levels: 6.4 and 5.5 (ig/dL, respectively) (Boscolo et al., 2000).

32                  Results from studies of occupationally-exposed adults  also suggested that Pb exposure
33                  may be associated with decreases in Thl cytokines and increases in Th2 cytokines (Di
34                  Lorenzo et al.. 2007; Valentino et al.. 2007; Yiicesoy et al.. 1997a). Valentino  et al.
35                  (2007) had the most rigorous statistical methods comprising regression analyses with
36                  adjustment for age, BMI, smoking status, and alcohol consumption status but not other
37                  occupational exposures. Regression coefficients describing the concentration-response
38                  functions were not reported; however, 44 male foundry workers in Italy (mean blood Pb
      November 2012                            5-472                  Draft - Do Not Cite or Quote

-------
 1                   levels: 21.7 (ig/dL) and 14 pottery workers (mean blood Pb level: 9.7 (ig/dL, ages of all
 2                   workers 30-61 years) had higher plasma IL-10 (ANOVA, p <0.05) than did the 59
 3                   unexposed controls (mean blood Pb level: 3.9 (ig/dL, ages 25-61 years). Levels of Th2
 4                   cytokines IL-2, IL-6, and IL-10 also increased from the lowest to highest blood Pb group
 5                   (ANOVA, p >0.05).  In contrast with most other studies, both exposed worker groups had
 6                   lower IL-4 levels compared with controls (ANOVA, p >0.05). In Yucesoy et al. (1997a).
 7                   serum levels of the Thl cytokines, IL-1(3 and IFN-y, were lower in 20 Pb-exposed
 8                   workers (mean blood Pb level: 59.4 (ig/dL, ages 19-49 years) than in the 12 age-matched
 9                   controls in Turkey (mean blood Pb level: 4.8 (ig/dL). Some (Di Lorenzo et al.. 2007;
10                   Valentino et al.. 2007) but not all (Yucesoy et al.. 1997a) studies found higher serum
11                   TNF-a in Pb-exposed workers. DiLorenzo et al. (2007) found a monotonic increase from
12                   the 28 unexposed (blood Pb levels not reported, mean age: 48.2 years) to 17 intermediate
13                   worker (9.1-29.4 (ig/dL) and 19 high worker (29.4-81.1  (ig/dL) blood Pb level groups
14                   (mean age of workers: 45.3 years) in Italy. Results also indicated a potential interaction
15                   between blood Pb level and smoking. Among current smokers (n = 9 to 20), a  12- to
16                   16-fold difference in TNF-a levels was observed among blood Pb groups. Among
17                   nonsmokers (n = 2 to 8), the differences were less than two fold.

18                   In summary, the few epidemiologic studies indicate associations of higher concurrent
19                   blood Pb level with higher levels of IL-4 and/or lower levels of IFN-y in children (Hsiao
20                   et al.. 2011; Lutzetal.. 1999) and occupationally-exposed adults (Di Lorenzo et al..
21                   2007; Valentino et al.. 2007; Yucesoy et al.. 1997a). Because quantitative results were
22                   not reported in each study of nonoccupationally-exposed adults, implications of findings
23                   are difficult to assess. Limitations of the epidemiologic evidence overall include the
24                   cross-sectional design of studies and lack of rigorous statistical analysis that considered
25                   potential confounding factors. Sufficient data were not reported to assess potential
26                   selection bias. Because of the many limitations, the epidemiologic evidence alone is not
27                   used to  draw conclusions about Pb-associated cytokine changes. However, they are
28                   useful in indicating the relevance of toxicological evidence to humans. Biological
29                   plausibility for an effect of Pb on cytokine production is provided by a large body of
30                   toxicological evidence that clearly demonstrates a Pb-induced shift to a Th2 phenotype
31                   with increases in the Th2 cytokine IL-4 and decreases in the Thl cytokine IFN-y.  Several
32                   of these observations were made in juvenile and adult animals exposed prenatally or
33                   postnatally via diet that resulted in blood Pb levels (upon cessation of Pb exposure)
34                   relevant to humans, 1.23-17 (ig/dL (lavicoli et al.. 2006b; Chen et al.. 2004; Dyatlov and
35                   Lawrence. 2002). While results were not uniform for other cytokines (i.e., Thl cytokine
36                   IL-2), most available results pointed to increases in Th2 cytokines, specifically, IL-6 and
37                   IL-10, in Pb-exposed animals. Several studies demonstrated Pb-induced increases in
38                   TNF-a but in animals with high prenatal dietary or postnatal air Pb exposure (e.g.,
      November 2012                             5-473                  Draft - Do Not Cite or Quote

-------
 1                   149 jig/dL) (Chen et al.. 1999; Miller et al.. 1998; Zelikoff etal. 1993). Gao et al. (2007)
 2                   described a role for dendritic cells in skewing T cells to Th2 cytokine production. In vitro
 3                   evidence indicates that Pb may induce increases in TNF-a via MAPK signaling pathways
 4                   (Khan etal.. 2011; Gao et al.. 2007; Cheng et al.. 2006). Toxicological evidence indicates
 5                   effects on cytokine production of prenatal and postnatal Pb exposures, whereas
 6                   epidemiologic studies examined only concurrent blood Pb level. Because concurrent
 7                   blood Pb level in children and adults reflects both previous and current Pb exposures,
 8                   associations with concurrent blood Pb level may reflect an effect of cumulative Pb
 9                   exposure. Overall, the consistent toxicological evidence for Pb-induced decreases in Thl
10                   cytokines and increases in Th2 cytokines and pro-inflammatory cytokines such as TNF-a
11                   provides clear mode of action support for the evidence indicating the effects of Pb on
12                   both increases in IgE and inflammation and decreased host resistance.
                    5.6.6.2     Decreased Nitric Oxide

13                  As described in the 2006 Pb AQCD (U.S. EPA. 2006b). key mode of action support for
14                  the effects of Pb on impairing macrophage function and decreasing host defense was
15                  provided by consistent toxicological findings for Pb-induced decreases in NO, which is
16                  involved in the cytotoxic activity of macrophages in host defense processes [see 2006
17                  Annex Table AX5.9.6 (U.S. EPA. 2006h)1. In adult rodents, decreases in NO from
18                  macrophages were observed with short-term Pb acetate exposures (1 or 6 days) during
19                  early gestation (BunnetaL 200 Ib: Lee etal.. 200 Ib) but not long-term exposures
20                  occurring during the full gestational period (Bunn etal.. 200 Ic; Chen etal.. 1999; Miller
21                  et al.. 1998). With short-term exposure, decreases in NO were found in Sprague-Dawley
22                  rats with blood Pb level 4.5 (ig/dL in males and 5.3 (ig/dL in females measured 2 weeks
23                  after Pb exposure in drinking water of dams was terminated (BunnetaL. 200 Ib) and in
24                  chicks with blood Pb levels that did not exceed  11 (ig/dL but with Pb injected into eggs
25                  embryos (Lee etal.. 200Ib).

26                  The  short-term in vivo findings are supported by several in vitro observations of
27                  decreases in NO in macrophages and splenocytes induced by a wide range  of Pb exposure
28                  concentrations (0.625-5 (iM) and durations (2 hours-6 days) (Tarrer et al.. 2008; Mishra
29                  et al.. 2006a: Krocova et al.. 2000; Chen etal.. 1997; Tian and Lawrence. 1996. 1995).
30                  Farrer et al.  (2008) further indicated that the mode of action for Pb (5 (iM)  may involve a
31                  decrease in inducible NO  synthase function in myeloid cells without a change in its gene
32                  expression. Additionally,  Pb glutamate abrogated the myeloid cell (CD1 lb+)-mediated
33                  suppression of CD4+ T cell proliferation, and exogenous NO restored suppression.
34                  Together, these findings indicated that Pb may indirectly enhance T cell proliferation
35                  through its effect on decreasing NO production. Combined with the observation that Pb


      November 2012                            5-474                  Draft - Do Not Cite or Quote

-------
 1                  can alter antigen processing (Farreretal.. 2005) and, hence, the quality and magnitude of
 2                  the acquired immune response against pathogen exposure, evidence indicated that
 3                  multiple arms of host defense against infectious challenge can be compromised.
 4                  Diminished production of NO in innate immune cells such as macrophages could affect
 5                  other physiological systems (e.g., neurological, cardiovascular, endocrine) that require
 6                  NO signaling cascades.

 7                  Consistent with the toxicological evidence, cross-sectional studies found associations
 8                  between concurrent blood Pb level with lower NO in populations living near Pb sources
 9                  (Barbosa et al.. 2006c: Pineda-Zavaleta et al.. 2004). These studies did not provide
10                  sufficient information on participation rates; however, examination of populations near
11                  Pb sources could limit generalizability of findings. Additional limitations include the
12                  cross-sectional design that does not permit determining the temporal sequence between
13                  Pb exposure and NO suppression and the limited consideration for potential confounding.
14                  Because of these limitations, epidemiologic evidence is not a major consideration in
15                  drawing conclusions about the effects of Pb on NO. However, they are useful in that they
16                  show Pb-associated decreases in NO in humans, similar to toxicological studies.

17                  In a study of 65 children (ages 6-11 years) in Lagunera, Mexico, mean concurrent blood
18                  Pb levels increased (7.02 to 20.6 to 30.38  (ig/dL) with increasing school proximity
19                  (650-8,100 meters) to a Pb smelter (Pineda-Zavaleta et al.. 2004). With adjustment for
20                  age and sex, a 1 (ig/dL higher blood Pb level was  associated  with a 0.00089 (95% CI:
21                  -0.0017, -0.00005) nmol/(ig protein lower NO release from macrophages activated by
22                  phytohemagglutinin (PHA). Because PHA activates macrophages indirectly through the
23                  activation of lymphocytes, the results indicated that Pb suppressed T cell-mediated
24                  macrophage activation. Blood Pb group comparisons indicated that associations were due
25                  largely to the lower NO  in the 23 children closest to the smelter who had blood Pb levels
26                  10.31-47.49 (ig/dL. Though not described in detail, higher blood Pb level was not
27                  associated  with lower NO in girls.

28                  Among 104 adults (ages 18-60 years) in Sao Paolo,  Brazil residing near a closed battery
29                  plant, Barbosa et al. (2006c) observed an association between higher concurrent blood Pb
30                  level and lower plasma NO in the 69 adults (mean blood Pb level: 6.4 (ig/dL) with the TC
31                  or CC eNOS genotype (r = 0.23, p = 0.048). The results are consistent with the reduced
32                  promoter activity and  potentially reduced gene expression of the TC/CC variants. Results
33                  were not adjusted for potential confounding factors, but subjects were nonsmoking,
34                  nonalcohol drinking with normal mean  BMI and SEP. The exclusion criteria may further
35                  limit the generalizability of findings. Because NO was measured in plasma, immune cells
36                  could not be identified as the  source of NO. In contrast, Valentino et al. (2007) found
37                  similar plasma NO levels in 44 male foundry workers (mean blood Pb level: 21.7 (ig/dL),
      November 2012                             5-475                  Draft - Do Not Cite or Quote

-------
 1                   14 male pottery workers (mean blood Pb level: 9.7 (ig/dL), and 59 male unexposed
 2                  workers of similar age (mean blood Pb level: 3.9 (ig/dL, ages 25-61 years). Quantitative
 3                  results were not reported, but blood Pb level was reported not to be correlated with NO.
 4                  Potential confounding factors, including other workplace exposures were not examined.

 5                  In summary, toxicological evidence indicates that short-term dietary Pb exposure early in
 6                  gestation but not long-term exposure for the full gestational period results in reduced NO
 7                  production by macrophages. Evidence consistently demonstrates Pb-induced decreases in
 8                  NO in cell cultures. The relevance of toxicological evidence is supported by observations
 9                  of an association between higher concurrent blood Pb level and lower release of NO from
10                  macrophages of children in Mexico. The association in children was due largely to lower
11                  NO in macrophages from children living near a Pb smelter with concurrent blood Pb
12                  levels >10 (ig/dL, higher than those in most of the current U.S. population. This study
13                  had limited consideration for potential confounding factors. However, the toxicological
14                  evidence provides clear mode of action support for the effects of Pb on decreasing host
15                  resistance given the role of NO in mediating cytotoxic activity of macrophages.
                    5.6.6.3     Increased Reactive Oxygen Species and Prostaglandins

16                  ROS are released from macrophages during phagocytosis and are involved in killing
17                  invading bacteria. ROS and PGE2 are important mediators of inflammation which can
18                  result in local tissue damage (Figure 5-34). The roles of ROS and PGE2 in both host
19                  defense and injury may explain some of the inconsistencies in the evidence as reported in
20                  the 2006 Pb AQCD. In activated macrophages undergoing phagocytosis, high
21                  concentration (10-1,000 (JVI, 15 minutes-20 hours) Pb chloride or acetate exposures were
22                  found to reduce release of ROS (Hilbertz et al., 1986; Castranova et al.. 1980). consistent
23                  with observations of Pb-induced decreased bacterial and viral resistance. In resting
24                  macrophages, Hilbertz et al. (1986) found that Pb acetate induced an increase in ROS one
25                  hour but not 20 hours after exposure, indicating a transient response. Chen et al. (1997)
26                  also found a Pb-induced (4 (iM Pb-glutamate, 18 hours) decrease in ROS but did not
27                  indicate the functional state of macrophages. Shabani and Rabbani (2000) found a
28                  Pb-induced (240 (iM Pb nitrate, 3 hours) increased ROS from alveolar macrophages that
29                  was linked to their apoptosis, also consistent with impaired host defense. Other studies
30                  reported depletion of antioxidants such as glutathione and catalase in conjunction with
31                  reduced macrophage function in Swiss mice treated with Pb  nitrate by oral gavage (40
32                  mg/kg/day, 30 days) (Lodi et al.. 2011) or increases in PGE2 and apoptosis in vitro with
33                  0.01-10 (iM Pb nitrate (3 hours) (Chettv et al.. 2005). While  several processes have been
34                  proposed to explain the mechanisms of Pb-induced oxidative damage, the exact
35                  combination of processes involved remains to be determined (Section 5.2.4).


      November 2012                            5-476                 Draft - Do Not Cite or Quote

-------
 1                  In adult animals, Pb exposure increased ROS release from macrophages immediately
 2                  upon cessation of exposure (Baykov et al.. 1996; Zelikoff etal.. 1993) but not 9-10 weeks
 3                  after exposure (Miller et al.. 1998). consistent with in vitro findings. These Pb exposures
 4                  occurred through relevant routes of exposure, i.e., diet or air, but with high
 5                  concentrations, 31 (ig/m3 Pb oxide in air for 3 hours/day for 4 days in rabbits (Zelikoff et
 6                  al.. 1993) and 1.5 mg/kg Pb acetate in diet for 30 days in Swiss mice (Baykov et al..
 7                  1996). Neither study measured the blood Pb levels of animals.

 8                  Pb-associated increases in ROS also were found in macrophages of humans. However,
 9                  the findings are based on cross-sectional design, higher blood Pb levels (means: 7.02,
10                  20.6, and 30.38 (ig/dL) than most of those in the current U.S. population,  and limited
11                  consideration for potential confounding. In addition to finding suppressed NO production
12                  (Section 5.6.6.2). Pineda-Zavaleta et al. (2004) found a  Pb-associated increase in ROS
13                  production from macrophages in children in Mexico living in varying proximities to a Pb
14                  smelter. With adjustment for age and sex, a 1 (ig/dL higher concurrent blood Pb level  was
15                  associated with a 0.00389 (95% CI: 0.00031, 0.00748) (imol/mg higher release of
16                  superoxide anion from macrophages directly activated by IFN-y/LPS. The blood
17                  Pb-associated superoxide anion release was larger from macrophages of males. Because
18                  IFN-y directly activates macrophages, these results indicated that Pb stimulated cytokine-
19                  induced macrophage activation. Blood Pb level was not associated with ROS from
20                  neutrophils in a study  of male Pb recycling workers (ages 19-45 years) in India. Despite
21                  large differences in blood Pb levels between 30 Pb workers (mean: 106 (ig/dL) and 27
22                  unexposed controls (mean: 4.5 (ig/dL), levels of ROS released from neutrophils
23                  (indicators of respiratory burst) were similar between groups (Mishra et al.. 2006a).
24                  Evidence does not clearly indicate that neutrophils are a major responding cell to Pb
25                  exposure (Section 5.6.2.5).

26                  PGE2 is produced from the metabolism of cell membrane phospholipids and may be
27                  released by macrophages to modulate their function in a paracrine or autocrine manner.
28                  Toxicological studies have found Pb-induced increases in PGE2 in macrophages with
29                  high concentration Pb exposures. Increases in PGE2 were found in turkeys exposed to
30                  Pb acetate in feed PND1-PND21 and with a mean blood Pb level of 42 (ig/dL (Knowles
31                  and  Donaldson. 1997). Dietary Pb  acetate exposure (250-2,000 ppm) of chicks
32                  PND1-PND19 resulted in an increase in serum arachidonic acid but not PGE2 or other
33                  prostaglandins (Knowles and Donaldson. 1990). In vitro studies also used high Pb
34                  exposure concentrations, >20 (iM Pb chloride (Flohe et al.. 2002; Lee and Battles. 1994).
35                  A recent in vitro study with human neuroblastoma cells found increases in PGE2 with
36                  lower Pb concentrations (0.01-1 (iM Pb acetate) than previously reported  (Chetty et al..
37                  2005).
      November 2012                            5-477                  Draft - Do Not Cite or Quote

-------
 1                  In summary, ROS and PGE2 function in modulating macrophage function, aiding in
 2                  bacterial killing, and inducing tissue damage as part of an inflammatory response.
 3                  Consistent with these diverse roles, toxicological studies have found both Pb-associated
 4                  increases and decreases in ROS and PGE2. Consistent with toxicological findings, a
 5                  cross-sectional study found an association between higher concurrent blood Pb level
 6                  (>10 (ig/dL) higher ROS release from macrophages in children in Mexico that was
 7                  adjusted for potential cofounding by only age and sex. In animals, Pb-induced increases
 8                  in macrophage production of ROS and PGE2 have occurred concomitantly with
 9                  functional alterations such as impaired macrophage phagocytosis and apoptosis.
10                  Although toxicological results were based on examination of high Pb concentrations, they
11                  nonetheless provide clear evidence for modes of action underlying the effects of Pb on
12                  reduced macrophage function and decreased host resistance.
                    5.6.6.4      Cellular Death (Apoptosis, Necrosis)

13                  The 2006 Pb AQCD reported contrasting effects of Pb on the apoptosis of macrophages
14                  in vitro. However, with recent studies in mice, evidence suggests that Pb exposure may
15                  induce apoptosis or mediators of apoptosis in immune cells. Xu et al. (2008) found that 4-
16                  week dietary exposure of juvenile ICR mice to Pb acetate (50-100 mg/kg) induced DNA
17                  damage in peripheral blood lymphocytes, increased p53 and Bax expression in the liver,
18                  but did not change Bel-2 expression (creating a Bax/Bcl-2 imbalance). Bax promotes
19                  apoptosis, whereas Bel-2 inhibits apoptosis. Concomitant increases in indicators of
20                  oxidative stress in liver homogenate suggested that oxidative stress mediated Pb-induced
21                  apoptosis. In Swiss mice, Bishayi and Sengupta (2006) found splenic macrophages to
22                  have elevated DNA fragmentation, a key event in apoptosis, but with i.p.  Pb acetate
23                  treatment (10 mg/kg). Consistent with in vivo findings, Gargioni et al. (2006) found 20
24                  and 40 (iM Pb nitrate to induce cell death in mouse peritoneal macrophages in vitro with
25                  a concomitant loss of cell membrane integrity, indicating that Pb primarily induced
26                  macrophage necrosis or cell lysis. While evidence for Pb-induced apoptosis of immune
27                  cells with routes of Pb exposure relevant to humans is sparse, the evidence suggests that
28                  the induction of cell death may be a potential mode of action for the effects of Pb on
29                  macrophage function and decreased host resistance.
      November 2012                            5-478                  Draft - Do Not Cite or Quote

-------
      5.6.7        Immune Effects of Pb within Mixtures

 1                  In the 2006 Pb AQCD (U.S. EPA. 2006b). the immune effects resulting from Pb within
 2                  metal mixtures were not well characterized; however some recent studies indicated that
 3                  immune effects may be observed with lower levels of Pb exposure when they occur in
 4                  conjunction with other metals. In Swiss mice treated with Pb acetate (10 mg/kg i.p., daily
 5                  for 15 days), As (0.5 mg/kg i.p., daily for 15 days), or both, Bishayi and Sengupta (2006)
 6                  reported a greater than additive effect of co-administered Pb and As on decreasing
 7                  bacterial resistance, decreasing macrophage myeloperoxidase release, and NO
 8                  production.

 9                  Epidemiologic studies have not widely examined interactions between Pb and other
10                  metals.  However, consistent with Bishayi and Sengupta (2006). Pineda-Zavaleta et al.
11                  (2004) (Section 5.6.6.2) found interactions between Pb and As among 65 children in
12                  Mexico ages 6-11 years. Contamination of drinking water by both Pb and As was a
13                  concern in the study area; however, urinary As levels were higher in children who had
14                  lower blood Pb levels. Higher urinary As was associated with lower NO release from
15                  macrophages (similar to blood Pb). Higher As and Pb internal dose was associated with a
16                  larger decrease in NO (p for interaction = 0.037) than was either metal alone. Higher
17                  urinary  As was associated with lower superoxide anion release (opposite  direction of Pb).
18                  However, higher Pb and As internal dose was associated with a larger increase in
19                  superoxide anion (p for interaction = 0.042) than was blood Pb level alone. Due to the
20                  high blood Pb in these children (means in three groups at varying distances from a Pb
21                  smelter: 7, 20.6, 30.4 (ig/dL), it is not clear whether these relationships would apply to
22                  the current U.S. population of children.

23                  Institoris et al. (2006) found that Cd or Hg co-exposure potentiated the effects of Pb.
24                  Lymph  node weight decreased in 4 week-old Wistar rats exposed to 20 mg/kg Pb acetate
25                  by drinking water plus another metal but not with Pb alone. In contrast, Fortier et al.
26                  (2008) did not find metal co-exposure to increase the effects of Pb. Pb chloride
27                  (7.5-20.7 (ig/dL) did not alter lymphocyte proliferation, monocyte phagocytosis, or NK
28                  cell activity in human leukocytes. A mixture of 20.7 (ig/dL Pb chloride plus  12.0 (ig/dL
29                  methylmercuric chloride (MeHgCl) decreased lymphocyte proliferation; however, these
30                  effects were attributed to MeHgCl, which singly had a stronger suppressive effect. Other
31                  toxicological studies found metal mixtures that included Pb to decrease antibody titers or
32                  increase neutrophil counts (Jadhav et al.. 2007; Massadeh et al.. 2007) but did not test
33                  each metal individually. The latter findings cannot be attributed to interactions between
34                  Pb and other components within the mixture. Overall, several results indicated that
35                  exposures to Pb-containing metal mixtures are associated with immune effects. However,
      November 2012                            5-479                  Draft - Do Not Cite or Quote

-------
 1                  not all results showed that co-exposures to metals such as As, Cd, or Hg produce increase
 2                  the immune effects of Pb.
      5.6.8       Summary and Causal Determination

 3                  The cumulative body of epidemiologic and toxicological evidence describes several
 4                  effects of Pb exposure on the immune system related to a shift from Thl- to Th2-type
 5                  responses, including an increase in atopic and inflammatory conditions and a decrease in
 6                  host resistance. Outcomes related to an increase in atopic and inflammatory conditions
 7                  include asthma, allergy, increased IgE, and mode of action endpoints such as selective
 8                  differentiation of Th2 cells, increased production of Th2 cytokines, B cell activation, and
 9                  inflammation. Outcomes related to decreased host resistance include enhanced
10                  susceptibility to bacterial and viral infection, suppressed DTH, and those describing
11                  mode of action, i.e., decreased production of Thl cytokines, reduced phagocyte function,
12                  and increased inflammation. A small body of studies indicates the effects of Pb exposure
13                  on autoimmunity. The sections that follow describe the evaluation of evidence for these
14                  three groups of outcomes, decreased host resistance, increased atopic and inflammatory
15                  conditions, and autoimmunity, with respect to causal relationships with Pb exposure
16                  using the framework described in Table II of the Preamble. The application of the key
17                  supporting evidence to the causal framework is summarized in Table 5-34.
                    5.6.8.1      Evidence for an Increase in Atopic and Inflammatory
                                 Conditions

18                  Collective epidemiologic and toxicological evidence indicates that a causal relationship is
19                  likely to exist between Pb exposure and atopic and inflammatory conditions. This
20                  relationship is supported by evidence for associations of blood Pb levels with asthma and
21                  allergy in studies in children (Jedrychowski et al., 2011; Pugh Smith and Nriagu. 2011;
22                  Joseph et al.. 2005). Pb-associated increases in IgE in children and animals, and evidence
23                  describing modes of action including increases in Th2 cytokines and inflammation.

24                  Recent studies on asthma and allergy expand upon the evidence presented in the
25                  2006 Pb AQCD by providing additional evidence from prospective analyses, and by
26                  better addressing uncertainties regarding potential confounding by factors such as SES,
27                  smoking  exposures, and residential allergen exposures (Table 5-34). Findings from
28                  studies that prospectively ascertained outcomes increase confidence that associations are
29                  not due to reverse causation (Jedrychowski et al.. 2011; Joseph et al.. 2005). In these
30                  studies, the lack of selective participation and objective assessment of outcomes of
      November 2012                            5-480                 Draft - Do Not Cite or Quote

-------
 1                  asthma and allergy through medical records and clinical testing, respectively, indicates
 2                  lack of biased reporting of asthma and allergy in children with higher blood Pb levels
 3                  (Section 5.6.5.2 and Table 5-34). Among children age 5 years in Poland, Jedrychowski et
 4                  al. (2011) found that a 1 (ig/dL increase in prenatal cord blood Pb level was associated
 5                  with an increased risk of allergic sensitization of 2.3 (95% CI: 1.1, 4.6). The magnitude
 6                  of risk did not differ with and without adjustment for maternal education or residential
 7                  allergen levels. An additional strength of this study was the adjustment for prenatal
 8                  cotinine levels and postnatal smoker in the home. A large study of 4,634 children in
 9                  Michigan ages 1-3 years found that compared with Caucasian children with blood Pb
10                  levels <5 (ig/dL measured up to 12 months before asthma assessment, Caucasian children
11                  with blood Pb levels > 5 (ig/dL had an increased risk of incident asthma of 2.7 (95% CI:
12                  0.9, 8.1) (Joseph et al.. 2005). Adjustment was made for census block average income,
13                  which may not adequately control for potential confounding by individual subject-level
14                  SES.

15                  Supporting evidence was provided by a cross-sectional study of 356 children ages 0-12
16                  years in Michigan, which found that compared with children with concurrent blood Pb
17                  levels <10 (ig/dL, children with concurrent blood Pb level > 10 (ig/dL had increased
18                  parental report of an asthma diagnosis in the previous 12 months with an OR of 7.5 (95%
19                  CI: 1.3, 42.9) (Pugh Smith and Nriagu, 2011). This study was cross-sectional and
20                  produced an imprecise effect estimate; however, a strength of the study was the relatively
21                  extensive consideration for potential confounding, including adjustment for family-level
22                  income. As with Jedrychowski et al. (2011). Pugh Smith and Nriagu (2011) found an
23                  association with adjustment for smoking exposures in the home plus other indicators of
24                  housing exposures and condition (Table 5-34). The studies of asthma and allergy differed
25                  in which and how potential confounding factors were considered, particularly SES. While
26                  there is no single complete measure of SES, the various indicators used across these few
27                  studies produces uncertainty regarding residual confounding. Residual confounding also
28                  is possible by  factors not examined. The examination of maternal education and exposure
29                  to smoking or allergens in Jedrychowski et al. (2011) and family income, smoking,
30                  housing conditions, pets, or pests in Pugh Smith and Nriagu (2011) increase confidence
31                  in the associations observed for blood Pb levels. However, because evidence is limited to
32                  a few populations, there is uncertainty regarding potential confounding by SES and other
33                  exposures well characterized in the literature to be associated with asthma and allergy.

34                  With respect to blood Pb levels associated with atopic and inflammatory conditions,
35                  Joseph et al. (2005) found elevated incidence of asthma in Caucasian children with earlier
36                  childhood blood Pb levels > 5 (ig/dL and in African American children with blood Pb
37                  levels > 10 (ig/dL. Pugh Smith and Nriagu (2011) found higher asthma prevalence in
38                  children with concurrent blood Pb levels >  10 (ig/dL. Jedrychowski et al. (2011) found
      November 2012                             5-481                  Draft - Do Not Cite or Quote

-------
 1                   increased allergic sensitization in association with cord blood Pb levels that were low
 2                   (geometric mean: 1.16 (ig/dL) but that may have been affected by maternal higher past
 3                   Pb exposures mobilized from bone to blood during pregnancy.

 4                   Biological plausibility for the relationships found between blood Pb levels and asthma
 5                   and allergy in children is provided by evidence characterizing modes of action, namely, a
 6                   Pb-associated shift in production from Thl cytokines (e.g., IFN-y) to Th2 cytokines (e.g.,
 7                   IL-4) and increase in Th2-dependent IgE levels (Table 5-34). A majority of this evidence
 8                   was available in the 2006 Pb AQCD (U.S. EPA. 2006b). The shift from Thl to Th2
 9                   cytokine production in animals was found with prenatal or postnatal (4 weeks in
10                   juveniles, 3 weeks or 8 weeks in adults) dietary Pb exposures. In the studies available in
11                   humans (Table 5-34), higher concurrent blood Pb levels were associated with higher
12                   serum IL-4 in children (Lutz et al.. 1999) and higher serum IL-6 in nonoccupationally-
13                   exposed adults (adjusted for age, BMI, and current smoking status and additional
14                   adjustment for income, physical activity, education, and history of inflammatory
15                   conditions in the large NHANES analysis) (Songdej etal., 2010; Kim et al., 2007).
16                   Because of the  limitations in the small body of epidemiologic studies, i.e., the cross-
17                   sectional design of studies and inconsistent consideration for potential confounding, the
18                   epidemiologic evidence is a lesser consideration in drawing conclusions about
19                   Pb-associated cytokine changes. However, epidemiologic evidence does not detract from
20                   the clear toxicological evidence for Pb-induced increases in Th2 cytokine production.
21                   Coherence for a shift from Thl to Th2 cytokine production is found in the in vitro
22                   evidence for Pb-induced selective differentiation of naive T cells to a Th2 subtype (Heo
23                   et al.. 1998; 1996; McCabe and Lawrence. 1991). A recent study in adult mice and in
24                   vitro provided new evidence that Pb may promote the shift to Th2 responses by
25                   increasing production of Th2 cytokines in dendritic cells, the major effector in antigen
26                   response (Gao et al.. 2007).

27                   Additional mode of action support is provided by associations observed between higher
28                   concurrent blood Pb levels and higher serum IgE in several different populations of
29                   children (Section 5.6.3. Table 5-34). While most studies found elevated IgE in groups of
30                   children with concurrent blood Pb levels >10 (ig/dL, Karmaus et al. (2005) found higher
31                   serum IgE in children ages 7-10 years in Germany with blood Pb levels 2.8-3.4 (ig/dL
32                   compared with children with lower blood Pb levels. Some studies found increasing IgE
33                   across increasing blood Pb groups, except in the highest group (Karmaus et al.. 2005;
34                   Lutz et al.. 1999); however, a monotonic concentration-response relationship was not
35                   found in a recent study of children in Egypt (Hegazy et al.. 2011). Lutz et al. (1999)
36                   recruited children in Michigan from a public assistance program, and  Karmaus et al.
37                   (2005) recruited schoolchildren but excluded those from homes where more than 12
38                   cigarettes were smoked per day. The nature  of recruitment may limit generalizability of
      November 2012                             5-482                  Draft - Do Not Cite or Quote

-------
 1                  findings. Sufficient information was not reported to assess biased participation by Pb
 2                  exposure and history of allergy or asthma. The limited consideration for potential
 3                  confounding comprised adjustment for age (Karmaus et al., 2005; Lutz et al., 1999).
 4                  smoking exposure, serum lipids, blood organochlorine levels, and previous infections
 5                  (Karmaus et al., 2005) but not SES or allergen exposure. Although these findings were
 6                  based on cross-sectional analyses and had limited consideration for potential
 7                  confounding, they were supported by similar findings in animals, which are not subject to
 8                  reverse causation and confounding bias. Despite clear evidence in animals overall, there
 9                  was some inconsistency for Pb-induced increases  in IgE in animals with gestational or
10                  gestational/lactational dietary Pb exposures that resulted in blood Pb levels 5-20 (ig/dL,
11                  which are more relevant to humans (Chen et al.. 2004; Snyder et al.. 2000). Miller et al.
12                  (1998) found elevated IgE in adult F344 rats after gestational Pb exposure via drinking
13                  water of dams whose blood Pb levels peaked at 30-39 (ig/dL. There is lack of coherence
14                  between the consistent results for IgE and the inconsistent findings for Pb-induced
15                  activation of B cells, which differentiate into allergic antibody-producing cells
16                  (Section 5.6.3). However, there is strong mode of action support in animals for
17                  Pb-induced increases in IL-4, which stimulates differentiation of B cells.

18                  Further support for the effects of Pb exposure on increasing risk of atopic and
19                  inflammatory conditions is provided by evidence of Pb-associated inflammation
20                  (Section 5.6.4 and Table 5-34). Coherence for this evidence is found with findings for
21                  Pb-induced increases in IgE which primes basophils and mast cells to release pro-
22                  inflammatory mediators. Pb-induced inflammation is clearly demonstrated by a large
23                  toxicological evidence base for the effects of Pb exposure on inducing macrophages into
24                  a hyperinflammatory state as characterized by enhanced production of TNF-a, PGE2, and
25                  ROS. Inflammation was observed in rabbits exposed to Pb via air for 4 days (31 (ig/m3)
26                  (Zelikoff et al.. 1993) and rodents exposed via diet (250 ppm drinking water during
27                  gestation, 1.5 mg/kg food postnatally for 30 days) (Miller et al., 1998; Bavkov et al.,
28                  1996). Consistent with previous toxicological evidence, a large analysis of adults
29                  participating in NHANES  found an association between concurrent blood  Pb levels and
30                  serum CRP, an indicator of systemic inflammation, in 4,278 men with adjustment for
31                  age, BMI, income, physical activity, education, history of inflammatory conditions,
32                  cardiovascular disease, diabetes, and smoking status (Songdei et al.. 2010). Because only
33                  concurrent blood Pb levels were examined, there is uncertainty regarding the temporal
34                  sequence between Pb exposure and inflammation  and the magnitude, timing, frequency,
35                  and duration of Pb exposures that contributed to the observed associations. Because of
36                  the sparse epidemiologic evidence, it is a lesser consideration in drawing conclusions
37                  regarding the effects of Pb exposure on inflammation.
      November 2012                             5-483                  Draft - Do Not Cite or Quote

-------
 1                  With respect to important lifestages of Pb exposure, gestational Pb exposures, producing
 2                  blood Pb levels 8 and 20 (ig/dL, were found to affect endpoints such as IgE and/or
 3                  cytokine levels in juvenile and adult rodents (Chen et al.. 2004; Snvder et al.. 2000).
 4                  However, increases in Th2 cytokines also were found in adult animals with lifetime Pb
 5                  exposures beginning in gestation and producing blood Pb levels 1-12 (ig/dL (lavicoli et
 6                  al.. 2006b). The Pb exposure lifestage, magnitude, frequency, and duration associated
 7                  with atopic and inflammatory conditions are not well characterized in humans. Cord
 8                  blood Pb level was associated with allergic sensitization in children (Jedrychowski et al..
 9                  2011). whereas other studies of children and adults examined only concurrent blood Pb
10                  levels. Neither toxicological nor epidemiologic evidence clearly identifies an individual
11                  critical lifestage or duration  of Pb exposure that is more strongly associated with atopic
12                  and inflammatory  conditions. In children and adults, concurrent blood Pb levels are
13                  influenced by cumulative (from remodeling of bone) and recent Pb exposures. The
14                  combined evidence indicates that gestational and cumulative postnatal Pb exposures may
15                  influence atopic and inflammatory conditions.

16                  In conclusion, prospective studies in a few populations of children indicate associations
17                  of prenatal cord and earlier childhood blood Pb levels with asthma and allergy, with a
18                  cross-sectional study providing supporting evidence with associations with concurrent
19                  blood Pb level. Prospective design, lack of selective participation of subjects, and
20                  objective assessment of outcomes reduce the likelihood of undue selection bias and
21                  reverse causation.  These few studies varied in their consideration for potential
22                  confounding by SES and exposure to smoking or allergens. Thus, uncertainty remains
23                  regarding residual confounding in associations observed between blood Pb levels and
24                  asthma and allergy in children. The evidence for asthma and allergy is supported by
25                  cross-sectional associations found between higher concurrent blood Pb levels in children
26                  and higher IgE, an important mediator of asthma and allergy. The biological plausibility
27                  for the effects of Pb on IgE is provided by consistent findings in animals with gestational
28                  or gestational-lactational Pb  exposures, with some evidence at blood Pb levels relevant to
29                  humans. In epidemiologic studies, higher IgE  and higher asthma prevalence were
30                  examined and found in children with concurrent blood Pb levels >10 (ig/dL. Coherence
31                  for the evidence of Pb-associated increases in  asthma, allergy, and IgE is found with
32                  evidence for most of the examined endpoints related to mode of action, i.e., Pb-induced
33                  increases in Th2 cytokine production and inflammation in animals. Neither toxicological
34                  nor epidemiologic evidence  clearly identifies an individual critical lifestage or duration of
35                  Pb  exposure  associated with atopic and inflammatory conditions but indicates that
36                  gestational and cumulative postnatal Pb exposures may influence atopic and
37                  inflammatory conditions. The strong toxicological evidence supporting modes of action
38                  for a shift to  a Th2 phenotype combined with the epidemiologic evidence for asthma and
39                  allergy in a few populations  with some uncertainty regarding potential confounding is

      November 2012                             5-484                  Draft - Do Not Cite or Quote

-------
 1                  sufficient to conclude that a causal relationship is likely to exist between Pb exposures
 2                  and an increase in atopic and inflammatory conditions.
                    5.6.8.2     Evidence for Decreases in Host Resistance

 3                  Evidence indicates that a causal relationship is likely to exist between Pb exposure and
 4                  decreased host resistance based on consistent observations that relevant Pb exposures
 5                  decrease responses to antigens (i.e., suppresses DTH) and increase bacterial titers and
 6                  subsequent mortality in rodents (Table 5-34. Sections 5.6.2.3 and 5.6.5.1). A majority of
 7                  this evidence was available in the 2006 Pb AQCD (U.S. EPA. 2006b). The studies that
 8                  reported blood Pb levels demonstrated increased bacterial titers and mortality with adult-
 9                  only 16 week Pb exposure via drinking water in adult mice with Salmonella infection and
10                  blood Pb level 20 (ig/dL (Fernandez-Cabezudo et al.. 2007) and with lactational
11                  (PND1-PND22) Pb exposure in juvenile mice with Listeria infection and blood Pb level
12                  25  (ig/dL (Dyatlov and Lawrence, 2002). DTH was suppressed in adult rats with blood
13                  Pb  levels 6 and 25 (ig/dL after gestational Pb exposure in drinking water (Chen et al..
14                  2004; Bunnetal. 200la). While a few epidemiologic studies found higher prevalence of
15                  respiratory infections in children with higher concurrent blood Pb levels (Karmaus et al..
16                  2005; Rabinowitz et al.. 1990) and Pb-exposed workers (Ewers etal.. 1982). the
17                  implications are limited by the lack of rigorous statistical analysis (i.e., regression) and
18                  consideration for potential confounding. These limitations also apply to the recent cross-
19                  sectional evidence of Pb-related increases in respiratory infections in children (Carreras
20                  et al.. 2009) (Table 5-34). These limitations produce uncertainty about the effects of Pb
21                  exposure on decreased host resistance in humans but do not detract from the  consistent
22                  evidence in animals.

23                  The effects of Pb on decreased host resistance are well supported by evidence describing
24                  underlying modes of action (Table 5-34). Evidence in animals indicates Pb-induced
25                  functional impairment of macrophages, which phagocytize pathogens. Decreased
26                  macrophage colony formation was found in rats after gestational Pb exposure (Bunn et
27                  al.. 2001b). and decreased phagocytic activity was found in mice and turkeys after
28                  lactational or 2-week juvenile Pb exposure (Knowles and Donaldson. 1997; Kowolenko
29                  et al.. 1991). Additional coherence for Pb-induced decreased host resistance is found with
30                  observations in animals that gestational Pb exposure suppressed macrophage production
31                  of NO which is involved in bacteria killing (Section 5.6.6.2) and postnatal Pb exposure
32                  (air for 4 days, food for 30 days) increased production of ROS and PGE2, which mediate
33                  tissue damage (Section 5.6.6.3). Similarly, a cross-sectional epidemiologic study found a
34                  smaller release of NO and larger release of superoxide anion from macrophages of
35                  children with higher concurrent blood Pb levels (10.3-47.5 versus <10.3 (ig/dL) (Pineda-


      November 2012                            5-485                  Draft - Do Not Cite or Quote

-------
 1                   Zavaleta et al.. 2004) after adjustment for age and sex. Because of the limited
 2                   consideration for potential confounding in this study and examination of higher blood Pb
 3                   levels than those in most of contemporary U.S. children, the results are a lesser
 4                   consideration in drawing conclusions about the effects of Pb on macrophages. However,
 5                   they do suggest the relevance of toxicological observations to humans. The killing
 6                   capability of macrophages is enhanced by the Thl cytokine IFN-y. Therefore, an effect of
 7                   Pb exposure on decreased host resistance is additionally supported by clear evidence in
 8                   animals  for the effects of Pb exposure on suppressing IFN-y production (Section 5.6.6.1).
 9                   A recent study in mice indicated that Pb-induced suppression of DTH may be mediated
10                   by a shift in production from Thl to Th2  cytokines specifically in dendritic cells (Gao et
11                   al.. 2007).

12                   Some evidence did not contribute strong  support for the mode of action for Pb-induced
13                   decreased host resistance. Pb-exposed workers were found to have reduced functionality
14                   of neutrophils, which respond to bacterial infection (Table 5-34. Section 5.6.2.5) but
15                   without  consideration for potential confounding or analogous toxicological evidence.
16                   Neither epidemiologic nor toxicological evidence clearly demonstrated an effect of Pb
17                   exposure on NK cells, which respond to viral infection (Section 5.6.2.7).

18                   With respect to important lifestages of Pb exposure, animal studies found that gestational
19                   Pb exposures, producing blood Pb levels  of 6 and 25 (ig/dL, resulted in decreases in Thl
20                   cytokines, suppression of DTH, and greater susceptibility to bacterial infection (Chen et
21                   al.. 2004; Bunnetal.. 200la).  However, these effects related to decreased host resistance
22                   in mice also were affected by postnatal lactational (Dyatlov and Lawrence. 2002). adult
23                   long-term (>4 weeks) (Fernandez-Cabezudo et al.. 2007). and lifetime Pb exposures
24                   beginning in gestation in adult mice (lavicoli et al.. 2006b) that produced blood Pb levels
25                   1-25 (ig/dL. Thus, the animal toxicological evidence does not clearly identify a particular
26                   lifestage of Pb exposure that is more strongly associated with decreased host resistance.

27                   In conclusion, decreased host resistance is demonstrated by several toxicological
28                   observations that dietary Pb exposure producing relevant blood Pb levels increased
29                   susceptibility to bacterial infection and suppressed DTH in rodents and by the coherence
30                   with evidence describing modes of action, including suppressed production of Thl
31                   cytokines and decreased macrophage function in animals. These effects were found with
32                   gestational, lactational, adult-only, and lifetime Pb exposures of animals. Cross-sectional
33                   epidemiologic evidence indicates Pb-associated increases in respiratory infections but
34                   limitations, including the lack of rigorous methodology and consideration for potential
35                   confounding produce uncertainty  in the epidemiologic evidence for decreased host
36                   resistance in humans. The consistent toxicological evidence in animals but uncertainty in
37                   the epidemiologic evidence for decreased host resistance in humans is sufficient to
      November 2012                             5-486                  Draft - Do Not Cite or Quote

-------
 1                  conclude that a causal relationship is likely to exist between Pb exposure and decreased
 2                  host resistance.
                    5.6.8.3     Evidence for Autoimmunity

 3                  Toxicological evidence indicates the potential of Pb to increase autoimmunity, with a few
 4                  previous studies showing Pb-induced generation of auto-antibodies (Hudson et al., 2003;
 5                  Bunn et al.. 2000; El-Fawal et al..  1999; Waterman et al.. 1994) and recent studies
 6                  providing indirect evidence by showing formation of neoantigens that could result in the
 7                  formation of auto-antibodies (Table 5-34). Several observations were made in animals
 8                  injected with Pb, which is a route of exposure with less relevance to humans. Higher
 9                  levels of auto-antibodies also were found in Pb-exposed battery workers; however,
10                  implications are limited because of the high blood Pb levels (range: 10-40 (ig/dL) of
11                  some of the workers and lack of consideration for potential confounding by several
12                  factors, including other occupational exposures (El-Fawal et al., 1999). Because results
13                  from available toxicological and epidemiologic studies do not sufficiently inform
14                  Pb-induced generation of auto-antibodies with relevant Pb exposures, the evidence is
15                  inadequate to determine if there is a causal relationship between Pb exposure and
16                  autoimmunity.
     November 2012                            5-487                  Draft - Do Not Cite or Quote

-------
Table 5-34   Summary of evidence supporting immune causal determinations.
Attribute in
Causal
Framework3
Key Supporting Evidence
References
                                                                          Pb Exposure or
                                                                          Blood Levels
                                                                          Associated with
                                                                          Effects0
Increase in Atopic and Inflammatory Conditions - Likely Causal
Associations
consistently
found in high-
quality
epidemiologic
studies with
relevant blood
Pb levels
Prospective studies indicate higher asthma and
allergy incidence in association with earlier
childhood or prenatal blood Pb levels in children
(ages 1-5 yr) in a few populations (U.S., Poland)



Joseph et al. (2005).
Jedrychowski et al. (2011).


Section 5.6.5.2


Children (ages 1-3
yr) with blood Pb
levels measured
earlier in childhood
>10 ug/dL
Prenatal (cord):
geometric mean
1.16 ug/dL
                 Some studies report high participation and/or
                 follow-up retention, not conditional on Pb
                 exposure or outcome.

                 Some studies objectively assessed outcomes
                 with clinical testing, medical records.

                 Adjustment or consideration for potential
                 confounding by SES, exposure to smoking,
                 and/or allergen  exposure.
                 Heterogeneity in evaluation of potential
                 confounding among the few available studies
                 produces uncertainty regarding potential
                 confounding.

                 Supporting cross-sectional evidence in children
                 (ages 6 mo-10 yr) for increases in IgE but with
                 limited consideration for potential confounding
                 factors, particularly SES. Associations observed
                 in studies in U.S., Europe, Asia; insufficient
                 information to assess potential selection bias.
                 Evidence for C-R varies for IgE. Some studies
                 show increasing IgE across blood Pb groups,
                 except in highest group.

                 Another study did not show monotonic C-R
                 relationship.
                                           Joseph etal. (2005).
                                           Jedrychowski et al. (2011)
                                           Jedrychowski et al. (2011),
                                           Pugh Smith and Nriagu (2011)

                                           Jedrychowski et al. (2011).
                                           Pugh Smith and Nriagu (2011)
                                           Children:
                                           Karmauset al. (2005).
                                           Hegazy et al. (2011).
                                           Lutz et al. (1999).
                                           Hon et al. (2010: 2009),
                                           Sun et al. (2003)
                                           Section 5.6.3

                                           Karmauset al. (2005).
                                           Lutz et al. (1999)
                                           Hegazy et al. (2011
                               Groups (ages 6 mo
                               - 10yr) with
                               concurrent blood Pb
                               levels >10 ug/dL
Epidemiologic
evidence
supported by
toxicological
evidence at
relevant
exposures






Most animal studies show elevated IgE in
animals with prenatal and postnatal dietary Pb
exposures. Some inconsistency in animals with
relevant Pb concentrations.









Increase in IgE:
Snyderetal. (2000).
Miller et al. (1998)






No IgE increase in
Chen etal. (2004)
Also see Section 5.6.3

Increased IgE with
gestational-
lactational Pb
exposure, blood Pb
means 5, 20 ug/dL
Gestational Pb
exposure producing
maternal blood Pb
peak: 30-39 ug/dL
No increase with
gestational Pb
exposure, blood Pb
means 7-8 ug/dL
November 2012
                               5-488
             Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Evidence clearly
describes mode
of action
Stimulation of
Th2 phenotype
Key Supporting Evidence13 References'1
Extensive, consistent evidence of increased Section 5.6.6.1
production of Th2 cytokines (e.g., IL-4, IL-6, &
Pb Exposure or
Blood Levels
Associated with
Effects0

                   IL-10) in animals with prenatal and postnatal
                   (4 weeks juvenile, 3, 8 weeks adults) dietary Pb
                   exposures. Recent evidence for the role of
                   dendritic cells in mediating Th2 shift.

                   Limited available cross-sectional evidence in
                   children, adults. Epidemiologic evidence has
                   limited consideration for potential confounding. Is
                   not a major consideration in conclusions.
                   The few available in vitro studies indicate
                   activation Th2 cells from naTve T cells or over
                   Th1 cells.

  Inflammation      Extensive evidence for increased production of
                   TNF-a, IL-6, ROS, PGE2 by macrophages from
                   animal with prenatal and postnatal (dietary
                   4 weeks juvenile, dietary 3, 8 weeks adults, air
                   4 days adults) Pb exposure. Supported by in
                   vitro evidence.

                   Cross-sectional association observed in children
                   living near Pb source, adjusted for confounding
                   by age and sex but not other factors such as
                   SES. Cross-sectional evidence in adults in
                   NHANES that adjust  for  inflammatory conditions,
                   smoking and SES. Is not a major consideration
                   in conclusions.
  B cell            Inconsistent toxicological evidence in animals for
  activation         B cell activation by Pb exposure concentration
                   and duration and strain.

                   Inconsistent epidemiologic evidence for B cell
                   abundance, B cell activation not examined.
             Table 5-7 of the 2006 Pb AQCD
             (U.S. EPA. 2006g)
             Children:
             Lutz et al. (1999)
             Adults:
             Kim et al.  (2007)
             Section 5.6.6.1
                                                                 Section 5.6.2.1
             Sections 5.6.6.1 and 5.6.6.3
             Children:
             Pineda-Zavaleta et al. (2004)
             Sections 5.6.6.2. and 5.6.6.3
             Adults:
             Songdej et al. (2010)
             Sections 5.6.4


             Section 5.6.3
             Table 5-32 and Section 5.6.3
                   Children (ages 6 mo-
                   6yr):
                   Concurrent blood Pb
                   group range
                   15-19ug/dL
                   Adults:
                   Group range
                   2.5-10.5 ug/dL
                   Children (ages
                   6-11 yr):
                   Concurrent blood Pb
                   group
                   >10 ug/dL
                   Adults:
                   Concurrent group
                   >1.16 ug/dL
Decreases in Host Resistance - Likely Causal
Consistent
toxicological
evidence with
relevant
exposures






Available
epidemiologic
evidence is not
sufficiently
informative






The few studies with relevant dietary Pb
exposures demonstrate increased bacterial
infection, sickness behavior, and mortality in
mice. Similar observations in several other
studies with higher Pb exposures.


The few studies with relevant prenatal dietary Pb
exposures show suppressed DTH in rodents.
Similar observations in several other studies with
higher Pb exposures.
Epidemiologic studies found associations with
increased respiratory infections but limitations
include lack of consideration for potential
confounding, rigorous statistical analysis, or Pb
biomarker assessment, and/or ecological study
design






Dyatlov and Lawrence (2002).
Fernandez-Cabezudo et al.
(2007)
Section 5.6.5.1



Chenetal. (2004),
Bunn et al. (2001 a: 200 1c)
Section 5.6.2.3

Children:
Karmauset al. (2005),
Rabinowitz et al. (1990).
Carreras et al. (2009)


Pb-exposed workers:
Ewers (1 982)
Section 5.6.5.1

Blood Pb means
20 ug/dL after adult
1 6-week
Pb exposure,
25 ug/dL after
lactational
Pb exposure
Blood Pb means:
6.75, 25 ug/dL after
gestational Pb
exposure
Children (ages 7-10
yr):
Group with
concurrent blood Pb
>3.34 ug/dL,
Group with cord
blood Pb>10 ug/dL
Pb-exposed
workers:
Concurrent blood Pb
21-85 ug/dL
November 2012
5-489
Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
References
                                                                               Pb Exposure or
                                                                               Blood Levels
                                                                               Associated with
                                                                               Effects0
Evidence clearly
describes mode
of action

  Decreased
  macrophage
  function
  Decreased
  Th1 cytokine
  (IFN-y)
  production
Decreased macrophage colony formation in
animals with dietary prenatal and postnatal Pb
exposure; not widely examined.

Decreased macrophage phagocytosis in animals
and in cell culture, not widely examined.

Several studies demonstrate decreased NO
production by macrophages from animals with
prenatal and postnatal Pb exposure. Supported
by in vitro evidence.

Cross-sectional association of decreased NO in
macrophages of children living near Pb source
with higher concurrent blood Pb level, adjusted
for age and sex but not SES.
Inconsistent evidence in Pb-exposed workers
but for macrophage abundance, not function.


Consistent evidence from a large body of
lexicological studies with prenatal and postnatal
(4 weeks juvenile, 3, 8 weeks adults) dietary Pb
exposures of animals.
                                              Section 5.6.2.4
                                                                Section 5.6.2.4
                                                                Section 5.6.6.2
                                                                Pineda-Zavaleta et al. (2004)
                                                                Section 5.6.6.2
Pinkerton et al. (1998).
Fischbein et al. (1993).
Conterato (In Press)

Section 5.6.6.1 and
Table 5-7 of the 2006 Pb AQCD
(U.S. EPA.2006g)
                                 Children (ages
                                 6-11 yr):
                                 Group with
                                 concurrent blood Pb
                                 >10 ug/dL
Autoimmunity - Inadequate
Available
toxicological and
epidemiologic
evidence is not
sufficiently
informative
A study in rats shows generation of auto-
antibodies with relevant adult-only dietary Pb
exposure for 4 days. Several other studies have
Pb exposure concentrations and/or routes
(e.g., i.p.) with uncertain relevance to humans.
Rats:
EI-Fawaletal. (1 999)
Section 5.6.5.4
Rats:
Blood Pb level
range
11-50 ug/dL
                  Evidence for increased auto-antibodies in
                  Pb-exposed workers with high blood Pb levels
                  and limited consideration for potential
                  confounding, including other workplace
                  exposures.
                                              Workers:
                                              EI-Fawaletal. (1999)
                                              Section 5.6.5.4
                                Workers:
                                Blood Pb level
                                range:
                                10-40ug/dL
Described in detail in Table II of the Preamble.
bDescribes the key evidence and references contributing the most heavily to causal determination. References to earlier sections
indicate where full body of evidence is described.
""Describes the blood Pb levels in children with which the evidence is substantiated and blood Pb levels in animals most relevant to
humans.
November 2012
                                 5-490
              Draft - Do Not Cite or Quote

-------
      5.7        Hematological Effects
      5.7.1       Introduction

 1                  The effects of Pb exposure on red blood cell function and heme synthesis have been
 2                  extensively studied over several decades in both human and animal studies. The 1978
 3                  National Ambient Air Quality Standard for Lead was set to protect blood Pb levels in
 4                  children from exceeding 30 (ig/dL as such levels were associated with impaired heme
 5                  synthesis, evidenced by accumulation of protoporphyrin in erythrocytes (U.S. EPA.
 6                  1978).

 7                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported that Pb exposure significantly decreases
 8                  several hematological parameters including hemoglobin (Hb), hematocrit (Hct), mean
 9                  corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean
10                  corpuscular hemoglobin concentration (MCHC). Further, the 2006 Pb AQCD reported
11                  that Pb affects developing red blood cells (RBCs) in children and occupationally exposed
12                  adults as noted by anemia observed with blood Pb >40 (ig/dL. Pb-induced anemia is
13                  thought to occur due to decreased RBC life span and effects on Hb synthesis. The  exact
14                  mechanism for these effects is not known, although Pb-induced changes on iron uptake or
15                  inhibition of enzymes in the heme synthetic pathway may be responsible. Once Pb enters
16                  the cells, it is predominantly found in protein-bound form, with Hb and aminolevulinic
17                  acid dehydratase (ALAD) both identified as targets.

18                  Consistent with the majority of human evidence that high Pb blood levels
19                  (i.e., >20 (ig/dL) are associated with decreased hematological indices, blood Pb levels
20                  >100 (ig/dL were associated with decreased RBC survival in laboratory animals. Effects
21                  on RBC membrane mobility were observed at blood Pb levels as low as 10 (ig/dL,
22                  although the  precise mechanisms underlying these effects of Pb are not known. Studies
23                  conducted in animal and in vitro models provide evidence of multiple other effects on
24                  RBC membranes, including altered microviscosity and fluidity, decreased sialic acid
25                  content, decreased lamellar organization, decreased lipid resistance to oxidation (possibly
26                  mediated by perturbations in RBC membrane lipid profiles), and increased permeability.
27                  These alterations to RBC membranes may indicate potential modes of action by which Pb
28                  induces RBC fragility, abnormal cellular function, RBC destruction, and ultimately
29                  anemic conditions. Pb exposure also has been shown to result in increased activation of
30                  RBC scramblase, an enzyme responsible for the expression of phosphatidylserine (PS) on
31                  RBC membranes. This expression of PS decreases the life span of RBCs via phagocytosis
32                  by macrophages. Pb  exposure has been observed to alter the phosphorylation profiles of
      November 2012                            5-491                 Draft - Do Not Cite or Quote

-------
 1                  membrane proteins, which may influence the activity of membrane enzymes and the
 2                  functioning of receptors and channels located on the membrane.

 3                  The 2006 Pb AQCD reported that Pb exposure affects heme synthesis in humans and
 4                  animals through the inhibition of multiple key enzymes, most notably ALAD, the enzyme
 5                  that catalyzes the second, rate-limiting step in heme biosynthesis (Figure 5-36 presents a
 6                  schematic representation of the heme biosynthetic pathway). The 2006 Pb AQCD (U.S.
 7                  EPA. 2006b) further reported that decreased RBC ALAD activity is the most sensitive
 8                  measure of human Pb exposure, in that measurement of ALAD activity is correlated with
 9                  blood Pb levels. Concentration-response changes in the ratio of activated/nonactivated
10                  ALAD activity  in avian RBCs were observed to be not dependent on the method of Pb
11                  administration.  The Pb-associated inhibition of the ALAD enzyme was consistently
12                  observed in RBCs from multiple species, including birds, cynomolgous monkeys, and
13                  humans. Pb was also observed to inhibit other enzymes responsible for heme
14                  biosynthesis, including ferrochelatase, porphobilinogen (PEG) deaminase, and
15                  coproporphyrinogen oxidase. Pb also potentially alters heme biosynthesis through
16                  inhibition of transferrin (TF) endocytosis and iron transport.

17                  Pb has been found to alter RBC energy metabolism through inhibition of enzymes
18                  involved in anaerobic glycolysis and the pentose phosphate pathway. Pb was also found
19                  to inhibit pyrimidine 5'-nucleotidase (P5N) activity, and the 2006 Pb AQCD indicated
20                  that this might be another biomarker of Pb exposure. Inhibition of P5N results in an
21                  intracellular increase in pyrimidine nucleotides leading to hemolysis and potentially
22                  ultimately resulting in anemic conditions. The 2006 Pb AQCD indicated that
23                  perturbations in RBC energy metabolism may be related to significant decreases in levels
24                  of nucleotide pools, including nicotinamide adenine nucleotide (NAD), possibly due to
25                  decreased NAD synthase activity, and nicotinamide adenine nucleotide phosphate
26                  (NADP) accompanying significant increases  in purine degradation products.

27                  The 2006 Pb AQCD identified oxidative stress as an important potential mode of action
28                  by which Pb exposure induced effects on RBCs. Increased lipid peroxidation and
29                  inhibition of antioxidant enzymes in RBCs (e.g., superoxide dismutase [SOD], catalase
30                  [CAT])  were observed following exposure to Pb.

31                  The epidemiologic and toxicological studies published since the 2006 Pb AQCD, largely
32                  support the reported Pb-associated effects on RBC function and heme synthesis.
33                  Epidemiologic studies support previous observations that occupationally-exposed adults
34                  with higher blood Pb levels than the current U.S. general population (>26 (ig/dL) have
35                  decreased RBC numbers. However, a few epidemiological studies investigating
36                  occupationally-exposed adults and pregnant women provide some evidence that more
37                  relevant blood Pb levels, <10 (ig/dL, are associated with decreased RBC numbers,


      November 2012                            5-492                  Draft - Do Not Cite or Quote

-------
 1                  possibly through decreased survival of the RBCs. Effects seen in children are largely
 2                  consistent with those observed in adults, and a number of toxicological studies support
 3                  findings observed in human populations. Recent epidemiologic and toxicological studies
 4                  also support previous findings that Pb exposure in adults, children, and laboratory
 5                  animals decreases ALAD activity, as well as the activity of other enzymes in the heme
 6                  biosynthetic pathway. Recent epidemiologic and toxicological studies expand upon the
 7                  evidence that Pb exposure results in oxidative stress in RBCs. Although the
 8                  epidemiologic studies included below are cross-sectional in study design, they do
 9                  improve upon earlier studies as more studies characterize the effects in children, and
10                  investigate effects in populations with blood Pb levels more comparable to those in the
11                  current U.S. population. Additionally, the associations observed in these cross-sectional
12                  studies are supported by a large number of animal toxicology studies.
      5.7.2        Red Blood Cell Function

13                  As stated in the 2006 Pb AQCD (U.S. EPA. 2006b). Pb poisoning in children has been
14                  associated with anemia. As of 2006, the mechanism for this was not clear, but it was
15                  determined not to be due to iron deficiency, which can be found to occur independently
16                  of Pb exposure. However, Zimmerman et al. (2006) found that blood Pb level differences
17                  were statistically significant, and lower in non-anemic (or mildly-anemic) iron-
18                  deficient 5- to 9-year-old children in India fed an iron-fortified diet for 30 weeks,
19                  compared to 14 weeks (mean [range]: 8.1 [3.1-219] (ig/dL versus 12.1 [3.7-26.8] (ig/dL;
20                  p <0.02); however, blood Pb levels were not significantly lower in children receiving the
21                  no-iron diet for 30 weeks compared to 14 weeks (mean [range]:  10.2 [4.4-25.3]  (ig/dL
22                  versus  12.0 [3.8-25.5] (ig/dL). Although a number of epidemiologic  studies found
23                  decreases in RBCs and/or Hct levels associated with higher blood Pb levels, it is not
24                  known whether this is due to reduced RBC survival or a decrease in RBC production.
25                  Regardless, decreased RBC survival and hematopoiesis can be expected to occur
26                  simultaneously, and any effect on RBC numbers is likely a combination of the two modes
27                  of action.
                    5.7.2.1      Pb Uptake, Binding, and Transport into Red Blood Cells

28                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported that Pb uptake into human RBCs occurs
29                  via passive anion transport mechanisms. Although Pb can passively cross the membrane
30                  in both directions, little of the Pb is found to leave the cell after entry. Simons (1993b)
31                  found that in vitro uptake of 203Pb (1-10 (iM) occurred via an anion exchanger while the
32                  efflux occurred via a vanadate-sensitive pathway. After entry into the RBC, radioactive

      November 2012                            5-493                  Draft - Do Not Cite or Quote

-------
 1                  Pb was found to partition with Hb at a ratio estimated to be about 6000:1 bound to
 2                  unbound (Simons. 1986). However, Bergdahl et al. (1997a) suggested that ALAD was
 3                  the primary Pb-binding protein and not Hb. The 2006 Pb AQCD also reported that the
 4                  majority (approximately 98%) of Pb accumulates in RBC cytoplasm bound to protein,
 5                  and only about 2% is found in the membrane. This is related to the high ratio of Pb in
 6                  RBCs compared to plasma Pb. Further information on Pb binding and transport in blood
 7                  can be found in the kinetics section of Chapter 4 (Section 4.2).

 8                  Although no recent studies were identified that examined transport of Pb into RBCs, Lind
 9                  et al. (2009) recently observed that several Zn ionophores (8-hydroxyquinoline
10                  derivatives and Zn and Na pyrithione) were able to effectively transport Pb out of RBCs
11                  into the extracellular space.
                    5.7.2.2     Red Blood Cell Survival, Mobility, and Membrane Integrity

12                  A number of cross-sectional studies have investigated the effect of exposure to Pb on
13                  various inter-connected and related hematological parameters in children and adults. As
14                  these studies were cross-sectional in design, there is uncertainty regarding the
15                  directionality of effects and the magnitude, timing, frequency, and duration of Pb
16                  exposure that contributed to the observed associations. Additionally, unless explicitly
17                  noted, potential confounding by subject characteristics and other workplace or residential
18                  exposures was not accounted for in these studies. Adults and children exposed to Pb may
19                  also have been co-exposed to other contaminants that can affect the hematological
20                  system, and the potential for co-exposure was not assessed in most studies.

21                  In an earlier cross-sectional study of children in Idaho (aged 1-9 years) with blood Pb
22                  levels ranging from 11 to 165 (ig/dL (approximately 40% were >40 (ig/dL),  a 10%
23                  probability of anemia (Hct <35%) was predicted (in association with blood Pb levels of
24                  ~20 (ig/dL [age 1 year], 50 (ig/dL [age 3 years] and 75 (ig/dL [age 5 years])  (Schwartz et
25                  al.. 1990). More recent studies have also demonstrated adverse effects on hematological
26                  parameters in children due to the Pb exposure. Ahamed et al.  (2006) studied 39 male
27                  urban adolescents in India who were separated into groups according to their blood Pb
28                  level (Group 1: <10 (ig/dL [mean 7.4 (ig/dL],  Group 2: >10 (ig/dL [mean 13.27 (ig/dL]).
29                  Although the groups were similar in age (mean [SD]: 16.59 [0.91] versus 16.76 [0.90]
30                  years, respectively), height, weight, and BMI (therefore, not considered to be potential
31                  confounders), Group  2 had a significantly lower packed cell volume (PCV) compared to
32                  Group 1. In a related  study, Ahamed et al. (2007) investigated the relationship between
33                  blood Pb level, anemia, and other hematological parameters in urban children in India
34                  (n = 75). Children were split into two groups as above: Group 1 had blood Pb levels
      November 2012                            5-494                 Draft - Do Not Cite or Quote

-------
 1                  <10 (ig/dL (mean [SD]: 6.89 [2.44] (ig/dL, n = 19), whereas Group 2 had blood Pb levels
 2                  >10 (ig/dL (mean [SD]: 21.86 [7.58] (ig/dL, n = 56). As with the earlier study, ages were
 3                  similar between the two groups: mean [SD]: 4.68 [1.49] and 4.11  [1.77] years,
 4                  respectively. Hb and Hct were significantly decreased in Group 2, compared to Group 1
 5                  after adjusting for age, sex, and area of residence. Children in Group 2 had an increased
 6                  odds ratio of anemia (OR:  2.87 [95%  CI: 1.60, 2.87]) compared to Group 1 after
 7                  adjustment for age, sex, and area of residence.

 8                  In a cross-sectional study measuring blood Hb as the independent variable, blood Pb
 9                  levels were observed to decrease with increasing blood Hb. Riddell et al.  (2007) found
10                  that 21% of children, who  were 6 months to 5 years of age living  in the rural Philippines,
11                  had concurrent blood Pb levels >10 (ig/dL (total population mean: 6.9 (ig/dL). After
12                  controlling for potential confounding  by age, sex, birth weight, and history of
13                  breastfeeding, Hb levels were inversely related to blood Pb level,  with a decrease of 3%
14                  blood Pb associated with every 1 g/dL increase in Hb. Among children aged 6-36 months
15                  (n = 222) living in Montevideo, Uruguay, 32.9% had blood Pb levels greater than
16                  10  (ig/dL (population mean [SD]: 9.0 [6.0] (ig/dL) (Oueirolo etal.. 2010V The mean
17                  [SD] Hb concentration was 10.5 [1.5] g/dL, and 44.1% of children were diagnosed as
18                  anemic (Hb <10.5 g/dL). Blood Pb levels were significantly higher in anemic  children
19                  compared to non-anemic (mean  [SD]: 10.4 [6.8] versus 7.9 [5.1] (ig/dL),  and anemic
20                  children were more likely to have elevated blood Pb after controlling for age and
21                  mouthing behavior (OR =  1.9, 95% CI: [1.098, 3.350]). The likelihood of elevated blood
22                  Pb  was more pronounced in anemic children younger than 18 months (OR = 3.1, 95% CI:
23                  [1.3,7.4]).

24                  Similarly, in a cross-sectional study of 340 children (aged 1-5 years) from Karachi,
25                  Pakistan, mildly-anemic and severely-anemic children (mean [SD] Hb levels:  8.9 [0.9]
26                  and 7.4 [0.5] g/dL, respectively) had lower Hb levels but higher blood Pb levels
27                  compared to non-anemic children (mean [SD] Hb: 12.1 [1.3] g/dL). Mean [SD] blood Pb
28                  levels in the mildly-anemic, severely-anemic, and non-anemic children were 14.9  [0.81],
29                  21.4 [2.7], and 7.9 [1.7] ng/dL, respectively (p <0.01) (Shah etal.. 2010). Additionally,
30                  Hct, RBC count, and MCV were all decreased in anemic children versus non-anemic
31                  children. Although statistical analyses were not reported, the levels of Hb, Hct, RBC
32                  count, and MCV in anemic children all fell outside of the reported normal range for these
33                  parameters, whereas the reported values in non-anemic children did not. Blood Pb level
34                  was negatively correlated with Hb level in all groups, with the magnitude of negative
35                  correlation increasing  with increasing severity of anemia: r = -0.315 (non-anemic
36                  children), -0.514 (mildly-anemic), and -0.685 (severely- anemic). In iron-deficient
37                  anemic children (n = 23) from Denizli, Turkey, mean (SD) serum Pb levels were
38                  statistically (p <0.05) increased compared to healthy children (n = 179): 0.013 (0.004)
      November 2012                            5-495                  Draft - Do Not Cite or Quote

-------
 1                  versus 0.008 (0.001) (ig/dL, respectively (Turgut et al., 2007). The iron-deficient children
 2                  were observed to have decreased Hb, MCV, RBC, and ferritin compared to controls, but
 3                  increased RDW. In 140 children from southern Brazil aged 2-11 years, living within 25
 4                  km of a Pb smelter, blood Pb levels were not observed to differ between anemic and
 5                  non-anemic children (mean [SD]:  10.36 [6.8] versus 9.73 [5.8] (ig/dL, p = 0.98) (Rondo
 6                  et al.. 2006). However, blood Pb levels were significantly negatively correlated with Hb
 7                  in anemic children (r = -0.41, p = 0.01); this relationship was not observed in non-anemic
 8                  children (r = 0.018, p = 0.84).

 9                  In children aged 5-9 years (n = 189) without anemia living in Cartagena, Columbia, a
10                  smaller percentage (4.7%) of children had blood Pb >10 (ig/dL (mean [SD]: 5.49
11                  [0.23] (ig/dL). The only hematological parameters that fell outside of their reference
12                  values were MCV and MCH, which were negatively correlated with blood Pb levels
13                  (r = -0.159 [p = 0.029] and -0.171  [p = 0.019], respectively) (Olivero-Verbel et al.. 2007).
14                  RBC count, which was not observed to differ from reference values, was positively
15                  correlated with blood Pb level (r = 0.208, [p =  0.004]). In a group of 268 Lebanese
16                  children, children aged 11-23 months with blood  Pb levels >10 (ig/dL had increased
17                  likelihood of having iron-dependent anemia and transferrin saturation (TF <12%)
18                  compared to age-matched children with blood  Pb  levels <10 (ig/dL (OR = 4.59, 95% CI:
19                  [1.51, 13.92]) (Muwakkit et al.. 2008). In children aged 24-35 months, higher blood Pb
20                  level was not associated with increased likelihood of either effect. Huo et al. (2007)
21                  found that children (less than 6 years of age) living near an area where electronic waste
22                  was recycled in China had significantly  higher mean blood Pb levels than did children in
23                  the neighboring town with no waste recycling  (15.3 versus 9.94 (ig/dL).  However,
24                  contrary to the findings above, no  difference was detected in the mean Hb levels of the
25                  children in the two towns (12.76 g/dL in children  from the waste recycling town versus
26                  12.35 g/dL in children from the town with no recycling).

27                  In adult, occupationally exposed populations, decreased erythrocyte numbers and Hb
28                  were observed in multiple, earlier  cross-sectional  studies investigating workers with
29                  blood Pb levels >40 (ig/dL (Solliwav et al.. 1996;  Horiguchi et al.. 1991; Poulos et al..
30                  1986). However, a larger, longitudinal study (Hsiao etal. 2001) observed that
31                  occupationally-exposed adults exhibited erythrocyte counts and Hct that were positively
32                  associated with blood Pb levels. Most of the recent occupationally-exposed groups
33                  represent populations highly exposed to Pb, with mean blood Pb levels ranging from 26-
34                  74 (ig/dL. Although effects observed within these groups may not be generalizable to the
35                  general population as a whole, they are useful  in demonstrating consistent effects on a
36                  number of hematological parameters, including Hb, MCV, MCH, MCHC, total RBCs,
37                  and packed cell volume (PCV) (Khan etal.. 2008: Patil et al.. 2006a: Patil et al.. 2006b:
38                  Karita et al.. 2005).
      November 2012                            5-496                  Draft - Do Not Cite or Quote

-------
 1                  A few recent cross-sectional occupational studies did investigate the effect of moderate
 2                  occupational Pb exposure on hematological parameters. In gas station attendants in
 3                  Sarajevo (Bosnia and Herzegovina), workers (mean [SD] duration of exposure:  12.1 [9.1]
 4                  years) had significantly increased blood Pb levels (mean: 5.96 (ig/dL) in 2008, compared
 5                  to the same population that were previously examined in 2003 (mean: 4.07 (ig/dL; mean
 6                  [SD] duration of exposure: 10.4 [5.5]). Levels of MCH and MCHC  were significantly
 7                  decreased when assessed in 2008, compared to the 2003  measurements, although RBC
 8                  numbers, Hb, Hct, and MCV were increased in 2008 compared to 2003. Positive
 9                  correlations were observed in all subjects between blood Pb and RBC count, Hb, and
10                  MCH (r = 0.241, 0.201, and 0.213, respectively;  p <0.05). No control group was included
11                  in this study (Cabaravdic et al.. 2010). Ukaejiofo et al. (2009) studied the hematological
12                  effects of Pb in 81  male subjects moderately exposed to Pb at three  different
13                  manufacturing companies in Nigeria for durations between six months and 20 years. The
14                  exposed individuals had a mean blood Pb level of 7.00 (ig/dL compared to 3 (ig/dL in
15                  controls drawn from industries not involved in Pb handling (control group I) and 2 (ig/dL
16                  in controls drawn from the general population (control group II). Pb-exposed workers had
17                  significantly reduced Hb and PCV levels and increased percentage of reticulocytes
18                  compared to controls. Although the differences were statistically significant between the
19                  exposed and control  subjects, the study authors stated that the levels in the exposed
20                  subjects were at the lower range of normal for Nigerians. The percent cell lysis did not
21                  significantly differ between controls and exposed workers; however, when workers and
22                  controls were stratified by  age, there was a significant increase in cell  lysis in workers
23                  under age 30 compared to similarly aged controls in group II (p <0.01). Similarly,
24                  stratification of subjects by duration of exposure revealed that MCHC was decreased in
25                  exposed workers (6-60 months of exposure). Conterato et al. (In Press) investigated
26                  hematological parameters in automotive painters exposed to Pb in Brazil. Exposed
27                  painters had a mean [SEM] blood Pb concentration of 5.4 [0.4] (ig/dL compared to  1.5
28                  [0.1] (ig/dL  in controls. The mean [SEM] duration of exposure to Pb in painters was
29                  133.9 [14.5] months, whereas the controls were not occupationally exposed to Pb.
30                  Although differences in Hct, Hb concentration, and the number of RBCs were
31                  significantly decreased in painters compared to controls, these differences were  not
32                  correlated with blood Pb levels; however, these parameters were correlated with blood
33                  Cd2+ levels,  which were also significantly elevated in painters compared to controls.

34                  Taken together, the above occupational studies provide consistent evidence that high
35                  (mean blood Pb >26 (ig/dL) occupational exposure to Pb reduces the number of RBCs in
36                  circulation. Additionally, the Ukaejiofo et al. (2009) study  suggests  that blood Pb levels
37                  below 10 (ig/dL (7.0 (ig/dL) may also result in decreased RBC survival. Although the
38                  decrease in RBCs observed in highly exposed worker populations may be explained by
39                  both decreased RBC survival  and/or disruption of hematopoiesis, the observation of

      November 2012                             5-497                  Draft - Do Not Cite or Quote

-------
 1                  increased reticulocytes in Ukaejiofo et al. (2009) seems to represent compensation for
 2                  decreased RBC survival due to Pb exposure.

 3                  In a non-occupational study, the associations between blood Pb levels, Ca2+, Fe, and Hb
 4                  were investigated in 55  pregnant Brazilian women (21.9% were 14-19 years old, 74.5%
 5                  were 20-34 years olds,  and 3.6% were > 35 years old) (Zentner et al.. 2008). The
 6                  majority of women (across all age groups) had concurrent blood Pb levels <5 (ig/dL
 7                  (58.2%), although the mean blood Pb level was not reported; only 5.4% of women had
 8                  blood Pb levels >10 (ig/dL. The vast majority of the women (78.2%) were also observed
 9                  to have adequate levels  of Hb (> 11 g/dL). In a multiple linear regression model, blood
10                  Pb level was observed to be negatively associated with Hb ((3 = -0.359), when controlling
11                  for age, BMI, income, energy intake, Ca2+ intake, vitamin C intake, and Fe intake.

12                  The associations of blood Pb levels with hematological parameters observed in
13                  epidemiologic studies are clearly supported by a number of animal toxicology studies
14                  reporting blood Pb levels relevant to humans, i.e., <10 (ig/dL. Hb concentrations in
15                  plasma (a marker of RBC hemolysis) was significantly increased in rats exposed to
16                  Pb acetate (1,000 ppm in drinking water for 9 months; blood Pb level: 7.1 (ig/dL)
17                  compared to controls (Baranowska-Bosiacka et al.. 2009). In a complementary in vitro
18                  experiment, a concentration-dependent increase in the amount of hemolysis was observed
19                  in human RBCs exposed to Pb at concentrations ranging from 0.1-100 (iM for 5-30
20                  minutes. Hemolysis was increased even at the lowest concentration tested (i.e., 0.1 (iM).
21                  Pb-induced hemolysis in these experiments  may be due to inhibition of RBC
22                  phosphoribosyltransferases (Section 5.7.2.5). In weanling rats (PND25 days, n = 10)
23                  whose dams were exposed to Pb acetate in drinking water (2.84 mg/mL, approximating
24                  mean [SD] daily exposures of 342.57 [28.11] and 744.47  [29.27] mg/kg [dam weight]
25                  during gestation and lactation, respectively), blood Pb level was significantly elevated
26                  compared to controls (mean [SE]: 69.8. [7.82] versus 0.54 [0.08] (ig/dL ). The only
27                  hematological parameter affected by Pb exposure was Hct, which was decreased in
28                  exposed rats (mean [SE]: 27.3 [0.5]%) versus controls (33.4 [0.3]%) (Molina et al.,
29                  2011). In rats treated with 25 mg Pb/kg by oral gavage for 4 weeks, average plasma Pb
30                  concentrations were 6.5 (ig/dL (9.6-fold higher than controls), and statistically significant
31                  decreases in Hct, Hb, and RBCs were observed (Lee et al.. 2005). Effects on erythrocyte
32                  survival were similar in mice treated with Pb nitrate (50 mg/kg via gavage for 40 days):
33                  mean [SD] blood Pb levels were 1.72 [0.02] (ig/dL versus 0.09 [0.011] (ig/dL in control
34                  mice, and exposed mice had significantly reduced total RBC counts, total leukocyte
35                  counts, Hb, lymphocytes, and monocytes compared to controls (p <0.001) (Sharma et al..
36                  2010b).
      November 2012                            5-498                 Draft - Do Not Cite or Quote

-------
 1                  A number of toxicological studies also reported similar hematological effects, but did not
 2                  report final blood Pb concentrations. Rats exposed to Pb acetate (2 g/L in drinking water
 3                  for 30 days) had significantly decreased RBCs, Hb, PCV, MCH, and MCHC compared to
 4                  controls (p <0.05) (Simsek et al.. 2009). but not a disruption of hematopoiesis. Mice
 5                  exposed to Pb acetate (1 g/L in drinking water for 90 days, but not those exposed for 15
 6                  or 45 days), had significantly decreased RBC counts and Hct compared to controls
 7                  (p<0.05) (Marques et al., 2006). Spleen weights were observed to  be increased relative to
 8                  body weight in animals exposed to Pb for 45 days. Mice injected daily with Pb acetate
 9                  (50 mg/kg subcutaneously) had significantly reduced Hb, MCV, MCH, and MCHC
10                  compared to controls injected with 5% dextrose (Wang et al., 2010g).

11                  Some toxicological studies found no evidence of hematological effects in animals
12                  following exposure to Pb. Male rats exposed to Pb acetate in their drinking water for
13                  4 weeks at concentrations ranging from 100-1,000 ppm had a concentration-dependent
14                  increase in blood Pb levels (range: 6.57-22.39 (ig/dL) compared to controls (0.36 (ig/dL),
15                  but there were no significant changes in any of the hematological parameters (complete
16                  blood cell count performed) measured at the end of treatment (Lee et al.. 2006b). Slight,
17                  statistically nonsignificant increases in PS expression on RBC membranes were also
18                  observed. Similarly, exposure of male rats to 5,000 ppm Pb nitrate in drinking water
19                  (blood Pb not reported) for three weeks had no affect on any measured hematological
20                  parameter (Gautam and Flora. 2010). In vitro experiments with rat and human blood did
21                  not demonstrate a significant increase in hemolysis after 4 hours of treatment with
22                  Pb acetate at concentrations up to 10
23                  Although Pb exposure has been consistently shown to shorten RBC life span and alter
24                  RBC mobility, as of the 2006 Pb AQCD, the mechanism of this was not well understood.
25                  While the mechanism is still not fully understood, there has been some indication for a
26                  role of free Ca2+. Occupational studies investigating highly Pb-exposed worker
27                  populations (mean blood Pb >28 (ig/dL) observed increased intracellular free Ca2+ levels
28                  ([Ca2+]0 in RBCs, and decreased RBC membrane Ca2+/Mg2+ATPase activity in workers
29                  compared with unexposed controls (Abam et al.. 2008; Quintanar-Escorza et al.. 2007).
30                  [Ca2+]j levels were highly correlated with blood Pb levels even among unexposed control
31                  populations with mean blood Pb levels of approximately 10 (ig/dL (9.9 ± 2 (ig/dL)
32                  (Quintanar-Escorza et al., 2007). Changes in [Ca2+]j were associated with increased
33                  fragility of the RBCs and dramatic morphological alterations, including the increased
34                  presence of echinocytes (cells without normal biconcave shape) and crenocytes
35                  (speculated cells) in Pb-exposed workers.

36                  Similar to the associations observed in Quintanar-Escorza et al. (2007). [Ca2+]j increased
37                  in a concentration-dependent manner when RBCs from healthy human volunteers were
      November 2012                            5-499                 Draft - Do Not Cite or Quote

-------
 1                  exposed (in vitro) to 0.2 or 0.4 (iM Pb nitrate for 24 or 120 hours (0.4 (iM Pb nitrate
 2                  roughly approximates 10 (ig/dL Pb, although concentrations in exposure media are not
 3                  directly comparable to blood Pb levels) (Quintanar-Escorza et al., 2010). The increase in
 4                  [Ca2+]j levels was observed to be related to increased Ca2+ influx and decreased efflux. As
 5                  was observed among highly Pb-exposed workers, changes in [Ca2+]j were associated with
 6                  increased fragility of the RBCs and dramatic morphological alterations following
 7                  exposure to 0.4 (iM Pb. Similarly, Ciubar et al. (2007) found that RBC morphology was
 8                  disrupted, with > 50% RBCs having lost the typical discocytic morphology and
 9                  displaying moderate to severe echinocytosis following exposure to Pb nitrate
10                  concentrations of 0.5 (iM or higher for 24 hours. Exposure of RBCs to higher
11                  concentrations (concentrations not stated) of Pb nitrate resulted in cell shrinkage. In rats
12                  exposed to 200 ppm Pb acetate via drinking water for three months (mean [SD] blood Pb
13                  level: 40.63 [9.21] (ig/dL), the cholesterol/phospholipid ratio of RBC membranes was
14                  increased, indicating that RBC membrane fluidity was decreased.

15                  Khairullina et al. (2008) observed that the surface profiles of RBC membrane shadows
16                  incubated with 0.5-10 (iM Pb acetate for three hours were much smoother than were
17                  untreated RBC membranes when examined by atomic force microscopy. The authors
18                  postulated that the observed smoothing in Pb-treated RBC membranes may be due to
19                  clusterization of band 3 protein. Band 3 (anion exchanger 1 [AE1]), is a
20                  chloride/bicarbonate (C1~/HCO3~) exchanger and is the most abundant protein in RBC
21                  membranes. AE1 is integral in carbon dioxide (CO2) transport and linkage of the cellular
22                  membrane to the underlying cytoskeleton (Akel et al.. 2007; Su et al.. 2007). The
23                  observed smoothing of the RBC membrane may due to Pb interfering with how the
24                  membrane attaches to the cytoskeletal structure of the RBC through perturbation of the
25                  normal activity of AE1.


                    Eryptosis

26                  Eryptosis is the suicidal death of RBCs. It is characterized by cell shrinkage, membrane
27                  blebbing, and cell membrane phospholipid scrambling associated with PS exposure on
28                  the cell membrane that leads to cell destruction via macrophages (Foller et al.. 2008;
29                  Lang et al.. 2008). As previously reported in the 2006 Pb AQCD (U.S. EPA. 2006bX
30                  Kempe et al. (2005) found that exposing human RBCs to Pb at concentrations ranging
31                  from 0.3 (iM to 3 (iM caused increased activation of K+ channels that led to cell
32                  shrinkage and scramblase  activation. The activation of scramblase increased the exposure
33                  to PS on the cell membrane, which causes an increase in the destruction of the RBCs by
34                  macrophages.
      November 2012                            5-500                  Draft - Do Not Cite or Quote

-------
 1                  Consistent observations were made in recent studies that included in vitro and in vivo
 2                  evidence. Shin et al. (2007) found that in vitro exposure of human RBCs to 1-5 (iM
 3                  Pb acetate increased PS expression in a time- and concentration-dependent manner. The
 4                  maximum mean [SE] increase in expression of PS was 26.8% [3.15] (compared to
 5                  deionized water), following exposure to 5 (iM Pb for four hours. Scramblase activity was
 6                  increased in Pb-exposed RBCs, and [Ca2+]l5 which regulates scramblase activation, was
 7                  also increased in exposed RBCs. Flippase, which translates PS exposure to inner
 8                  membranes, is inhibited by high levels of [Ca2+]j and was observed to exhibit reduced
 9                  activity following Pb exposure. The inhibition of flippase is additionally influenced by
10                  the depletion of cellular adenosine triphosphate  (ATP). ATP levels were decreased in a
11                  concentration-dependent manner following exposure to Pb. To corroborate these findings
12                  in vivo, Shin et al. (2007) treated male rats with Pb acetate (i.p. to 25, 50, or 100 mg/kg;
13                  blood Pb not reported). Expression of PS was observed to increase in a concentration-
14                  dependent manner at concentrations > 50 mg/kg, confirming the in vitro results. No
15                  hemolysis or microvesicle formation was observed in the in vitro or in vivo experiments.

16                  In a follow-up study, the same laboratory observed that in vitro exposure of human RBCs
17                  to much lower concentrations of Pb acetate (0.1, 0.25, and 0.5 (JVI) also induced PS
18                  expression. Most notably, exposure to 0.1 (iM Pb for 24 hours increased PS expression
19                  on RBC membranes by approximately 20% (Jang et al., 2011). Accompanying the
20                  increased expression of PS (associated with Pb exposure) was the presence of abnormal
21                  echinocytic RBCs. Unlike the Shin et al. (2007) study described above, incubation of the
22                  RBCs with low concentrations of Pb (0.1 (iM) induced the generation of microvesicles,
23                  which also expressed PS on their membranes in this (Jang et al.. 2011) study. At 0.5 (iM,
24                  Pb-exposed RBCs with externalized PS  were observed to be targeted and engulfed by
25                  differentiated macrophages. Similar ex vivo effects were observed in rat erythrocytes four
26                  hours after oral exposure (0, 10 and 50 mg/kg) to Pb, although higher concentrations
27                  were generally required. PS expression on the rat erythrocytes was also observed. To
28                  corroborate these in vitro and ex vivo findings, rats were also exposed in vivo to 0, 50,
29                  250, or  1,000 ppm Pb acetate in drinking water for 4 weeks. At 1,000 ppm, Hb and Hct
30                  were significantly decreased relative to control,  and liver and spleen weights were
31                  increased. At the two highest doses, iron accumulation was observed in the spleen, a clear
32                  sign of increased RBC  clearance via phagocytosis.
      November 2012                            5-501                  Draft - Do Not Cite or Quote

-------
 1                  Ciubar et al. (2007) also found that exposure to Pb nitrate (0.5-2 (iM) resulted in an
 2                  increase in PS exposure in RBCs and cell shrinkage, which the authors stated were
 3                  indicators of cell apoptosis. As reported above, Khairullina et al. (2008) observed
 4                  Pb-induced RBC membrane smoothing that may be due to alterations in AE1 activity.
 5                  Disruptions in AE1 activity may also result in enhanced PS exposure and premature cell
 6                  death. Akel et al. (2007) observed that in AE1"" knockout mice, Pb-induced PS exposure
 7                  was much greater than that in wild type mice. Decreased RBCs and increased
 8                  reticulocytes were also  observed, an indication of high cell turnover.
                    5.7.2.3     Red Blood Cell Hematopoiesis

 9                  Erythropoietin (EPO) is a glycoprotein hormone excreted by the kidney to promote the
10                  development of RBCs in the bone marrow. As reported in the 2006 Pb AQCD, analyses
11                  of the cohort of children in Yugoslavia observed that EPO was increased in children aged
12                  4.5 and 6.5 years of age living in a town near Pb sources  (blood Pb levels >30 (ig/dL)
13                  compared to children living in more distant town (blood Pb levels <10 (ig/dL), when
14                  stratified by Hb concentrations (Graziano et al.. 2004; Factor-Litvak et al.. 1999; Factor-
15                  Litvak et al., 1998). These differences were not observed in children aged 9.5 or 12 years.
16                  With adjustment for Hb concentrations, blood Pb levels were observed to be significantly
17                  associated with EPO levels at ages 4.5 and 6.5 years when considering all children
18                  together. No significant association was observed at ages 9.5 and 12 years. Hb was not
19                  observed to differ at any age between towns, thus possibly indicating that
20                  hyperproduction of EPO is necessary to maintain Hb levels in young children living near
21                  Pb sources. The authors postulated that increases in EPO in younger children reflect bone
22                  marrow hyperactivity to counteract RBC destruction, whereas the lack of EPO elevation
23                  in older children may reflect a transitional period where increasing renal and bone
24                  marrow toxicity leads to decreases in EPO observed later in life, as observed in anemic,
25                  pregnant women (Graziano et al.. 1991). Decreased EPO concentrations were also
26                  observed in association with Pb exposure in adults in two cross-sectional studies cited in
27                  the 2006 Pb AQCD (Osterode et al..  1999; Romeo etal.  1996).

28                  Consistent with findings that EPO is  negatively associated with blood Pb levels in adults,
29                  Sakata et al. (2007) observed that non-anemic tricycle taxi drivers (n=27) working in
30                  Kathmandu, Nepal (blood Pb level: 6.4 (ig/dL) had significantly lower levels of EPO
31                  (12.7 versus 18.8 mU/mL) compared to non-driver controls (blood Pb level: 2.4 (ig/dL).
32                  In taxi drivers, there was an inverse relationship between the level of serum
33                  erythropoietin and blood Pb level (r = -0.68, p <0.001). Blood Pb level was not associated
34                  with any other hematological effects.
      November 2012                            5-502                 Draft - Do Not Cite or Quote

-------
 1                  Recent toxicology studies of cytotoxicity and genotoxicity in RBC precursor cells
 2                  support the observations that Pb exposure disrupts normal hematopoiesis. Cytotoxicity
 3                  and genotoxicity in RBC precursor cells are strong indications of altered hematopoiesis in
 4                  bone marrow. Celik et al. (2005) observed that treatment of female rats with Pb acetate
 5                  (140, 250, or 500 mg/kg via gavage once per week for 10 weeks; blood Pb not reported)
 6                  resulted in decreased numbers of polychromatic RBCs (PCE) and increased numbers of
 7                  micronucleated PCEs, compared to controls (p <0.001).  Alghazal et al. (2008b) exposed
 8                  male and female rats to 100 mg/L Pb acetate daily in drinking water for 125 days (blood
 9                  Pb not reported) and observed increases in micronucleated PCEs in female rats (p = 0.02)
10                  but no significant reduction in the ratio of PCEs to normochromic RBCs (NCE). In male
11                  rats, an increase in micronucleated PCEs was observed (p <0.001) along with a decrease
12                  in the PCE/NCE ratio (p = 0.02). While the results from Alghazal  et al. (2008b) indicate
13                  that Pb is cytotoxic in male rats only, but is genotoxic in both sexes, results from Celik et
14                  al. (2005) indicate that Pb is cytotoxic in female rats as well. Mice exposed to Pb acetate
15                  (1 g/L in drinking water for 90 days; blood Pb not reported) had statistically significant
16                  increases in micronucleated PCEs; a small, but statistically  nonsignificant decrease in the
17                  PCE/NCE ratio was also observed (Marques et al.. 2006).
                    5.7.2.4     Membrane Proteins

18                  While there have been few studies, evidence included in the 2006 Pb AQCD indicated
19                  there are effects of Pb on changes in RBC proteins. Huel et al. (2008) found that newborn
20                  hair and cord blood Pb levels (mean [SD]: 1.22 [1.41] (ig/g and 3.54 [1.72] (ig/dL
21                  respectively) were negatively associated with Ca2+ATPase activity in plasma membranes
22                  of RBCs isolated from cord blood after controlling for gestational age and maternal Ca
23                  pump activity. However, newborn hair Pb levels were more strongly associated with cord
24                  Ca2+-pump activity than were cord blood Pb (p <0.0001 versus p <0.05). Maternal blood
25                  Pb levels were not correlated with Ca2+-pump activity in maternal or newborn cord blood.
26                  Pb-induced disruptions in Ca2+ homeostasis in RBCs can lead to cytotoxicity and
27                  necrosis, and these effects may be representative of cellular dysfunction in other organ
28                  systems.

29                  In RBC membranes from Pb-exposed workers, Fukumoto et al. (1983) used
30                  polyacrylamide electrophoresis analysis and found increased levels of polypeptides in
31                  bands 2, 4, 6, and 7 and decreased levels of polypeptides in band 3. Apostoli et al. (1988)
32                  found changes in RBC membrane polypeptides, including a significant decrease in band
33                  3, in occupationally exposed workers with blood Pb levels greater than 50 (ig/dL.
34                  Apostoli et al. (1988) suggested that band 3 may represent an anion channel protein,
35                  whereas, Fukumoto et al. (1983) suggested that the changes in the RBC membrane


      November 2012                            5-503                  Draft - Do Not Cite or Quote

-------
 1                  polypeptides may cause changes in membrane permeability. Exposure to Pb acetate at
 2                  concentrations above 0.1 (iM for 60 minutes has also been found to increase the
 3                  phosphorylation of proteins in human RBC membranes in vitro (Belloni-Olivi et al.,
 4                  1996). Phosphorylation did not occur in cells depleted of protein kinase C (PKC),
 5                  indicating a PKC-dependent mechanism.
                    5.7.2.5     Red Blood Cell Energy Metabolism Enzymes

 6                  RBCs use high energy purine nucleotides (i.e., ATP and guanine triphosphate [GTP]) to
 7                  support basic metabolic functions. In mature RBCs, these nucleotides are synthesized via
 8                  salvage reactions through either an adenine pathway, which requires adenine
 9                  phosphoribosyltransferase (APRT), or an adenosine pathway, which requires adenosine
10                  kinase. The 2006 Pb AQCD (U.S. EPA. 2006b) reported that Pb significantly reduces the
11                  nucleotide pool (including NAD and NADP, as well as increases purine degradation
12                  products) resulting in altered RBC energetics. Since the 2006 Pb AQCD was published,
13                  there have been few studies examining Pb effects on energy metabolism. Baranowska-
14                  Bosiacka et al. (2009) examined the effects of Pb on RBC APRT and hypoxanthine-
15                  guanine phosphoribosyltransferase (HPRT). In an in vitro experiment, APRT and HPRT
16                  were measured in lysate of human RBCs after exposure to Pb at a concentration range
17                  from 0.1 to 100 (iM for 5-30 minutes. Complementary in vivo tests measured APRT and
18                  HPRT in RBC lysate from rats exposed to Pb acetate (1,000 ppm) in drinking water for
19                  9 months. Both the in vivo and vitro studies found a significant decrease in both HPRT
20                  and APRT levels. The levels in human RBCs were significantly decreased in vitro after
21                  only 5 minutes of exposure to the 0.1 (iM concentration, but the decrease was also
22                  concentration-dependent.  However, the study authors considered the inhibition moderate
23                  (30-35%) even with the highest Pb levels used in vitro. Shin et al. (2007) found a
24                  concentration-dependent decrease in intracellular ATP in human RBCs in vitro with
25                  significant decreases, found even with the lowest concentration (i.e., 1
                    5.7.2.6     Other Enzymes

26                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported that K+ permeability was increased by
27                  Pb exposure due to altered sensitivity of the membrane Ca2+-binding site that caused
28                  selective efflux of K+ ions from the RBC membrane. However, inhibition of the RBC
29                  Na+/K+ATPase is more sensitive to Pb exposure than is the inhibition of
30                  Ca2+/Mg2+ATPase. Few recent studies were found that examined the effects of Pb
31                  exposure on other enzymes. Ekinci et al. (2007) tested the effects of Pb exposure on two
32                  carbonic anhydrase isozymes (I and II) isolated from human RBCs. Carbonic anhydrases

      November 2012                            5-504                  Draft - Do Not Cite or Quote

-------
 1                  are metalloproteins that use Zn to catalyze the equilibrium between CO2 and bicarbonate
 2                  in the cells of higher invertebrates. Although investigators found that Pb nitrate inhibited
 3                  both carbonic anhydrase isozymes in a concentration-dependent manner, the
 4                  concentrations used (i.e., 200-1,000 (iM) were above those that would be physiologically
 5                  relevant. Inhibition of isozyme I was noncompetitive, while the inhibition for isozyme II
 6                  was uncompetitive. Bitto et al. (2006) examined the mechanisms of action of Pb-induced
 7                  inhibition of P5N, an enzyme important in the pyrimidine salvage pathway that requires
 8                  Mn for normal activity. Pb was observed to bind directly to the active site of the enzyme
 9                  in a different position than the Mn, thus possibly resulting in improper protein folding
10                  and inhibition of activity.
                    5.7.2.7     Red Blood Cell Oxidative Stress

11                  It has been suggested that the Pb-associated decreases in ALAD activity result in
12                  increased oxidative stress, owing to the buildup of ALA. ALA can act as an electron
13                  donor in the formation of reactive oxygen species (ROS) (Nemsadze et al.. 2009;
14                  Ahamed and Siddiqui. 2007). Many epidemiologic and toxicological studies have found
15                  an association between the level of blood Pb and lipid peroxidation, antioxidant levels, or
16                  indicators of ROS production. The same limitations regarding cross-sectional  studies
17                  listed in Section 5.7.2.2 (including uncertainty in directionality of effects and specific
18                  information regarding exposure) apply to the epidemiologic studies investigating RBC
19                  oxidative stress. Additionally, potential confounders and co-exposures were not
20                  considered in the majority of these studies. However, in studies were confounders were
21                  considered, they are explicitly delineated in the text.


                    Oxidative Stress, Lipid Peroxidation, and  Antioxidant Enzymes

22                  Malondialdehyde (MDA) is an end product of lipid peroxidation and is commonly
23                  measured as an indicator of oxidative stress. Evidence of lipid peroxidation has been
24                  observed in children moderately exposed to Pb. Ahamed et al.(2008; 2006. 2005)
25                  investigated the relationship between blood Pb levels and antioxidant enzyme levels and
26                  lipid peroxidation in children in India. In children (n = 62) aged 4-12 years in  Lucknow,
27                  India, children with mean blood Pb levels of 11.39 (SD: 1.39) (ig/dL had increased
28                  measures of lipid peroxidation and decreased GSH levels compared to children with
29                  mean blood Pb  levels of 3.93 (SD: 0.61) or 7.11 (SD: 1.25) fig/dL (Ahamed et al.. 2005).
30                  Catalase activity was decreased in children with a mean 7.11 (SD: 1.25) (ig/dL blood Pb
31                  level, compared to children with mean 3.93 (SD:  0.61) (ig/dL blood Pb level.
32                  Additionally, blood Pb levels were found to be significantly positively correlated with
      November 2012                            5-505                  Draft - Do Not Cite or Quote

-------
 1                  MDA and CAT, and negatively correlated with GSH. In a similar study, Ahamed et al.
 2                  (2006) observed significantly higher levels of CAT and MDA in children with a mean
 3                  13.27 (ig/dL blood Pb level compared to children with a mean 7.40 (ig/dL blood Pb level;
 4                  with other characteristics such as age, height, wieght, and BMI not differing between the
 5                  two groups and thus, not considered as potential confounders. Examining all the study
 6                  subjects together, investigators found a correlation between blood Pb level and blood
 7                  MDA and RBC CAT levels, as well as an inverse relationship between ALAD activity
 8                  and MDA and CAT levels. Among Indian children with neurological disorders, blood Pb
 9                  levels were significantly increased compared to healthy control children (18.60 versus
10                  10.37 (ig/dL respectively) (Ahamed et al., 2008). Potential confounding characteristics
11                  such as age, sex, area of residence, and SES were not observed to be statistically different
12                  between the two groups, and therefore, were not included in statistical analyses. In
13                  addition, the following indicators of oxidative stress were elevated among case children:
14                  increased blood MDA, RBC SOD and CAT levels, and decreased blood GSH levels.  GPx
15                  levels were similar between the two groups. Typical indicators of Pb exposure
16                  (active/nonactive ALAD ratio) were found to be correlated with lipid peroxidation and
17                  oxidative stress. Children aged 3-6 years old living near a steel refinery in China with
18                  blood Pb levels > 10 (ig/dL also had a significant increase in plasma MDA compared to
19                  children with blood Pb levels <10 (ig/dL. However, levels of RBC SOD, GSH, and GPx
20                  were  not different from those in controls (Jin et al., 2006).

21                  Evidence of lipid peroxidation was also observed in occupational cohorts moderately
22                  exposed to Pb. In auto repair apprentices in Turkey (mean [SD]: 16.8 [1.2] years of age,
23                  3.8 [1.8] years duration of exposure) with minimum blood Pb levels of 7.9 (ig/dL
24                  (Ergurhan-Ilhan et al.. 2008). increases in glutathione peroxidase (GPx) and MDA, as
25                  well as decreases in a-tocopherol and (3-carotene were observed compared with controls
26                  (mean [SD]  age: 16.3 [1.0] years, mean blood Pb level: 2.6  (ig/dL). Decreases  were
27                  observed in  SOD and CAT, but the results did not attain statistical significance.
28                  Statistically significant alterations in measures of oxidative  stress were also observed in
29                  other occupationally exposed populations. SOD, glutathione (GSH), and CAT were
30                  decreased; while oxidized GSH (i.e., GSSG) and thiobarbituric acid reactive species
31                  (TEARS, expressed in terms of MDA) were increased in painters in India (mean [SD]
32                  duration of exposure: 126.08  [49.53 months], mean blood Pb level: 21.92  (ig/dL,
33                  compared to 3.06 (ig/dL in controls) (Mohammad et al., 2008). Glutathione-S-transferase,
34                  GPx, and SOD were positively correlated with blood Pb levels (mean: 5.4 (ig/dL,
35                  r = 0.34, 0.38, and 0.32, respectively; p <0.05) in automotive painters in Brazil
36                  (Conterato et al.. In Press).
      November 2012                             5-506                  Draft - Do Not Cite or Quote

-------
 1                  Numerous cross-sectional, occupational studies have also demonstrated increased lipid
 2                  peroxidation in highly-exposed worker populations (blood Pb levels ranging from 29.0 to
 3                  74.4 (ig/dL) (Kasperczyk et al.. 2009; Khan et al. 2008; Quintanar-Escorza et al.. 2007;
 4                  Patil et al.. 2006a: Patil et al.. 2006b). There was a correlation between MDA levels and
 5                  blood Pb levels, even in the unexposed workers who had lower (i.e., <12 (ig/dL) blood Pb
 6                  levels, although the magnitude of correlation in exposed workers was greater (Quintanar-
 7                  Escorza et al.. 2007). Increases in C-reactive protein and decreases in RBC SOD,
 8                  catalase, and plasma ceruloplasmin were also observed in these workers, further
 9                  indicating increased RBC oxidative stress due to higher Pb exposure.

10                  Oral administration of Pb (25 mg/kg) to rats once a week (i.e., bolus gavage) for 4 weeks,
11                  which produced a blood Pb level of about 6.5 (ig/dL, caused a significant increase in
12                  RBC MDA levels (Lee et al.. 2005). Other indications of Pb-induced oxidative stress
13                  included significant increases in RBC SOD and CAT levels accompanied by significant
14                  decreases in GSH and GPx. Exposure of rats to Pb acetate  (750 mg/kg in drinking water
15                  for 11  weeks)  resulted in decreased concentrations of plasma vitamin C, vitamin E,
16                  nonprotein thiol, and RBC-GSH, with  simultaneous increased activity of SOD and GPx
17                  (Kharoubi et al.. 2008b). CAT activity was also slightly elevated in RBCs from  the
18                  Pb-exposed rats, but the increase failed to reach statistical significance. Exposure of male
19                  rats to 5,000 ppm Pb nitrate in drinking water (blood Pb not reported) for three weeks
20                  decreased GSH levels compared to that in controls (mean [SE]: 1.91 [0.02] versus 2.44
21                  [0.09]  mg/mL, respectively) (Gautam and Flora. 2010). SOD activity was significantly
22                  decreased in rats injected with Pb acetate (15 mg/kg, i.p. for seven days, but not rats
23                  injected with 5 mg/kg) (Berrahal et al.. 2007).  GPx activity and MDA concentrations
24                  were slightly elevated in the two  exposed groups, but differences with the control
25                  (15 mg Na acetate/kg) group failed to reach statistical significance. Effects on indices of
26                  oxidative stress were also observed in in vitro  studies: increased MDA and decreased
27                  SOD and CAT in RBCs exposed to 2 (iM Pb (Ciubar et al.. 2007). decreased glutathione
28                  reductase (GR) activity in human RBCs incubated with 5-18 (iM Pb (Coban et al.. 2007).
29                  and decreased GSH and increased GSSG and lipid peroxidation in RBCs from healthy
30                  volunteers (with no history of Pb exposure) incubated with 0.4 (iM Pb for 24-120 hours
31                  (Quintanar-Escorza et al.. 2010).


                    Antioxidant  Defense

32                  In addition to the studies listed above that examined lipid peroxidation and oxidative
33                  stress, there have been toxicological studies that indicate that the use of antioxidants and
34                  free radical reactions is protective against Pb-induced RBC oxidative stress. Rats treated
35                  with 500 ppm  Pb acetate in drinking water for 15 or 30 days had  a significant increase in
36                  free RBC protoporphyrin and TEARS  levels that was related to length of exposure and

      November 2012                             5-507                  Draft - Do Not Cite or Quote

-------
 1                  blood Pb levels (Rendon-Ramirez et al. 2007). Vitamin E administration after exposure
 2                  to Pb significantly reduced the rat RBC TEARS levels and increased ALAD activity,
 3                  compared to exposure to Pb alone. Co-exposure to vitamin E and Pb simultaneously and
 4                  exposure to vitamin E before Pb exposure also prevented Pb-induced oxidative stress. In
 5                  vitro studies by Casado et al. (2007) found that Pb-induced hemolysis using blood from
 6                  non-occupationally exposed volunteers indicated that RBC membrane damage was
 7                  mediated via oxidative stress. The in vitro studies demonstrated a concentration- and
 8                  time-dependent formation in lipid peroxide that was inhibited with a number of
 9                  antioxidants, including desferrioxamine (iron chelator), trolox (chain breaking
10                  antioxidant), and mannitol and Na formate (OH scavengers). Results suggested the role
11                  of singlet oxygen in Pb-mediated membrane damage and hemolysis of exposed RBCs. In
12                  rats exposed to 2,000 ppm Pb in drinking water for 5 weeks, MDA levels were
13                  significantly increased, whereas vitamin E concentrations were significantly decreased
14                  (Caylak et al., 2008). In the case of MDA, co-exposure to Pb and a number of sulfur-
15                  containing antioxidants (e.g., L-methionine, N-acetylcysteine, and L-homocysteine)
16                  reduced concentrations to a level not significantly different from that in controls, but
17                  were significantly smaller than concentrations observed with Pb  alone. Exposure to L-
18                  methionine and N-acetylcysteine also reduced Pb-induced depletion of vitamin E.
                    5.7.2.8     Summary of Effects on RBC Survival and Function

19                  In summary, Pb exposure has been shown to affect multiple hematological outcomes that
20                  are related to RBC survival and function, as demonstrated in both cross-sectional
21                  epidemiologic studies and toxicological studies. Pb exposure has been shown to decrease
22                  RBC survival, either through direct effects on RBC membranes leading to increased
23                  fragility, or through the induction of eryptosis and eventual phagocytosis by
24                  macrophages. Limited evidence that Pb can negatively affect hematopoiesis is also
25                  available. Consistent evidence also exists demonstrating that Pb exposure increases
26                  oxidative stress in exposed adults and children, as well as in laboratory animals. The
27                  epidemiologic studies demonstrating these effects are cross-sectional in design, therefore
28                  there is some uncertainty regarding the direction of effects and the magnitude, timing,
29                  frequency, and duration of Pb exposure that contributed to the observed observations.
30                  Also, the majority of epidemiologic studies did not account for potential confounding,
31                  although the effects observed in these studies are consistent with effects from studies that
32                  did consider potential  confounding. The coherence with effects observed in animal
33                  toxicology studies supports the conclusion that Pb exposure affects both the survival and
34                  function of RBCs.
      November 2012                            5-508                  Draft - Do Not Cite or Quote

-------
      5.7.3
Red Blood Cell Heme Metabolism
 i
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11
12
13
  Pb exposure has been found to inhibit several enzymes involved in heme synthesis,
  namely ALAD (a cytoplasmic enzyme catalyzing the second, rate-limiting, step of the
  heme biosynthesis pathway),  coproporphyrinogen oxidase (catalyses the sixth step in
  heme biosynthesis converting coproporphyrinogen III into protoporphyrinogen IX), and
  ferrochelatase (catalyses the terminal [eighth] step in heme synthesis converting
  protoporphyrin IX into heme) (Figure 5-36). The observations of decreased Hb
  (measured as total Hb, MCH, or MCHC) in occupationally-exposed adults (Ukaejiofo et
  al.. 2009; Khan et al.. 2008; Patil et al.. 2006b: Karita et al.. 2005) and Pb-exposed
  experimental animal models (Sharma et al.. 2010b: Baranowska-Bosiacka et al.. 2009;
  Simsek et al.. 2009; Marques et al.. 2006; Lee et al..  2005) and associations with blood
  Pb levels in children (Oueirolo et al.. 2010; Shahet al.. 2010; Olivero-Verbel et al.. 2007;
  Riddell et al.. 2007) are supporting lines of evidence for decreased heme synthesis due to
  Pb exposure.
          Cytosol
                2x6-aminolevulpnic acid
                     1
     ALA dehycfratase
     (porpfiobitinagei
     synthase
                Porphobilinogen 1PBG)
                   3  | | P8G deomtnase
                hydroxymethylbila
                             Uroporphyrin ogen
                             fit syinht-'tase
                                       ^>  Uroporphyrinogen II
                                                                    Copiopotphyrinogcn III
                                    Uroporphyrinogen
                                    decarboxylase
      Note: Steps in the pathway potentially affected by Pb are indicated with curved arrows pointing to the affected enzyme, and the
      directions of effects are represented by f and \, arrows.
      Figure 5-36      Schematic representation of the enzymatic steps involved in the
                        heme synthetic pathway.
      November 2012
                             5-509
Draft - Do Not Cite or Quote

-------
                    5.7.3.1      Red Blood Cell 5-Aminolevulinic Acid Dehydratase

 1                  Decreases in RBC 5-aminolevulinic acid dehydratase (ALAD) levels are strongly
 2                  associated with Pb exposure in humans; to such an extent that RBC ALAD activity is
 3                  used as a biomarker to assess Pb toxicity. Several epidemiologic studies published since
 4                  the 2006 Pb AQCD evaluated the relationship between Pb exposure, blood Pb levels and
 5                  ALAD activity in adults and children (see below). These studies were cross-sectional in
 6                  nature. This limits their utility in assessing the direction of effects and the magnitude,
 7                  timing, frequency, and duration of Pb exposure necessary to contribute to the observed
 8                  associations. In studies which considered potential confounders, those confounding
 9                  variables are listed in the test. However, potential confounding was not accounted for in
10                  the majority of these studies.

11                  Wang et al. (201 Of) found that, after controlling for sex, age, alcohol consumption and
12                  smoking (adults only), there was also a concentration-dependent decrease in ALAD
13                  activity in both children (4-13 years old) and adults (16-77 years old) (mean blood Pb
14                  levels: 7.1 and 6.4 (ig/dL, respectively) in rural southwest China. Further, Wang et al.
15                  (201 Of) observed that the relationship between blood Pb level and ALAD activity was
16                  nonlinear and exponential, with larger decreases in ALAD activity occurring with blood
17                  Pb levels >10 (ig/dL. No correlation was observed between urinary ALA levels and blood
18                  Pb levels. Ahamed et al. (2006) studied male urban adolescents in India. The 39
19                  adolescents were separated into two groups according to their blood Pb levels (Group 1:
20                  <10 (ig/dL [mean 7.4 (ig/dL], Group  2: >10 (ig/dL [mean 13.27 (ig/dL]). Although
21                  Groups 1 and 2 were similar in age (mean [SD]: 16.59 [0.91] versus 16.76 [0.90] years,
22                  respectively), height, weight, and BMI (therefore not considered potential confounders),
23                  Group 2 (with the higher blood Pb levels) had lower ALAD activity than did Group 1
24                  (p <0.001). When all 39 adolescents were examined together, an inverse relationship was
25                  found between blood Pb and ALAD activity. Similar decreases in ALAD  activity were
26                  observed in other populations of children from India (aged 4-12 and 1-7 years) with
27                  elevated blood Pb levels (mean [SD]: 11.39 [1.39] and 21.86 [7.58] (ig/dL respectively)
28                  compared to the two age ranges of the children with lower blood Pb levels (mean [SD]:
29                  3.93 [0.61] and 6.89 [2.44] (ig/dL respectively) (Ahamed et al.. 2007: Ahamed et al..
30                  2005). While Ahamed et al. (2005) did not address potential confounding, Ahamed et al.
31                  (2007) observed decreases in ALAD activity after controlling for age, sex, and area of
32                  residence. Decreases in ALAD activity were also observed in children 3-6 years of age
33                  with Pb blood levels >10 (ig/dL, compared to children <10 (ig/dL (mean blood Pb
34                  concentration for groups not reported) in northeastern China (Jin et al.. 2006).

35                  As was seen with epidemiologic studies investigating Pb-associated deficits in
36                  hematological parameters, most occupational studies investigating ALAD levels may not
      November 2012                            5-510                 Draft - Do Not Cite or Quote

-------
 1                  be generalizable to the population as a whole; however, they are useful in demonstrating
 2                  consistent and negative effects of Pb exposure on the activity of this enzyme (Quintanar-
 3                  Escorzaetal.. 2007; Patil et al. 2006a: Patil et al.. 2006b: Ademuyiwa et al.. 2005b).
 4                  Occupationally-exposed adults had levels of inhibition of ALAD that were as great as
 5                  90% relative to  control (Quintanar-Escorza et al.. 2007). There were few studies that
 6                  investigated Pb-associated decrements in ALAD levels among moderately-exposed
 7                  workers. Painters in India with a mean blood Pb level of 21.92 (ig/dL (mean [SD]
 8                  duration of exposure: 126.08 [49.53] months) had lower ALAD levels (p <0.01)
 9                  compared to controls whose mean blood Pb level was 3.06 (ig/dL (Mohammad et al..
10                  2008). Stoleski  et al. (2008) observed that workers in a Pb smelter in Macedonia (mean
11                  [SD]: 16.4 [8.5] (ig/dL blood Pb; 18.8 [7.5] years employment) had lower ALAD activity
12                  (p <0.001) and higher ALA levels (p <0.0005) compared  to workers with no history of
13                  exposure to Pb (mean [SD] blood Pb: 7.0 [5.4] (ig/dL). In automotive painters exposed to
14                  Pb in Brazil (mean [SD]: 5.4 [0.4] (ig/dL blood Pb level;  133.9 [14.5] months duration of
15                  exposure), the ALAD reactivation index was increased over that in controls, although
16                  ALAD activity  did not differ between groups (Conterato et al.. In Press). However,
17                  ALAD activity  was negatively correlated with blood Pb levels (r = -0.59, p <0.05) but not
18                  blood Cd levels, whereas ALAD reactivation index was positively correlated with blood
19                  levels of both metals (Pb: r = 0.84, p <0.05; Cd: r = 0.27,  p <0.05). In a benchmark dose
20                  (BMD)-based analysis (BMR = 5% using the hybrid approach and a 5% adversity cut-off
21                  value), Murata et al. (2009) calculated the BMD and 95% lower confidence limit of the
22                  BMD (BMDL)  for decreased ALAD activity in RBCs of exposed Pb workers. The
23                  calculated BMD and BMDL values for Pb blood levels of 2.7 and  2.3 (ig/dL,
24                  respectively, were substantially lower than the BMDs (28.7-44.2 (ig/dL) and BMDLs
25                  (19.4-29.6 (ig/dL) for decreased Hb, Hct, and RBC count in similarly exposed workers,
26                  indicating decreases in ALAD activity can occur at blood Pb levels that do not decrease
27                  RBC survival.

28                  Decreased ALAD activity in response to Pb exposure has also been observed in
29                  toxicological studies. Rats administered 500 ppm Pb acetate in drinking water for 15 or
30                  30 days had decreased blood ALAD activity, which was related to duration of exposure
31                  and blood Pb levels (Rendon-Ramirez et al.. 2007). Oral administration of Pb (25 mg/kg)
32                  to rats once a week for 4 weeks achieved a blood  Pb level of 6.5 (ig/dL, which was
33                  associated with statistically significant decreases (approximately 50% lower than control
34                  levels) in RBC ALAD activity (Lee et al.. 2005).  Exposure of male Wistar rats to
35                  5,000 ppm Pb acetate via drinking water for three weeks significantly decreased ALAD
36                  activity by 72% (mean [SD]: 7.35 [0.35] versus controls:  26.14 [2.19] nmol/min/mL
37                  RBCs [nanomoles of porphobilinogen (PEG) formed per minute, per 1  mL blood])
38                  (Gautam and Flora. 2010).
      November 2012                            5-511                  Draft - Do Not Cite or Quote

-------
                    5.7.3.2      Other Heme Metabolism Enzymes

 1                  The 2006 Pb AQCD (U.S. EPA. 2006b) indicated that Pb affects RBC PEG synthase
 2                  (Simons. 1995; Farant and Wigfield. 1990. 1987). PEG deaminase (Tomokuni and
 3                  Ichiba. 1990). and TF endocytosis and iron transport across membranes (Qian and
 4                  Morgan. 1990). all of which are directly or indirectly involved in heme synthesis.
 5                  Although there are no recent studies that examine the effect Pb has on the activities of
 6                  other heme metabolism enzymes, a number of studies investigated associations of blood
 7                  Pb level with concentrations of various intermediate products in the heme biosynthetic
 8                  pathway.

 9                  Pb intoxication has been shown to inhibit the function of ferrochelatase, the enzyme that
10                  catalyzes the last (eighth) step in the heme biosynthetic pathway. Under normal
11                  conditions, ferrochelatase incorporates ferrous iron  (Fe2+) into protoporphyrin IX,
12                  converting it into a heme molecule (Figure 5-36). However, Pb has been shown to inhibit
13                  this insertion of Fe2+ into the protoporphyrin ring and instead, Zn is inserted into the ring
14                  creating ZPP. A number of recent studies have shown that blood Pb level is significantly
15                  associated with increased RBC ZPP levels in adults occupationally exposed to high levels
16                  of Pb (blood Pb levels: 27-54 jig/dL)  (Patil et al.. 2006b: Ademuyiwa et al.. 2005b).
17                  workers exposed to moderate levels of Pb (blood Pb level = 21.92 (ig/dL) (Mohammad et
18                  al.. 2008). children aged 1-21 years (blood Pb levels: 18-23 (ig/dL) (Counter et al.. 2009.
19                  2008; Counter et al.. 2007). and animals exposed to 500 ppm Pb via drinking water for 15
20                  or 30 days (Rendon-Ramirez et al.. 2007). Interestingly, Wang et al. (201 Of) found that in
21                  children and adults living in a rural area of Southwest China, ZPP levels were negatively
22                  correlated with blood Pb at blood Pb levels <10  (ig/dL and were only positively
23                  correlated with blood Pb at higher blood Pb concentrations (i.e., >10 (ig/dL). The authors
24                  suggested that this may be representative of ALAD activities at low blood Pb levels,
25                  which contributes to lower ZPP levels. Scinicariello et al. (2007) performed a meta-
26                  analysis and observed that Pb-exposed individuals who carried the ALAD2 allele had
27                  slightly lower concentrations of blood ZPP levels compared to carriers of the ALAD1
28                  allele (overall pooled standardized mean estimate: -0.09 [units not specified]; 95% CI:
29                  -0.22, 0.03, p = 0.13).
                    5.7.3.3      Hematological Effects

30                  In summary, Pb exposure has been shown in both cross-sectional epidemiologic studies
31                  and toxicological studies to alter heme synthesis. Pb exposure has been shown to inhibit
32                  the activities of two major enzymes in the heme biosynthetic pathway: ALAD and
33                  ferrochelatase. Evidence for the inhibition of ALAD comes from direct measurements of
      November 2012                            5-512                 Draft - Do Not Cite or Quote

-------
 1                  its activity in exposed human populations, whereas evidence for inhibition of
 2                  ferrochelatase comes from the observation of increased ZPP following exposure. Animal
 3                  toxicology and ecotoxicology studies provide evidence of coherent effects in animals.
 4                  The epidemiologic studies demonstrating these effects are cross-sectional in design,
 5                  therefore there is some uncertainty regarding the direction of effects and the magnitude,
 6                  timing, frequency, and duration of Pb exposure that contributed to the observed
 7                  observations. Also, the majority of epidemiologic studies did not account for potential
 8                  confounding, although the effects observed in these studies are consistent with effects
 9                  from studies that did account for confounding. The coherency  of effects observed in
10                  animal toxicology and ecotoxicology studies support the conclusion that Pb exposure
11                  alters the synthesis of heme in RBCs.
      5.7.4       Summary and Causal Determination

12                  Recent toxicological and epidemiologic evidence substantiates evidence presented in the
13                  2006 Pb AQCD that exposure to Pb affects hematological endpoints, and supports a
14                  causal relationship between Pb exposure and decreased RBC survival and function and
15                  altered heme synthesis. Outcomes related to decreased RBC survival and function
16                  included alterations in multiple hematological parameters (e.g., Hb, Hct, PCV, MCV,
17                  MCH), oxidative stress (altered antioxidant enzyme activities [SOD,  CAT, GPx],
18                  decreased cellular GSH, and increased lipid peroxidation), increased  cytotoxicity in RBC
19                  precursor cells, and mode of action endpoints such as decreased intracellular calcium
20                  concentrations, decreased ATPase activity, and increased phosphatidylserine expression.
21                  Outcomes related to altered heme synthesis included decreased activities of ALAD and
22                  ferrochelatase, and decreased levels of Hb. The sections that follow describe the
23                  evaluation of evidence for decreased red blood cell (RBC) survival and function and
24                  heme synthesis, with respect to causal relationships with Pb exposure using the
25                  framework described in Table II of the Preamble. The application of the key supporting
26                  evidence to the causal framework is summarized in Table 5-35.
                    5.7.4.1      Evidence for Decreased RBC Survival and Function

27                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported that Pb exposure is associated with
28                  multiple measures of decreased RBC survival and function. Epidemiologic evidence
29                  included the observation of a 10% probability of anemia with blood Pb levels of
30                  approximately 20 (ig/dL at age 1 year, and perturbed hematopoiesis in children and adults
31                  at blood Pb levels below 40 (ig/dL. Oxidative stress was also identified by the
32                  2006 Pb AQCD as a potential mode of action for Pb-induced effects in RBCs. A causal

      November 2012                            5-513                  Draft - Do Not Cite or Quote

-------
 1                  relationship between Pb exposure and decreased RBC survival and function is strongly
 2                  supported by the available, recent toxicological and epidemiological data. Among the
 3                  strongest evidence for Pb-induced decreases in RBC survival and function is the
 4                  consistent observation of alterations in hematological parameters (e.g., Hb, Hct, PCV,
 5                  MCV, MCH), oxidative stress (altered antioxidant enzyme activities [SOD, CAT, GPx],
 6                  decreased cellular GSH, and increased lipid peroxidation), and increased cytotoxicity in
 7                  RBC precursor cells in rodents exposed to various forms of Pb via drinking water (Jang
 8                  et al.. 2011; Molina et al. 2011; Gautam and Flora.  2010; Baranowska-Bosiacka et al..
 9                  2009; Simsek et al.. 2009; Alghazal et al.. 2008b: Kharoubi et al.. 2008b: Marques et al..
10                  2006). Some of these effects have been observed in  toxicological studies reporting blood
11                  Pb levels <10 (ig/dL, and therefore occur at blood Pb levels that are relevant to humans.
12                  These effects at relevant blood Pb levels were  found primarily in adult animals with Pb
13                  exposure durations of 4 weeks to 9 months. Although not as representative of potential
14                  human exposure pathways as exposure via drinking water, numerous toxicological
15                  studies utilizing oral gavage have also observed effects on hematological parameters,
16                  oxidative stress, and hematopoiesis (Sharma et al., 2010b; Celik etal.. 2005; Lee et al..
17                  2005). The animal toxicological evidence for decreased RBC survival and function is
18                  particularly important to the weight of evidence as it establishes clear temporality of
19                  exposure to Pb and induction of effects  on red blood cells.

20                  Associations between increased Pb blood levels and decreased RBC survival and
21                  function, are also evident in diverse populations  of human adults and children. Cross-
22                  sectional studies in children measuring concurrent blood Pb levels are consistent
23                  regarding effects on hematological  parameters (Oueirolo et al.. 2010; Shah et al.. 2010;
24                  Ahamed et al.. 2007; Huo et al.. 2007; Olivero-Verbel et al.. 2007; Riddell et al.. 2007;
25                  Turgut et al.. 2007; Ahamed et al.. 2006; Jin et al.. 2006; Rondo et al.. 2006).
26                  Associations between altered indices  of RBC oxidative stress and blood Pb levels were
27                  also seen in adolescents and children  (Ahamed et al.. 2008; Ahamed et al.. 2006; Jin et
28                  al.. 2006). The blood Pb levels observed in cross-sectional studies of children tended to
29                  be lower than those observed in adult populations (see below),  with the majority of
30                  studies in children (ages 5 months to 5 years old) reporting mean blood Pb levels
31                  <15 (ig/dL (range: 6.9 - 21.86 (ig/dL). The difference in blood Pb levels may reflect the
32                  comparatively shorter duration and lower magnitude of Pb exposure experienced by
33                  children compared to adults.

34                  For adult populations, the largest body of evidence consists of occupationally-exposed
35                  workers in which measures of RBC survival (e.g., Hb, Hct, PCV, MCV, MCH) are
36                  altered when compared with unexposed control populations in cross-sectional studies
37                  (Cabaravdic et al.. 2010; Ukaeiiofo et al.. 2009; Khan et al.. 2008; Patil et al.. 2006a;
38                  Patil et al.. 2006b; Karitaet al.. 2005; Conterato  et al.. In Press). Only one
      November 2012                             5-514                  Draft - Do Not Cite or Quote

-------
 1                   non-occupational study was found investigating the association of Pb with hematological
 2                   parameters; in pregnant women, concurrent blood Pb levels were found to be negatively
 3                   correlated with Hb concentrations. Cross-sectional studies have also observed consistent
 4                   increases in lipid peroxidation in occupationally-exposed adult populations (Ergurhan-
 5                   Ilhan et al.. 2008; Khan et al.. 2008; Mohammad et al.. 2008; Quintanar-Escorza et al..
 6                   2007; Patil et al.. 2006a: Patil et al.. 2006b). and have observed changes in oxidative
 7                   stress parameters, including lowered activities of antioxidant enzymes such as SOD, GR,
 8                   and CAT. Recent evidence of disrupted hematopoiesis, including the observation of
 9                   decreased serum EPO in occupationally-exposed adults with a mean blood Pb level of
10                   6.4 (ig/dL (Sakata et al., 2007). was consistent with previous  findings of decreased EPO
11                   in exposed adults reported in the 2006 Pb AQCD. Although the mean blood Pb level in
12                   most occupationally exposed populations was >20 (ig/dL, multiple studies observed
13                   adverse effects in occupationally-exposed populations with mean blood Pb levels
14                   <10 (ig/dL, including significant correlations PCV (7 (ig/dL), significant correlations
15                   between RBC distribution width and MCHC (5.4 (ig/dL), and decreased EPO
16                   (6.4 (ig/dL). Any differences in the effects on specific hematological and oxidative stress
17                   parameters between adult populations and children may reflect differences in exposure
18                   durations or patterns of exposure, although there is greater uncertainty regarding the
19                   timing and duration of exposure associated with these effects in adults.

20                   The evidence for Pb-associated decrements in RBC function and survival in adults and
21                   children comes from cross-sectional studies measuring concurrent blood Pb levels, and
22                   thus, the temporality of effects and the timing and duration of exposure associated with
23                   altered RBC survival and function in RBCs is unclear. This uncertainty is greatest in
24                   adults and older children as concurrent blood Pb levels also reflect higher past Pb
25                   exposures. Additional limitations of the epidemiologic database include the general lack
26                   of controlling for potential  confounders or other possible co-exposures to contaminants
27                   that can affect the hematological system. Although most studies did not control for
28                   potential confounders, a few studies investigating effects in children did adjust for
29                   potential confounders such as age, sex, area of residence, breastfeeding, mouthing
30                   behavior, family structure,  and SES-related variables, and still observed negative effects
31                   on RBC survival and function. However, no studies controlled for nutritional status,
32                   including iron intake. Further, while the epidemiologic database may be limited for the
33                   above reasons, the findings in these studies demonstrated coherence with findings from
34                   multiple toxicological studies that either reported blood Pb levels that are relevant to
35                   humans, i.e., <10 (ig/dL (drinking water and gavage studies) or utilized a relevant route
36                   of exposure (drinking water), and reported clear evidence for decreased RBC survival
37                   and function.
      November 2012                             5-515                  Draft - Do Not Cite or Quote

-------
 1                  The causal relationship between Pb exposure and decreased RBC survival and function is
 2                  further supported by epidemiologic and toxicological evidence characterizing mode of
 3                  action and biological plausibility. Pb was shown to reduce Ca2+ATPase and
 4                  Ca2+/Mg2+ATPase activities in RBC membranes, which leads to an increase in RBC
 5                  [Ca2+]l5 increased membrane fragility, and abnormal morphological changes in studies of
 6                  occupationally exposed adults (Quintanar-Escorza et al.. 2007) and in in vitro studies
 7                  (Quintanar-Escorza et al., 2010; Ciubar et al., 2007). Heul et al. (2008) observed a
 8                  reduction in plasma membrane Ca2+ATPase pump activity in newborn children's RBC
 9                  membranes in association with a concurrent group mean newborn cord blood Pb level of
10                  3.54 (ig/dL. Pb exposure has also been observed to increase PS expression on RBC
11                  membranes, leading to cell shrinkage, erythropoiesis, and destruction of the RBCs by
12                  macrophages (Jang  et al.. 2011; Ciubar etal.. 2007; Shin et al.. 2007).

13                  Experimental animal studies demonstrate that Pb exposures via drinking water and
14                  gavage, resulting in blood Pb levels relevant to humans, alter several hematological
15                  parameters, increase measures of oxidative stress, and increase cytotoxicity in RBC
16                  precursor cells. These effects were found primarily in adult animals with Pb exposure
17                  durations of 4 weeks to 9 months. Support for these findings is provided by biologically
18                  plausible modes  of action, including decreased intracellular calcium concentrations,
19                  decreased ATPase activity, and increased phosphatidylserine expression. Epidemiologic
20                  studies demonstrate evidence in both adults and children that concurrent blood Pb levels
21                  are associated with altered hematological endpoints and increased measures of oxidative
22                  stress, and altered hematopoiesis. However, the majority of these studies are limited by
23                  the lack of rigorous methodology and consideration for potential confounding. While
24                  some studies in children did control for or considered potential confounding and effects
25                  in adults and children are coherent with effects observed in exposed animals, there
26                  remains some uncertainty regarding the evidence for altered RBC  survival and function
27                  in human populations. Because epidemiologic evidence is limited to associations with
28                  concurrent blood Pb levels, there is uncertainty regarding the timing, duration,
29                  magnitude, and frequency of Pb exposure  associated with decreased RBC survival and
30                  function. Collectively, the strong evidence from toxicological studies that is supported by
31                  findings  from mode of action and epidemiologic studies is sufficient to conclude that
32                  there is a causal  relationship between Pb exposures and decreased RBC survival and
33                  function.
      November 2012                             5-516                  Draft - Do Not Cite or Quote

-------
                    5.7.4.2     Evidence for Altered Heme Synthesis

 1                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported that Pb exposure affects heme synthesis
 2                  in humans and animals through the inhibition of multiple key enzymes in the heme
 3                  biosynthetic pathway, including ALAD and ferrochelatase. A causal relationship between
 4                  Pb exposure and altered heme synthesis is strongly supported by the available
 5                  toxicological, ecotoxicological, and epidemiologic data (Table 5-35). The greatest weight
 6                  of evidence for Pb-induced alterations in heme synthesis lies primarily in the
 7                  toxicological and ecotoxicological literature. A small, but coherent, body of recent
 8                  toxicological evidence demonstrates decreased ALAD activity (Gautam and Flora. 2010;
 9                  Lee et al.. 2005) and ferrochelatase (Rendon-Ramirez et al.. 2007) in adult rats exposed
10                  to Pb via drinking water and oral gavage for 3-4 weeks. Lee et al. (2005) observed effects
11                  on ALAD activity at mean blood Pb levels of 6.5 (ig/dL after Pb administration by oral
12                  gavage once per week for 4 weeks. Evidence from previous studies cited in the
13                  2006 Pb AQCD consistently observed Pb-induced ALAD inhibition in multiple species,
14                  including birds, primates, and humans, further supporting the causal association between
15                  Pb exposure and altered heme synthesis.

16                  Similar to the earlier and more recent toxicological studies that demonstrate an
17                  association between Pb exposure and hematological effects in humans and laboratory
18                  animals, the ecological literature has consistently reported on hematological responses in
19                  aquatic and terrestrial invertebrates and vertebrates (Sections 7.3.12.5. 7.4.12.5. and
20                  7.4.21.5). The most consistently observed effect in metal impacted environments is
21                  decreased RBC ALAD activity. This effect has been observed across a wide range of
22                  taxa, including bivalves, fish, amphibians, birds, and mammals. More limited evidence
23                  exists regarding deleterious effects of Pb exposure on serum enzyme levels and white
24                  blood cell counts in birds and mammals.

25                  Consistent associations between increased Pb blood levels and decreased activity of
26                  multiple enzymes involved in the heme synthetic pathway have also been observed in
27                  diverse populations of adults and children. The strongest evidence for altered heme
28                  synthesis in adults and children come from cross-sectional epidemiological studies
29                  measuring concurrent blood Pb and reporting decreases in RBC ALAD levels and
30                  activity (Wang et al..  2010f; Mohammad et al.. 2008; Ahamed et al.. 2007; Quintanar-
31                  Escorza et al.. 2007; Ahamed et al.. 2006; Patil et al.. 2006a; Patil et al.. 2006b;
32                  Ademuyiwa et al.. 2005b; Ahamed et al.. 2005; Conterato et al.. In Press). In addition to
33                  ALAD inhibition, recent studies have also shown that Pb exposure inhibits the activity of
34                  ferrochelatase, leading to increased RBC ZPP levels in children and occupationally-
35                  exposed adults (Counter et al.. 2009. 2008; Mohammad et al.. 2008; Counter et al.. 2007;
36                  Patil et al.. 2006b; Ademuyiwa et al.. 2005b). Although the mean blood Pb levels in most
      November 2012                            5-517                 Draft - Do Not Cite or Quote

-------
 1                   of the studies investigating these effects in adults and children were >20 (ig/dL, two
 2                   studies did observe adverse effects in populations with mean blood Pb levels <10 (ig/dL:
 3                   increased ALAD reactivation index in exposed painters (5.4 (ig/dL), and statistically
 4                   significant, positive associations between ALAD and blood Pb level in children and the
 5                   elderly (7.1  and 6.4 (ig/dL, respectively).

 6                   The cross-sectional nature of the above epidemiologic studies in adults and children, and
 7                   the measurement of concurrent blood Pb, introduces some uncertainty regarding the
 8                   temporality of effects and the timing and duration of exposure associations with altered
 9                   heme synthesis. Although most studies did not control for potential confounders, a few
10                   studies investigating effects in children, and one study investigating effects in adults, did
11                   adjust for confounders such as age, sex, urban/rural residence, height, weight, BMI,
12                   smoking status, and alcohol use, and still observed negative effects on heme synthesis.
13                   However, no studies controlled for nutritional status, including iron intake. Further, while
14                   the epidemiologic database may be limited for the above reasons, the findings in these
15                   studies demonstrated coherence with findings from multiple toxicological  and
16                   ecotoxicological studies.

17                   The causal relationship between Pb exposure and altered heme synthesis is further
18                   supported by cross-sectional studies observing decreased Hb (measured as total Hb,
19                   MCH, or MCHC) in occupationally-exposed adults (Ukaejiofo et al.. 2009; Khan et al..
20                   2008; Patil et al.. 2006b: Karita et al.. 2005) and in children (Queirolo etal. 2010; Shah
21                   etal.. 2010; Olivero-Verbel et al.. 2007; Riddell et al.. 2007). Several recent toxicological
22                   studies also observed decreased Hb levels in laboratory animals exposed to Pb (Sharma et
23                   al..2010b: Baranowska-Bosiacka et al.. 2009: Simsek et al.. 2009: Marques et al.. 2006:
24                   Lee et al.. 2005). Decreased Hb levels are a direct indicator of decreased heme synthesis
25                   due to Pb exposure.

26                   In summary, altered heme synthesis is demonstrated by a small, but coherent, body of
27                   studies in adult animals reporting that Pb exposures via drinking water and gavage
28                   (resulting in blood Pb levels relevant to humans) for 15 days to 9 months decreased
29                   ALAD and ferrochelatase activities. Supporting this toxicological evidence is a larger
30                   body of ecotoxicological studies that demonstrate decreased ALAD activity across a wide
31                   range of taxa exposed to Pb. Epidemiologic studies demonstrate  evidence  in both adults
32                   and children that concurrent blood Pb levels are associated with decreased ALAD and
33                   ferrochelatase activities. However, the majority of these studies are limited by the lack of
34                   rigorous methodology and consideration for potential confounding. While  some studies
35                   in children did control for or considered potential confounding and effects in adults and
36                   children are coherent with effects observed in exposed animals, there remains some
37                   uncertainty regarding the evidence for altered heme synthesis in human populations.
      November 2012                             5-518                  Draft - Do Not Cite or Quote

-------
1

2

3

4

5

6

7
Because epidemiologic evidence is limited to associations with concurrent blood Pb

levels, there is uncertainty regarding the timing, duration, magnitude, and frequency of
Pb exposure associated with decreased RBC survival and function. Evidence for altered

heme synthesis is also provided by a large body of toxicological and epidemiologic
studies that report decreased Hb concentrations due to Pb exposure. Collectively, the

strong evidence from toxicological and ecotoxicological studies, which is supported by
findings from epidemiologic studies, is sufficient to conclude that there is a causal

relationship between Pb exposures and altered heme synthesis.
     Table 5-35   Summary of evidence supporting RBC survival and heme synthesis
                    causal determinations.
Attribute in
Causal
Framework3 Key Supporting Evidence13 Recent References'3
Pb Exposure or
Blood Pb Levels
Associated with
Effects0
     Decreased RBC Survival and Function: Causal
     Consistent       Large body of studies with
     toxicological      consistent findings for decreased
     evidence with     RBC survival and function
     relevant         (decreased Hb, Hct, PVC,
     exposures       increased eryptosis, decreased
                     hematopoiesis, increased oxidative
                     stress) in rodents with relevant
                     concentrations of Pb and routes of
                     exposure
                                 Baranowska-Bosiacka et al.
                                 (2009).
                                 Lee et al. (2005).
                                 Sharma et al. (2010b).
                                 Simsek et al. (2009),
                                 Marques et al. (2006),
                                 Molina et al. (2011),
                                 Jang et al (2011),
                                 Celik et al. (2005).
                                 Alghazal et al. (2008b).
                                 Kharoubi et al. (2008b).
                                 Gautam and Flora (2010)
            Rodents:
            Blood Pb level:
            1.7—7.1 ug/dL
            Exposures:
             Drinking water
            50-2,000 ppm,
            21—270 days as
            adults
            Oral gavage
            25—500 mg/kg,
            28—70 days
     Associations     Cross-sectional studies that
     consistently      considered potential confounding
     found in         factors found blood Pb-associated
     multiple         decreases in Hb,  increases in
     epidemiologic    anemia prevalence, increased
     studies with      oxidative stress in children ages 6
     relevant blood    mo-5 yr
     Pb levels
                     Association with Hb found in
                     children with concurrent blood Pb
                     levels with consideration for
                     potential  confounding by age, sex,
                     mouthing behavior, anemia, dairy
                     product consumption, maternal
                     age, education, employment,
                     marital status, family structure,
                     SES-related variables

                     Other studies of Hb, oxidative
                     stress adjusted for factors such as
                     age, sex, birthweight, breastfeeding
                     history, urban/rural residence
                                  Riddell et al. (2007),
                                 Queirolo et al. (2010),
                                 Ahamed et al. (2007),
                                 Ahamed et al. (2008)
                                 Queirolo et al. (2010)
                                 Riddell et al. (2007),
                                 Ahamed et al. (2008),
                                 Ahamed et al. (2007)
            Children: majority of
            concurrent blood Pb
            levels <15 ug/dL
     November 2012
                            5-519
Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
Recent References
Pb Exposure or
Blood Pb Levels
Associated with
Effects0
                Consistent evidence in large body
                of cross-sectional studies without
                consideration for potential
                confounding in occupationally-
                exposed adults and in children of
                associations of blood Pb levels with
                decreases RBC survival, interferes
                with hematopoiesis, and increases
                oxidative stress
                                  Karita et al. (2005). Khan et al.
                                  (2008), Patil et al. (2006a), Patil
                                  et al. (2006b), Ukaejiofo et al.
                                  (2009), Conterato et al. (]n
                                  Press), Cabaravdic et al. (2010),
                                  Ergurhan-llhan et al. (2008).
                                  Mohammad et al. (2008).
                                  Quintanar-Escorza et al. (2007).
                                  Sakata et al. (2007). Riddell et al.
                                  (2007), Queirolo  et al. (2010),
                                  Olivero-Verbel et al. (2007),
                                  Ahamed et al. (2006), Ahamed et
                                  al. (2007), Ahamed et al. (2008),
                                  Turgut et al. (2007). Huo et al.
                                  (2007). Shah et al. (2010).
                                  Rondo et al. (2006). Jin et al.
                                  (2006)
                                Adults (occupational
                                exposures): majority of
                                blood Pb levels
                                >20 ug/dL, some
                                studies observed
                                effects in the range of
                                5-7 ug/dl_
Evidence
clearly
describes
Mode of Action

  Altered RBC
  membrane
  ion transport
  Phosphatidyl
  serine (PS)
  expression
Evidence of increased [Ca2+]i and
decreased Ca2+/Mg2+ATPase
activity in the RBCs of exposed
workers. [Ca2+]i levels highly
correlated with blood Pb even
among unexposed controls.
[Ca2+]i levels increased in RBCs
from healthy volunteers when
exposed in vitro to Pb
[Ca2+]i associated with increased
RBC fragility and alterations in
RBC morphology

Consistent evidence from in  vivo
and in vitro  studies that Pb
exposure increases PS expression
on RBC membranes via modulation
of [Ca2+]i concentrations. Increased
PS expression leads to eryptosis
and phagocytosis by macrophages
See Section 5.7.2.2
November 2012
                             5-520
                    Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Key Supporting Evidence
Recent References
Pb Exposure or
Blood Pb Levels
Associated with
Effects0
Altered Heme Synthesis: Causal
Consistent       A small, but coherent toxicology
toxicological     database indicates decreased
evidence with    heme synthesis in rodents with
relevant         relevant Pb concentrations and
exposures       routes of exposure
                                  Rendon-Ramirez et al. (2007),
                                  Lee et al. (2005),
                                  Gautam and Flora (2010)
                                Blood Pb levels
                                6.5 ug/dL
                                Exposures: 500-
                                5,000 ppm Drinking
                                water, 15-30 days as
                                adults
Consistent       Pb-induced decreased ALAD
ecotoxicologica  activity observed across many taxa
I evidence       (bivalves, fish, amphibians, birds,
                and mammals) in multiple
                ecotoxicity studies
                                   Birds:
                                  Berglund et al. (2010).
                                  Gomez-Ramirez et al. (2011).
                                  Hansen  et al. (2011 a),
                                  Martinez-Haro et al. (2011)

                                  Freshwater Invertebrates:
                                  Aisemberg et al. (2005)
                                                  Fish:
                                                  Schmitt et al. (2005),
                                                  Schmitt et al. (2007b),
                                                  Heier et al. (2009).

                                                   Bivalves:
                                                  Kalman et al. (2008).
                                                  Company  et al. (2011).
                                 Birds:
                                6->100ug/dL
                                 Freshwater
                                Invertebrates
                                (48-h exposure in
                                aquaria)
                                 0.2-300 ug/g wet
                                tissue

                                 Fish:
                                6-14 ug/g
                                (gill or liver
                                concentrations)

                                 Bivalves:
                                0.38-3.50 ug/g dry
                                weight
Associations
found in
epidemiologic
studies with
relevant blood
Pb levels







Cross-sectional studies that Ahamed et al. (2006),
considered potential confounding Ahamed et al. (2007)
by age, sex, urban/rural residence,
height, weight, BMI found
associations with lower ALAD and
ferrochelatase activities in children.

Concurrent blood Pb level Wang et al. (201 Of)
associated with lower ALAD and
higher ZPP in adults with
consideration for potential
confounding by age, sex, smoking
status, and alcohol use.
Adults (occupational
exposure) and
children: Majority of
concurrent blood Pb
levels >20 ug/dL,
Two studies observed
associations in the
range of concurrent
blood Pb levels 5—
7 ug/dL.
T~ty' *" •— •


                Associations found in several
                studies, mostly in occupationally-
                exposed adults, that did not
                consider potential confounding
                                  Children:
                                  Ahamed et al. (2005)
                                  Occupational:
                                  Ademuyiwa et al. (2005b).
                                  Mohammad et al. (2008).
                                  Patil et al. (2006a). (2006b).
                                  Quintanar-Escorza et al. (2007),
                                  Conterato et al. (In Press)
November 2012
                             5-521
                   Draft - Do Not Cite or Quote

-------
Attribute in
Causal
Framework3
Epidemiologic
and
toxicological
evidence for
altered heme
synthesis
supported by
consistent
evidence of
decreased Hb,
a direct marker
for decreased
heme
synthesis.
Key Supporting Evidence13
Consistent evidence in animals for
decreases in Hb with relevant Pb
exposures.





Association found in children with
concurrent blood Pb levels with
consideration for potential
confounding by age, sex, mouthing
behavior, anemia, dairy product
consumption, maternal age,
Recent References'3
Animals:
Baranowska-Bosiacka et al.
(2009),
Lee et al. (2005),
Marques et al. (2006),
Sharma et al. (2010b),
Simsek et al. (2009)

Queirolo et al. (2010)





Pb Exposure or
Blood Pb Levels
Associated with
Effects0
Adult animals: Blood
Pb levels 1.7-
7.1 ug/dL after 15 day-
9 month Pb exposure




Children: Majority of
concurrent blood Pb
<15ug/dL



                 status, family structure, SES-
                 related variables

                 Other studies in children  had
                 limited or no consideration for
                 potential confounding.

                 Associations found in adults and,
                 as well as coherent findings in
                 animal toxicological studies, for
                 decreased Hb.
     Shah et al. (2010),
     Olivero-Verbel et al. (2007),
     Riddell et al. (2007)

     Adults: Karita et al. (2005).
     Khan et al. (2008).
     Patil et al. (2006b).
     Ukaejiofo et al. (2009)
             Adults (occupational
             exposure): Majority of
             blood Pb >20 ug/dl_
aDescribed in detail in Table II of the Preamble.
Describes the key evidence and references contributing most heavily to causal determination. Also noted are the
sections where full body of evidence is described.
cDescribes the blood  Pb levels in humans with which the evidence is substantiated and blood Pb levels in animals
most relevant to humans.
November 2012
5-522
Draft - Do Not Cite or Quote

-------
      5.8        Reproductive and Developmental Effects

 1                  The effect of Pb on reproductive and developmental outcomes has been of interest for
 2                  years, starting in cohorts of occupationally-exposed individuals. More recently,
 3                  researchers have begun to focus on reproductive and developmental effects in
 4                  populations without occupational exposures, with more environmentally-relevant levels
 5                  of Pb exposure. The toxicological and epidemiologic literature on reproductive effects of
 6                  Pb include research on female and male reproductive function such as hormone levels,
 7                  fertility, spontaneous abortions, effects on sperm, estrus, and effects on reproductive
 8                  organs. Evaluation of effects on the developing organism includes effects on puberty,
 9                  postnatal growth, and effects on the development of the teeth, sensory organs, and other
10                  systems. Research on birth outcomes includes birth defects, infant mortality, preterm
11                  birth, and low birth weight. A few studies of pregnancy-induced hypertension and
12                  eclampsia have been  conducted and are reported on in the section on hypertension
13                  (Section 5.4.2.1). Briefly, the relatively small number of studies found consistently
14                  positive associations between blood Pb levels and pregnancy-induced hypertension.
15                  Biomarkers of Pb exposure, including blood Pb and bone Pb, are used in the
16                  epidemiologic studies reviewed in this section. Bone Pb typically indicates  cumulative
17                  exposure to Pb, whereas, blood Pb may indicate more recent exposure. However, Pb can
18                  also be remobilized from the bone during times of active bone remodeling,  such as
19                  pregnancy or lactation. Toxicological studies typically report exposure using blood Pb.
20                  More detailed discussion of these measures and Pb transfer via umbilical cord blood Pb
21                  across the placenta, and via lactation is given in Section 4.2.2.4 on Pb Toxicokinetics.

22                  Overall, the recent literature on reproductive effects of Pb exposure continues to support
23                  associations reported in earlier Pb AQCDs between Pb exposure and effects on various
24                  parameters of sperm (function, motility, count, integrity, histology). The toxicological
25                  and epidemiologic literature of developmental effects of Pb exposure also indicates that
26                  Pb exposure is associated with delayed onset of puberty in both males and females.
27                  Associations between Pb exposure and other reproductive and developmental effects
28                  have less consistent findings. The recent information from epidemiologic and
29                  toxicological studies is integrated with conclusions from previous Pb AQCDs below.
      5.8.1        Effects on Development

30                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported Pb-associated developmental effects on
31                  teeth, sensory organs, the GI system, the liver, and postnatal growth as well as delayed
32                  puberty (U.S. EPA, 2006b). There was recognition that Pb is transferred across the
33                  placenta and through the breast milk, contributing to exposure during development. The

      November 2012                            5-523                  Draft - Do Not Cite or Quote

-------
 1                  2006 Pb AQCD reported delayed puberty in both male and female populations in animal
 2                  toxicology studies showing associations with dam blood Pb levels of ~40 (ig/dL and pup
 3                  blood Pb levels of 26 (ig/dL. The research reported in this ISA continues to find delayed
 4                  puberty with Pb exposure at even lower Pb doses in animal toxicology studies as is
 5                  detailed below. Mechanistic understanding of delayed puberty is also reported in this
 6                  ISA. Lower dose Pb exposure studies in animal toxicology are also reported in studies of
 7                  sensory organ function and postnatal growth in this ISA. Studies included in this ISA
 8                  expand upon evidence reported in previous Pb AQCDs for the aforementioned systems
 9                  sensitive to developmental effects with recent studies showing effects at lower doses of
10                  Pb.
                    5.8.1.1      Effects on Puberty among Females

11                  Recent toxicological studies of rodents have examined the effects of Pb on pubertal and
12                  reproductive organ development and on biomarkers of pubertal development among
13                  females. There have also been recent epidemiologic studies examining associations
14                  between blood Pb levels and onset of puberty among girls, which are summarized in
15                  Table 5-36 and in the text below. All of the epidemiologic studies examined concurrently
16                  measured blood Pb and puberty and are reported below. Additionally, while there was a
17                  longitudinal investigation by Naicker et al. (2010). who followed girls to determine their
18                  age of menarche, blood Pb levels were measured once at 13 years of age.
     November 2012                            5-524                 Draft - Do Not Cite or Quote

-------
Table 5-36  Summary of recent epidemiologic studies of associations between Pb levels and puberty for
            females.
Reference
(Studies are
presented
in order of
first
appearance
in the text of Study Location
this section) and Years
Wu et al. U.S.A.
(2003b) 1988-1994









Selevanetal. U.S.A.
(2003) 1988-1994
Out- Methodological
come Study Population Details
Tanner Girls ages 8-16 from the Cross-sectional
staging and NHANES III study study using
age at logistic regression
menarche with weighting
N=1706









Tanner Girls ages 8-18 from the Cross-sectional
staging and NHANES III study study using ordinal
age at N =600 logistic regression
menarche ' ,NHwhlle ° and Cox
NNHbiack-805 proportional
NMexican-American=781 hazards
Mean Pb
Pb (SD)
Biomarker in ug/dL
Blood Pb 2.5 (2.2)
Weighted
proportion of
the sample
with blood Pb
5.0-21 .7: 5.9%









Blood Pb Geometric
mean
NHWhites: 1.4
NHBIacks: 2.1
Mexican-
Americans: 1.7
Adjusted
Effect
Estimates
OR (95% Cl)
Breast
development
0.7-2.0 ug/dL:
1 .00 (Ref)
2.1-4.9 ug/dL:
1.51 (0.90,2.53)
5.0-21 .7 ug/dL:
1.20(0.51, 2.85)
Pubic hair
development
0.7-2.0 ug/dL:
1 .00 (Ref)
2.1-4.9 ug/dL:
0.48 (0.25, 0.92)
5.0-21 .7 ug/dL:
0.27 (0.08, 0.93)
Menarche
0.7-2.0 ug/dL:
1.00 (Ref)
2.1-4.9 ug/dL:
0.42(0.18,0.97)
5.0-21. 7 ug/dL:
0.19(0.08, 0.43)
OR (95% Cl)
Breast
development
NH Whites:
1 ug/dL: 1.00
Potential
Confounders
Adjusted for in
Analysis
Race/ethnicity,
age, family size,
residence,
poverty income,
ratio, BMI









For breast
development:
Age, age2, height,
BMI, family
income, ever
smoked>100
cigarettes, dietary
November 2012
5-525
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented
in order of
first
appearance
in the text of Study Location
this section) and Years






































Mean Pb
Out- Methodological Pb (SD)
come Study Population Details Biomarker in ug/dL

Blood Pb
levels>5|jg/dL:
NHWhites:
2.7%
NHBIacks:
11.6%
Mexican-
Americans:
12.8%
Blood Pb
levels
>10|jg/dL:
NHWhites:
0.3%
NHBIacks:
1.6%
Mexican-
Americans:
2.3%
























Adjusted
Effect
Estimates
(Ref)
3 |jg/dL: 0.82
(0.47, 1 .42)
NH Blacks:
1 |jg/dL: 1 .00
(Ref)
3 |jg/dL: 0.64
(0.42, 0.97)
Mexican
Americans:
1 |jg/dL: 1 .00
(Ref)
3 |jg/dL: 0.76
(0.63,0.91)

Pubic hair
development
NH Whites:
1 ug/dL: 1 .00
(Ref)
3 ug/dL: 0.75
(0.37,1.51)
NH Blacks:
1 ug/dL: 1 .00
/D _ f\
(Ref)
3 ug/dL: 0.62
(0.41 , 0.96)
Mexican
Americans:
1 ug/dL: 1 .00
(Ref)
3 ug/dL: 0.70
(0.54,0.91)
HR (95% Cl)
'included only
girls 8-1 6
Age at menarche
NH Whites:
1 ug/dL: 1 .00




Potential
Confounders
Adjusted for in
Analysis
Fe, dietary
vitamin C, dietary
Ca .

For pubic hair
development:
Age, age2, height,
family income,
ever smoked>100
cigarettes,
anemia, dietary
Fe, dietary
vitamin C
For age at
menarche:
Height, BMI,
family income,
anemia, dietary
Ca2+.

Considered in all
models: age,
smoking, dietary
Ca , dietary Fe,
dietary vitamin C,
dietary total fat,
anemia, urban
residence, family
income












November 2012
5-526
Draft - Do Not Cite or Quote

-------
 Reference
(Studies are
 presented
 in order of                                                                                                                            _ .   ..  .
    first                                                                                                                               Potential
appearance                                                                                          Mean Pb        Adjusted      n(:onf?"Jnder?
in the text of    Study Location       Out-                             Methodological       Pb          (SD)           Effect       Adjusted for in
this section)	and Years	come	Study Population	Details	Biomarker    in ug/dL	Estimates	Analysis
                                                                                                                  (Ref)
                                                                                                                  3 |jg/dL: 0.74
                                                                                                                  (0.55, 1.002)
                                                                                                                  NH Blacks:
                                                                                                                  1 |jg/dL: 1.00
                                                                                                                  (Ref)
                                                                                                                  3 |jg/dL: 0.78
                                                                                                                  (0.63, 0.98)
                                                                                                                  Mexican
                                                                                                                  Americans:
                                                                                                                  1 |jg/dL: 1.00
                                                                                                                  (Ref)
                                                                                                                  3 |jg/dL: 0.90
                                                                                                                  (0.73,1.11)
November 2012                                5-527                   Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented
in order of
first
appearance
in the text of Study Location
this section) and Years
Gollenberg et U.S.A.
al. (2010) 1988-1994































Potential
Mean Pb Adjusted n(:onf?"Jnder?
Out- Methodological Pb (SD) Effect Adjusted for in
come Study Population Details Biomarker in ug/dL Estimates Analysis
Luteinizing Girls ages 6-11 from the Cross-sectional Blood Pb Median 2.5 OR (95% Cl) for Age,
hormone NHANES III study study using survey (range 0.07, exceeding race/ethnicity,
(LH) and logistic regression 29.4) pubertal inhibin B BMI, census
inhibin B blood Pb cutoff (>35pg/mL) region, poverty -
>10ug/dL:5% <1 ug/dL: 1.00 income ratio
(Ref)
1-4.9ug/dL: 0.38
(0.12, 1.15)
> 5 ug/dL: 0.26
(0.11,0.60)
OR (95% Cl) for
exceeding
pubertal LH cutoff
(>0.4mlU/mL)
<1 ug/dL: 1.00
(Ref)
1 -4.9 ug/dL: 0.98
(0.48, 1.99)
> 5 ug/dL: 0.83
(0.37, 1.87)
*Note: a
sensitivity
analysis including
only those with
blood Pb
<10 ug/dLhad
similar results but
ORs were slightly
attenuated
November 2012
5-528
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented
in order of
first
appearance
in the text of Study Location Out-
this section) and Years come
Denham et Akwesasne Mohawk Age at
a 1. (2005) Nation (boundaries of menarche
New York, Ontario,
and Quebec
NS
Potential
Mean Pb Adjusted ,£?nfou"ders
Methodological Pb (SD) Effect Adjusted for in
Study Population Details Biomarker in ug/dL Estimates Analysis
10- to 16.9-yr-old girls in Cross-sectional Blood Pb 0.49(0.905) Coefficients for Age, SES, BMI
the Akwesasne study using probit binary logistic
community and logistic . regression
regression Median: 1 .2 predicting
menarche with Pb
N=138 centered at the
mean:
log blood Pb
-1 .29 (p-value
0.01)
log blood Pb -
squared: -1.01 (p-
value 0.08)
Non-linear
relationship
observed and Pb
below the mean
did not appear to
affect the odds of
menarche.
Increasing blood
Pb from 0.49 to
0.98 ug/dL
decreased the
odds of menarche
attainment by
72%
November 2012
5-529
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented
in order of
first
appearance
in the text of Study Location
this section) and Years
Naicker et al. Johannesburg/Soweto,
(2010) South Africa
Born in 1990








Den Hond et Flanders
al. (2011) 2003-2004








Out-
come
Self-
reported
Tanner
staging at
age 13 and
age at
menarche








Tanner
staging, age
at
menarche,
regular
menses








Study Population
Girls of blacker mixed
ancestry who were
enrolled in the Birth to
Twenty (Bt20) cohort
(born in 1990) that lived
in Johannesburg/Soweto
for at least 6 mo after
birth
N=682






Girls ages 14 and 15, in
their 3rd year of
secondary education and
living in the same study
areas for at least 5 years
N=792








Mean Pb
Methodological Pb (SD)
Details Biomarker in ug/dL
Cross-sectional Blood Pb at 4.9 (1 .9)
and longitudinal 1 Syr of age blood Pb
study using ,eve|s
logistic regression >10ug/dL
1%








Cross-sectional Blood Pb Median: 1 .81
study using 10th
logistic regression percentile:
0.88
90th
percentile:
3.81








Adjusted
Effect
Estimates
OR (95% Cl)
Delay in breast
development at
age 13
<5 ug/dL: 1 .00
(Ref)
> 5 ug/dL: 2.34
(1 .45, 3.79)
Delay in pubic
hair development
at age 13
<5 ug/dL: 1 .00
(Ref)
> 5 ug/dL: 1 .81
(1.15, 2.84)
Delay in
attainment of
menarche at age
13
<5 ug/dL: 1 .00
(Ref)
> 5 ug/dL: 2.01
(1 .38, 2.94)
OR (95% Cl) for
pubic hair
development with
doubling of
exposure
0.65 (0.45, 0.93)
'Association was
no longer
statistically
significant when
PCB marker
included in the
model
No association
between Pb and
breast
development
(results not given)

Potential
Confounders
Adjusted for in
Analysis
BMI








Age, BMI,
smoking, oral
contraceptive use
Considered but
did not include:
food intake and
lifestyle
parameters






November 2012
5-530
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented
in order of
first
appearance
in the text of Study Location
this section) and Years
Tomoum et Cairo, Egypt
al. (2010) 2007





















Wolff et al. New York City, NY
(2008): Wolf 1 gg^ 997
and Daley
(2007)








Out-
come
Hormones
and pubertal
development




















Pubertal
stages
defined
using
standard
drawings







Study Population
Healthy children aged
10-13 yr; seeking
treatment for minor
health problems and
living in one of two
designated areas (one
with high-risk for Pb
contamination and one
with no Pb source)

N=20















9-yr old girls from the
study hospital and
nearby pediatric offices

N=192







Methodological
Details
Cross-sectional
study using Chi-
square




















Cross-sectional
study using
Poisson
multivariate
regression with
robust error
variance







Mean Pb
Pb (SD)
Biomarker in ug/dL
Blood Pb NS for girls
only
(combined
with boys in
the study the
mean was
9.46 [3.08])


















Blood Pb Median: 2.4











Adjusted
Effect
Estimates
Breast
Development
<10 ug/dL:
Stage 2: 36.4%
Stage 3: 63.6%
> 10 ug/dL:
Stage 2: 100%
Stage 3: 0%
Chi-square p-
value<0.01
Pubic Hair
Development
<10 ug/dL:
Stage 2: 36.4%
Stage 3: 63.6%
>10ug/dL:
Stage 2: 77.8%
Stage 3: 22.2%
Chi-square p-
value>0.05
'Quantitative
results for
hormones not
provided
PR (95% Cl) (unit
not given,
assume results
are per 1 ug/dL)
Breast stage:
1.01 (0.79, 1.30)
Pubic hair stage:
1.25(0.83, 1.88)




Potential
Confounders
Adjusted for in
Analysis
None






















For breast
development:
Age, BMI, race

For hair stage:
Height, private
clinic, race

November 2012
5-531
Draft - Do Not Cite or Quote

-------
 1                  Multiple studies have been performed examining blood Pb levels and puberty using
 2                  NHANES III data (Gollenbera etal.. 2010: Selevan etal. 2003: Wu et al.. 2003b).  A
 3                  study that included girls aged 8-16 years reported an association between increased blood
 4                  Pb and delayed attainment of menarche and pubic hair development, but not for breast
 5                  development (Wu et al., 2003b). The associations were observed even at blood Pb levels
 6                  of 2.1-4.9 (ig/dL compared to girls with blood Pb levels <2.1 (ig/dL. Another NHANES
 7                  III study included girls  8-18 years of age and reported the results stratified by race
 8                  (Selevan et al.. 2003). This study also included many important potential confounders,
 9                  such as nutritional information. Higher blood Pb levels were associated with lower
10                  Tanner stage of breast and pubic hair development and later age at menarche among
11                  African Americans and with lower stage of breast and pubic hair development among
12                  Mexican Americans. For whites, the associations were in the same directions, but none
13                  reached statistical significance. In a study of girls aged 6-11 years old from NHANES III
14                  data, higher blood Pb levels were associated with lower inhibin B, a protein that inhibits
15                  FSH production, but no association was observed for LH. (Gollenberg et al.. 2010). The
16                  inverse association between blood Pb and inhibin B was greater among girls with iron
17                  deficiency compared to those with high Pb but sufficient  iron levels. Inhibin B and LH
18                  were chosen for this study because, as the authors indicated, these hormones are,
19                  "believed to be relevant for younger girls... near the onset of puberty and... serve as
20                  markers for hypothalamic-pituitary-gonadal functioning."

21                  A study of girls aged 10-16.9 years of age in the Akwesasne Mohawk Nation reported a
22                  nonlinear association between higher blood Pb and greater age at menarche (Denham et
23                  al., 2005). No association was observed below blood Pb of 0.49 (ig/dL in a nonlinear
24                  model of the Pb-menarche relationship. A study conducted in South Africa reported an
25                  association between increased blood Pb levels and older age at first menarche and
26                  pubertal development (Naicker et al.. 2010). Another study reporting on girls with low
27                  blood Pb concentrations observed an association between higher blood Pb and less  pubic
28                  hair development but not breast development (DenHond E. 2011). The association was
29                  no longer statistically significant when a marker for polychlorinated biphenyl exposure
30                  was included in the model. A study among girls aged 10-13 years (median: 12 years)
31                  reported lower levels of FSH and LH levels in the group with blood Pb of at least
32                  10 (ig/dL compared to the group with blood Pb less than  10 (ig/dL (Tomoum et al..
33                  2010). In addition, there were some indications of lower Tanner stages of breast
34                  development associated with Pb levels of at least 10 (ig/dL, but this relationship was not
35                  present for stages of pubic hair development and there was no control for potential
36                  confounders. A study performed in NYC among 9-year old girls reported no association
37                  between Pb levels and pubertal development (Wolff etal.. 2008). but this age group may
38                  be too young to study when investigating delayed puberty as the outcome.
      November 2012                            5-532                 Draft - Do Not Cite or Quote

-------
                Puberty; Neonate/adult; Mouse;
                  Female; lavicoli et al. (2006)
               Neurotransmitter; Adult; Mouse;
                  Both; Leasure et al. (2008)
            Physical development; Adult; Mouse;
                Male; Leasure et al. (2008)
            Eye; Ad lilt; Rat; Both; Fox et al. (2008)
              Redox-oxidative stress; Adult; Rat;
              Male; Nava-Hernandez et al. (2009)
           Sperm; Adult; Rabbit; Male; Moorman
                      etal. (1998)
           Neurobehavioral; Adult; Mouse; Male;
                  Leasure et al. (2008)
           Hematological parameters; Adult; Rat;
                 Both; Teijon et al. (2006)
             Histology; Adult; Rat; Both; Teijon et
                       al. (2006)
           Biomarkers; Adult; Rat; Both; Teijon et
                      al, (2006)
               Physical development; Adult; Rat;
                  Both; Teijon et al. (2006)


J

^
<




o Highest Concentration
* Lowest Cone, with Response
A Highest Cone, with No Response
o Lowest Concentration

«o
i An
fc— *



                                                             10                 100
                                                         Blood Lead Level (|jg/dL)
                                                   1000
Note: This figure illustrates reproductive and developmental effects associated with Pb exposure in studies that examined multiple
exposure concentrations. Dosimetric representation reported by blood Pb level. (Studies are described in Table 5-37).

Figure 5-37       Toxicological concentration-response array for reproductive and
                      developmental effects of Pb.
November 2012
5-533
Draft - Do Not Cite or Quote

-------
     Table 5-37  Toxicological concentration-response array summary for
                   reproductive and developmental effects of Pb presented in
                   Figure 5-37.
Reference
lavicoli et al. (2006a)
Leasure et al. (2008)
Fox et al. (2008)
Nava-Hernandez et al. (2009)
Moorman et al. (1998)
Teijon et al. (2006)
Fox et al. (2008)
Nava-Hernandez et al. (2009)
Moorman et al. (1998)
Teijon et al. (2006)
Fox et al. (2008)
Blood Pb level with Effect (|jg/dL)
8&13
10&42
10, 24 & 42
10&42
12
19.5
25-130
40 & 100
40 & 100
40 & 100
100
12
19.5
25-130
40 & 100
40 & 100
40 & 100
100
12
Altered Outcome
Delayed onset female puberty
Neurotransmitter, Dopamine homeostasis
Physical Development, Adult obesity
(males)
Aberrant response to amphetamine
Retinal aberrations
Sperm affected via redox imbalance
Semen quality affected
Hematology
Histology-Offspring renal & hepatic
Biomarker-Offspring renal function
Physical development: birth weight
Retinal aberrations
Sperm affected via redox imbalance
Semen quality affected
Hematology
Histology-Offspring renal & hepatic
Biomarker-Offspring renal function
Physical development: birth weight
Retinal aberrations
 1
 2
 o
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
Earlier studies showed that prenatal and lactational exposures to Pb can cause a delay in
the onset of female puberty in rodents. Recent studies corroborate these findings and
show that puberty onset is one of the more sensitive markers of effects of Pb exposure as
is demonstrated in the exposure response array (Figure 5-37 and Table 5-37; including
outcomes described in sections that follow). Dumitrescu et al. (2008c) exposed adult
Wistar female rats to varying doses of Pb acetate (50-150 ppb) in drinking water for
3 months before mating and during pregnancy. Vaginal opening, an indicator of sexual
maturation, was statistically significantly delayed in pups from all Pb treated groups
when compared to pups from non-treated dams. The age at vaginal opening in female
pups from the Pb treated groups increased, in a concentration-dependent manner, from
39 days to 43-47 days. The authors also observed a correlation between body weight and
age at vaginal opening meaning that as body weight decreased the age at vaginal opening
increased. This effect also exhibited a concentration-dependent relationship.
     November 2012
                          5-534
Draft - Do Not Cite or Quote

-------
 1                   In another recent study, lavicoli et al. (2006a) reported a statistically significant delay in
 2                   several indicators of sexual maturity in offspring (Swiss mice, FI generation) born to
 3                   dams that ingested 3.5-40 ppm Pb in their daily diet; offspring had continuous dietary
 4                   exposure until the termination of the experiment at puberty. Maternal ingestion of Pb at
 5                   the various doses resulted in female pup blood Pb levels of 3.5-13 (ig/dL. For all diet
 6                   groups in this range (3.5-13 (ig/dL), there was a delay in age at vaginal opening, age of
 7                   first estrus, age of vaginal plug formation, and age of first parturition when compared to
 8                   the group at background Pb concentration (2 (ig/dL). A novel finding in the lavicoli study
 9                   was that very low dose Pb (blood Pb of 0.7 (ig/dL, food concentration of 0.02 ppm
10                   continuous through gestation, lactation and until the termination of the experiment)
11                   induced statistically significant acceleration of markers of sexual maturation in female
12                   offspring versus background Pb level animals (blood Pb of 2 (ig/dL). There were
13                   statistically significant increases in time of vaginal opening (30% earlier), first estrous,
14                   first vaginal plug formation, and first parturition at the very low Pb exposure versus
15                   2 (ig/dL animals. Thus, the timing of puberty is delayed in a concentration-dependent
16                   fashion with very low dose Pb having a statistically significant earlier  onset of puberty
17                   than the background Pb animals (2 (ig/dL). Also, the animals exposed  to the higher dose
18                   of Pb (blood Pb up to 13 (ig/dL) had statistically significant delays in onset of puberty
19                   when compared to the other dose groups.

20                   In addition, Pb-induced shifts in sexual maturity were observed in the  subsequent
21                   generation (F2 generation) across that dose range. These F2 animals continued to be
22                   exposed to same concentrations of Pb over multiple generations through the diet. Results
23                   in the F2 generation closely resembled those of the FI generation, as both generations
24                   received Pb exposure. The authors concluded that a modest elevation in blood Pb level
25                   (13 (ig/dL) over background (2-3 (ig/dL) can result in a profound delay in the onset of
26                   puberty (15-20%). In the F2 generation, reduction in blood Pb (0.7 (ig/dL) below
27                   background (2-3  (ig/dL) was associated with an earlier onset of sexual maturity (30%
28                   increase) above background.

29                   In the 2006 Pb AQCD (U.S. EPA. 2006b). it was reported that a statistically significant
30                   reduction in the circulating levels of insulin-like growth factor 1 (IGF-1), LH, and
31                   estradiol (E2) was associated with Pb-induced delayed puberty in Fischer 344 pups.
32                   Subsequently, Pine et al. (2006) evaluated whether IGF-1 replacement could reverse the
33                   effects of Pb on delayed female puberty onset. The authors reported that offspring from
34                   dams exposed to  Pb during gestation and lactation (daily oral gavage of dam with 1.0 mL
35                   solution of Pb acetate 12 mg/mL; mean maternal blood Pb level 40 (ig/dL) exhibited a
36                   marked increase in LH and luteinizing hormone releasing hormone (LHRH) secretion
37                   after IGF-1 administration (200 ng3/(iL i.p. injection twice daily from PND23 until the
38                   appearance of vaginal opening which appears in control animals at -PND40) resulting in
      November 2012                             5-535                  Draft - Do Not Cite or Quote

-------
 1                  restored timing of vaginal to that of control animals. It should be noted that, IGF-1
 2                  replacement in Pb-exposed animals did not cause advanced puberty over non-Pb-exposed
 3                  controls. The results of this study provide support to the theory that Pb-induced delayed
 4                  onset of puberty may be due to disruption of pulsatile release of sex hormones (U.S.
 5                  EPA. 2006b) and not necessarily due to a direct toxic effect on the hypothalamic-
 6                  pituitary-gonadal axis (Salawu et al.. 2009). and IGF-1 may play a prominent role in the
 7                  process.

 8                  In summary, epidemiologic studies consistently show an association between higher
 9                  concurrent blood Pb and delayed pubertal development in girls. This association is
10                  apparent even at low blood Pb levels. Most of the studies had good sample sizes and
11                  controlled for some potential confounders. Nutritional information was rarely controlled
12                  for although this could be important, especially in populations where malnutrition is
13                  prevalent. These epidemiologic studies are cross-sectional, which does not allow for the
14                  study of temporality between Pb levels and pubertal onset nor does it consider the
15                  influence of past blood Pb levels. New evidence from the toxicology literature continues
16                  to indicate Pb-induced delays in the onset of puberty.  Further, the biological plausibility
17                  of delayed puberty is expanded with the toxicological literature that shows this pathway
18                  is mediated by IGF-1.
                    5.8.1.2     Effects on Puberty among Males

19                  Recent epidemiologic studies examining the association between blood Pb and onset of
20                  puberty in males are summarized in Table 5-38. The majority of studies used concurrent
21                  measures of blood Pb and puberty (DenHond E. 2011; Tomoum et al.. 2010; Hauser et
22                  al.. 2008). but Williams et al. (2010) performed a longitudinal analysis of blood Pb levels
23                  measured at ages 8-9 years and pubertal onset, following the participants for 3 years.
24                  Little epidemiologic information was available regarding pubertal onset in the
25                  2006 Pb AQCD (U.S. EPA. 2006b).
      November 2012                            5-536                 Draft - Do Not Cite or Quote

-------
Table 5-38  Summary of recent epidemiologic studies of associations between Pb levels and puberty for males.
Reference
(Studies are
presented in
order of first
appearance in
the text of this Study
section) Location Outcome
Mauser etal. Chapaevsk, Pubertal
(2008) Russia stages
2003-2005 defined using
standard
drawings
Pb Adjusted _ . .. , _ ,
Methodological Bio- Mean Pb (SD) Effect Potential Confounders
Study Population Details marker in ug/dL Estimates Adjusted for in Analysis
Healthy boys aged 8-9 Cross-sectional Blood Median: 3 (IQR OR (95% Cl) Gestational age, height, BMI,
study using Pb 2-5) Pubertal onset age at exam
M-489 multivariable blood Pb based on
IOglStlCregreSSIOn >10ug/dL3% testicular volume considered but did not
<5 ug/dL: 1.00 include: parental education,
(Ref) household income
> 5 ug/dL: 0.83
(0.43, 1 .59)
*after adjustment
for
macronutrients,
the OR (95% Cl)
became 0.66
(0.44, 1 .00)
Genital
development
<5 ug/dL: 1.00
(Ref)
> 5 ug/dL: 0.57
(0.34, 0.95)
*after adjustment
for
macronutrients,
the OR (95% Cl)
became 0.52
(0.31 , 0.88)
Pubic hair
development
<5 ug/dL: 1.00
(Ref)
> 5 ug/dL: 0.74
(0.34, 1 .60)
November 2012
5-537
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this Study
section) Location Outcome
Williams et al. Chapaevsk, Pubertal
(2010) Russia stages
2003-2008 defined using
standard
drawings

















Tomoum et al. Cairo, Hormones
(2010) Egypt and pubertal
2007 development


















Study Population
Healthy boys aged 8-9
at enrollment who had
3 annual follow-up
evaluations
N=481
















Healthy children aged
10-1 3 seeking
treatment for minor
health problems and
living in one of two
designated areas (one
with high-risk for Pb
contamination and
one with no Pb
source)

N=21











Pb
Methodological Bio- Mean Pb (SD)
Details marker in ug/dL
Longitudinal cohort Blood Median: 3 (IQR
using Cox Pb at 2-5)
proportional ages B,ood pb |eve|
hazards 8-9 >lOug/dL:3%

















Cross-sectional Blood NS for boys only
study using Chi- Pb (combined with
square girls in the study
the mean was
9.46 [3.08])

















Adjusted
Effect
Estimates
HR (95% Cl)
Pubertal onset
based on
testicular volume
<5 ug/dL: 1 .00
(Ref)
> 5 ug/dL: 0.73
(0.55, 0.97)
Genital
development
<5 ug/dL: 1.00
CRpn
\,na\)
> 5 ug/dL: 0.76
(0.59, 0.98)


Pubic hair
development
<5 ug/dL: 1.00
(Ref)
> 5 ug/dL: 0.69
(0.44, 1 .07)
Testicular size
<10 ug/dL:
Stage 1 : 0%
Stage 2: 44.4%
Stage 3: 55.6%
> 10 ug/dL:
Stage 1 : 33.3%

Stage 2: 66.7%
Stage 3: 0%
Chi-square p-
value<0.01
Pubic Hair
Development
<10 ug/dL:
Stage 1 : 0%
Stage 2: 55.6%




Potential Confounders
Adjusted for in Analysis
Birthweight, gestational age,
energy intake, proportion of fat
consumption, proportion of
protein consumption, maternal
alcohol consumption during
pregnancy, height at study
entry, BMI at study entry,
household income, parental
education
NOTE: exclusion of BMI and
height, in case they were part
of the causal pathway,
resulted in very similar
estimates


Considered but not included:
parity, maternal or household
smoking during pregnancy,
maternal age at birth




None
















November 2012
5-538
Draft - Do Not Cite or Quote

-------
  Reference
  (Studies are
  presented in
  order of first
 appearance in
 the text of this      Study
   section)	Location
Outcome     Study Population
Methodological
    Details
  Pb
 Bio-
marker
Mean Pb (SD)
   in ug/dL
Ad justed
  Effect
Estimates
                                                                                                Potential Confounders
                                                                                               Adjusted for in Analysis
                                                                                                           Stage 3: 44.4%
                                                                                                           > 10ug/dL:
                                                                                                           Stage 1: 33.3%
                                                                                                           Stage 2: 66.7%
                                                                                                           Stage 3: 0%
                                                                                                           Chi-square p-
                                                                                                           value<0.05

                                                                                                           Penile staging
                                                                                                           <10ug/dL:
                                                                                                           Stage 1: 11.1%
                                                                                                           Stage 2: 44.4%
                                                                                                           Stage 3: 44.4%
                                                                                                           > 10ug/dL:
                                                                                                           Stage 1: 58.3%
                                                                                                           Stage 2: 41.7%
                                                                                                           Stage 3: 0%
                                                                                                           Chi-square p-
                                                                                                           value<0.05

                                                                                                           Mean
                                                                                                           testosterone level
                                                                                                           <10ug/dL:
                                                                                                           4.72 (SD 1.52)
                                                                                                           > 10ug/dL:
                                                                                                           1.84 (SD 1.04)

                                                                                                           •Quantitative
                                                                                                           results for  LH and
                                                                                                           FSH not provided
November 2012
                  5-539
         Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Den Hond et al.
(2011)










Study
Location Outcome Study Population
Flanders Tanner Boys ages 14 and 15,
2003-2004 stagir|g and in their 3rd year of
gynecomastia secondary education
and living in the same
study areas for at
least 5 years


N=887






Methodological
Details
Cross-sectional
study using logistic
regression










Pb
Bio- Mean Pb (SD)
marker in ug/dL
Blood Median: 2.50
Pb 10thpercentile:
1.20
90th percentile:
5.12









Adjusted
Effect
Estimates
OR (95% Cl) for
gynecomastia
with doubling of
exposure
1.84(1.11, 3.05)

No association
between Pb and
pubic hair or
genital
development
(results not given)


Potential Confounders
Adjusted for in Analysis
Parental education, age, BMI,
smoking status
Considered but not included:
food intake, lifestyle
parameters

NOTE: results were the same
when hexachlorobenzene was
included in the model


November 2012
5-540
Draft - Do Not Cite or Quote

-------
 1                  Studies were performed among a cohort of Russian boys enrolled between ages 8-9 years
 2                  ("Williams et al.. 2010; Hauser et al.. 2008). The area where these studies were performed
 3                  had various environmental contaminants, such as dioxin, polychlorinated biphenyls, and
 4                  other metals, present but these were not included in the analyses (although preliminary
 5                  analyses found no correlation between blood Pb levels and serum dioxin levels). Both the
 6                  cross-sectional study (Hauser et al.. 2008) and the prospective study with annual follow-
 7                  ups (Williams et al.. 2010) demonstrated an association; higher blood Pb levels at
 8                  8-9 years of age was associated with later onset of puberty. In a study of boys in Egypt,
 9                  boys with higher blood Pb had delayed pubertal onset compared to those with lower
10                  levels (Tomoum et al., 2010). In addition, compared to the low blood Pb group, those
11                  boys with higher blood Pb had lower testosterone, FSH, and LH levels but there was no
12                  control for confounding. A study in Flanders reported no associations between blood Pb
13                  concentration and pubertal development among 14- and 15-year old boys (Den Hond E.
14                  2011). However, higher blood Pb levels were associated with an increased odds of
15                  gynecomastia.

16                  No recent toxicological studies address Pb-induced male sexual maturation and
17                  development, but earlier studies do provide support to findings in epidemiologic cohorts.
18                  Pb exposure resulted in delayed sexual maturity as measured by prostate weight in male
19                  Sprague-Dawley pups at PND35. These pups were exposed chronically to 1,500 or
20                  4,500 ppm Pb acetate in dam or their own drinking water from GD5 until PND85 and had
21                  blood Pb ranges from low to high of 88-196 and 120-379 ug/dL, respectively (Ronis et
22                  al.. 1998b). Cynomolgus monkeys exposed to Pb over a lifetime (an oral capsule of
23                  1,500 (ig/kg body weight/day for 10 years, blood Pb levels ranging from 30-60 ug/dL)
24                  had altered pituitary and Sertoli cell function along with decreases in inhibin/FSH ratio
25                  and reduced gonadotropin-releasing hormone (GnRH) stimulation of LH release in
26                  adulthood (Foster et al.. 1993). all indicators that are  important in proper sexual
27                  maturation. Further mechanistic understanding  of the effect of Pb can be gleaned from
28                  studies in adult male Wistar rats exposed to Pb  for 1 month (starting at PND56, 1,000 or
29                  3,000 ppm Pb acetate in drinking water, respective blood Pb levels of 34 or 60 ug/dL)
30                  that showed significant decreases in FSH, ventral prostate weight and serum testosterone
31                  but no change in serum LH (Sokol  et al., 1985). These Pb-exposed adult male rats
32                  (3,000 ppm Pb acetate in drinking water starting at PND56 for 30 days) demonstrated an
33                  impaired pituitary release of LH in response to  challenge of the hypothalamic-pituitary-
34                  adrenal (HPA) axis with the opiate antagonist naloxone, an enhanced release of LH from
35                  the pituitary in response to direct stimulation of the pituitary with luteinizing hormone-
36                  releasing hormone (LHRH), an enhanced response to human chorionic gonadotropin
37                  (hCG) by the testes, increased pituitary LH stores, and increased GnRH mRNA levels in
38                  the hypothalamus  (Klein etal. 1994; Sokol.  1987). Thus, Pb likely interferes with the
39                  male HPA axis, contributing to its reproductive toxicity.

      November 2012                            5-541                  Draft - Do Not Cite or Quote

-------
 1                  In summary, recent epidemiologic studies have demonstrated an inverse effect of Pb on
 2                  pubertal development among boys at low concurrent blood Pb levels. These studies were
 3                  mostly cross-sectional, but associations were observed between Pb levels and delayed
 4                  puberty in a longitudinal study as well (Williams et al. 2010). The larger studies
 5                  controlled for some potential confounders, with a few studies at least considering the
 6                  inclusion of dietary factors, which may be an important confounder, especially in
 7                  populations with high prevalence of malnutrition. Some populations, such as the Russian
 8                  boys cohorts, had potential exposures to dioxins and polychlorinated biphenyls, but these
 9                  were not considered in the analyses. No recent toxicological studies were found that
10                  addressed the effect of Pb on male sexual development and maturation; however, the
11                  2006 Pb AQCD (U.S. EPA. 2006b) supported earlier findings that Pb exposure may
12                  result in delayed onset of male puberty and altered reproductive function later in life in
13                  experimental animals.
                    5.8.1.3      Effects on Postnatal Stature and Body Weight

14                  Findings from previous toxicological studies of rodents and primates have demonstrated
15                  Pb induced impairment of postnatal growth (U.S. EPA. 2006b). Little epidemiologic
16                  evidence was available in the 2006 Pb AQCD on postnatal growth. Several recent
17                  epidemiologic studies examining the association of various biomarkers of Pb exposure
18                  with stature and body weight have been conducted and the evidence reported is mixed.
     November 2012                            5-542                 Draft - Do Not Cite or Quote

-------
Table 5-39
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Afeiche et al.
(2011)




Schelletal.
(2009)
























Summary of recent epidemiologic


Study
Location Study Methodological
and Years Population Details
Mexico City, n=523 boys Longitudinal cohort
Mexico n-477 nirls Usin9 varying
Children born 9 coefficient models
between witn random effects
1994 and
2005


Albany, New n=244 Longitudinal cohort
York study using
1986-1992 multivariate
and 1992-' regression
1998























studies of associations between Pb levels and postnatal growth.


Pb Bio-
marker
Maternal bone Pb
1 month
postpartum




Maternal blood Pb
during second
trimester, third
trimester, and
delivery; Infant
blood Pb at
delivery,
6 months, and
12 months






















Mean Pb (SD)
Patella: 10.4
(11.8)ug/g




Maternal blood Pb
during second
trimester 2. 8
(2.6) ug/dL,
maternal blood Pb
during third
trimester: 2.6
(2.2) ug/dL,
maternal blood Pb
at delivery: 2.8
(2.4) ug/dL
Infant blood Pb at
delivery: 2.3
(2.7) ug/dL, infant
blood Pb at
6 months: 3.2
(3.3) ug/dL, and
infant blood Pb at
1 2 months' 6 3

(4.8) ug/dL











Adjusted Effect
Estimates
Change in weight at 5
years of age (g) per 1 SD
increase in maternal bone
Pb (95% Cl):
Girls: -171 .6 (-275.2, -68.0)
Boys: -35.0 (-132.4, 62.3)


P (p-value) for maternal
second trimester Pb
Length for age:
6 month: 0.1 49 (0.05)
12 month: 0.073(0.38)
Weight for age:
6 month: 0.01 3 (0.89)
12 month: 0.124(0.25)
Weight for length:
6 month: -0.1 58 (0.1 6)
12 month: 0.084(0.45)
Head circumference for
age:
6 month: -0.242 (0.01)
12 month: -0.220 (0.05)
Upper arm circumference
for age:
12 month: -0.1 32 (0.25)

Note: When examining
second trimester maternal
Pb a 3 ug/dL, associations
were observed for 6 mo
weight for age, 6 mo weight
for length, 6 and 12 mo
head circumference, and
12 mo upper arm
circumference for age


Potential Confounders
Adjusted for in Analysis
Cohort, maternal age, calf
circumference, height,
education, number of
pregnancies, brea+st feeding
for 6 months, Ca + treatment,
child's gestational age at
birth, height, repeated
measures of concurrent child
blood Pb
Infant sex, infant birth
weight, infant nutrition,
maternal age, marital status,
employment, race, height,
parity, second trimester
smoking, and education.























November 2012
5-543
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in Study
the text of this Location Study Methodological Pb Bio-
section) and Years Population Details marker Mean Pb (SD)
Lamb et al. Kosovo, n=309 Longitudinal cohort Maternal blood Pb Pristina: 5.60
(2008) Yugoslavia mother child study using linear measured mid- (1.99)ug/dL
1985-1986 Pairs regression pregnancy Mitrovica- 20 56
(7.38) ug/dL
Adjusted Effect
Estimates
Regression coefficients
relating maternal blood Pb:
To Height (95% Cl):
Pristina
1 yr: -0.61 (-2.24, 1 .03)
4 yr: 0.79 (-1.71, 3.29)
6.5 yr: 0.1 5 (-2.43, 2.74)
10yr: -0.09 (-3.69, 3.52)
Mitrovica
1 yr: -0.30 (-2.55, 1 .96)
4 yr: -0.72 (-3.26, 1 .82)
6.5 yr: -1 .87 (-4.38, 0.64)
10yr: -2.87 (-6.21, 0.47)
To BMI (95% Cl):
Pristina
1 yr: 0.61 (-0.28, 1 .50)
4 yr: 0.1 7 (-0.67, 1.00)
6.5 yr: 0.61 (-0.09, 1.30)
10yr: -0.49 (-1.45, 0.46)
Mitrovica
1 yr: 0.23 (-0.84, 1 .30)
4 yr: 0.1 6 (-0.66, 0.98)
6.5 yr: -0.12 (-0.90, 0.66)
10yr: 1.31 (-0.95,3.57)
Potential Confounders
Adjusted for in Analysis
Infant sex, ethnicity, parity,
maternal height or maternal
BMI, maternal education,
gestational age at delivery,
gestational age at blood
sample, HOMES score
November 2012
5-544
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Ignasiak et al.
(2006)
















Mauser et al.
(2008)




Little et al.
(2009)







Study
Location
and Years
South-
western
Poland
1995

(Industrial
area with Cu
smelters and
refineries)










Chapaevsk,
Russia
May 2003 -
May 2005



Dallas, Texas
1 980-1 989
and 2002







Study
Population
school
children 7-15
years
n=463 boys
n= 436 girls













n=489 boys
8-9 yrs old




n=196
(1980s)
n=169
(2002)
2-1 2 yrs old







Methodological Pb Bio-
Details marker Mean Pb (SD)
Cross-sectional Concurrent blood 7.7 (3.5) ug/dL
study using stepwise Pb
multiple regression
analysis














Cross-sectional Concurrent blood 3 (2-5) ug/dL
study using multiple Pb Median (25.75
linear regression percentile)




Cross-sectional Concurrent blood 1980s: 23.6 (1.3
study using Pb SE) ug/dL
MANOVA, 2002' 1 6 (0 2
MANCOVA.and SE) ug/dL
regression models







Adjusted Effect
Estimates
Estimated decrement per
10 ug/dL increase in blood
Pb (p-value)
Weight:
Boys: 2.8 kg (0.002)
Girls: 3.5 kg (0.007)
Height:
Boys: 3.2cm (0.10)
Girls: (0.001)4.0 cm
Trunk length:
Boys: 1 .2 cm (0.02)
Girls: 1.1 cm (>0.01)
Leg length:
Boys: 2.1 cm (0.002)
Girls: 2.9 cm (0.0001)
Arm length:
Boys: 1.8cm (0.0001)
Girls: 1 .9 cm (0.008)
Regression coefficient
(95% Cl)
Height (cm): -1 .439 (-2,25,
-0.63)
Weight (kg): -0.761 (-1 .54,
0.02)
BMI: -0.107 (-0.44, 0.23)
Changes in mean scaled
measure per 10ug/dL Pb
increase (95%CI):
Height (cm): -2.1 (-1.9,
-2.3)
Weight (kg): -1.9 (-1 .7, -2.1)
BMI (kg/m2): -0.5 (-0.4,
-0.7)




Potential Confounders
Adjusted for in Analysis
Age, age2, education

















Birth weight, gestational age,
age at exam




Age, age2, sex and cohort
effect



November 2012
5-545
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Min et al.
(2QQ8b)





Study
Location
and Years
Seoul, South
Korea
Date(s) not
specified





Study
Population
n=62 boys
n= 46 girls
5-13yrs





Methodological Pb Bio-
Details marker Mean Pb (SD)
Cross-sectional Concurrent blood 2.4 (0.7) ug/dL
study using multiple Pb
linear regression





Adjusted Effect
Estimates
Linear model estimate (SE;
P)
Height: -1 .449 (0.639;
p=0.026)
Total arm length: -1 .804
(0.702;p=0.012)
Body weight: -0.646(0.718;
p=0.370)
BMI: -0.006 (0.272;
p=0.982)


Potential
Adjusted


Confounders
for in Analysis
Age, sex, and father's
education






Sanna and Sardinia, n=825 Cross-sectional Pb in hair
Vallascas (2011) Italy children study using multiple
Data 11-14yrsold regression analysis
collected in
1 998, 2002
and 2007









1998: 5.84
(6.56) ug/g
2002: 1 .49
(1.72) ug/g
2007: 0.78
(0.93) ug/g









Height Age, sex
1998: plog Pb=-0.121
(p=0.0021)
2002: p log Pb= -0.1 15
(p=0.0349)
2007: plog Pb= 0.011
(p=0.8665)
Sitting Height
1998:plogPb=-0.117
(p=0.0017)
2002: p log Pb=-0.036
(p=0.5149)
2007: p log Pb=0.028
(p=0.6633)
ELL
                                                                                                  1998:plogPb=-0.103
                                                                                                  (p=0.0209)
                                                                                                  2002: plog Pb=-0.164
                                                                                                  (p=0.0057)
                                                                                                  2007: p log Pb=-0.008
                                                                                                  (p=0.9058)
November 2012
5-546
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Zailina et al.
(2008)










Tomoum et al.
(2010)









Study
Location Study Methodological Pb Bio-
and Years Population Details marker Mean Pb (SD)
Kuala n=269 Cross-sectional Concurrent blood Industrial:
Lumpur, children study using Pb 3.75 ug/dL
Malaysia 6.5-8.5 yrs correlations Urban: 3.56 ug/dL
n=169 urban
n=100
industrial







Cairo, Egypt n=41 boys Cross-sectional Concurrent blood 9.46 (3.08) ug/dL
Jan-Jun and 9irls study using t-test or Pb
2007 10-13 yrs old Mann-Whitney U-
' test









Adjusted Effect Potential Confounders
Estimates Adjusted for in Analysis
Correlation with blood Pb: N/A
Height for age:
Urban: -0.095 (p=0.21 9)
Industrial: -0.037 (p=0.71 6)
Weight for age:
Urban: 0.01 9 (p=0.806)
Industrial: -0.063 (p=0.535)
Weight for height:
Urban: 0.1 36 (p=0.079)
Industrial: -0.069 (p=0.493)
Left arm circumference:
Urban: 0.041 (p=0.595)
Industrial: -0.055 (p=0.587)
Percentage of the median N/A
(SD):
Pb<10ug/dL
Weight:
Boys: 127.56(16.26)
Girls: 114.8(10.8)
Height:
Boys: 98.06 (3.1 9)
Girls: 96.75 (2.91)
                                                                                                       Pb> 10 ug/dL
                                                                                                       Weight:
                                                                                                       Boys: 122.0(16.71)
                                                                                                       Girls: 123.11 (12.52)
                                                                                                       Height:
                                                                                                       Boys: 99.5 (5.04)
                                                                                                       Girls: 100.33 (4.53)
                                                                                                       p-value for all comparisons
                                                                                                       >0.05
November 2012
5-547
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Olivero-Verbel et
al. (2007)


Study
Location
and Years
Cartegena,
Columbia
Jun-Aug
2004


Study
Population
n=189
children 5-9
yrs old


Methodological
Details
Cross-sectional
study using
Spearman
correlations


Pb Bio-
marker Mean Pb (SD)
Concurrent blood 5.49 (0.23) ug/dL
Pb


Adjusted Effect
Estimates
Spearman correlation
coefficient (p-value)
between blood Pb and
body size: -0.224 (0.002)


Potential Confounders
Adjusted for in Analysis

                                                                                                           weight: -0.126(0.087)
                                                                                                           *no significance in partial
                                                                                                           correlation between blood
                                                                                                           Pb and size when
                                                                                                           controlled forage:
                                                                                                           -0.096(0.189)
                  Guiyu, China
                  Chendian,
                  China
                  Jan-Feb
                  2008
n=303
3-7 yrs old
Cross-sectional
study using sample
t-tests
Concurrent blood
Pb
Guiyu: 13.2
(4.0-48.5) ug/dL
Chendian: 8.2
(0-21.3) ug/dL
Median (range)
Mean chest circumference
among girls:
<10ug/dL: 50.31 +1-3.22
cm
> 10 ug/dL: 49.03+/-2.27
cm
(p-value <0.05)


Mean chest circumference
among children >6 years
old
<10 ug/dL: 51.70+/-3.35
cm
> 10 ug/dL: 52.87+/-
2.49 cm
(p-value <0.05)
                                                                                                           Mean head circumference
                                                                                                           among children >6 years
                                                                                                           old
                                                                                                           <10 ug/dL: 48.71 +/-1.66
                                                                                                           cm
                                                                                                           > 10 ug/dL: 50.04+/-1.29
                                                                                                           cm
                                                                                                           (p-value <0.01)
November 2012
                  5-548
                               Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Mahram et al.
(2007)






Study
Location
and Years
Zanjan
province,,
Iran
Date(s) not
specified



Study
Population
n=42 boys
n~ 39 girls

n-45 cases
n-36
controls
7-11 yrs


Methodological Pb Bio-
Details marker
Case-control study Concurrent blood
using t-tests Pb






Mean Pb (SD)
Area with Pb
smelters: 37.0
(24.7) ug/dL

Area without Pb
smelters: 15.6
(13.4)ug/dL


Adjusted Effect
Estimates
Comparison of control and
study groups Height,
standardized for age: p-
value 0.52
Weight, standardized for
age: p-value 0.8



Potential Confounders
Adjusted for in Analysis
N/A





*Estimated Lower Limb Length
November 2012
5-549
Draft - Do Not Cite or Quote

-------
 1                   Results from recent epidemiologic studies of postnatal growth are summarized in Table
 2                   5-39. Longitudinal epidemiologic studies have had inconsistent findings regarding the
 3                   association between Pb levels and post-natal growth. Afeiche et al. (2011) conducted a
 4                   longitudinal study of children in Mexico City, born between 1994 and 2005. Maternal
 5                   bone Pb during pregnancy was associated with a statistically significant decrease in
 6                   weight at age 5 years in girls but not in boys. The findings were robust to additional
 7                   adjustment for child's concurrent blood Pb level. A study in New York reported an
 8                   inverse association between maternal blood Pb during the second trimester of pregnancy
 9                   and various measures of growth, especially among those mothers with blood Pb levels of
10                   at least 3 (ig/dL (Schell et al., 2009). These associations did not persist for those with
11                   maternal blood Pb levels less than 3 (ig/dL. Among infants, 6 month blood Pb levels were
12                   not associated with measures of growth at 12 months. In comparisons of changes in blood
13                   Pb levels over time, high maternal blood Pb combined with low 12 month blood Pb
14                   among infants (indicating a decrease in blood Pb over time) resulted in the greatest
15                   growth, even compared to those with both low or both high maternal and infant blood Pb
16                   measures. In a prospective  study of 309 mother-child pairs  from Yugoslavia, the
17                   relationship between maternal blood Pb measured mid-pregnancy and attained height in
18                   children was investigated in those living in a highly exposed town with a smelter and
19                   battery plant and those living in a relatively lower exposed  town (Lamb et al.. 2008). In
20                   multivariate adjusted regression models, neither attained height (at birth,  1, 4, 6.6, or
21                   10 years age) nor rate of height change per month (at birth-1 year, 1-4 years, 4-6.5 years,
22                   6.5-10 years  age) was associated in a consistent direction with maternal pregnancy blood
23                   Pb levels in either the industrial or less exposed town. Weight was also not associated
24                   with maternal blood Pb in this study.

25                   Multiple cross-sectional studies reported an association between Pb levels and  impaired
26                   growth. Ignasiak et al.  (2006) studied school children aged  7-15 years living close to Cu
27                   smelters and refineries in Poland to assess the impact of Pb exposure on their growth
28                   status. There was a statistically significant linear relationship between concurrent blood
29                   Pb and reduced weight, height, trunk, leg and arm lengths.  This decrease in height was
30                   more influenced by decreases in leg length than trunk length. These results also indicated
31                   that there was attenuation in osteoblast activity associated with higher blood Pb levels,
32                   consistent with animal  toxicological studies (Long etal. 1990). Hauser et al. (2008)
33                   investigated the relationship between blood Pb and height in boys living in Chapaevsk,
34                   Russia, an area contaminated with multiple pollutants including dioxins and metals. In a
35                   multivariate adjusted regression analysis, height significantly decreased with increasing
36                   blood Pb. Statistically nonsignificant decreases in  weight and BMI were also observed.
37                   The association of blood Pb with height, weight, and BMI was examined among two
38                   cohorts of children living near Pb smelters in Texas (Little  et al.. 2009). The first cohort
39                   included children 2-12 years old in 1980 and the second cohort included children of the

      November 2012                             5-550                  Draft - Do Not Cite or Quote

-------
 1                   same age in 2002 when blood Pb levels were substantially lower. Decreases in height,
 2                   weight, and BMI with increasing blood Pb levels were observed among children in both
 3                   cohorts and increases in height and weight were observed comparing children from the
 4                   2002 cohort to those from the 1980 cohort. In a study with Korean children, Min et al.
 5                   (2008b) observed that height and total arm length decreased significantly with increasing
 6                   blood Pb in multivariate adjusted regression models. A statistically nonsignificant
 7                   decrease in body weight was observed with increasing blood Pb while no effect on BMI
 8                   was reported. In a study of children in Sardinia Italy, Sanna and Vallascas (2011)
 9                   measured Pb in hair at three points in time (1997, 2002, and 2007) and reported cross-
10                   sectional results from regression analyses for each of these time periods. Pb in hair
11                   decreased over time and significant associations of Pb in hair with height were observed
12                   only in earlier time periods when hair Pb levels were relatively high. However, Pb in hair
13                   samples is not a well-characterized biomarker (see Chapter_4 and Section 4.3.4.2).

14                   Contrary to the results summarized above, several cross-sectional studies do not observe
15                   associations between blood Pb levels and impaired growth. In a study with a similar
16                   design, Zailina et al. (2008) studied the relationship of blood Pb and height in 7 year-old
17                   Malaysian school children comparing those attending two schools in an urban setting to
18                   those attending a school near an industrial area. After adjustment for age no statistically
19                   significant associations between concurrent blood Pb and physical development were
20                   observed. Tomoum et al. (2010) investigated the association between blood Pb and height
21                   in pubertal children in Cairo, Egypt. Neither boys nor girls with concurrent blood Pb
22                   levels >10 (ig/dL differed significantly in height or weight when compared to those with
23                   blood Pb <10 (ig/dL. In a simple correlation analysis of children aged 5-9 years in
24                   Colombia, Olivero-Verbel et al. (2007) reported that concurrent blood Pb levels were
25                   negatively associated with body size (r = -0.224, p <0.002). However, when a partial
26                   correlation analysis was performed controlling for age, the association between blood Pb
27                   and body size was no longer statistically significant. In a study of school children in
28                   China, chest and head circumference were found to differ between high (>10(ig/dL) and
29                   low concurrent blood Pb level groups; however, the direction of the difference was not
30                   consistent (Liu et al.. 20 lib). Among girls, in comparison of those with high and low
31                   blood Pb levels, a reduction in head circumference was observed.  Among children greater
32                   than 6 years of age, those with higher blood Pb levels were  reported to have greater head
33                   and chest circumferences. In a study of children aged 7-11 years and living in an area of
34                   Iran with or without Pb smelters, age-standardized weight and height did not vary by
35                   study area (Mahram et al.. 2007).

36                   Evidence from previous toxicological studies has shown an association between
37                   gestational Pb exposure and impaired postnatal growth (U.S. EPA. 2006b). Recent
38                   toxicological studies report significant changes in postnatal or adult body weight after Pb
      November 2012                             5-551                  Draft - Do Not Cite or Quote

-------
 1                  exposure during different developmental windows. Masso-Gonzalez and Antonio-Garcia
 2                  (2009) found Pb-induced decreased body weights at weaning (PND21) in rat pups from
 3                  dams exposed to Pb during pregnancy and lactation (drinking water, 300 mg/L). Blood
 4                  Pb level in the control group was 1.43 ug/dL, in the Pb group it was 22.8 ug/dL. Dong et
 5                  al. (2009) reported decreased body weight in adult Kunming mice after exposure to
 6                  6,000 ppm Pb acetate in drinking water for 8 weeks. In contrast, Leasure et al. (2008)
 7                  reported a statistically significant inverse relationship between Pb exposure and body
 8                  weight for male mice exposed to lower  (27 ppm), moderate (55 ppm) and higher levels
 9                  (109 ppm) levels of Pb during gestation and lactation (dam drinking water, 2 weeks
10                  before mating, through gestation and to PND10) with those exposed to the lowest dose
11                  having the highest adult body weight among the overweight Pb-exposed animals. Male
12                  mice exposed to the lower and higher Pb concentrations during gestation were 26% and
13                  13% heavier than were controls at 1 year of age, respectively. In this study, dams were
14                  administered 27 ppm (low), 55 ppm (moderate), and 109 ppm (high) Pb in drinking water
15                  beginning which resulted in respective blood Pb levels from 10 ug/dL or less in the
16                  low-exposure offspring to 42 ug/dL in the high-exposure offspring at PND10. Leasure et
17                  al. (2008) also exposed a separate group of mice to Pb only during the postnatal period
18                  (PNDO-PND21, lactation only exposure) and mice exposed to the same aforementioned
19                  low or high dose of Pb did not exhibit a difference in body weight when compared to
20                  control offspring. Wang et al. (2009e) observed a statistically significant decrease in fetal
21                  body weight and body length of Wistar rats at GD20 after maternal exposure to 250 ppm
22                  Pb acetate during gestation days 1-10, 11-20, or 1-20. Also, associations were reported
23                  between elevated maternal blood Pb levels (0.6, 1.3, or 1.74 (iM, respectively or -12.4,
24                  26.9, or 36.0 (ig/dL, respectively) compared to control (0.04 (iM or -0.83 (ig/dL)
25                  decreased pup body length, and placental weight in Wistar rats at GD20. The greatest
26                  decrease in fetal body weight and length was observed in the group exposed to Pb during
27                  gestation days 1-20 followed by the group exposed to Pb during gestation days 11-20.
28                  Teijon et al.  (2006) observed reductions in birthweight of litters administered 200 ppm or
29                  400 ppm Pb  acetate in drinking  water (Wistar rats, Pb to dams from GD1 through
30                  lactation to 1 and 3 months postweaning to pups), but found that this effect did not persist
31                  in the postnatal growth of the rats.

32                  Notably, previous toxicological studies  observed reductions in postnatal weight as well as
33                  birth weight after exposure to Pb, albeit often at higher concentrations of Pb exposure.
34                  Ronis et al. (2001; 1998a: 1998b: 1996) have published a series of papers exposing rats to
35                  Pb over different developmental windows, showing associations between Pb exposure
36                  and deficits in growth. Sprague-Dawley rats with lifetime Pb exposure to 6,000 ppm
37                  Pb acetate in drinking water (gestational-termination of experiment Pb exposure,
38                  maximum blood Pb of 316 (ig/dL in males and 264 (ig/dL in females) had sex-
39                  independent pre-pubertal growth suppression, male-specific suppression of pubertal

      November 2012                           5-552                 Draft - Do Not Cite or Quote

-------
 1                  growth and loss of growth effects postnatally but still maintained an overall decreased
 2                  body size out to PND60 due to earlier deficits. In a follow up study using the same
 3                  exposure duration with a dose of 4,500 ppm Pb acetate (resulting in blood Pb of
 4                  263 (ig/dL at PND85) yielded the same results (Ronis etal.  1996) with mechanistic
 5                  insight showing decrements in insulin-like growth factor 1 (IGF1) accompanying the
 6                  decreases in growth rates.

 7                  In summary, the body of toxicological literature on postnatal growth with Pb exposure
 8                  indicates that Pb exposure can induce decrements in both height/body length and BW that
 9                  may be persistent and differ by sex. However, findings from epidemiologic studies of
10                  postnatal growth are not consistent. Many of these studies were limited by their cross-
11                  sectional design. A few studies used longitudinal cohorts and controlled for multiple
12                  potential confounders, such as parity, but the results of these studies are inconsistent.
13                  Animal toxicology studies give insight to mechanistic changes that may contribute to this
14                  Pb-induced decrement and to the windows of exposure that may contribute greatest to
15                  these decrements.
      5.8.2       Toxicological Studies of Other Developmental Effects
                    5.8.2.1      Developmental Effects on Blood and Liver

16                  The 1986 and 2006 Pb AQCDs [(U.S. EPA. 1986b) and (U.S. EPA. 2006b)1 reported
17                  studies that suggest Pb may alter hematopoietic and hepatic function during development.
18                  Some recent studies provide evidence that support these findings; however recent results
19                  are not consistent among the studies.

20                  Masso et al. (2007) reported a decrease in liver weights of pups born to dams that
21                  consumed 300 mg/L Pb  in drinking water during gestation and lactation. They also
22                  reported an increase in the number of erythrocytes; however the erythrocyte size was
23                  diminished by 62%. Pb produced microcytic anemia as evidenced by decreased
24                  hemoglobin content and hematocrit values without changes in mean corpuscular
25                  hemoglobin (MCH) concentration. Alkaline phosphatase (ALP) activity, CAT activity, or
26                  thiobarbituric acid reactive substances (TEARS) production did not change in pups at
27                  postnatal 0, but increased statistically significantly by PND21 indicating reactive oxygen
28                  generation. No change in acid phosphatase (ACP) activity was observed in the livers of
29                  pups at PNDO or PND21.
      November 2012                            5-553                 Draft - Do Not Cite or Quote

-------
 1                  Masso-Gonzalez and Antonia-Garcia (2009) reported normochromic and microcytic
 2                  anemia and a significant decrease in hematocrit values and blood 5-aminolevulinic acid
 3                  dehydratase (ALAD) activity (90% reduction) in pups from dams administered 300 mg/L
 4                  Pb acetate in drinking water during gestation. The authors also reported that erythrocyte
 5                  osmotic fragility was four times greater in Pb-exposed pups than in control pups.
 6                  Masso-Gonzalez and Antonia-Garcia (2009) reported increases in TEARS and CAT
 7                  activity in the liver after Pb exposure. Intoxication with Pb also resulted in decreased
 8                  liver protein concentrations and Mn-dependent SOD activity. Abnormalities in liver
 9                  function were further exemplified by increases in liver concentrations of ALP and ACP.

10                  Teijon et al. (2006) observed that gestational exposure to Pb caused a decrease in
11                  erythrocytes, hemoglobin, and MCH at weaning; however, by 1 and 3 months
12                  postweaning, these parameters had returned to normal values. The authors observed a
13                  slight increase in serum ALP, alanine aminotransferase  (ALT), and aspartate
14                  aminotransferase (AST) levels after Pb exposure in the absence of liver histological
15                  changes.

16                  Pb-induced effects on SOD activity in the liver of fetuses after Pb intoxication was
17                  supported by a study by Uzbekov et al. (2007). The authors reported an initial increase in
18                  SOD activity in livers of pups exposed to 0.3 mg/L and 3.0 mg/L Pb nitrate in drinking
19                  water during gestation for 1 month (mean daily consumption 27 (ig/kg). In contrast, long-
20                  term exposure (5 months) to the same concentrations of Pb nitrate concentration during
21                  gestation resulted in decreased hepatic SOD activity.

22                  Effects on hepatic Phase I and Phase II enzymes after early developmental exposure of
23                  offspring to Pb during gestation and lactation was evaluated by Pillai et al. (2009). In the
24                  study, pregnant Charles Foster rats were administered 0.05 mg/kg body weight Pb
25                  subcutaneously throughout gestation until PND21. Pups were evaluated on PND56.
26                  Results of the study show that Phase I xenobiotic-metabolizing enzymes (NADPH- and
27                  NADH cytochrome c reductase) and Phase II xenobiotic- and steroid-metabolizing
28                  enzymes (5-glutamyl transpeptidase, UDPGT, glutathione-s-transferase, and  17(3-
29                  hydroxysteroid oxidoreductase) were reduced in both male and female pups by PND56.
30                  Only inhibition in glutathione-s-transferase and 17(3-hydroxysteroid oxidoreductase
31                  activities demonstrated a sex-specific pattern (glutathione-s-transferase inhibition in
32                  males; 17|3-hydroxysteroid oxidoreductase inhibition greater in females). Observed
33                  Pb-induced histological changes included massive fatty degeneration in hepatocytes,
34                  large vacuoles in cytoplasm, appearance of pyknotic nuclei, and infiltration of
35                  lymphocytes in the liver. Activities of antioxidant enzymes (SOD, CAT, glutathione
36                  peroxidase, and glutathione reductase) were also reduced after Pb intoxication.
37                  Alterations in biochemical parameters included decreased DNA, RNA, and cholesterol
      November 2012                            5-554                  Draft - Do Not Cite or Quote

-------
 1                  content, although it was not clear whether these changes were related to genetic
 2                  expression of xenobiotic-metabolizing enzymes or changes in steroid hormone
 3                  homeostasis.
                    5.8.2.2      Developmental Effects on Skin

 4                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported a study that demonstrated Pb-induced
 5                  abnormalities in skin development. No current studies were identified that addressed
 6                  Pb-induced skin alterations.
                    5.8.2.3      Developmental Effects on the Retina

 7                  The 2006 Pb AQCD (U.S. EPA. 2006b) concluded that Pb exposure during early
 8                  postnatal development (resulting in blood Pb levels ~20 (ig/dL) impaired retinal
 9                  development in female Long-Evans hooded rats. A more recent study (Fox et al., 2008)
10                  exposed female Long-Evans hooded rats to low (27 ppm), moderate (55 ppm), and high
11                  (109 ppm) levels of Pb acetate in drinking water beginning 2 weeks before mating,
12                  throughout gestation, and until PND10. Blood Pb levels measured in these pups on
13                  postnatal days 0-10 were 10-12 (ig/dL (low), 21-24 (ig/dL (moderate), and 40-46 (ig/dL
14                  (high). Results of the study demonstrated supernormal persistent rod photoreceptor-
15                  mediated (scotopic) electroretinograms (ERGs) [(Fox et al.. 2008). and Table 5-131 in
16                  adult rats similar to ERG findings in male and female children in association with
17                  maternal first trimester blood Pb levels 10.5-32 (ig/dL [(Rothenberg et al., 2002b), and
18                  Table 5-131. In rats, low- and moderate-levels of Pb increased neurogenesis of rod
19                  photoreceptors and rod bipolar cells without affecting Miiller glial cells and statistically
20                  significantly increased the number of rods in central and peripheral retina. High-level Pb
21                  exposure (109 ppm) statistically significantly decreased the number of rods in central and
22                  peripheral  retina. Pb-exposure induced concentration-dependent decreases in adult rat
23                  retinal dopamine synthesis and utilization/release.
                    5.8.2.4      Developmental Effects on Teeth

24                  Pb has been associated with multiple health effects including dental caries, however,
25                  there is very limited information available on the temporal and spatial incorporation of Pb
26                  in dental tissue (Arora et al. 2005). Arora et al. (2005) demonstrated that Wistar rat pups
27                  exposed to Pb during gestation and lactation (40 mg/L of Pb nitrate in drinking water of
28                  pregnant dams) had higher concentrations of Pb on the surface of enamel and in the
      November 2012                             5-555                 Draft - Do Not Cite or Quote

-------
 1                  dentine immediately adjacent to the pulp. The authors concluded that additional research
 2                  is needed on the intracellular uptake of Pb during tooth development to fully understand
 3                  the spatial distribution of Pb in teeth.
      5.8.3        Effects on Birth Outcomes

 4                  The 2006 Pb AQCD reported on multiple studies of adverse birth outcomes such as, fetal
 5                  mortality, birth defects, preterm birth, and low birth weight/fetal growth (U.S. EPA.
 6                  2006b). The toxicological studies reviewed in the 2006 Pb AQCD concluded that Pb
 7                  exposure can increase fetal mortality and produce sublethal effects, smaller litters, and
 8                  fewer implantation sites. Epidemiologic studies using occupational histories reported the
 9                  possibility of small associations between increased Pb exposure and birth defects, and
10                  toxicological studies demonstrated associations between exposure to high doses of Pb
11                  and increased incidences of teratogenic effects in experimental animals. Epidemiologic
12                  studies on preterm birth and low birth weight/fetal growth included in the
13                  2006 Pb AQCD reported inconsistent findings. Evidence from previous toxicological
14                  studies has shown an association between gestational Pb exposure and reduced birth
15                  weight and decreased litter size or number of pups. Continued research on adverse birth
16                  outcomes is described in the sections that follow.
                    5.8.3.1      Infant Mortality and Embryogenesis

17                  No recent epidemiologic or toxicological studies have reported on the relationship
18                  between Pb levels and infant mortality. The 2006 Pb AQCD (U.S. EPA. 2006b)
19                  concluded that Pb exposure can increase fetal mortality and produce sublethal effects
20                  (disrupt growth and development) in offspring of Pb exposed dams at concentrations that
21                  do not result in clinical toxicity to the dams by disrupting implantation and pregnancy,
22                  particularly at the blastocyst stage of development. In rodent studies gestational exposure
23                  to Pb (blood Pb levels 32 to >70 ug/dL) resulted in smaller litters and fewer implantation
24                  sites and in non-human primates pre- and perinatal mortality was reported in squirrel
25                  monkeys exposed to Pb (mean dam blood Pb level of 54  ug/dL) in the last two-thirds of
26                  gestation (U.S. EPA. 2006b). There is substantial evidence to show that there is no
27                  apparent maternal-fetal barrier to Pb and it can easily cross the placenta and accumulate
28                  in fetal tissue during gestation (Pillai et al., 2009; Wang et al., 2009e; Uzbekov et al.,
29                  2007).
      November 2012                            5-556                  Draft - Do Not Cite or Quote

-------
                    5.8.3.2      Birth Defects

 1                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported the possibility of small associations
 2                  between high Pb exposure and birth defects, but many of the epidemiologic studies used
 3                  occupational histories instead of actual measures of blood Pb levels. Among the studies
 4                  included in the 2006 Pb AQCD, a couple reported associations between parental
 5                  exposure to Pb and neural tube defects (Irgens et al.. 1998; Bound et al..  1997). Recent
 6                  studies also examined indicators of Pb exposure and neural tube defects (Table 5-40). No
 7                  other recent epidemiologic studies of Pb exposure and birth defects were identified in the
 8                  literature. No recent toxicological studies were found that  investigated Pb-induced
 9                  changes in morphology, teratology effects, or skeletal malformations of developing
10                  fetuses as a result of maternal Pb exposure; however,  in the 2006 Pb AQCD toxicological
11                  studies demonstrated associations between exposure to high doses of Pb and increased
12                  incidences of teratogenic effect in experimental animals.
     November 2012                            5-557                 Draft - Do Not Cite or Quote

-------
Table 5-40 Summary of recent epidemiologic studies of associations between

Reference
(Presented
in order of
appearance
in the text)
Zeyrek et
al. (2009)









Brender et
al. (2006)



Huang et
al. (201 1b)










Pb levels and

Study Study
Location Population
Turkey Infants with
NS gestational
age of at
least
20 weeks
NNTD=74
NControls=70





Texas Infants of
1995-2000 Mexican-
American
women
NNTD=184
Ncontrols=225

China Live and
2002-2004 sti" birtns of
women
living in the
study area
(villages in
the Lvliang
region of
Shanxi
province)
N=112
villages
neural tube

Methodological
Details
Case-control
study using
Student's t-test
and Mann-
Whitney U-test







Case-control
study using
logistic
regression


Ecologic











defects.
Pb Biomarkers
and Exposure
Measurement
Maternal and
umbilical cord
blood Pb taken
0.5h afterbirth








Maternal blood
Pb
taken 5-6 weeks
post-partum


2 soil samples
from each
village










Mean Pb
(SD)
in ug/dL
Cases:
Maternal:
15.5(15.0)
Umbilical
cord' 182
(17.8)
Controls:
Maternal:
12.5(12.7)
Umbilical
cord: 16.5
(16.1)
Cases: 2.4
(1.9)
Controls: 2.5
(1.6)


56.14ug/g
(11.43ug/g)











Adjusted
Effect
Estimates
P-values for
differences of
Student's t-test
or Mann-
Whitney U test
(dependent on
distribution)
were 0.35 for
maternal blood
Pb and 0.63 for
umbilical cord
blood Pb

OR (95% Cl):
Blood
Pb<6.0 ug/dL:
1 .0 (Ref)
Blood
Pb> 6.0 ug/dL:
1.5(0.6, 4.3)
N/A












Potential
Confounders
Adjusted for
in Analysis
N/A










Inclusion of
breast feeding
in the model
changed the
OR (95% Cl)
to 3 8 (0 8
19.5)













 1
 2
 o
 J
 4
 5
 6
 7
 8
 9
10
11
12
13
Among the recent epidemiologic studies (described in Table 5-40). a study of women in
Turkey detected no difference between the blood Pb of mothers or the umbilical cord
blood Pb of the newborns for healthy infants compared with infants with neural tube
defects (cases of spina bifida occulta were excluded, but other forms of spina bifida were
included) (Zevrek et al.. 2009). Brender et al. (2006) performed a study of Mexican-
American women living in Texas. Measurements were taken 5-6 weeks postpartum,
which is a limitation of this study because the blood Pb levels may be different from
those during the developmental period of gestation. The  OR comparing women with at
least 6 (ig/dL blood Pb to those with less than 6 (ig/dL blood Pb was 1.5 (95% Cl: 0.6,
4.3). This increased after adjusting for breast feeding, although this variable was not a
confounder because it cannot be associated with neural tube defects. For these women,
neither occupational exposure to Pb nor proximity of residence to a facility with Pb air
emissions at the time of conception was associated with increased odds of neural tube
     November 2012
                           5-558
Draft - Do Not Cite or Quote

-------
 1                   defects. A study with an ecologic design was performed in China and did not use
 2                   individual-level biomarkers to determine Pb levels (Huang et al.. 201 Ib). A positive
 3                   association between Pb levels in soil samples and neural tube defects was reported.
 4                   Exposure to multiple other trace elements also demonstrated a positive association but no
 5                   control for co-exposures was included in the models for Pb.

 6                   In summary, previous studies included in the 2006 Pb AQCD observed associations
 7                   between Pb and neural tube defects but were limited due to the lack of biologically
 8                   measured Pb [Pb was measured in drinking water (Bound etal.. 1997)  and estimated
 9                   from occupational reports drgens et al.. 1998)1. A recent ecologic study reported an
10                   association between Pb in the soil and neural tube defects but was also limited by its lack
11                   of biological samples, as well as a lack of individual-level data and the prevalence of
12                   several other metals (Huang et al.. 201 Ib). Other recent epidemiologic studies of
13                   maternal blood Pb levels and neural tube defects observed no statistically significant
14                   associations (Zeyrek et al.. 2009; Brender et al.. 2006).  These studies also have
15                   limitations, including the timing of Pb measurements and  lack of control for potential
16                   confounders.
                     5.8.3.3      Preterm Birth

17                   Epidemiologic studies on preterm birth included in the 2006 Pb AQCD (U.S. EPA.
18                   2006b) reported inconsistent findings regarding the relationship between Pb and
19                   gestational age. Recent studies examined this potential association and again mixed
20                   results were reported (Table 5-41). Of these studies, the ones that categorized births as
21                   preterm or term all defined preterm birth as less than 37 weeks of gestation. One
22                   limitation to note for these  studies is that if Pb affects spontaneous abortion and length of
23                   gestation via a similar pathway, then the studies that only collect data at delivery and not
24                   at earlier stages of pregnancy would be biased toward the null.
      November 2012                             5-559                  Draft - Do Not Cite or Quote

-------
Table 5-41 Summary of recent epidemiologic studies of associations between Pb levels and preterm birth.
Reference
Presented in
order of
appearance
in the text)
Jelliffe-
Pawlowski et
al. (2006)













Vigeh et al.











Cantonwine
et al. (2010a)





Study
Location Outcome Study Population
California Preterm birth Singleton births to
1995-2002 (<37 . . . . . non-smoking mothers
completed week) with blood Pb
measures during
pregnancy from
either the California
Childhood Lead
Poisoning Prevention

Dl Ctl 1UI 1 Ul LI 1C
California
Occupational Lead
Poisoning Prevention
Program
Npreterm birth=30
Nterm birth=232


Tehran, Preterm birth Singleton births from
Iran (20-37 week) non-smoking, non-
2006 obese mothers aged
16-35 and referred
for prenatal care
during the
8th-12th week of
gestation

Npreterm birth~44
Nterm birth=304



Mexico Preterm birth Births to mothers with
City (<37wk), at least 1 blood Pb
1997-1999 Gestational age measurement during
pregnancy and no
chronic diseases
requiring medication


Methodological
Details
Longitudinal
cohort study using
logistic regression













Longitudinal
cohort study using
logistic regression










Longitudinal
cohort study using
linear regression




Pb Biomarkers or Mean Pb
Exposure (SD)
Measurement in ug/dL
Maximum maternal blood a 10 ug/dL:
Pb during pregnancy 30.9%













Maternal blood Pb at 3.8 (2.0)
8-12 weeks gestation











Maternal blood Pb during Blood Pb
pregnancy visit gt
<20wks
pregnant 7.2
(5 2)
\<->.*-i
Visit at


Adjusted
Effect
Estimates
Odd Ratios:
< 5 ug/dL: 1 .00
(Ref)
V *<-'y
6-9 ug/dL: 0.8
(0.1 , 6.4)
10-19 ug/dL: 1.1
(0.2, 5.2)

20-39 ug/dL: 4.5
(1 .8, 10.9)
> 40 ug/dL: 4.7
(1.1, 19.9)
<10 ug/dL: 1.00
(Ref)
> 10ug/dL: 3.2
(1.2,7.4)
Mean blood Pb
(SD):
Preterm birth:
4.52 (1 .63)

Term birth:
3.72 (2.03)
p-value for
difference: <0.05
OR (95% Cl)
1.41 (1.08, 1.84)
(unit not given,
assume per
1 ug/dL)
Linear regression
P (95% Cl) per
SD increase in
centered log-Pb
concentration
Blood Pb

Potential
Confounders
Adjusted for in
Analysis
In the 10ug/dL
model: race,
insurance, maternal
age, parity, infant
sex, low birthweight












Age, infant sex,
education, passive
smoking exposure,
pregnancy weight
gain, parity,
hematocrit, type of
delivery






Infant sex, maternal
age, maternal
education, history
of adverse birth
outcomes, cigarette
smoking, parity
November 2012
5-560
Draft - Do Not Cite or Quote

-------
Reference
Presented in
order of
appearance Study
in the text) Location Outcome






































Pb Biomarkers or Mean Pb
Methodological Exposure (SD)
Study Population Details Measurement in ug/dL
20-28 weeks
Npre«ermbirth=22 pregnant 6.3
N,ermbirth=213 ^ g(
>28 weeks
pregnant 6.8
(4.5)

Plasma Pb
Visit at
<20wks
pregnant
0.17(0.16)
Visit at
20-28 weeks
pregnant
0.13(0.10)
Visit at
>28 weeks
pregnant
0.16(0.26)





















Potential
Adjusted 9°Pfc?J"1clers-
Effect Adjusted for in
Estimates Analysis
Visit at <20wks:
-2.76 (-5.21,
-0.31)
Visit at
20-28 weeks:
-1.77 (-3.39,
-0.15)
Visit at
>28 weeks: -0.47
(-1 .78, 0.84)

Average: -1.49
(-3.63, 0.64)

Plasma Pb
Visit at <20wks:
-2.38 (-4.97,
0.21)
Visit at
20-28 weeks:
-1.34 (-2.98,
0.29)
Visit at
>28 weeks: -1 .28
(-2.63, 0.06)
Average: -0.28
(-2.81, 2.25)
Plasma-to-blood
Pb ratio
Visit at <20wks:
-3.23 (-6.01 ,
-0.44)
Visit at
20-28 weeks:
-1.41 (-3.10,
0.29)
Visit at
>28 weeks: -1 .30
(-2.67, 0.07)
Average: -1 .27
(-3.89, 1 .35)
Cord blood Pb
-0.68 (-2.37,
                                                                                                        1.00)
November 2012
5-561
Draft - Do Not Cite or Quote

-------
Reference
Presented in
order of
appearance
in the text)
Zhuetal.
(2010)






Chen et al.
(2006a)

















Study
Location Outcome Study Population
New York Preterm birth Singleton births to
2003-2005 (<37 , , . .. m°'h.f.rs ^1 5-48
completed week) with blood Pb
measures before or
on the date of
delivery and blood Pb
measuring <10 ug/dL

Npreterm birth=351 9
Nterm birth=39,769
Taiwan Preterm birth Infants born to at
1993-1997 (<37 week) least one parent who
was part of the
Program to Reduce
Exposure by
Surveillance System
- Blood Lead Levels
cohort that monitored
workers
occupationally
exposed to Pb
Npreterm birth=74
N,ermbirth=1537

*738 births had
maternal Pb
information and 967
had paternal Pb
information


Methodological
Details
Retrospective
cohort study using
logistic regression
with fractional
polynomials




Occupational
cohort study using
regression models

















Pb Biomarkers or
Exposure
Measurement
Maternal blood Pb







Maternal blood Pb during
pregnancy (or if that
wasn't available, the 1
year prior to fertilization)
and/or paternal blood Pb
during spermatogenesis
(the 64 days before
fertilization, or if that
wasn't available, the 1
year prior to
spermatogenesis)










Mean Pb
(SD)
in ug/dL
2.1







Maternal
blood Pb
10.1 (10.4)


Paternal
blood Pb
12.9(13.8)













Adjusted
Effect
Estimates
Odd Ratios:
<1.0ug/dL: 1.00
(Ref)
V *cly
1.1-2.0ug/dL:
1.03 (0.93, 1.13)
2.1-3.0ug/dL:
1.01 (0.92, 1.10)
3.1-9.9ug/dL:
1 .04 (0.89, 1 .22)
Risk Ratios

Maternal blood
Pb
<10 ug/dL: 1.00
10-19 ug/dL:
1 .97 (0.92, 3.86)
>20ug/dL: 1.86
(0.68, 4.28)
Paternal blood
Pb
<10 ug/dL: 1.00
10-19 ug/dL:
1.17 (0.53, 2.32)
> 20 ug/dL: 0.55
(0.19, 1.28)


Potential
Confounders
Adjusted for in
Analysis
Timing of Pb test,
maternal age, race,
Hispanic ethnicity,
smoking status,
drug abuse, marital
status, special
financial program
participation, parity,
infant sex
Parental age,
parental education,
parity, infant sex















November 2012
5-562
Draft - Do Not Cite or Quote

-------
Reference
Presented in
order of
appearance
in the text)
Patel and
Prabhu
(2009)












Jones et al.
(2010)







Wells et al.
(201 la)









Study
Location Outcome Study Population
Nagpur, Gestational age Consecutive births at
India the study hospital
NS
N=205 (mean
gestational age 39
+1-2 weeks)











Tennessee Gestational Age: Singleton births
2006 preterm (<37wk) > 27 week gestation
, term from mothers aged
(37-40 week), 1 6-45 living in the
post-term Shelby County area
(>40 week) for at least 5 mo
during pregnancy
Npreterm birth=10
Nterm birth=81
Nposttermbirth = 11
Baltimore, Gestational age Singleton births from
MD the Baltimore
2004-2005 Tracking Health
Related to
Environmental
Exposures (THREE)
study
Nnraf-arm hir+h=39
preterm birth ^^
Nterm birth=261


Pb Biomarkers or
Methodological Exposure
Details Measurement
Cross-sectional Umbilical cord blood Pb
study using linear
regression












Cross-sectional Umbilical cord blood Pb
study comparing
across geometric
means (test not
specified)





Cross-sectional Umbilical cord Pb
study using
multivariable linear
regression








Mean Pb
(SD)
in ug/dL
Umbilical
cord blood
Pb: 4.7
(12.1)











2.4 (4.3)
Geometric
mean: 1 .3






0.84 (95%:
Cl 0.72,
0.96)

a 5 ug/dL:
0.7%





Adjusted
Effect
Estimates
>5 ug/dL: mean
gestational age
38 weeks
< 5 ug/dL: mean
gestational age
39 weeks
Linear
regression:
gestational age
decreased
1 week with
every 1 ug/dL
increase in
umbilical cord
blood Pb (exact
values and 95%
Cl: not given)
Geometric Mean:
Preterm birth: 1.4
Term birth: 1.2
Post-term birth:
1.3
p-value for
difference: >0.10


Ratio for Pb
concentration per
10 days of
gestation: 0.99
(0.93, 1.06)






Potential
Confounders
Adjusted for in
Analysis
Not specified














None








Maternal age, race,
insurance, pre-
pregnancy BMI,
smoking status,
gestational age,
birthweight,
average year of
neighborhood
home construction

November 2012
5-563
Draft - Do Not Cite or Quote

-------
Reference
Presented in
order of
appearance
in the text)
Study
Location
Outcome
Study Population
Methodological
Details
Pb Biomarkers or
Exposure
Measurement
Mean Pb
(SD)
in ug/dL
Adjusted
Effect
Estimates
                                                                                                                Potential
                                                                                                                Confounders
                                                                                                                Adjusted for in
                                                                                                                Analysis
Berkowitz et
al. (2006)
Shoshone
County,
Idaho
1970-1981
Preterm birth
(<37wk)
Singleton births
with 28-45 week
gestation


Npreterm birth=7843
N,ermbirth= 162,035
Cohort study
using logistic
regression
Three time periods of two
locations (unexposed
and exposed/near
smelter): pre-fire, "high-
exposure period" (when
a fire happened at the
smelter and resulted in
damages leading to high
air Pb concentrations for
6 mo), and "post-fire"
During the
time of the
fire,
estimates of
Pbin
ambient air
were as high
as 30 ug/m
OR (90% Cl)
(unexposed
location is
referent group):
Pre-fire 0.93
(0.67, 1.28)
High exposure
0.68(0.34, 1.35)
Post-fire 1.17
(0.95, 1.45)
Maternal age, infant
sex, first birth,
previous
miscarriage or
abortion
Orun et al.
(2011)
Turkey
NS
Preterm birth
(<37 week)
Births to mothers not
occupationally
exposed to toxic
metals and living in a
suburban but non-
industrial area
Cohort study
using Mann-
Whitney U-test
Breast milk 2 months
post-partum
Median:
20.6 ug/L
>WHO limit
(5 ug/L):
Median Pb (IQR)
>37 week:20.6
(11.2, 29.2) ug/L
< 37 week: 20.4
(14.4, 27.9) ug/L
None
                                                                                                           87%
                                           Npreterm birth" 17

                                           Nterm birth=127
                                                                                                          p-value for
                                                                                                          Mann-Whitney U
                                                                                                          test: > 0.05
November 2012
                                   5-564
                                                 Draft - Do Not Cite or Quote

-------
 1                  In a study taking place in California, women with information on blood Pb levels during
 2                  pregnancy based on their participation in a surveillance program (reason for participation
 3                  in the surveillance program was unknown but the authors speculate it was likely because
 4                  of potential Pb exposure due to occupational or environmental exposures or a family
 5                  member was identified as exposed to Pb) were matched with the birth certificates of their
 6                  infants (Jelliffe-Pawlowski et al.. 2006). Almost 70% of women had maximum blood Pb
 7                  measurements <10 (ig/dL with the majority being <5 (ig/dL. Preterm birth was associated
 8                  with higher blood Pb when comparing women with maximum pregnancy blood Pb levels
 9                  >  10 (ig/dL to women with blood Pb levels <10 (ig/dL in adjusted analyses. In analyses
10                  of maximum Pb levels further refined into additional categories, the odds of preterm birth
11                  were elevated among women with maximum blood Pb measurement > 20 (ig/dL
12                  compared with women with maximum blood Pb levels  < 5 (ig/dL. A study in Iran also
13                  reported higher maternal blood Pb for preterm births than for term births fVigeh et al..
14                  2011). The women in this study had lower blood Pb levels than did those observed in the
15                  Jelliffe-Pawlowski et al. (2006). Higher maternal blood Pb level was associated with
16                  higher odds of preterm birth. Another study examining  blood Pb and gestational age
17                  among women with lower blood Pb levels reported an inverse association between
18                  maternal blood Pb concentration and gestational age, especially for blood Pb levels early
19                  in pregnancy (Cantonwine et al.. 2010a). However, a study conducted in New York
20                  among women with lower blood Pb levels (inclusion criteria mandated that blood Pb
21                  levels be less than 10 (ig/dL), no association was observed between blood Pb levels and
22                  preterm birth (Zhu et al.. 2010). Similarly, a study of maternal and paternal blood Pb
23                  concentrations reported no association between maternal or paternal blood Pb levels and
24                  preterm birth (Chen et al.. 2006a).

25                  In another study, measurements of umbilical cord blood were taken after birth at a
26                  hospital in Nagpur, India (Patel and Prabhu. 2009). A sample of women had their blood
27                  Pb measured and among this sample, maternal blood Pb was correlated with the umbilical
28                  cord Pb levels. Mean gestational age  differed between infants with >5 (ig/dL cord blood
29                  Pb and infants with < 5  (ig/dL cord blood Pb. In a linear regression model, gestational
30                  age was found to decrease with increasing umbilical cord Pb levels. A study of women in
31                  Tennessee consisted primarily of African American women living in an urban setting
32                  (Jones et al.. 2010). The mean level of umbilical cord blood Pb was slightly higher
33                  among infants born preterm but the difference was not statistically significant. Using
34                  umbilical cord blood Pb measures, a study reported no association between cord blood Pb
35                  levels and gestational age. The concentrations of cord blood Pb among study participants
36                  were overall low (99.3% had umbilical cord blood Pb < 5 (ig/dL) (Wells et al.. 201 la).
      November 2012                            5-565                  Draft - Do Not Cite or Quote

-------
 1                  A study of preterm birth included women living in two different residential areas over
 2                  three different time periods (Berkowitz et al.. 2006). One residential area had consistently
 3                  lower exposures but the other had a period of high Pb emissions due to damage at a local
 4                  factory (Pb measured in ambient air was up to 30 (ig/m3). Preterm birth rates were
 5                  examined during three time periods: before, during, and after the time of higher Pb
 6                  exposure. No association was observed between women living in the high exposure area
 7                  compared to those in the low exposure area during any of the exposure time periods, but
 8                  the number of preterm infants born during the period of higher exposure was small.

 9                  A study of breast milk in the  second month postpartum reported no difference in breast
10                  milk Pb levels for those infants born preterm or term; however, a limitation of this study
11                  is that Pb levels were not measured until two months after the birth (Oriin et al., 2011).

12                  In summary, as  in the 2006 Pb AQCD, (U.S. EPA. 2006b) recent epidemiologic studies
13                  report inconsistent findings for a relationship between indicators of Pb exposure and
14                  preterm birth. No patterns were apparent within type of exposure measurement or Pb
15                  level. Many of these studies are limited by the small number of preterm births and their
16                  cross-sectional design (i.e., studies of umbilical cord blood may not adequately
17                  characterized blood Pb levels earlier in pregnancy).A few studies utilized a longitudinal
18                  cohort design (Vigeh etal.. 2011; Cantonwine etal.. 2010a; Chen et al.. 2006a: Jelliffe-
19                  Pawlowski et al.. 2006). and although results among these studies were mixed some did
20                  report an association between maternal blood Pb during pregnancy and preterm birth.
21                  Most studies controlled for important confounders, such as maternal age and smoking.
                    5.8.3.4     Low Birth Weight/Fetal Growth

22                  The 2006 Pb AQCD reported inconsistent epidemiologic study results for the
23                  associations between Pb and birth weight/fetal growth but concluded that there could be a
24                  small effect of Pb exposure on birth weight and fetal growth (U.S. EPA. 2006b). Since
25                  then, multiple epidemiologic studies on the relationship between Pb exposure and birth
26                  weight and fetal growth have been published using various measures of exposure, such as
27                  air levels, umbilical cord blood, and maternal blood and bone. These studies are
28                  summarized in Table 5-42 below (organized in the text and table by the type of Pb
29                  measurement and then by study design). Additionally, there have been a few recent
30                  toxicological studies evaluating the effect of Pb exposure during gestation on birth
31                  weight.
      November 2012                            5-566                 Draft - Do Not Cite or Quote

-------
Table 5-42  Summary of recent epidemiologic studies of associations between Pb levels and low birth weight
            and fetal growth.
Reference
(Presented
in order of
appearance
in the text)
Gundacker
et al. (201 0)




















Zhuetal.
(2010)













Study
Location Outcome
Vienna, Birth length,
Austria birth weight,
2005 nead
circumference


















New York Birth weight,
2003-2005 sma" for
gestational age
(birth weight for
gestational age
<10th percentile
based on
national birth
weight by
gestational week
from weeks
25-42





Study
population
Infants of women
recruited during their
second trimester

N=53

















Singleton births to
mothers aged 15-49
with blood Pb
measures before or
on the date of
delivery and blood
Pb measuring
<10 ug/dL

NLBw=2744
N. ~.,=40 544
normal BW ^w,^-r-r
NSGA=4092



Methodological
Details
Cohort study
using categorical
regression



















Retrospective
cohort using linear
regression with
factional
polynomials for
birth weight and
logistic regression
with fractional
polynomials for
SGA





Pb Biomarkers
and Exposure
Measurement
Maternal blood Pb
between week 34-38
of gestation, whole
placentas and
umbilical cord Pb
shortly after birth,
meconium samples
in first five days after
birth

















Maternal blood Pb
before or at delivery












Mean Pb
(SD)
in ug/dL
Median
(IQR):
Maternal
blood Pb'
2.5
(1.8,3.5)
Umbilical
cord blood
Pb:1.3
(0.8, 2.4)
Placenta
Pb:
25.8 ug/kg
(21.0,
36.8 ug/kg)
Meconium
Pb:
1 5.5 ug/kg
(9.8,
O~7 O i I/N l\fit\
27.9 ug/kg)





2.1














Adjusted Effect
Estimates
Regression
coefficients (units
not given, assume
results are per
10 ug/dLor
1 ug/kg)
Birth length:
Placenta Pb:
0.599
(SE0.154, p-value
<0.001)
Meconium Pb:
-0.385
(SE 0.157, p-value
0.012)
Birth weight:
Placenta Pb:
0.658
(SE 0.136, p-value
<0.001)
Maternal blood
Pb:
-0.262
(SE0.131, p-value
0.058)
Difference in
birthweight in
grams:
0 ug/dL: Ref
1 ug/dL: -27.4
(-37.8, -17.1)
2 ug/dL: -38.8
(-53.4, -24.1)
3 ug/dL: -47.5
(-65.4, -29.6)
4 ug/dL: -54.8
(-75.5, -34.2)

Potential
Confounders
Adjusted for in
Analysis
Model for birth length:
placenta Pb, gestational
age, meconium Pb

Model for birth weight:
gestational age,
placenta Pb, maternal
blood Pb

















Model for birth weight:
Timing of Pb test,
maternal age, race,
Hispanic ethnicity,
education, smoking
status, alcohol use, drug
abuse, marital status,
financial assistance
program participation,
parity, infant sex

Model for SGA: Timing
of Pb test, maternal age,
November 2012
5-567
Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance
in the text)
Study
Location
Outcome
                                      Pb Biomarkers      Mean Pb
Study              Methodological   and Exposure       (SD)
population	Details	Measurement	in ug/dL
Adjusted Effect
Estimates
                                                                                                            Potential
                                                                                                            Confounders
                                                                                                            Adjusted for in
                                                                                                            Analysis
                                          NnoSGA=39,084
                                                                                         5 ug/dL:-61.3
                                                                                         (-84.4, -38.2)
                                                                                         6 ug/dL: -67.2
                                                                                         (-92.5, -41.8)
                                                                                         7 ug/dL: -72.5
                                                                                         (-99.9, -45.2)
                                                                                         8 ug/dL: -77.6
                                                                                         (-106.8,-48.3)
                                                                                         9 ug/dL: -82.3
                                                                                         (-113.3,-51.2)
                                                                                         10 ug/dL: -86.7
                                                                                         (-119.4,-54.0)


                                                                                         After exclusion of
                                                                                         blood Pb
                                                                                         <1 ug/dL, a
                                                                                         1 ug/dL increase
                                                                                         in blood Pb was
                                                                                         associated with a
                                                                                         7.0 g decrease in
                                                                                         birthweight


                                                                                         Odd Ratios for
                                                                                         small for
                                                                                         gestational age:
                                                                                         < 1.0 ug/dL: 1.00
                                                                                         (Ref)
                                                                                         1.1-2.0 ug/dL:
                                                                                         1.07(0.98,1.17)
                                                                                         2.1-3.0ug/dL:
                                                                                         1.06(0.98, 1.16)
                                                                                         3.1-9.9 ug/dL:
                                                                                         1.07 (0.93, 1.23)
                                                                                                                       race, education,
                                                                                                                       smoking status, drug
                                                                                                                       abuse, marital status,
                                                                                                                       financial assistance
                                                                                                                       program participation,
                                                                                                                       parity, infant sex
November 2012
                                   5-568
                                                 Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance
in the text)
Chen et al.
(2006a)




























Study
Location Outcome
Taiwan Low birth weight
1993-1997 (<2,500g),
small for
gestational age
(birth weight
< 10th percentile
of sex- and
gestational week
weights for
singletons in
1993-1996)

























Study
population
Infants born to at
least one parent
who was part of the
Program to Reduce
Exposure by
Surveillance System
- Blood Lead Levels
cohort that
monitored workers
occupationally
exposed to Pb

NLBW=72
Nnormal BW/I ,539 =
NSGA=135
NnoSGA=1,476


*738 births had
maternal Pb
information and 967
had paternal Pb
information














Pb Biomarkers Mean Pb
Methodological and Exposure (SD)
Details Measurement in ug/dL
Occupational Maternal blood Pb Maternal
cohort study using during pregnancy (or blood Pb
regression models if that wasn't 10.1(10.4)
available, the 1 year
prior to fertilization)
and/or paternal blood Paternal
Pb during blood pb
spermatogenesis 12.9(13.8)
(the 64 days before
fertilization, or if that
wasn't available, the
1 year prior to
spermatogenesis)























Adjusted Effect
Estimates
Risk Ratios
Low birth weight
Maternal blood Pb
<10 ug/dL: 1.00
(Ref)
10-19ug/dL:2.22
(1 .06, 4.26)
> 20 ug/dL: 1 .83
(0.67, 4.20)

Paternal blood Pb
<10ug/dL:1.00
(Ref)
10-19ug/dL:0.83
(0.34, 1 .75)
> 20ug/dL: 0.42
(0.12, 1 .06)

SGA
Maternal blood Pb
<10 ug/dL: 1.00
(Ref)
10-19ug/dL:1.62
(0.91 , 2.75)
>20ug/dL:2.15
(1.15,3.83)
Paternal blood Pb
<10ug/dL:1.00
(Ref)
10-19 ug/dL:0.94
(0.49, 1 .66)
> 20 ug/dL: 0.94
(0.51 , 1 .62)
Potential
Confounders
Adjusted for in
Analysis
Low birth weight models:
parental age, parental
education, infant sex,
parity
SGA models: parental
age, parental education



























November 2012
5-569
Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance
in the text)
Jelliffe-
Pawlowski
et al. (2006)






















Study
Location Outcome
California Low birth weight
1995-2002 (<2,500g)
Small for
gestational age
(birth weight for
gestational age
<10th percentile
of race- and
gender- specific
norms






















Pb Biomarkers
Study Methodological and Exposure
population Details Measurement
Singleton births to Longitudinal Maximum maternal
non-smoking cohort study using blood Pb during
mothers with blood logistic regression pregnancy
Pb measures during
pregnancy from
either the California
Childhood Lead
Poisoning
Prevention Branch
or the California
Occupational Lead
Poisoning
Prevention Program
and matched to birth
records

NLBw=9
Nnormal BW=253
NsGA=17
NnoSGA=245














Mean Pb
(SD) Adjusted Effect
in ug/dL Estimates
>10ug/dL: Odd Ratios:
30.9% Low bjrth weight
<5 ug/dL:
1.00(Ref)
6-9 ug/dL:
10-19 ug/dL:
2.7(0.5,14.8)
20-39 ug/dL:
1.5(0.3,7.7)
> 40 ug/dL:
—

<10 ug/dL:
1.00(Ref)
> 10 ug/dL:
3.6 (0.3, 40.0)

Small for
gestational age
<5 ug/dL:
1.00(Ref)
6-9 ug/dL:
10-19ug/dL:
2.3(0.6,9.2)
20-39 ug/dL:
2.1 (0.7,6.7)
> 40 ug/dL:
<10 ug/dL:
1.00(Ref)
> 10 ug/dL:
4.2(1.3,13.9)
Potential
Confounders
Adjusted for in
Analysis
Adjusted for in 10 ug/dL
model for birth weight:
preterm birth, race,
insurance, parity,
maternal age, infant sex
Adjusted for in 10 ug/dL
model for SGA:
insurance, parity,
maternal age






















November 2012
5-570
Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance
in the text)
Lambet al.
(2008)











Iranpouret
al. (2007)


























Study Study
Location Outcome population
Mitrovica Height and BMI Participants of the
and at birth Yugoslavia Study of
Pristina, Environmental Lead
Yugoslavia Exposure,
1985-1986 Pregnancy
Outcomes, and
Childhood
Development

n=292





Isfahan, Low birth weight Full-term infants
Iran (<2,500g, born at a hospital
2005 >37wk) affiliated with
Isfahan University

NLBW32
N normal BW=34





















Pb Biomarkers
Methodological and Exposure
Details Measurement
Population-based Mid-pregnancy blood
prospective cohort Pb
study using linear
regression










Cross-sectional Umbilical cord and
study using t-tests maternal blood Pb
and Spearman's within 12 h of
correlations delivery
























Mean Pb
(SD)
in ug/dL
Mitrovica:
20.56 (7.38)
Pristina'
5.60 (1 .99)










Maternal
blood Pb:
Cases: 12.5
(2.0)
Controls:
13.5(2.7)
Umbilical
cord blood
Pb:

Cases: 10.7
(1.7)
Controls:
11 3 (1 9)
\ /














Adjusted Effect
Estimates
Regression
Coefficients (95%
Cl) for 1 ug/dL
increase in Pb:
BMI
Mitrovica:
-0.1 8 (-0.69, 0.33)
Pristina:
-0.1 4 (-0.69, 0.42)
Height
Mitrovica:
0.43 (-0.83, 1 .69)
Pristina:
0.35 (-0.64, 1 .34)
P-values fort-
tests:
Maternal blood
Pb: 0.07
Umbilical cord
blood Pb:
0.20


P-values for
correlations:
Maternal blood Pb
and Birth weight:

Low birth weight:
0.17
Normal birth
weight: 0.3
P-values for
correlations:
Umbilical cord
blood Pb and birth
weight:
Low birth weight:
0.84
Normal birth
weight: 0.26

Potential
Confounders
Adjusted for in
Analysis
Infant sex, ethnicity,
parity, maternal height or
BMI, maternal
education, gestational
age at blood sample,
gestational age at birth,
quality of home
environment







None

























November 2012
5-571
Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance Study
in the text) Location Outcome
Kordasetal. Mexico Head
(2009) City, circumference,
Mexico birth weight,
1994-1995 birth length

















Afeiche et Mexico Birth weight
al. (2011) City
1 994-2005

















Study
population
Infants of mothers
receiving antenatal
care at hospitals
serving low-to-
middle income
populations (cross-
sectional study of
baseline info from
Ca supplementation
trial)


N=474










Term, singleton
births, at least 2,500
grams enrolled in
one of three birth
cohorts recruited for
other longitudinal
studies


N=1,000











Pb Biomarkers
Methodological and Exposure
Details Measurement
Cross-sectional Umbilical cord and
study using linear maternal blood Pb
regression within 12 h of
delivery; maternal
tibia Pb 1 month
post-pa rtum















Cross-sectional Maternal patella and
study using tibia Pb measured at
varying coefficient 1 month postpartum
models with
random effects














Mean Pb
(SD)
in ug/dL
Maternal
tibia Pb:
9.9 ug/g
(9.8 ug/g)

Maternal
blood Pb
> 10ug/dL:
27%
Umbilical
cord blood
Pb
> 10ug/dL:
13.7%








Patella Pb
10.4
(11. 8) ug/g

Tibia Pb 8.7
(9.7) ug/g















Adjusted Effect
Estimates
Regression
coefficients (SE)
for each 1 ug/g
increase in tibia
Pb:
Birth weight: -4.9
C1 81
V ' -°f
Birth length: -0.02
(0.01)
Head
circumference:
-0.01 (0.01;
p-value <0.05)

Women with 4th
quartile tibia Pb
(15.6-76.5 ug/g)
delivered infants
1 40 g less than
women with tibia
Pb in the lowest
quartile
P (95% Cl) for 1
SD increase in
maternal patella
Pb
Girls:
-45.7 (-131 .7,
40.2)

Boys:
72.3 (-9.8, 154.4)
No association for
birth weight and
tibia Pb among
girls. A positive
association was
observed for tibia
Pb and birth
weight among
boys, (results not
given)

Potential
Confounders
Adjusted for in
Analysis
Maternal age, pre-
pregnancy BMI,
maternal height,
education, parity, marital
status, ever smoker,
postpartum calf
circumference,
gestational age, infant
sex














Birth cohort, maternal
age, maternal calf
circumference, maternal
height, education, parity,
breast feeding,
Ca + treatment group
assignment, gestational
age, height at birth,
repeated concurrent
child blood Pb measures










November 2012
5-572
Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance
in the text)
Cantonwine
etal.
(201 Ob)













Wells et al.
(2011 a)








Al-Saleh et
al. (2008b)







Study
Location Outcome
Mexico Birthweight
City
1994-1995













Baltimore, Birth weight
MD
2004-2005







Saudi Head
Arabia circumference
2004






Study
population
Infants who were
part of a clinical trial
to assess maternal
Ca2+-
supplementation on
bone Pb
mobilization during
lactation

N=533









Singleton births from
the Baltimore
Tracking Health
Related to
Environmental
Exposures (THREE)
study

Ni_Bw=33
Nnormal BW=267
Infants with a
gestational age of at
least 34 weeks born
to healthy mothers
aged 1 7-46 years
and non-
occupationally
exposed to Pb

N=653


Methodological
Details
Cross-sectional
study using linear
regression













Cross-sectional
study using
multivariable
linear regression






Cross-sectional
study using linear
regression





Pb Biomarkers
and Exposure
Measurement
Umbilical cord blood
Pb
Maternal tibia and
patella Pb one month
after delivery













Umbilical cord Pb









Umbilical cord blood
Pb






Mean Pb
(SD)
in ug/dL
Umbilical
cord blood
Pb varied by
genotype
from 6.3 to
6.9

Umbilical
cord blood
Pb
£ 10 ug/dL'
12.6%








0.84 (95%:
Cl 0.72,
0.96)


> 5 ug/dL:
0.7%



2.210
(1.691)


Umbilical
cord blood
Pb
>10 ug/dL:
1 .23%




Adjusted Effect
Estimates
Regression
models
P (95% Cl)
Umbilical cord
blood Pb: -31.1
(-105.4, 43.3)

Maternal tibia Pb
Overall: -4.4 (-7.9,
-0.9)
<1-4.1 ug/g: Ref
4.1-9.2 ug/g: 17.2
(-75.6, 110.1)
9.2-1 5.4 ug/g:
-19.1 (-112.1,
73.9)
15.4-43.2 ug/g:
-95.4 (-189.9, -0.8)
Ratio for Pb
concentration per
100g birth weight:
1 .01 (0.99, 1 .02)






Regression
models for those
above the 75th
percentile of cord
blood Pb levels
P (SE) per unit of
log-transformed
Pb
-0.158(0.718), p-
value: 0.036

Potential
Confounders
Adjusted for in
Analysis
Maternal age, education,
infant sex, maternal arm
circumference,
gestational age, smoking
status during pregnancy,
marital status, maternal
hemoglobin first month
postpartum, parity











Maternal age, race,
insurance, pre-
pregnancy weight,
smoking status,
gestational length, birth
weight, average year of
neighborhood home
construction


BMI, gestational age


Considered but not
included: prenatal
supplements, location
of residence


November 2012
5-573
Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance
in the text)
Atabeketal.
(2007)










Llanos and
Ronco
(2009)



Zentner et
al. (2006)















Janjua et al.
(2009)








Study
Location Outcome
Turkey Birth weight,
NS birth length,
head
circumference,
mid-arm
circumference







Santiago, Fetal growth
Chile restriction
NS (1 ,000-2,500g)
"note normal
birth weights
were >3,000g

Santo Birth weight and
Amaro, length
Brazil
2002













Karachi, Low birth weight
Pakistan (< 2,500g)
2005








Study
population
Term, singleton
infants born to
healthy mothers
living in urban areas
and assumed to
have high Pb
concentrations

N-E.A
— \Jt



Term births
(37-40 weeks) from
non-smoking
mothers

N -9D
'^growth restricted ^^*
N normal BW=20
Singleton births with
maternal residence
within 5 km of Pb
smelter
N=55












Infants of randomly
selected women
who planned to
deliver between
37-42 week


NLBw=100
Nnormal BW=440

Pb Biomarkers
Methodological and Exposure
Details Measurement
Cross-sectional Umbilical cord blood
study using linear Pb
regression









Cross-sectional Placenta Pb
study using Mann-
Whitney U-test



Cross-sectional Umbilical cord blood
study using linear Pb from delivery
regression














Cross-sectional Umbilical cord blood
study using Pb
binomial
regression






Mean Pb
(SD)
in ug/dL
14.4(8.9)

Umbilical
cord blood
Pb>
10 ug/dL:
53.7%
Umbilical
cord blood
Pb
> 25 ug/dL:
9.2%
Fetal growth
restricted:
0.21 ug/g
(0.04 ug/g)
Controls:
0.04 ug/g
(0.009 ug/g)
Umbilical
cord blood
Pb:
3.9 (3.6)













Umbilical
cord blood
Pb:
10.8(0.2)







Adjusted Effect
Estimates
Regression
models
P (p-value)
Birth weight:
-0.81 (0.01)
Birth length:
0.41 (0.05)
Mid-arm
circumference:
0.30 (0.05)


P-value for Mann-
Whitney U-test
<0.01



Linear regression
coefficient with
umbilical cord
blood Pb as the
dependent
variable in model
with only length
and weight (unit
not given, assume
per 1 ug/dL):
Length -0.46 (p-
value 0.003)
and Weight -0.275
(0.048)
(i.e., in this study,
Pb is assessed as
the outcome)
Prevalence ratio:
<10 ug/dL:
1 .00 (Ref)

a 10 ug/dL:
0.82
(0.57, 1.17)



Potential
Confounders
Adjusted for in
Analysis
Age, sex

Note: inclusion of SES
did not change the
results







None





No other variables
besides length and
weight were included in
the model













None







November 2012
5-574
Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance
in the text)
Jones et al.
(2010)







Orun et al.
(2011)
















Study Study
Location Outcome population
Tennessee Low birth weight Singleton births
2006 (<2.500g) > 27 weeks
gestation from
mothers aged 16-45
living in the Shelby
County area for at
least 5 mo during
pregnancy
NLBw=10
N normal BW=92
Turkey Birth weight and Births to mothers
NS head not occupationally
circumference exposed to toxic
metals and living in
a suburban but non-
industrial area

NLBw=9
NnormalBW=135









Pb Biomarkers
Methodological and Exposure
Details Measurement
Cross-sectional Umbilical cord blood
study comparing Pb
across geometric
means (test not
specified)





Cohort study Breast milk 2 months
using Pearson post-partum
correlation
coefficients













Mean Pb
(SD)
in ug/dL
2.4 (4.3)
Geometric
mean' 1 3







Median:
20.6 ug/L


>WHO limit
(5 ug/L):
87%
Median
(IQR)
<2500g:
20.4 (8.5,
27.1) ug/L
> 2500g:
20.6(11.8,
29.5) ug/L



Potential
Confounders
Adjusted Effect Adjusted for in
Estimates Analysis
Geometric Mean: None
Low birth weight:
1 .2

Normal birth
weight: 1.3
p-value for
difference: >0.10


Correlations for None
breast milk Pb and
z-scores of head
circumference
Girls: 0.087
Boys: 0.029
Correlations for
breast milk Pb and
z-scores of birth
weight
Girls: 0.097
Boys: 0.045


*AII p-values for
correlations>0.05
November 2012
5-575
Draft - Do Not Cite or Quote

-------
Reference
(Presented
in order of
appearance
in the text)
Williams et
al. (2007)















Berkowitz et
al. (2006)

















Study
Location Outcome
Tennessee Birth weight
2002















Idaho Low birth weight
1970-1981 (<2,500gand
> 37 week)
Small for
gestational age
(birth weight
£ 5th percentile
of sex- and
gestational week
weights for
singletons in
Idaho)














Study
population
Infants from
singleton births or
the firstborn infant in
a set of multiples

N=not specified












Singleton infants
with 28-45 week
gestation
NLBW=4297
NnormalBW= 162,035
NSGA=7020
NnoSGA=1 62,035














Pb Biomarkers
Methodological and Exposure
Details Measurement
Longitudinal Air Pb levels during
cohort study using first trimester of
hierarchical linear pregnancy
models














Cohort study Three time periods of
using logistic two locations
regression (unexposed and
exposed/near
smelter): pre-fire,
"high-exposure
period" (when a fire
happened at the
smelter and resulted
in damages leading
to high air Pb
concentrations for 6
mo), and "post-fire"













Mean Pb
(SD)
in ug/dL
0.12 ug/m3
(0.04 ug/m3)















During the
time of the
fire,
estimates of
Pbin
ambient air
were as high
as 30 ug/nr















Adjusted Effect
Estimates
p-value for
multilevel
regression of Pb
with birth weight:
0.002
Increase of Pb
from 0 to 0.04
relates to a 38g
decrease in birth
weight
Increase of Pb
from 0 to 0.13
(maximum) relates
to a 1 24g
decrease in birth
weight




Term Low birth
weight:
OR (90% Cl)
(unexposed
location is referent
group):
Pre-fire: 0.81
(0.55, 1 .20)
High exposure:
2.39(1.57,3.64)
Post-fire: 1.28
(0.95, 1 .74)
Small for
gestational age:
OR (90% Cl)
(unexposed
location is referent
group):
Pre-fire: 0.98
(0.73, 1 .32)
High exposure:
1.92(1.33,2.76)
Post-fire: 1.32
(1.05,1.67)

Potential
Confounders
Adjusted for in
Analysis
Previous preterm birth,
previous birth>4000g,
pregnancy-induced
hypertension, chronic
hypertension,
oligohydramios, other
maternal risk factors,
education,
cigarettes/day, black
race, Hispanic ethnicity,
other race/ethnicity,
plurality, infant sex, first
trimester SO2, within
5km of an air monitor,
poverty, interaction of
poverty and other
maternal risk factors,
percentage of previous
pregnancies that
resulted in non-live
births
Maternal age, infant sex,
first birth, previous
miscarriage or abortion














November 2012
5-576
Draft - Do Not Cite or Quote

-------
 1                  Multiple studies were conducted that examined the association between maternal blood
 2                  Pb and birth weight/fetal growth. A study in Vienna, Austria reported an inverse
 3                  association between maternal blood Pb levels and birth weight but no associations for
 4                  birth length or head circumference (Gundacker et al.. 2010). Similarly, increased
 5                  maternal blood Pb was associated with decreased birth weight among infants in a study
 6                  performed in New York (Zhu et al.. 2010). No association was observed between
 7                  maternal blood Pb levels and SGA. A study in Taiwan examined both maternal and
 8                  paternal blood Pb levels among those occupationally exposed to Pb and their associations
 9                  with birth weight and SGA (Chen et al.. 2006a). Paternal blood Pb levels were not
10                  associated with increased risk of low birth weight or SGA. Higher maternal blood Pb
11                  concentration was associated with higher risk of low birth weight and SGA,  although not
12                  all of the associations were statistically significant. There were small numbers of infants
13                  with low birth weight or SGA, especially at the highest blood Pb levels (> 20 (ig/dL). In
14                  California,  blood Pb measurements of women during pregnancy were matched with the
15                  corresponding birth certificates (Jelliffe-Pawlowski et al.. 2006). The adjusted OR for
16                  low birth weight that compared women with blood Pb levels > 10 (ig/dL to women with
17                  levels <10 (ig/dL was elevated. However, it was difficult to draw conclusions about the
18                  relationship between blood Pb and birth weight due to small numbers (n = 9 for low birth
19                  weight) and the subsequently wide 95% CI. An association was detected for increased
20                  blood Pb and having an infant who was small for his/her gestational age (SGA). Women
21                  residing in two different towns in Yugoslavia (one with  a Pb smelter and one without a
22                  Pb smelter) were recruited during their first prenatal visit (Lamb et al., 2008) (study
23                  based on previous work by Factor-Litvak et al. (1991)).  The mid-pregnancy  blood Pb
24                  levels were greater in women from the town with a Pb smelter. No association was
25                  reported between maternal blood Pb and height or BMI  at birth for the infants of these
26                  women despite the differences in maternal blood Pb between the two towns. A study of
27                  term births in Iran reported no difference in blood  Pb levels of women giving birth to a
28                  normal weight infant and women giving birth to an infant with low birth weight (Iranpour
29                  et al.. 2007).

30                  A study examining the association between Pb biomarker levels and birth weight used
31                  tibia bone measurements one month post-partum from mothers living in Mexico City
32                  (Kordas et al.. 2009). Tibia Pb levels were inversely associated with birth weight but not
33                  with birth length. This association between Pb and birth weight was not modified by
34                  maternal folate consumption or maternal or infant  MTHFR genotype, although the
35                  association between tibia Pb levels and birth weight was greater in magnitude among
36                  women with certain MTHFR SNPs (statistical tests not reported). Another study in
37                  Mexico City reported no association between maternal tibia Pb levels and birth weight
38                  among girls but reported a positive association for boys  (Afeiche et al.. 2011). No
39                  associations were observed with maternal patella Pb concentration, although among boys,

      November 2012                            5-577                 Draft - Do Not Cite or  Quote

-------
 1                  the relationship was positive but not statistically significant. One of the cohorts used by
 2                  Afeiche et al. (2011) was also evaluated in another study (Cantonwine et al.. 2010b). An
 3                  inverse association was observed between tibia Pb and birth weight, especially at higher
 4                  levels (over 15.4 (ig/dL). This association was stronger among those mothers with
 5                  variants of the hemochromatosis iron gene (HFE).

 6                  Multiple studies examined the relationship between Pb level and birth weight using Pb
 7                  measured from the placenta or umbilical cord. A study performed in Baltimore, MD
 8                  reported no association between umbilical cord blood Pb concentration and birth weight
 9                  (Wells etal.. 201 la). This study had low blood Pb levels, with only 0.7% of participants
10                  having umbilical cord blood Pb measuring >5 (ig/dL. In Saudi Arabia, a study was
11                  conducted among non-occupationally exposed women (Al-Saleh et al., 2008b). Umbilical
12                  cord blood Pb concentrations were low and an association was observed between
13                  umbilical cord Pb and head circumference. A study with high Pb concentrations in
14                  umbilical cord blood reported an inverse association between Pb levels and birth weight
15                  (Atabek et al., 2007). However, no correlation was detected in an analysis restricted to
16                  umbilical cord Pb less than 10 (ig/dL. No associations with other measures of growth,
17                  such as birth length and mid-arm circumference, were detected.  Researchers in Chile
18                  collected the placentas from term births and compared the Pb levels for those born with
19                  normal birth weights to those with low birth weights (Llanos and Ronco. 2009). Pb levels
20                  were greater in the placentas of infants with low birth weights. In addition, the authors
21                  note that 3 low birth weight infants had extremely high Pb levels in the placentas
22                  (>1.5 (ig/g) and were excluded from these analyses. A study in Brazil examined Pb levels
23                  in umbilical cord blood from term births of women residing within 5 km  of a Pb smelter
24                  (Zentner et al.. 2006). The cord blood Pb level was found to be inversely correlated with
25                  length and weight of the infants. Another study recruited women in Pakistan (Janjua et
26                  al.. 2009). Umbilical cord blood Pb levels were not associated with low birth weight. The
27                  study by Iranpour et al. (2007) discussed above investigated the  association with
28                  umbilical cord blood Pb levels in addition to their examination of maternal whole blood
29                  Pb. They again report no difference in levels between term infants of normal and low
30                  birth weight. A study comparing geometric mean  umbilical cord blood Pb levels reported
31                  no difference in the levels for normal and low birth weight infants born to women living
32                  primarily in urban areas of Memphis, TN (Jones etal.. 2010).  A study previously
33                  mentioned that observed an inverse association between maternal tibia Pb and birth
34                  weight in Mexico City reported no association between umbilical cord blood Pb
35                  concentration and birth weight (Cantonwine et al., 2010b). Finally, a study in Vienna
36                  measured Pb in the placenta (Gundacker et al.. 2010). A positive correlation was
37                  observed between placenta Pb and birth length and weight; however, in the same study,
38                  maternal blood Pb was inversely related to birth weight.
      November 2012                             5-578                  Draft - Do Not Cite or Quote

-------
 1                  A study performed in Turkey examined the relationship between Pb levels in breast milk
 2                  two months postpartum and size at birth (Orim et al.. 2011). No association was observed
 3                  between breast milk Pb concentration and birth weight or head circumference.

 4                  A few studies examined air exposures and reported inverse associations between air Pb
 5                  concentrations and birth weight. However, a limitation of these studies is the difficulty in
 6                  assessing if the measured concentrations represent population exposures (see
 7                  Section 3.5.3). Williams et al. (2007) examined Pb concentrations in the air during the
 8                  first trimester.  The purpose of their study was to demonstrate the use of hierarchical
 9                  linear models and they used the example of air pollution and birth weight in Tennessee.
10                  The model results showed an association between ambient Pb concentration and birth
11                  weight, with an estimated decrease in birth weight of 38 grams for every 0.04 (ig/m3
12                  (i.e., one standard deviation) increase in Pb concentration. Another study of air Pb levels
13                  was conducted in Idaho and included two areas over three time periods. One study area
14                  was affected by damage to a local factory that led to high Pb emissions during one of the
15                  time periods under study (Berkowitz et al.. 2006). During the time of the fire, estimates
16                  of Pb in ambient air were as high as 30  (ig/m3. Mean birth weight forterm births was
17                  decreased among infants born to women living in the high exposure area during the
18                  period of high exposure compared to those living in the lower exposure area. The
19                  difference in birth weight of term births remained, but was reduced, between the two
20                  areas during the time period after the exposure ended. During the period of higher
21                  exposure, the odds of low birth weight among term births was increased among those
22                  living in the higher exposed area compared to those in the lower exposed area, but the
23                  odds were not  different between the two study areas during the time periods before or
24                  after the high level of exposure. An increase in SGA infants (defined as infants with
25                  weights less than or equal to the lowest 5th percentile of birth weight for their sex and
26                  age) was also associated with living in the higher exposed area during the time period of
27                  higher exposure. The odds  of SGA infants decreased during the time period after the
28                  exposure but the  odds were still elevated compared to those residing in the lower exposed
29                  area.

30                  Evidence from previous toxicological studies has shown an association between
31                  gestational Pb  exposure and reduced birth weight (U.S. EPA. 2006b). More recent studies
32                  have reported conflicting results. Wang et al. (2009e) demonstrated a statistically
33                  significant decrease in fetal body weight and body length of Wistar rats after maternal
34                  exposure to 250 ppm Pb acetate during gestation days 1-10, 11-20, or 1-20. The greatest
35                  decrease in fetal  body weight and length was observed in the group exposed to Pb during
36                  gestation days 1-20 followed by the group exposed to Pb during gestation days 11-20.
37                  Teijon et al. (2006) observed that when pregnant dams were administered 200 ppm or
38                  400 ppm Pb acetate in drinking water, litter weight was significantly decreased (400 ppm
      November 2012                            5-579                 Draft - Do Not Cite or Quote

-------
 1                  Pb only) versus controls due to significant decrements in female pup birth weight; male
 2                  birth weight was unaffected. The results of these studies indicate that as Pb exposure
 3                  increases, the body weight of exposed offspring decreases. Masso-Gonzalez and
 4                  Antonia-Garcia (2009) also observed an 8-20% decrease in body weight of pups from rat
 5                  dams given 300 mg/L Pb acetate in drinking water (exposure during gestation and
 6                  lactation resulting in mean blood Pb level of 22.8 (ig/dL), but no changes in body length
 7                  were reported.

 8                  In summary, associations were observed between Pb and low birth weight in
 9                  epidemiologic studies of maternal bone Pb and studies of Pb air exposures and birth
10                  weight. The associations were  less consistent when using maternal blood Pb or umbilical
11                  cord and placenta Pb as the exposure measurement although some studies did
12                  demonstrate associations. Epidemiologic studies of Pb and fetal growth face multiple
13                  limitations. One limitation is the cross-sectional nature of many studies. These do not
14                  allow an understanding of the temporality for Pb and fetal growth. In addition, some
15                  studies suffer from small sample size. The studies of air Pb levels and birth weight
16                  demonstrate positive associations but are limited in that individual exposure levels are
17                  unknown. Also, many of the studies controlled for important confounders, such as parity
18                  and gestational age, but adjustment in some studies was lacking. Previous toxicological
19                  studies observed an association between gestational Pb exposure and reduced birth
20                  weight with moderate to high dose Pb.
      5.8.4        Effects on Male Reproductive Function

21                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported on male Pb exposure or biomarker
22                  levels and reproductive functions in males as measured by sperm
23                  count/motility/morphology, time to pregnancy, reproductive history, and chromosomal
24                  aberrations. Despite limitations, most of the studies found slight associations between
25                  high blood Pb levels (i.e., > 45 (ig/dL) and reduced male fecundity or fertility (U.S. EPA.
26                  2006b). Evidence reviewed in the 1986 Pb AQCD  (U.S. EPA. 1986a) also demonstrated
27                  that Pb exposure affects male reproductive function in humans and experimental animals.
28                  Recently published research has continued to support an association between Pb and
29                  sperm/semen production, quality, and function. Studies of Pb and male reproductive
30                  function are described in the sections below.
      November 2012                            5-580                  Draft - Do Not Cite or Quote

-------
                   5.8.4.1      Effects on Sperm/Semen Production, Quality, and
                               Function

1                  Multiple epidemiologic and toxicological studies have examined the relationship between
2                  Pb and sperm and semen production, quality, and function. These studies are summarized
3                  in the text below. In addition, recent epidemiologic studies are included in Table 5-43.
4                  All epidemiologic studies were cross-sectional with concurrent measurements of Pb
5                  levels in biological  samples and sperm-related outcomes.
    November 2012                           5-581                 Draft - Do Not Cite or Quote

-------
Table 5-43   Summary of recent epidemiologic studies of associations between Pb levels and effects on sperm
               and semen.
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Hsuetal.
(2009b)
Study
Location
and
Years
Taiwan
NS
Study Population
Men working at a
battery plant
N=80
Methodological
Details
Occupational cohort
study (cross-sectional)
using ANOVA and
linear regression
Pb Biomarker
or Exposure
Measurement
Blood Pb
Categorized into 3
groups:
<25 ug/dL,
25-45 ug/dL,
>45 ug/dL
Mean Pb
(SD)
in ug/dL
40.2
Adjusted Effect Estimates
p-values for difference across the
three groups were <0.05 for:
sperm head abnormalities, sperm
neck abnormalities, sperm
chromatin structure assay
(aT, COMPaT)
p-values for difference across the
Potential
Confounders
Adjusted for in
Analysis
Smoking status
                                                                                            three groups were >0.05 for:
                                                                                            semen volume, sperm count,
                                                                                            motility, sperm tail abnormalities,
                                                                                            sperm immaturity, computer-
                                                                                            assisted semen analysis, % sperm
                                                                                            with ROS production


                                                                                            Coefficients for regression
                                                                                            analysis with blood Pb:
                                                                                            Morphologic abnormality 0.271 (p-
                                                                                            value O.0001)
                                                                                            Head abnormality 0.237 (p-value
                                                                                            0.0002)
                                                                                            aT 1.468 (p-value 0.011)
                                                                                            COMPaT 0.233 (p-value 0.21)
Kasperczyk et Poland
al. (2008) NS






Healthy, non-smoking,
fertile men that worked
at the Zn and Pb
Metalworks

NControls=14
' *low exposure"*^
l^high exposure"*^*?


Occupational cohort
study (cross-sectional)
using Kruskal-Wallis
ANOVA and
Spearman's coefficient
for non-parametric
correlation



Blood Pb; seminal
fluid Pb

Categorized as:
high exposure
workers
(blood Pb
40-81 ug/dL),
low exposed
workers
(blood Pb
Blood Pb
High
exposure
workers:
53.1 (2.05)
Low
exposure
workers:
34.7 (0.83)
Controls:
8.47 (0.54)
Mean (SE) None
Sperm volume (ml)
Controls:
2.94 (0.32)
Low exposure:
2.89 (0.22)
High exposure:
2.98 (0.22)
(p-value for ANOVA: 0.993)

November 2012
5-582
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)	
Study
Location
and
Years	Study Population
Methodological
Details
Pb Biomarker
or Exposure
Measurement
Mean Pb
(SD)
in ug/dL
Adjusted Effect Estimates
                                Potential
                                Confounders
                                Adjusted for in
                                Analysis
                                                                         25-40 ug/dL),
                                                                         and controls
                                                                         (office workers
                                                                         with no history of
                                                                         occupational Pb
                                                                         exposure)
                                                                           Seminal
                                                                           plasma Pb
                                                                           High
                                                                           exposure
                                                                           workers:
                                                                           2.02 (0.23)
                                                                           Low
                                                                           exposure
                                                                           workers:
                                                                           2.06 (0.40)
                                                                           Controls:
                                                                           1.73(0.16)
                                                      Sperm cell count (mln/mL)
                                                      Controls:
                                                      43.1 (7.0)
                                                      Low exposure:
                                                      44.6(10.1)
                                                      High exposure:
                                                      42.2 (5.86)
                                                      (p-value forANOVA: 0.400)


                                                      Normal morphology (%)
                                                      Controls:
                                                      63.3 (2.7)
                                                      Low exposure:
                                                      57.3 (2.5)
                                                      High exposure:
                                                      58.4(2.1)
                                                      (p-value forANOVA: 0.266)
                                                                                                         Progressively motile sperm
                                                                                                         after 1 h (%)
                                                                                                         Controls:
                                                                                                         16.4(3.2)
                                                                                                         Low exposure:
                                                                                                         14.8(2.6)
                                                                                                         High exposure:
                                                                                                         10.5(1.9)
                                                                                                         (p-value forANOVA: 0.217)


                                                                                                         Motile sperm after 24 h (%)
                                                                                                         Controls:
                                                                                                         4.4(1.8)
                                                                                                         Low exposure:
                                                                                                         7.3(1.7)
                                                                                                         High exposure:
                                                                                                         3.1 (0.8)
                                                                                                         (p-value forANOVA: 0.188)
                                                                                                         p-value for correlation between
                                                                                                         blood Pb and sperm cell motility
                                                                                                         after! h: 0.011
November 2012
                                5-583
                       Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Naha and
Manna (2007)

















Naha and
Chowdhury
(2006)





















Study
Location
and
Years Study Population
Bangalore, Non-occupationally
India exposed controls and
NS occupationally exposed
workers

NControls=50
'^low exposure"^"-"
N-9D
nigh exposure *-**











Kolkata, Men aged 31 -45 that
India were non-
NS occupationally exposed
controls and
occupationally exposed
workers)


NControls=50
'^low exposure"^"-"
N-^n
nigh exposure ^u













Methodological
Details
Occupational cohort
study using ANOVA,
Student's t-test, and
Scheffe's F-test















Occupational cohort
study using ANOVA,
Student's t-test, and
Scheffe's F-test





















Pb Biomarker
or Exposure
Measurement
Categorized by
work history as
controls, low
exposure (7-10 yr
of exposure for 8
h/day) and high
exposure (>10 yr
of exposure for 8
h/day)












Categorized by
work history as
controls, low
exposure (7-1 0 yr
of exposure for 8
h/day) and high
exposure (>10yr
of exposure for 8
h/day)
















Mean Pb
(SD)
in ug/dL
Blood Pb
measurement
Controls
10.25(2.26)
Low
exposure
50.29 (3.45)
High
exposure
68.26 (2.49)
Semen Pb
measurement
Controls 2.99
(0.76)
Low
exposure
15.85(1.95)
High
exposure
25.30 (2.28)
Blood Pb
measurement
Controls
13.62 (2.45)

Low
exposure
48.29 (4.91)
High
exposure
77.22 (1 .25)

Semen Pb
measurement
Controls 3.99
(1.36)
Low
exposure
10.85(0.75)
High
exposure
18.30(2.08)


Potential
Confounders
Adjusted for in
Adjusted Effect Estimates Analysis
p-values for difference across the None
three groups for mean values of
semen profiles were <0.01 for:
liquefaction time, seminal volume,
sperm count, sperm DNA hyploidy,
sperm morphological abnormality,
sperm motility, sperm ATPase
activity, seminal plasma fructose,
seminal plasma total protein,
seminal plasma free amino acid,
seminal plasma cholesterol










p-values for difference across the None
three groups for mean values of
semen profiles were <0.01 for:
sperm count, sperm protein,
sperm DNA hyploidy, sperm DNA,
sperm RNA, sperm viability, sperm
membrane lipid peroxidation,
seminal plasma total ascorbate,
seminal plasma DHAA, sperm
ATPase activity, sperm motility,
sperm velocity, seminal plasma
fructose










November 2012
5-584
Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)	
Study
Location
and
Years	Study Population
Methodological
Details
Pb Biomarker
or Exposure
Measurement
Mean Pb
(SD)
in ug/dL
Adjusted Effect Estimates
                                Potential
                                Confounders
                                Adjusted for in
                                Analysis
Telisman et al.    Croatia      Men aged 19-55, never
(2007)           2002-2005   occupationally exposed
                            to metals and going to a
                            clinic for infertility
                            examination or for
                            semen donation to be
                            used for artificial
                            insemination
                            N=240
                                   Cross-sectional study
                                   using linear multiple
                                   regression
                                                                          Blood Pb
                                         Median: 4.92   Standardized regression
                                         (range         coefficients for log blood Pb (units
                                         1.13-14.91)     not given)
                                                       Immature sperm: 0.13 (p-value
                                                       <0.07)
                                                       Pathologic sperm: 0.31 (p-value
                                                       <0.0002)
                                                       Wide sperm: 0.32 (p-value
                                                       O.0001)
                                                       Round sperm: 0.16 (p-value
                                                       <0.03)
                                                                 Cd, Cu, Zn, Se,
                                                                 age, smoking
                                                                 status, alcohol use
                                                                                                           Coefficients and p-values not
                                                                                                           given if not statistically significant:
                                                                                                           semen volume, sperm
                                                                                                           concentration, slow sperm, short
                                                                                                           sperm, thin sperm, amorph sperm
November 2012
                                 5-585
                        Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)	
Study
Location
and
Years	Study Population
Methodological
Details
Pb Biomarker
or Exposure
Measurement
Mean Pb
(SD)
in ug/dL
Adjusted Effect Estimates
                                 Potential
                                 Confounders
                                 Adjusted for in
                                 Analysis
Meeker et al.      Michigan     Men aged 18-55 going
(2008)           NS          to infertility clinics
                             (distinction not made
                             between clinic visits for
                             male or female fertility
                             issues)


                             N=219
                                    Cross-sectional study
                                    using multiple logistic
                                    regression
                                                                           Blood Pb
                                          Median: 1.50   OR (95% Cl) for having below
                                          (IQR 1.10,      reference-level semen parameters
                                          2-°°)          Concentration
                                                        Istquartile: 1.00 (ref)
                                                        2nd quartile: 0.88 (0.32, 2.44)
                                                        3rd quartile: 2.58 (0.86, 7.73)
                                                        4th quartile: 1.16(0.37,3.60)
                                                                  Age, smoking
                                                                  status
                                                                                                                             Models with
                                                                                                                             multiple metals
                                                                                                                             included: smoking
                                                                                                                             status, Mo, Mn,
                                                                                                                             Cd, and Hg
                                                                                                             Motility
                                                                                                             Istquartile: 1.00 (ref)
                                                                                                             2nd quartile: 1.04(0.43,2.53)
                                                                                                             3rd quartile: 1.95(0.70,5.46)
                                                                                                             4th quartile: 1.66(0.64,4.29)
                                                                                                                             Considered but did
                                                                                                                             not include: BMI,
                                                                                                                             race
                                                                                                             Morphology
                                                                                                             Istquartile: 1.00 (ref)
                                                                                                             2nd quartile: 0.83 (0.37, 1.87)
                                                                                                             3rd quartile: 1.41 (0.54,3.67)
                                                                                                             4th quartile: 1.18(0.50,2.79)


                                                                                                             Models with adjustment for
                                                                                                             multiple metals
                                                                                                             Concentration
                                                                                                             Istquartile: 1.00 (ref)
                                                                                                             2nd quartile: 0.89 (1.57, 2.89)
                                                                                                             3rd quartile: 3.94(1.15, 13.6)
                                                                                                             4th quartile: 2.48(0.59, 10.4)
November 2012
                                 5-586
                        Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Slivkova et al.
(2009)
Study
Location
and
Years
NS
Study Population
Men aged 22-48
undergoing semen
analysis at an infertility
clinic
Methodological
Details
Cross-sectional study
using correlation
Pb Biomarker
or Exposure
Measurement
Semen Pb
Mean Pb
(SD)
in ug/dL
1 .49 mg/kg
(0.40 mg/kg)
Adjusted Effect Estimates
Correlation between Pb and
flagellum ball : -0.39 (p-value not
given)
Potential
Confounders
Adjusted for in
Analysis
None
                                                                                                              'correlations not given for any
                             N=47                                                                            other sperm pathological changes
                                                                                                              (therefore assume not statistically
                                                                                                              significant): broken flagellum,
                                                                                                              separated flagellum, separated
                                                                                                              flagellum, small heads, retention
                                                                                                              of cytoplasmic drop, other
                                                                                                              pathological spermatozoa, large
                                                                                                              heads, acrosomal changes, and
                                                                                                              knob twisted flagellum
November 2012                                   5-587                      Draft - Do Not Cite or Quote

-------
Reference
(Studies are
presented in
order of first
appearance in
the text of this
section)
Mendiola et al.
(2011)

























Study
Location
and Methodological
Years Study Population Details
Spain Men attending infertility Case-control study
2005-2007 clinics and classified as using multiple linear
either normal sperm regression
(controls) or oligo-
astheno-
teratozoospermia
(cases) based on WHO
semen quality criteria

NControls=31
NCases=30



















Pb Biomarker Mean Pb
or Exposure (SD)
Measurement in ug/dL
Seminal plasma Seminal
Pb plasma: 2.90
OQR2.70,
Whols blood*
9.50 (IQR
7.50, 11.90)

Blood
plasma: 2.90
(IQR 2.70,
3.10)

Cases:
Seminal
plasma* 3 0
(0.3)
Whole blood:
9.8 (2.3)
Blood
plasma: 2.9
(0.2)

Controls:
Seminal
plasma: 2.9
(0.3)
Whole blood:
9.7 (2.3)
Blood
plasma: 2.9
(0.3)
Potential
Confounders
Adjusted for in
Adjusted Effect Estimates Analysis
P (95% Cl) Age, BMI, number
Sperm concentration of cigarettes/day
Seminal plasma: -1 .0 (-3.1 , 2.3)
Whole blood: -0.2 (-1.7, 1.6)
Blood plasma: 0.08 (-4.1 , 5.2)

% Immotile sperm
Seminal plasma: 1 .5 (0.37, 1 .9)
Whole blood: 0.05 (-0.32, 0.43)
Blood plasma: -0.49 (-1 .8, 0.62)

% Morphologically normal sperm

Seminal plasma: -0.54 (-3.1 , 2.0)
Whole blood: -0.31 (-1.5, 0.89)
Blood plasma: -0.08 (-3.5, 3.4)

"Units not given (assume 1 ug/dL)
Note: No correlation in Pb levels
among bloods or seminal plasma.
There was correlation between Pb
and other metals (Cd and Hg)
within each body fluid. Other
metals were not controlled for in
models



November 2012
5-588
Draft - Do Not Cite or Quote

-------
 1                  International epidemiologic studies of men occupationally exposed to Pb have reported
 2                  on associations between Pb exposure or biomarker levels and sperm count and quality
 3                  and semen quality. In most of these occupational studies, mean blood Pb levels over
 4                  40 (ig/dL have been reported for individuals occupationally exposed to Pb. In addition,
 5                  they did not control for other potential occupational exposures. A study performed in
 6                  Taiwan among men with high levels of blood Pb reported that men with higher blood Pb
 7                  levels had increased sperm head abnormalities, increased sperm DNA denaturation, and
 8                  increased sensitivity to denaturation compared to men with lower blood Pb levels (Hsu et
 9                  al.. 2009b). No difference was detected between three Pb exposure groups and semen
10                  volume, sperm count, motility, velocity, and reactive oxygen species production. A
11                  similar study in Poland included employees exposed to Pb and compared them with a
12                  group of male  office workers (Kasperczyk et al., 2008). Pb levels measured in seminal
13                  fluid were slightly higher among those in the exposed groups although they were not
14                  statistically different from the levels in the control group. No difference was observed for
15                  semen volume, sperm count,  or sperm morphology among the groups. Sperm motility
16                  was lower in the highest exposure group compared to both the control and moderate
17                  exposure groups. Lipid peroxidation, which can induce tissue damage in sperm via
18                  reactive oxygen species, was greater in the highest exposure group compared to the
19                  controls. Studies performed in India (Naha and Manna. 2007; Naha and Chowdhury.
20                  2006) reported that men in the highest exposure group (men working  in battery or paint
21                  manufacturing plants for 10-15 years for 8 hours/day) had mean blood Pb levels of
22                  77.22 (ig/dL (Naha and Chowdhury. 2006) and 68.26 (ig/dL (Naha and Manna. 2007).
23                  Control groups in these studies (those without occupational Pb exposure) had mean blood
24                  Pb levels below 15 (ig/dL. Increases in levels of Pb in semen were also noted across
25                  exposure groups. Both studies report decreases in sperm count and in sperm velocity and
26                  motility with increasing Pb exposure. Higher Pb exposure was also associated with
27                  greater hyploidy of sperm DNA and morphologic abnormalities (Naha and Manna. 2007;
28                  Naha and Chowdhury. 2006). Decreased viability and increased lipid peroxidation were
29                  detected (Naha and Chowdhury. 2006).

30                  A few studies examined blood or seminal plasma Pb levels and semen quality of men at
31                  infertility clinics (Mendiolaet al., 2011; SlivkovaetaL 2009; Meeker etal., 2008)
32                  (Telisman et al.. 2007). In general, these men had lower levels of Pb biomarkers than
33                  men who were occupationally exposed, but the studies are limited by the strong
34                  possibility of selection bias related to the recruitment of men attending infertility clinics.
35                  A study performed in Croatia recruited men attending a clinic for infertility examination
36                  or to donate semen for use in artificial inseminations, who had never been occupationally
37                  exposed to metals and therefore had lower blood Pb levels than the occupational studies
38                  (although leaded gasoline was still sold during the study period) (Telisman et al.. 2007).
39                  Increased blood Pb was associated with increased percentages of pathologic sperm, wide

      November 2012                            5-589                  Draft - Do Not Cite or Quote

-------
 1                   sperm, and round sperm. There was also a slight increase in immature sperm although it
 2                   was not statistically significant. Similar results were seen when other biomarkers for Pb
 3                   (erythrocyte protoporphyrin and 5-aminolevulinic acid dehydratase [ALAD]) were used
 4                   instead. This study controlled for multiple potential confounders, including other metals.
 5                   Meeker et al. (2008) detected no associations between higher blood Pb and semen
 6                   concentration, morphology, or motility (although a slight positive trend was observed
 7                   between higher Pb levels and motility in unadjusted models). In models that include
 8                   multiple metals, blood Pb was associated with being below the WHO limit of sperm
 9                   concentration levels (less than 20 million sperm/mL), although the 95% CI: was wide for
10                   the 4th quartile of Pb levels and included the null. The precision of estimates in this study
11                   was extremely low. Slivkova et al. (2009) reported a negative correlation between semen
12                   Pb and pathological changes in sperm (specifically, flagellum ball), but no correlations
13                   were observed for other alterations in the sperm. Another study reported a positive
14                   association between seminal plasma Pb concentration and percentage of immotile sperm,
15                   but this analysis did not adjust for exposure to other metals reported to be correlated with
16                   Pb concentration in the seminal plasma (Mendiola et al.. 2011). No association was
17                   observed for seminal plasma Pb concentration and sperm concentration or percentage of
18                   morphologically normal sperm. Additionally, neither Pb levels in whole blood nor
19                   plasma were associated with sperm concentration, percentage of immotile sperm, or
20                   percentage of morphologically normal sperm.

21                   Extensive evidence in the toxicological literature demonstrates that Pb exposure is
22                   detrimental to the quality and overall health  of testicular germ cells, affecting sperm and
23                   semen quality and production. Earlier Pb AQCDs contained studies of Pb-induced
24                   decreased sperm counts, decreased sperm production rate, and dose-dependent
25                   suppression of spermatogenesis in adult rodents with 30 day drinking water Pb exposure
26                   lYSokol and Berman. 1991). blood Pb level 35 and 37 (ig/dL; (Sokoletal.. 1985). blood
27                   Pb level 34 (ig/dL; (Sokol. 1989). blood Pb levels <43 (ig/dL]. Chronic Pb exposure
28                   (15 weeks) in adult male rabbits,  resulting in blood Pb of 24  (ig/dL, induced statistically
29                   significant decrements in semen quality and greater testicular pathology (Moorman et al..
30                   1998) with dosing by subcutaneous injection, loading dose of 0.2-3.85 mg/kg BW
31                   Monday (M), Wednesday (W) and Fridays (F) weeks 6-10, followed by maintenance
32                   dose of 0.13-2.0 mg/kg BW Pb acetate MWF over weeks 11-20  of the  study. The
33                   2006 Pb AQCD also cited studies in which sperm from Pb exposed rats yielded lower
34                   rates of fertilization when used for in vitro fertilization of eggs harvested from unexposed
35                   females [(Sokoletal.. 1994). blood Pb level 33-46 (ig/dL].

36                   Recent studies corroborate earlier findings that Pb alters sperm parameters such as sperm
37                   count, viability, motility, and morphology. Anjum et al. (2010) exposed 50 day old male
38                   albino Wistar/NIN rats to Pb acetate (273 or 819 mg/L in drinking water, 500 or
      November 2012                             5-590                  Draft - Do Not Cite or Quote

-------
 1                   1,500 ppm, respectively) for 45 days. Affected endpoints included reduced epididymal
 2                   sperm count, motile sperm, and viable sperm, indicating decreased sperm production and
 3                   quality. Anjum did not report blood Pb values. Wistar/NIN rats (1,500 ppm Pb acetate in
 4                   drinking water for 70 days) supplemented with the herb Centella asiatica (Sainath et al..
 5                   2011) had significant attenuation of the Pb-induced changes observed by Anjum et al.
 6                   (2010).  Pillai et al. (2012) found gestational and lactational treatment with Pb acetate in
 7                   Charles Foster rats (subcutaneous injection of 0.05 mg/kg BW/day) induced effects on
 8                   sperm in adults (PND65) including significant decreases in testicular sperm count,
 9                   epididymal sperm count, and sperm motility. These findings are consistent with
10                   Pb-associated  effects on sperm and male reproductive organs in wildlife from the
11                   ecological literature including deer, Asian earthworms, rainbow trout, marine worms, and
12                   the fathead minnow (see Sections 7.3.12.1 and 7.4.12.1  from the ecology terrestrial and
13                   aquatic  reproduction sperm sections).

14                   Pb exposure has been shown to affect the male reproductive organs, as is seen with
15                   histological or morphological changes. Studies included in previous Pb AQCDs showed
16                   that histological and ultrastructural damage to the testes or seminiferous tubules was seen
17                   in non-human primates with chronic oral Pb exposure (daily Pb exposure, gelatin
18                   capsule; control plus 3 treatment groups: (1) infancy exposure group [PNDO-PND400,
19                   resulting in maximum blood Pb level of 36 ug/dL], (2) post-infancy exposure group
20                   [PND300 up to 10 years of age, resulting in maximum blood Pb level of 33  ug/dL], and
21                   (3) lifetime exposure group [PNDO up to 10 years of age, resulting in maximum blood Pb
22                   level of 32 ug/dL]) (Foster et al..  1998; Singh etal.. 1993a). Rodent studies using i.p.
23                   injections of Pb also showed ultrastructural damage to structures involved in
24                   spermatogenesis (blood Pb level after i.p. injection treatment for 16 days: 7.4 ug/dL)
25                   (Murthy et al.. 1995). More recently, Salawu et al. (2009) observed a decrease in absolute
26                   testicular weight after Pb exposure (adult male SD rats, 10,000 ppm Pb acetate in
27                   drinking water for 8 weeks). Anjum  et al. (2010) reported decreased testicular and
28                   epididymal weights of male rats exposed to Pb acetate (500 or 1,500 ppm Pb acetate in
29                   drinking water for 45 days) which were significantly attenuated with Pb co-exposure to
30                   the herb Centella asiatica (Sainath et al.. 2011). Pb induced morphological abnormalities
31                   in sperm in a concentration-dependent manner (Allouche et al.. 2009; Oliveiraet al..
32                   2009: Salawu  et al.. 2009: Shan et al.. 2009: Tapisso et al.. 2009: Massanvi  et al.. 2007:
33                   Wang et al.. 2006a). Sperm abnormalities reported after Pb exposures were  amorphous
34                   sperm head, abnormal tail, and abnormal neck. Dong et al. (2009) reported decreased
35                   epididymis and body weights in mice after an eight-week exposure to 6,000 ppm
36                   Pb acetate in drinking water (adult male Kunming mice, 8-week exposure).  Oliveira et al.
37                   (2009) observed a negative correlation between Pb dose and intact acrosomes (8 week
38                   old ICR-CD1 mice, subcutaneous injection of 74 and 100 mg PbCl2/kg body weight for
39                   four consecutive days). Rubio et al. (2006)  (adult mice treated with i.p. injections of 8, 16

      November 2012                            5-591                 Draft - Do Not Cite or Quote

-------
 1                  or 24 mg/kg of Pb acetate for 35 days), Biswas and Ghosh (2006) (adult male Wistar rats,
 2                  i.p. injection of 8 mg/kg body weight Pb acetate for 21 days). Rubio et al. (2006) and
 3                  Biswas and Ghosh (2006) also observed a Pb-induced decrease in seminal vesicle and
 4                  ventral prostate weights. Rubio et al. (2006) reported that Pb acetate, in an exposure
 5                  concentration-dependent manner (8-24 mg/kg body weight), reduced the length of certain
 6                  stages of the spermatogenic cycle of rat seminiferous tubules and thus affected
 7                  spermatogenesis. Oral Pb acetate exposure (25 mg/kg bw in drinking water for 3 months,
 8                  resulting in blood Pb level of 5.3 (ig/dL) to adult male albino rats produced significant
 9                  histological seminiferous tubule damage (epithelium, spermatocytes, acrosomes) that was
10                  attenuated with ascorbic acid treatment (Pb exposure + 100 mg/kg bw/day ascorbic acid,
11                  resulting in blood Pb level of 4.7 (ig/dL) (El Shafai et al.. 2011). However, the majority
12                  of studies did not observe a statistically significant difference in body weight or
13                  reproductive organ weights after Pb exposure at the doses used in the studies. Not all of
14                  the aforementioned studies observed changes in every parameter. This may be due to the
15                  use of different strains or species, chemical form of the Pb compound administered,
16                  dosage schedule, duration of exposure, and age of animals at the time of the study
17                  (Oliveira et al.. 2009).Data from recent studies suggested that the generation of reactive
18                  oxygen species (ROS) in the male reproductive tissues, which can then affect antioxidant
19                  defense systems of cells (Pandya et al.. 2010) (adult male Charles Foster rats, Pb acetate
20                  0.025 mg/kg body weight/day i.p. for 8 weeks) contributes to the MOA of Pb damage to
21                  the male reproductive organs and sperm or semen. Salawu et al. (2009) observed a
22                  statistically significant increase in malondialdehyde (MDA, oxidative stress marker) and
23                  a significant decrease in the activity of antioxidant enzymes superoxide dismutase (SOD)
24                  and catalase (CAT) in plasma and testes of adult male Sprague Dawley rats after
25                  administration of 10,000 ppm Pb acetate in drinking water for 8 weeks. Supplementation
26                  with tomato paste (used as a source of antioxidants) reduced Pb-induced ROS production
27                  and prevented the Pb-induced increase  in MDA formation and decrease in SOD and CAT
28                  activity. Furthermore, co-treatment of Pb with substances that are known to have
29                  antioxidant properties [i.e., tomato paste, Maca (Lepidium meyenii), and ascorbic acid]
30                  prevented the Pb-induced reduction in sperm count, sperm motility, and sperm viability
31                  (Salawu et al.. 2009: Shan et al.. 2009: Madhavi et al.. 2007: Rubio et al.. 2006: Wang et
32                  al.. 2006a).

33                  Recent studies continue to demonstrate that Pb may be directly toxic to mature
34                  spermatozoa (adult Algerian mice, Pb acetate 21.5 mg/kg BW/every other day i.p. for 11
35                  or 21 days) (Tapisso et al., 2009; Hernandez-Ochoa et al.. 2006) (adult NMRI mice,
36                  6,000 ppm Pb chloride  in drinking water for 16 weeks) as well as primary spermatocytes
37                  (adult male Wistar rats  were treated with drinking water  containing 250mg/L Pb acetate
38                  for) (Nava-Hernandez  et al.. 2009: Rafique et al.. 2009)  (adult albino rats, 10 mg/kg BW
39                  Pb chloride i.p. once daily for 8 weeks). Nava-Hernandez et al. (2009) exposed two

      November 2012                            5-592                 Draft - Do Not Cite or Quote

-------
 1                  groups of adult male rodent to Pb via drinking water (LI and L2, 250mg/L or 500 mg/L
 2                  Pb acetate starting at PND60 for 90 days). They found significant increases in spermatid
 3                  DNA damage with Pb exposure. In their study, all Pb-treated animals had blood Pb levels
 4                  statistically significantly higher than controls (Ll:19.54 (ig/dL and L2:21.90 (ig/dL); no
 5                  statistically significant difference in blood Pb levels existed between the two Pb exposure
 6                  groups likely because the L2 group drank less water than did the LI group. Piao et al.
 7                  (2007) reported that Pb exposure caused DNA damage to sperm; the Pb exposed group
 8                  had a blood Pb of 67 (ig/1. Piao et al. (2007) also examined the effect of Zn
 9                  supplementation on Pb-induced sperm aberrations and found that the proportion of
10                  abnormal sperm was statistically significantly higher in the Pb group and the Pb+Zn
11                  group than in controls (25 mg/kg Pb acetate i.p., 4 mg/kg Zn  acetate i.p., both Pb acetate
12                  and Zn acetate, once every two days, for 2 weeks). However, the proportion of abnormal
13                  sperm in Pb+Zn group was statistically significantly lower than in Pb alone group.
14                  Hernandez-Ochoa et al. (2006) reported that Pb reaches the sperm nucleus in the
15                  epididymis of mice chronically exposed (16 weeks in adult animals) to Pb (resulting in
16                  mean blood Pb of 75.6 (ig/dL) by binding to nuclear sulfhydryl groups from the
17                  DNA-protamine complex, increasing sperm chromatin condensation, and thereby
18                  interfering with the sperm maturation process without altering sperm quality parameters.
19                  Tapisso et al. (2009) observed a statistically significant increase in the number of
20                  micronuclei and frequency of sister chromatid exchange with increasing treatment
21                  duration in adult male mice administered 21.5 mg/kg body weight Pb acetate by i.p.
22                  injection (adult Algerian mice, Pb acetate 21.5 mg/kg BW/every other day for 11 or
23                  21 days). Nava-Hernandez (2009) reported a concentration-dependent increase in DNA
24                  damage in rat primary spermatocytes after a 13-week exposure period to Pb acetate in
25                  drinking water (resulting  in mean blood Pb levels between 19.5 and 21.9 (ig/dL). Rafique
26                  et al.  (2009) reported degenerative changes from pyknosis to apoptosis in primary
27                  spermatocytes (adult albino rats, 10 mg/kg BW Pb chloride i.p. once daily for 8 weeks).
28                  Hepatic expression of spermatogenic genes was transiently down-regulated in 8 week old
29                  male Wistar-Kyoto (WKY) rats in response to Pb nitrate (100 (imol single i.v. injection)
30                  3 hours after injection and recovered to baseline by 12 hours  (Nemoto et al.. 2011); this
31                  effect was not seen in the stroke-prone spontaneously hypertensive rats, which are from a
32                  WKY background, or in Sprague-Dawley rats, demonstrating strain specificity.

33                  Pb-induced apoptosis in germ cells within the seminiferous tubules is another suggested
34                  mechanism by which Pb exerts its toxic effects on sperm production and function (Wang
35                  et al.. 2006a) (Kunming mice, 2,000 ppm Pb acetate in drinking water for 14-42 days).
36                  Dong et al. (2009) reported a exposure concentration-related  increase in apoptosis in
37                  spermatogonia and spermatocytes of Kunming mice after exposure to  1,500-6,000 ppm
38                  Pb acetate in drinking water. Pb-induced testicular germ cell  apoptosis was associated
39                  with up-regulation of genes involved in the signal pathway of MAPK and death receptor

      November 2012                            5-593                  Draft - Do Not Cite or Quote

-------
 1                   signaling pathway of FAS. For instance, up-regulation of K-ras and Fas expressions was
 2                   concomitant with activation of c-fos and active caspase-3 proteins. Wang et al. (2006a)
 3                   observed an exposure concentration-dependent increase in the expression of apoptotic
 4                   markers TGF(31 and caspase-3 in spermatogenic cells, Sertoli cells, and Leydig cells.
 5                   Shan et al. (2009) (20 mg/kg BW intragastric Pb acetate for 6 weeks) also reported a
 6                   statistically significant increase in mRNA expression and protein levels of Fas, Fas-L and
 7                   caspase-3 after Pb exposure. Supplementation with ascorbic acid inhibited or reduced the
 8                   Pb-induced apoptosis in germ cells and protected testicular structure and function (El
 9                   ShafaietaL 2011;  Shan et al.. 2009; Wang et al.. 2006a) suggesting ROS generation is a
10                   major contributing  factor in decreased male fertility observed after chronic  Pb exposure.

11                   Similar to the results summarized in previous Pb AQCDs, recent epidemiologic and
12                   toxicological studies indicate that Pb exposure has effects on sperm, semen, and male
13                   reproductive organs. Consistent toxicological evidence from multiple labs with multiple
14                   species of animals showed decrements in sperm or semen quality with Pb exposure
15                   including decreased sperm counts, decreased sperm production rate, and a dose-
16                   dependent suppression of spermatogenesis. Histological damage to rodent sperm and
17                   ultrastructural damage to rodent and non-human primate seminiferous tubules has been
18                   reported. Sperm from Pb-exposed male rodents used for in vitro fertilization of eggs from
19                   unexposed females yielded a lower rate of fertilization. Also, direct effects  of Pb on
20                   rodent sperm DNA have been reported in rodents with drinking water exposure. The
21                   toxicological findings cross species and are seen in wildlife including deer, earthworms,
22                   rainbow trout, marine worms and fathead minnow.  In studies of men exposed to Pb in
23                   occupational settings, associations were observed between blood Pb levels of at least
24                   25 (ig/dL and sperm count and quality. Multiple epidemiologic studies of occupational
25                   cohorts included control populations with high blood Pb levels (close to or greater than
26                   10 (ig/dL), which makes identification of effects at lower levels difficult.  Occupational
27                   studies had limited consideration for potential confounding factors, such as other
28                   workplace exposures. An epidemiologic study of men attending a clinic for purposes of
29                   infertility exam or semen donation demonstrated an inverse relationship between Pb
30                   levels and sperm and semen quality (Telisman et al.. 2007).  This study also controlled for
31                   other metals in the  analyses. Other studies of men at infertility clinics had greater
32                   imprecision in their estimates, less control for confounding (such as other metals), and/or
33                   small sample sizes. Additionally, studies limited to men at infertility clinics may suffer
34                   from selection bias and are not generalizable. Future epidemiologic studies are warranted
35                   to determine whether this association is observed at lower Pb levels.
      November 2012                             5-594                  Draft - Do Not Cite or Quote

-------
                   5.8.4.2      Effects on Hormone Levels

1                  The 2006 Pb AQCD (U.S. EPA. 2006b) provided evidence that Pb acts as an endocrine
2                  disrupter in males at various points along the hypothalamic-pituitary-gonadal axis. The
3                  2006 Pb AQCD also reported inconsistencies in the effects of Pb exposure on circulating
4                  testosterone levels. Recent epidemiologic and toxicological studies are reported below.
5                  Epidemiologic studies are summarized in Table 5-44. Epidemiologic studies were cross-
6                  sectional; biological samples used for the measurement of Pb were measured
7                  concurrently with hormone levels. One study estimated cumulative blood Pb.
     November 2012                            5-595                 Draft - Do Not Cite or Quote

-------
Table 5-44   Summary of recent epidemiologic studies of associations between Pb levels and hormones for
            males.
Reference
(Presented in
order of
appearance in
the text)
Telisman et al.
(2007)














Meeker et al.
(2010)

















Study
Location Pb Biomarker
and Methodological or Exposure
Years Outcome Study Population Details Measurement
Croatia FSH, LH, Men aged 19-55, never Cross-sectional Blood Pb
2002-2005 testosterone, occupationally exposed study using linear
estradiol, to metals and going to multiple
prolactin a clinic for infertility regression
examination or for
semen donation to be
used for artificial
insemination


N=240






Michigan FSH, LH, Men aged Cross-sectional Blood Pb
NS inhibin B, 18-55 going to infertility study using
testosterone, clinics (distinction not multiple linear
SHBG, FAI, made between clinic regression
testosterone visits for male or
/LH female fertility issues)

N=219













Mean Pb
(SD) Adjusted Effect
in ug/dL Estimates
Median: 4.92 Standardized regression
(range coefficients for log blood Pb
1.13-14.91) (units not given)
T6stost6ron6 '
0.21
(p-value <0.003)
Estradiol:
0.22
(p-value <0.0008)
Prolactin'
-0.18
(p-value <0.007)
Note: Coefficients and p-
values not given if not
statistically significant (LH,
FSH)
Median: Regression coefficients
1 .50 (95% Cl)
(IQR1.10, FSH
2-°°) Istquartile:
O(ref)
2nd quartile:
0.13
(-0.10, 0.37)
3rd quartile:
0.10
(-0.15, 0.35)
4th quartile:
0.07
(-0.18,0.31)
LH
1st quartile:
O(ref)
2nd quartile:
0.004
(-0.20, 0.21)

Potential
Confounders
Adjusted for
in Analysis
Age, smoking
status, alcohol
use, Cd, Cu,
Zn, Se

An interaction
between Pb
and Cd was
included in
models for
testosterone





Age, BMI,
current
smoking
Considered but
did not include:
race, income,
season












November 2012
5-596
Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order of
appearance in
the text)
Study
Location
and
Years      Outcome
Study Population
Methodological
Details
Pb Biomarker
or Exposure
Measurement
Mean Pb
(SD)
in ug/dL
Adjusted Effect
Estimates
                                                                                                  Potential
                                                                                                  Confounders
                                                                                                  Adjusted for
                                                                                                  in Analysis
                                                                                                                  3rd quartile:
                                                                                                                  0.13
                                                                                                                  (-0.09, 0.35)
                                                                                                                  4th quartile:
                                                                                                                  0.88
                                                                                                                  (-0.14,0.29)


                                                                                                                  Inhibin B
                                                                                                                  1st quartile:
                                                                                                                  O(ref)
                                                                                                                  2nd quartile:
                                                                                                                  -6.45
                                                                                                                  (-27.2, 14.3)
                                                                                                                  3rd quartile:
                                                                                                                  -4.62
                                                                                                                  (-26.6,17.4)
                                                                                                                  4th quartile:
                                                                                                                  -7.79
                                                                                                                  (-29.0,13.4)


                                                                                                                  Testosterone
                                                                                                                  1st quartile:
                                                                                                                  O(ref)
                                                                                                                  2nd quartile:
                                                                                                                  28.6
                                                                                                                  (-6.82,64.1)
                                                                                                                  3rd quartile:
                                                                                                                  15.8
                                                                                                                  (-21.8, 53.3)
                                                                                                                  4th quartile:
                                                                                                                  39.9
                                                                                                                  (3.32, 76.4)


                                                                                                                  SHBG
                                                                                                                  1st quartile:
                                                                                                                  O(ref)
                                                                                                                  2nd quartile:
                                                                                                                  -0.01
                                                                                                                  (-0.16,0.15)
                                                                                                                  3rd quartile:
                                                                                                                  0.04
                                                                                                                  (-0.12,0.21)
                                                                                                                  4th quartile:
                                                                                                                  0.07
                                                                                                                  (-0.10,0.23)
November 2012
                                 5-597
                                Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order of
appearance in
the text)
Study
Location
and
Years     Outcome
Study Population
Methodological
Details
Pb Biomarker
or Exposure
Measurement
Mean Pb
(SD)
in ug/dL
Adjusted Effect
Estimates
                                                                                               Potential
                                                                                               Confounders
                                                                                               Adjusted for
                                                                                               in Analysis
                                                                                                              FAI
                                                                                                              1st quartile:
                                                                                                              O(ref)
                                                                                                              2nd quartile:
                                                                                                              0.8
                                                                                                              (-0.04, 0.20)
                                                                                                              3rd quartile:
                                                                                                              0.03
                                                                                                              (-0.10,0.17)
                                                                                                              4th quartile:
                                                                                                              0.08
                                                                                                              (-0.05,0.21)


                                                                                                              Testosterone /LH
                                                                                                              1st quartile: 1.00 (ref)
                                                                                                              2nd quartile:
                                                                                                              0.07
                                                                                                              (-0.16,0.30)
                                                                                                              3rd quartile:
                                                                                                              -0.05
                                                                                                              (-0.29,0.19)
                                                                                                              4th quartile:
                                                                                                              0.07
                                                                                                              (-0.17,0.31)
November 2012
                                5-598
                               Draft - Do Not Cite or Quote

-------
Reference
(Presented in Study
order of Location
appearance in and
the text) Years Outcome
Mendiola et al. Spain FSH, LH,
(2011) 2005-2007 testosterone


























Pb Biomarker
Methodological or Exposure
Study Population Details Measurement
Men attending infertility Case-control Seminal plasma
clinics and classified as study using Pb
either normal sperm multiple linear Blood Pb
(controls) or oligo- regression
astheno-
teratozoospermia
(cases) based on WHO
semen quality criteria
NCases=30
N-^1
controls"0 '






















Mean Pb
(SD)
in ug/dL
Seminal
plasma: 2.90
(IQR 2.70,
3.20)
Whole blood:
9.50 (IQR
7.50,11.90)
Blood
plasma: 2.90
(IQR 2.70,
3.10)


Cases:
Seminal
plasma' 3 0
(0.3)
Whole blood:
9.8 (2.3)

Blood
plasma: 2.9
(0.2)


Controls:
Seminal
plasma: 2.9
(0.3)
Whole blood:
9.7 (2.3)
Blood
plasma: 2.9
(0.3)



Adjusted Effect
Estimates
Linear regression p (95%
Cl)
FSH
Seminal plasma:
0.05
(-0.24, 0.39)
Whole blood:
0 04
(-0.03, 0.04)
Blood plasma:
-0.20
(-0.64, 0.25)

LH

Seminal plasma:
0.14
(-0.13,0.41)
Whole blood:
0 05
(-0.05, 0.07)
Blood plasma:
-0.07
(-0.49,0.31)

Testosterone
Seminal plasma:
0 11
(-0.10,0.31)
Whole blood:
0 01
(-0.05, 0.02)
Blood plasma:
-0.12
(-0.40,0.14)
Potential
Confounders
Adjusted for
in Analysis
Age, BMI,
number of
cigarettes/day


























                                                                                                        "Units not given (assume
                                                                                                        1 ug/dL)
November 2012
5-599
Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order of
appearance in
the text)
Hsieh et al.
(2009a)
Study
Location
and
Years Outcome
Taiwan FSH, LH,
1991-NS testosterone,
mhibm B
Study Population
Workers at a Pb-acid
battery factory with
annual blood Pb
measures
N=181
Methodological
Details
Longitudinal
occupational
cohort study using
multivariate linear
regression
Pb Biomarker
or Exposure
Measurement
Current blood Pb,
cumulative blood
Pb, time-
weighted
cumulative blood
Pb
Mean Pb
(SD)
in ug/dL
Current blood
Pb:
<10 ug/dL:
11.6%
>40 ug/dL:
17.1%
Adjusted Effect
Estimates
P from linear regression
Inhibin B
Current blood Pb:
0.40 (p-value 0.40)
Cumulative blood Pb:
0.05 (p-value 0.02)
Time-weighted cumulative
blood Pb:
1 .33 (p-value 0.007)
Potential
Confounders
Adjusted for
in Analysis
LH, FSH,
testosterone,
age, smoking
status, alcohol
use, BMI
                                                                                                                      Pearson's correlations
                                                                                                                      detected no correlations
                                                                                                                      between current blood Pb
                                                                                                                      levels and FSH, LH, or
                                                                                                                      testosterone. Cumulative
                                                                                                                      blood Pb levels were
                                                                                                                      correlated with FSH and
                                                                                                                      LH, but not testosterone.
                                                                                                                      Time-weighted cumulative
                                                                                                                      blood Pb levels were
                                                                                                                      correlated with LH, but not
                                                                                                                      FSH or testosterone.
November 2012
5-600
Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order of
appearance in
the text)
Study
Location
and
Years
Outcome     Study Population
                      Methodological
                      Details
                  Pb Biomarker
                  or Exposure
                  Measurement
                  Mean Pb
                  (SD)
                  in ug/dL
              Adjusted Effect
              Estimates
                                                                                                                 Potential
                                                                                                                 Confounders
                                                                                                                 Adjusted for
                                                                                                                 in Analysis
Naha and
Manna (2007)
Bangalore,
India
NS
FSH, LH,
testosterone
Non-occupationally
exposed controls and
occupationally exposed
workers
                                          NControls=50

                                          NIC
Occupational
cohort study using
ANOVA, Student's
t-test, and
Scheffe's F-test
                                          ''high exposure'
                                >sure=30

                                   .=20
Categorized by
work history as
controls, low
exposure (7-10 yr
of exposure for 8
h/day) and high
exposure (>10yr
of exposure for 8
h/day)
Blood Pb
measurement
Controls
10.25(2.26)
Low
exposure
50.29 (3.45)
High
exposure
68.26 (2.49)


Semen Pb
measurement
Controls 2.99
(0.76)
Low
exposure
15.85(1.95)
High
exposure
25.30 (2.28)
Mean FSH (SD)
Control:
2.69(1.22)
Low exposure:
2.58 (1.94)
High exposure:
2.16(0.99)
p-values for difference
>0.05
                                                                                                                  Mean LH (SD)
                                                                                                                  Control:
                                                                                                                  5.14(2.35)
                                                                                                                  Low exposure:
                                                                                                                  4.27 (2.52)
                                                                                                                  High exposure:
                                                                                                                  3.9 (1.69)
                                                                                                                  p-values for difference
                                                                                                                  >0.05


                                                                                                                  Mean Testosterone (SD)
                                                                                                                  Control:
                                                                                                                  5.24 (2.40)
                                                                                                                  Low exposure:
                                                                                                                  4.83(1.21)
                                                                                                                  High exposure:
                                                                                                                  4.59(1.27)
                                                                                                                  p-values for difference
                                                                                                                  >0.05
                                                                                                                                            None
November 2012
                                 5-601
                                               Draft - Do Not Cite or Quote

-------
 1                  Hormone levels were measured in a few recent epidemiologic studies. In a study of men
 2                  non-occupationally exposed to Pb in Croatia, increased blood Pb level was associated
 3                  with increasing serum testosterone and estradiol but decreasing serum prolactin
 4                  (Telisman et al.. 2007). In addition, the analysis of an interaction term for blood Pb and
 5                  blood Cd levels demonstrated a synergistic effect on increasing serum testosterone levels.
 6                  No association was observed between blood Pb and FSH or LH. This study controlled for
 7                  multiple potential confounders, including other metals. Among men recruited from
 8                  infertility clinics in Michigan, median blood Pb levels were much lower than those
 9                  observed in the other studies of Pb and hormone levels among men (Meeker et al., 2010).
10                  No association was detected between blood Pb and levels of FSH, LH, inhibin B, sex
11                  hormone-binding globulin (SHBG), free androgen index (FAI), or a measure of Leydig
12                  cell function (T/LH). A positive association between the highest quartile of blood Pb and
13                  testosterone was present, but this association did not persist when other metals were
14                  included in the model. Similarly, another study of men recruited from infertility clinics
15                  observed no association between Pb concentrations from seminal plasma, whole blood, or
16                  blood plasma and FSH, LH, or testosterone (Mendiola et al.. 2011).

17                  A study of occupationally-exposed men in Taiwan reported an association between
18                  measures of cumulative  blood Pb levels and inhibin B levels, but no association was
19                  detected when using  current blood Pb levels (Hsieh et al., 2009a). A correlation between
20                  cumulative blood Pb measures and LH levels was detected but correlations were not
21                  present when examining FSH or testosterone levels. No correlations were apparent
22                  between FSH, LH, or testosterone and current blood Pb levels. Another study of men
23                  with high blood Pb levels reported no difference in serum FSH, LH, and testosterone
24                  among the three groups  (controls: mean blood Pb 10.25 (ig/dL, low exposure: mean
25                  blood Pb 50.29 (ig/dL, high exposure: mean blood Pb 68.26 (ig/dL) (Naha and Manna.
26                  2007). This study did not assess any potential confounding factors.

27                  In a recent toxicological study, Rubio et al. (2006) observed a decrease in testosterone
28                  levels in Pb acetate-treated rats in a exposure concentration-related fashion (8-24 mg/kg
29                  body weight), and this decrease correlated with reduced lengths of spermatogenic cycle
30                  stages VII-VIII (spermiation) and IX-XI (onset of spermatogenesis).  Anjum et al. (2010).
31                  who dosed 50 day old male rats with 273 or 819 mg/L Pb acetate in drinking water (500
32                  or 1,500 ppm, respectively; blood Pb not reported), found significant decreases in  serum
33                  testosterone and testicular 3(3-HSD and  17(3-HSD levels in Pb-exposed animals versus
34                  controls. Pandya et al. (2010) reported altered hepatic steroidogenic enzyme activity.
35                  Pillai et al. (2012) found gestational and lactational exposure to Pb acetate in Charles
36                  Foster rats (subcutaneous injection of 0.05 mg/kg BW/day, blood Pb not reported)
37                  induced significant decreases in testicular 17(3-HSD and serum testosterone. Biswas and
38                  Ghosh (2006) reported a Pb-induced decrease in serum testosterone and  gonadotropins
      November 2012                            5-602                  Draft - Do Not Cite or Quote

-------
 1                   (FSH, LH) with inhibition of spermatogenesis, however, there was a statistically
 2                   significant increase in adrenal steroidogenic enzyme, A5-3(3-HSD, activity and serum
 3                   corticosterone levels indicating disruption of the adrenocortical process. Exposure
 4                   concentration-dependent decreases in serum testosterone were reported in Pb-exposed
 5                   male rats (Anjum et al., 2010). In contrast, Salawu et al. (2009) did not observe a
 6                   decrease in serum testosterone between control animals and animals administered
 7                   10,000 ppm Pb acetate in drinking water for 8 weeks. Allouche et al. (2009) not only did
 8                   not observe any statistically significant changes in serum FSH or LH, but reported an
 9                   increase in serum testosterone levels after 500-3,000 ppm Pb acetate treatment in
10                   drinking water (only statistically significant in animals administered 500 ppm Pb acetate).
11                   The results of these recent studies further support the theory that compensatory
12                   mechanisms in the hypothalamic-pituitary-gonadal axis may allow for the adaptation of
13                   exposed animals to the toxic endocrine effects of Pb (Rubio et al.. 2006; U.S. EPA.
14                   2006b).

15                   Overall, recent epidemiologic and toxicological studies report mixed findings regarding
16                   hormone aberrations in males associated with Pb exposure or Pb biomarker levels. These
17                   results are  similar to those from the 2006 Pb AQCD (U.S. EPA. 2006b)on the effects of
18                   Pb exposure on circulating testosterone levels. Epidemiologic studies are limited by their
19                   sample populations, often occupational cohorts or men at infertility clinics, which may
20                   not be generalizable. Occupational cohorts may have other exposures that confound the
21                   associations, and studies at infertility clinics are subject to selection bias. A few of the
22                   recent epidemiologic studies include important confounding  factors, such as smoking, but
23                   other factors, such as exposure to other metals, were often absent. Additionally, most
24                   studies examine concurrent Pb and hormone levels which may not reflect changes
25                   resulting from long-term exposures, as demonstrated by the longitudinal occupational
26                   cohort study. Further, in cross-sectional studies, the temporality of effects cannot be
27                   established.
                     5.8.4.3      Fertility

28                   Epidemiologic studies have been performed comparing Pb and infertility in men. The
29                   SMART study is a longitudinal study that examined the success of IVF treatment for
30                   women and their partners starting their first round of treatment (Bloom et al.. 20 lib;
31                   Bloom et al.. 2010). A small number of the male partners participated (n=16). Their mean
32                   (SD) blood Pb level was 1.50 (0.80) (ig/dL. Higher blood Pb levels were associated with
33                   greater oocyte fertilization (OR 1.08 [95% CI: 0.97, 1.21] per 1 (ig/dL increase in blood
34                   Pb when adjusted for Cd, Hg, age, cigarette smoking, race/ethnicity, and creatinine),
35                   which is not the  expected direction (Bloom et al.. 2010). However, higher blood Pb was


      November 2012                             5-603                  Draft - Do Not Cite or Quote

-------
 1                   associated with lower embryo cell number, a predictor of IVF success, and with higher
 2                   embryo fragmentation score, an inverse predictor of IVF success (OR for embryo cell
 3                   number: 0.58 [95% CI: 0.37, 0.91]; OR for embryo fragmentation score: 1.47 [95% CI:
 4                   1.11, 1.94] per 1 (ig/dL, controlled for age, race/ethnicity, cigarette smoking, creatinine,
 5                   Cd, and Hg, plus day of embryo transfer for embryo cell number) (Bloom et al.. 20 lib).
 6                   A case-control study conducted in Turkey reported that blood and seminal plasma Pb
 7                   levels were different in fertile (n=45; mean [SD] blood Pb: 23.16 [5.59] (ig/dL) and
 8                   infertile men (n=50; mean [SD] blood Pb: 36.82 [12.30] jig/dL) (p <0.001, ANOVA)
 9                   (Kiziler et al., 2007). There was no control for potential confounding factors although the
10                   relationship persisted when limited to non-smokers. Another case-control study examined
11                   occupational Pb exposure (determined by self-report of occupational exposure in the
12                   past month) and detected no difference in reported exposure for infertile (n=650) versus
13                   fertile men (n=698) (unadjusted OR 0.95 [95% CI: 0.6, 1.6]) (Gracia et al.. 2005). Blood
14                   Pb was not measured but approximately 5.0% of infertile men and 5.3% fertile men
15                   reported occupational exposure to Pb. A limitation present in these studies is that the
16                   cases included are men who are seeking help at fertility clinics; the study populations are
17                   not a sample of the general population regarding fertility. The results could be biased due
18                   to the recruitment of individuals going to an infertility clinic, who may be different than
19                   individuals suffering from infertility without knowing it or without going to a clinic.

20                   Recent animal toxicology studies assessed paternal-mediated reproductive fitness by
21                   examining the reproductive success of Pb-exposed males with non-exposed control
22                   females. Anjum et al.  (2010) found that adult male rats who were exposed to 273 or
23                   819 mg/L Pb acetate in drinking water (500 or 1,500 ppm, respectively; blood Pb not
24                   reported) spent a significantly longer time copulating than did their control littermates.
25                   The Pb-exposed males were less successful copulators with only 73% of the 0.05%
26                   Pb acetate exposed males, and 53% of the 1,500 ppm exposed males generating
27                   copulatory plugs in the unexposed female mates. While the number of pregnant females
28                   did not significantly differ from controls, Pb exposed males contributed to the formation
29                   of significantly fewer implantations/dam, and significantly fewer fetuses/dam.
30                   Pb-exposed males were able to sire offspring, but produced fewer offspring per litter. In a
31                   group of males rats  with co-exposure to Pb and the herb Centella asiatica, these
32                   reproductive decrements were  attenuated relative to rats exposed to Pb alone (adult albino
33                   male rats, 1,500 ppm Pb acetate in drinking water for 70 days) (Sainath et al., 2011).
      November 2012                             5-604                  Draft - Do Not Cite or Quote

-------
 1                  Overall, large, well-conducted epidemiologic studies of Pb exposure and fertility in males
 2                  are lacking. The few available studies reported inconsistent findings. Toxicological
 3                  studies demonstrated paternal associated subfecundity (fewer pups sired per pregnancy)
 4                  with altered mating behavior (longer time spent copulating), albeit in studies with no
 5                  blood Pb levels reported. Supplementation with antioxidants in a separate study showed
 6                  restoration of this subfecundity, possibly contributing MOA support to this decreased
 7                  fertility in male rodents.
                    5.8.4.4     Effects on Morphology and Histology of Male Sex Organs

 8                  Recent toxicological studies further support historical findings that showed an association
 9                  between Pb exposure and changes in the sex organs as well as germ cells. Histological
10                  changes of testes in Pb nitrate-treated adult animals (a single i.p. dose of 12.5, 25, or
11                  50 mg/kg of BW and were sacrificed 48 hours later) included seminiferous tubule
12                  atrophy, Sertoli cell and Leydig cell shrinkage with pyknotic nuclei (Shan et al.. 2009;
13                  Wang et al.. 2006a). dilatation of blood capillaries in the interstitium, undulation of basal
14                  membrane, and occurrence of empty spaces in seminiferous epithelium (adult male
15                  Wistar rats, single i.p. dose of 12, 25 or 50 mg/kg BW Pb acetate) (Massanyi et al..
16                  2007). Pillai et al. (2010)  found gestational and lactational exposure to Pb acetate in
17                  Charles Foster rats (subcutaneous injection of 0.05 mg/kg BW/day) induced significant
18                  decreases in absolute organ weight (testes and epididymis) and significant decreases in
19                  relative epididymal weight. Anjum et al. (2010). who exposed 50 day old male albino
20                  Wistar/NIN rats to Pb acetate (273 or 819 mg/L in drinking water, 500 or 1,500 ppm,
21                  respectively, blood Pb levels not reported) for 45 days, reported significant decreases in
22                  relative reproductive organ weight (epididymis, testis, vas deferens, and seminal vesicle)
23                  in Pb-exposed animals.
                    5.8.4.5     Summary of Effects on  Male Reproductive Function

24                  Evidence of associations between Pb exposure and male reproductive function vary by
25                  outcome. The strongest evidence of an association is the relationship observed between
26                  Pb and negative effects on sperm and semen in both recent epidemiologic and
27                  toxicological studies and studies reviewed in previous Pb AQCDs. Decrements in sperm
28                  count, sperm production rate and semen quality were reported in animal toxicological
29                  studies in rodents with drinking water Pb exposure rodents [(Sokol and Berman. 1991;
30                  Sokol et al.. 1985). (blood Pb level 34-37 (ig/dL)] and rabbits exposed to subcutaneous
31                  Pb [blood Pb levels of 25  (ig/dL, (Moorman  et al.. 1998)]. Rodents exposed to Pb had
32                  direct effects of Pb on sperm DNA, i.e., elevated levels of DNA damage [(Nava-

      November 2012                            5-605                 Draft - Do Not Cite or Quote

-------
 1                  Hernandez et al.. 2009). blood Pb levels 19 and 22 (ig/dL]. Histological or ultrastructural
 2                  damage to the male reproductive organs were reported in studies from rodents [(El Shafai
 3                  et al.. 2011). blood Pb level 5.1 (ig/dL] and non-human primates [(Singh et al.. 1993a).
 4                  blood Pb level 43 (ig/dL]. Subfecundity has been reported in unexposed females mated to
 5                  Pb exposed males (decreased number of pups born/litter). Also, sperm from Pb-exposed
 6                  rats (blood Pb 33 to 46 ug/dL) used for in vitro fertilization of eggs harvested from
 7                  unexposed females yielded lower rates of fertilization (Sokol et al.. 1994). Many of the
 8                  epidemiologic studies included occupational cohorts, which had high blood Pb levels
 9                  (> 25(ig/dL), or men attending infertility clinics, which have potential selection bias.
10                  Additionally, control for confounding factors, such as other workplace exposures, was
11                  not often performed. A study of men (who were attending a clinic for an infertility exam
12                  or to donate semen for use in artificial insemination) did control for multiple factors,
13                  including other metals and smoking status,  and reported an association between blood Pb
14                  levels and some indicators of poor sperm quality (Telisman et al.. 2007). Recent
15                  toxicological studies also reported an association between Pb exposure and decreases in
16                  reproductive organ weight, organ histological changes in the testes and germ cells. Male
17                  rats exposed to Pb also showed subfecundity, in that they produced smaller litters when
18                  mated with unexposed females (Anjum et al.. 2010). Further coherence for these findings
19                  in laboratory animal models  is found in with findings in the ecological literature for the
20                  effects of Pb exposure on reduced fecundity in terrestrial and aquatic animal species
21                  (Sections 7.4.5.2. 7.3.4.2. and 7.4.5.3). Similar to the 2006 Pb AQCD (U.S. EPA. 2006b).
22                  recent epidemiologic and toxicological studies reported inconsistent results regarding
23                  hormone aberrations associated with Pb exposure. Mixed findings were also apparent
24                  among epidemiologic studies of fertility among men.
      5.8.5        Effects on Female Reproductive Function

25                  The epidemiologic studies on Pb and female reproductive function presented in the
26                  2006 Pb AQCD (U.S. EPA. 2006b) provided little evidence for an association between
27                  Pb biomarkers and effects on female reproduction and fertility. However, the 1986 and
28                  2006 Pb AQCDs (U.S.  EPA. 2006b. 1986a) reported toxicological findings that Pb
29                  exposure was associated with effects on female reproductive function that can be
30                  classified as alterations in female sexual maturation, effects on fertility and menstrual
31                  cycle, endocrine disruption, and changes in morphology or histology of female
32                  reproductive organs including the placenta. Since the 2006 Pb AQCD, many
33                  epidemiologic studies have been published regarding Pb biomarker levels in women and
34                  reproductive effects. In addition, recent toxicological studies add further knowledge of
35                  Pb-related effects on the female reproductive system.
      November 2012                            5-606                  Draft - Do Not Cite or Quote

-------
                   5.8.5.1      Effects on Female Sex Endocrine System and Estrus
                               Cycle

1                  Multiple epidemiologic studies have examined the association between blood Pb levels
2                  and hormone levels and the estrus cycle. Epidemiologic studies (characterized in Table
3                  5-45) were cross-sectional in design, analyzing measures of Pb and hormones that were
4                  collected either concurrently or close in time. These studies support the toxicological
5                  findings, which are the major body of evidence on endocrine effects of Pb.
    November 2012                           5-607                 Draft - Do Not Cite or Quote

-------
Table 5-45   Summary of recent epidemiologic studies of associations between Pb levels and hormones for
                females.
Reference
(Presented in
order
appearance in
the text)
Study,
Location,
and
Years
Outcome      Study Population
                             Mean
Methodological   Pb          Pb (SD)   Adjusted Effect
Details	Biomarker  in ug/dL  Estimates	
                                                                                              Potential Confounders
                                                                                              Adjusted for in
                                                                                              Analysis
                U.S.
                1988-1994
           FSH, LH       Women aged 35-60 from
                         the NHANES III study
                         N=3375
                                     Cross-sectional     Blood Pb      2.8        Linear regression
                                     study using linear                          slope (95% Cl)
                                     regression                                for log-transformed
                                                                             Pb
                                                                             FSH:
                                                                             Post-menopausal
                                                                             22.2(13.5, 30.8)
                                                                             Pregnant
                                                                             0.1 (-0.1,0.3)
                                                                             Menstruating at
                                                                             time of exam
                                                                             2.1 (-2.1,6.3)
                                                                             Both ovaries
                                                                             removed
                                                                             32.6(10.1, 55.1)
                                                                             Birth control pills
                                                                             being used
                                                                             -6.3 (-10.0,-2.5)
                                                                             Pre-menopausal
                                                                             8.3(3.8, 12.7)
                                                                             LH:
                                                                             Post-menopausal
                                                                             6.2 (3.0, 9.5)
                                                                             Pregnant
                                                                             -0.8 (-1.9, 0.4)
                                                                             Menstruating at
                                                                             time of exam
                                                                             -0.3 (-1.8, 1.3)
                                                                             Both ovaries
                                                                             removed
                                                                             10.0(1.1, 18.9)
                                                                             Birth control pills
                                                                             being used:
                                                                             -0.6 (-2.9, 1.6)
                                                                             Pre-menopausal
                                                                             1.7 (-0.6, 4.1)
                                                         Age, total bone mineral
                                                         density, serum cotinine,
                                                         alcohol use, current breast
                                                         feeding, hysterectomy, one
                                                         ovary removed, Norplant
                                                         use, radiation or
                                                         chemotherapy, hormone pill
                                                         use, vaginal cream use,
                                                         hormone patch use
November 2012
                              5-608
                                           Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order
appearance in
the text)
Pollack et al.
(2011)



















Study,
Location,
and
Years Outcome Study Population
Buffalo, FSH, Healthy, premenopausal
NY estradiol, LH, women aged 18-44 with
onnc 9m7 progesterone, menstrual cycle length of
and cycle 21-35 days, BMI of 18-35
length kg/m , not recently using
birth control, not planning
to become pregnant, and
not breast feeding
N=252



















Mean
Methodological Pb Pb (SD) Adjusted Effect
Details Biomarker in ug/dL Estimates
Longitudinal Blood Pb 0.93 Mean % Estradiol
cohort using |op. 0.30-0.72 ug/dL:
nonlinear mixed 0 68' 1 2rj Ref
models with ' ' ' 0.73-1.10 ug/dL:
harmonic terms a 9 c 1 9 IRK^
andwei9hted ?~(«™ 1
linear mixed 1.11-6.20 ug/dL:
models 4.7 (-4.7, 15.2)
Amplitude
Estradiol
0.30-0.72 ug/dL:
Ref
0.73-1. 10ug/dL:
-0.01
(-0.06, 0.04)
1.11-6.20 ug/dL:
-0.02
(-0.7, 0.03)
Phase Shift
Estradiol
0.30-0.72 ug/dL:
Ref
0.73-1.10 ug/dL:
-0.09 (-0.24, 0.05)
1. 11-6.20 ug/dL:
0.1 4 (-0.01, 0.29)
Mean % FSH
0.30-0.72 ug/dL:
Ref
0.73-1. 10ug/dL:
8.0 (-0.9, 17.7)
1.11-6.20 ug/dL:
3.6 (-5.3, 13.3)
Amplitude FSH
0.30-0.72 ug/dL:
Ref
0.73-1.10 ug/dL:
-0.01 (-0.03, 0.02)
1. 11-6.20 ug/dL:
-0.02 (-0.04, 0.01)
Phase Shift FSH
0.30-0.72 ug/dL:
Ref
Potential Confounders
Adjusted for in
Analysis
Age, BMI, race
Also examined, but did not
include: smoking, income,
education, physical activity,
parity, dietary Fe, fish
consumption, shellfish
consumption, vegetable
consumption, total calories



















                                                                                                    0.73-1.10ug/dL:
November 2012
5-609
Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order
appearance in
the text)
Study,
Location,
and
Years
Outcome      Study Population
Methodological
Details
Pb
Biomarker
Mean
Pb (SD)
in ug/dL
Adjusted Effect
Estimates
                                                                                                      Potential Confounders
                                                                                                      Adjusted for in
                                                                                                      Analysis
                                                                                                                 -0.06 (-0.25, 0.12)
                                                                                                                 1.11-6.20 |jg/dL:
                                                                                                                 -0.02 (-0.21, 0.18)


                                                                                                                 Mean % LH
                                                                                                                 0.30-0.72 |jg/dL:
                                                                                                                 Ref
                                                                                                                 0.73-1.10|jg/dL:
                                                                                                                 5.1 (-5.1,16.4)
                                                                                                                 1.11-6.20 |jg/dL:
                                                                                                                 -0.5 (-10.5, 10.7)
                                                                                                                 Amplitude LH
                                                                                                                 0.30-0.72 |jg/dL:
                                                                                                                 Ref
                                                                                                                 0.73-1.10 |jg/dL:
                                                                                                                 -0.01 (-0.03, 0.02)
                                                                                                                 1.11-6.20 |jg/dL:
                                                                                                                 -0.02 (-0.04, 0.01)
                                                                                                                 Phase Shift LH
                                                                                                                 0.30-0.72 |jg/dL:
                                                                                                                 Ref
                                                                                                                 0.73-1.10|jg/dL:
                                                                                                                 -0.16 (-0.36, 0.03)
                                                                                                                 1.11-6.20 |jg/dL:
                                                                                                                 -0.11 (-0.32, 0.10)
                                                                                                                 Mean %
                                                                                                                 Progesterone
                                                                                                                 0.30-0.72 |jg/dL:
                                                                                                                 Ref
                                                                                                                 0.73-1.10|jg/dL:
                                                                                                                 7.5(0.1,15.4)
                                                                                                                 1.11-6.20 |jg/dL:
                                                                                                                 6.8 (-0.8, 14.9)
                                                                                                                 Amplitude
                                                                                                                 Progesterone
                                                                                                                 0.30-0.72 |jg/dL:
                                                                                                                 Ref
                                                                                                                 0.73-1.10|jg/dL:
                                                                                                                 0.07(0.01,0.15)
                                                                                                                 1.11-6.20 |jg/dL:
                                                                                                                 -0.06 (-0.13, 0.01)
                                                                                                                 Phase Shift
                                                                                                                 Progesterone
                                                                                                                 0.30-0.72 |jg/dL:
November 2012
                                5-610
                                              Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order
appearance in
the text)
Study,
Location,
and
Years
Outcome     Study Population
Methodological
Details
Pb
Biomarker
Mean
Pb (SD)
in ug/dL
Adjusted Effect
Estimates
                                                                                                 Potential Confounders
                                                                                                 Adjusted for in
                                                                                                 Analysis
                                                                                                           Ref
                                                                                                           0.73-1.10|jg/dL:
                                                                                                           0.04 (-0.06, 0.15)
                                                                                                           1.11-6.20 |jg/dL:
                                                                                                           0.15(0.05, 0.26)


                                                                                                           Linear models
                                                                                                           P (95% Cl)
                                                                                                           Estradiol
                                                                                                           0.03 (-0.05, 0.11)
                                                                                                           FSH
                                                                                                           -0.01  (-0.07, 0.06)
                                                                                                           LH
                                                                                                           0.02 (-0.06, 0.10)
                                                                                                           Progesterone
                                                                                                           0.06 (-0.04, 0.17)


                                                                                                           OR (95% Cl) for
                                                                                                           anovulation per
                                                                                                           1  |jg/dL
                                                                                                           1.20(0.62, 2.34)
November 2012
                               5-611
                                            Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order
appearance in
the text)
Jackson et al.
(2011)























Chang et al.
(2006)





Study,
Location,
and
Years Outcome
Buffalo, FSH,
NY estradiol, LH,
2005-2007 progesterone,
and cycle
length




















Kaohsiung Estradiol
City,
Taiwan
1999
2000-2001



Study Population
Healthy, pre-menopausal
women aged 1 8-44 with
menstrual cycle length of
21 -35 days, BMI of 18-35
kg/m , not recently using
birth control, not planning
to become pregnant, and
not breast feeding

N=252
















Women receiving care at a
infertility clinic in
2000-2001 or delivering a
normal infant at a nearby
medical center in 1 999
N=147


Methodological
Details
Longitudinal
cohort study using
linear regression
and logistic
regression




















Case-control
study using
multivariate linear
regression



Mean
Pb Pb (SD) Adjusted Effect
Biomarker in ug/dL Estimates
Blood Pb Median: Adjusted percent
0.87 change (95% Cl)
IQR. in serum hormone
0 68 1 20 level for Cnan9e in
' ' blood Pb
FSH'
-2.5 (-11. 2, 7.0)
Estradiol:
4.9 (-5.0, 15.9)
LH:
2.5 (-12.3, 19.9)
Progesterone:
4.6 (-12.2, 24.6)
Cycle length:
0.2 (-2.8, 3.3)
OR (95% Cl) per
unit Pb
<25 day vs.
25-35 day cycle
length:
0.9 (0.4, 2.3)
>35 day vs.
25-35 day cycle
length:
0.5(0.1, 1.9)
Blood Pb 3.12 Linear regression
(0.19) p(SE)forPb
1.18(0.60)
p-value: 0.049



Potential Confounders
Adjusted for in
Analysis
Cd, Hg, age, race /ethnicity
























Not specified





November 2012
5-612
Draft - Do Not Cite or Quote

-------
 1                  An epidemiologic study using the NHANES III data and including women aged
 2                  35-60 years old examined the relationship between blood Pb levels (mean 2.8 (ig/dL) and
 3                  serum follicle stimulating hormone (FSH) and luteinizing hormone (LH) (Krieg. 2007).
 4                  Deviation from normal FSH and LH levels may indicate endocrine disruption related to
 5                  ovary functioning. Researchers found that higher blood Pb levels were associated with
 6                  higher levels of serum FSH and LH among both postmenopausal women and women
 7                  with both ovaries removed. There was also a trend of increasing serum FSH with blood
 8                  Pb levels for pre-menopausal women who were not menstruating at the time of the exam
 9                  or pregnant, although the association was not statistically significant for LH. A limitation
10                  of this portion of the study is that FSH and LH were measured without attention to day of
11                  a woman's menstrual cycle and LH and FSH are known to vary throughout the cycle of
12                  non-menopausal, cycling women who are not taking birth control pills. Higher blood Pb
13                  levels were associated with lower levels of serum  FSH among women taking birth
14                  control pills. The inverse association was also present for LH, but it was not statistically
15                  significant. No associations between blood Pb and FSH or LH were apparent for women
16                  who were menstruating at the time  of the exam or were pregnant. Further analysis
17                  indicated that the lowest level of blood Pb for which a statistically significant association
18                  between blood Pb and FSH could be observed was 1.7 (ig/dL among women with their
19                  ovaries removed. For LH, the lowest level of blood Pb for which a statistically significant
20                  association between blood Pb and LH could be observed was 2.8 (ig/dL among
21                  postmenopausal women. Associations between hormones and blood Pb level were  also
22                  investigated using the BioCycle study cohort (Jackson et al., 2011; Pollack et al., 2011).
23                  These women were premenopausal with normal cycles and not on birth control. Blood Pb
24                  was measured at enrollment and hormones were measured multiple times throughout the
25                  menstrual cycle. Neither study detected an association between unit change in blood Pb
26                  and hormone levels. However, when examining tertiles of Pb, women in the highest
27                  tertile blood Pb (1.11-6.20 (ig/dL) had higher mean progesterone and longer length of a
28                  phase shift compared to women in the lowest tertile (0.30-0.72 (ig/dL) (Pollack et al..
29                  2011). Other associations were observed but were not statistically significant (Pollack et
30                  al., 2011). No associations were detected for anovulation (Pollack et al., 2011) or for
31                  cycle length (Jackson et al.. 2011).  Another epidemiologic study was performed in
32                  Kaohsiung City, Taiwan  among two groups of women aged 23-44 years: those who were
33                  seeking help at a fertility clinic after one year of trying to conceive, and those who had
34                  previously delivered an infant and were identified from medical records of a postpartum
35                  care unit (Chang et al.. 2006). The mean (SD) concurrent blood Pb level in this study was
36                  3.12 (0.19) (ig/dL. The study reported apositive association between blood Pb levels and
37                  serum estradiol concentrations during the early follicular phase, which reflects ovary
38                  activity.
      November 2012                             5-613                  Draft - Do Not Cite or Quote

-------
 1                  The effect of Pb exposure on the female endocrine system was demonstrated in
 2                  toxicological studies reviewed in the 1986 and 2006 Pb AQCD (U.S. EPA. 2006b.
 3                  1986a). However, the mechanism by which Pb affects the endocrine system has not been
 4                  fully elucidated. Several recent articles continue to demonstrate that Pb alters the
 5                  concentration of circulating hormones in female experimental animals. As mentioned in
 6                  the previous AQCD, Pine et al. (2006) observed that maternal Pb exposure (during
 7                  gestation and lactation) caused a decrease in basal LH levels in pre-pubertal female
 8                  Fischer 344 rat pups as compared to control, non-Pb exposed pups. Dumitrescu et al.
 9                  (2008a) observed alteration of hormone levels in adult female Wistar rats after ingesting
10                  Pb acetate (50, 100, 150 ppb) in drinking water for 6 months; measurements were made
11                  during the pro-estrous stage of the estrous cycle to allow for consistent timing for
12                  comparison of cyclic hormonal variation. The authors reported decreases in FSH,
13                  estradiol,  and progesterone levels with increases in LH and testosterone levels.
14                  Nampoothiri and Gupta (2008) administered Pb acetate at a concentration that did not
15                  affect reproductive performance, implantation or pregnancy outcome (0.05 mg/kg body
16                  weight) to Charles Foster female rats 5 days before mating and during the gestational
17                  period. They observed a decrease in steroidogenic enzymes, 3(3- hydroxysteroid
18                  dehydrogenase (HSD) and 17J3-HSD, activity in reproductive organs, as well as  a
19                  decrease in steroid hormones (progesterone and estradiol), suggesting that chronic
20                  exposure to low levels of Pb may affect reproductive function of mothers and their
21                  offspring. Similarly, Pillai et al.  (2010) reported impaired ovarian steroidogenesis in
22                  Charles Foster adult female rats (PND56) from  dams exposed gestationally and
23                  lactationally to Pb acetate (subcutaneous daily injections of 0.05 (ig/kg BW). Pillai
24                  observed a decrease in steroidogenic enzymes, 3(3-HSD and 17(3-HSD, but saw no
25                  changes in ovarian steroidogenic acute regulatory protein (StAR) or CYP11 mRNA
26                  levels indicating Pb-induced inhibition of ovarian steroidogenesis.

27                  Kolesarova et  al. (2010) conducted an in vitro study to examine the secretory activity of
28                  porcine ovarian granulose cells after Pb administration for 18 hours. The results of the
29                  study showed that Pb acetate concentrations of 0.046 mg/mL and 0.063 mg/mL
30                  statistically significantly inhibited insulin-like growth factor-1 (IGF-1) release, but
31                  concentrations of 0.25 mg/mL and 0.5 mg/mL did not influence  IGF-1 release.
32                  Progesterone release was not affected by Pb treatment; however, Pb caused a reduction in
33                  LH and FSH binding in granulose cells and increased apoptosis as evidenced by
34                  increased  expression of caspase-3 and cyclin Bl, suggesting a Pb-induced alteration in
35                  the pathways of proliferation and apoptosis of porcine ovarian granulose cells. Decreased
36                  gonadotropin binding was also observed in rats  after Pb exposure subcutaneously
37                  administered Pb (0.05 mg/kg body weight daily before mating and during pregnancy)
38                  with a resulting blood Pb of 2.49 (ig/mL) (Nampoothiri and Gupta. 2006).
      November 2012                            5-614                  Draft - Do Not Cite or Quote

-------
 1                   No recent toxicological studies were found that examined Pb-induced effects on the
 2                   estrus cycle.

 3                   Overall, toxicological studies report alterations in hormone levels related to blood Pb
 4                   concentration. Similarly, epidemiologic studies reported associations between
 5                   concurrent/closely timed blood Pb levels and hormone levels in female adults. Although
 6                   Pb-associated changes in hormone levels are observed, there are discrepancies and the
 7                   hormones examined vary by study. One explanation for the inconsistent findings is that
 8                   changes could vary based on current hormonal and reproductive status of the participants.
 9                   Also, the covariates included in statistical models as potential confounders varied among
10                   studies, which could contribute to between study heterogeneity. This is also a limitation
11                   of the epidemiologic studies; not all of the studies investigated important confounders,
12                   such as other metal exposures or smoking. Additionally, the cross-sectional design of
13                   these studies leaves uncertainty regarding Pb exposure timing, duration, and frequency
14                   that contributed to the observed associations.
                     5.8.5.2      Effects on Fertility

15                   Previous studies indicated that Pb exposure does not produce total sterility, but it can
16                   disrupt female fertility (U.S. EPA, 2006b). Recent epidemiologic studies and studies in
17                   experimental animals have inconsistent results. The epidemiologic studies are
18                   summarized in Table 5-46. Most of these studies examining biological measures of Pb
19                   collected at or during the period of possible fertilization or start of fertility treatment,
20                   although Bloom et al. (2011 a) measured blood Pb at baseline and followed women for at
21                   least 12 menstrual cycles (or until pregnancy).
      November 2012                             5-615                  Draft - Do Not Cite or Quote

-------
Table 5-46 Summary of recent epidemiologic studies of associations between Pb levels and
Reference
(Presented in
order of
appearance in
the text)
Bloom et al.
(2011 a)







Chang et al.
(2006)










Study
Location,
and
Years Outcome
New York Achieving
1996-1997 pregnancy







Kaohsiung Infertility
City,
Taiwan
1999,
2000-2001










Study Population
Women who were aged
18-34 years, were
previously part of a study
about fish consumption,
and were not currently
pregnant and were
followed for 12 menstrual
cycles or until pregnant
N=80
Women receiving care at
a infertility clinic in
2000-2001 or delivering
a normal infant at a
nearby medical center in
1999
N:
Cases
=64
N:
Controls
=83



Methodological
Details
Longitudinal
cohort using Cox
proportional
hazards





Case-control
study using
unconditional
logistic regression









Mean Pb
Pb (SD)
Biomarker in ug/dL
Blood Pb at No positive
baseline pregnancy
test: 1 .55
(0.16)
Positive
pregnancy
test: 1 .54
(0.12)

Blood Pb 3.12(0.19)












Adjusted
Effect
Estimates
P (95% Cl)
-0.031
(-1 .066, 1 .004)
per 0.6 ug/dL





OR (95% Cl)
Infertility
< 2.5 ug/dL:
1.00
(Referent group)


>2.5 ug/dL:
2.94
(1.18,7.34)


fertility for females.



Potential Confounders
Adjusted for in Analysis
Baseline As, baseline Cd,
baseline Mg, baseline Ni,
baseline Se, baseline Zn, total
serum lipids, age, parity,
frequency of intercourse
during fertility window, alcohol
use, cigarette use


Age, BMI, active smoking, use
of Western medicine
Considered but did not
include: irregular
menstruation, age at first
menses, marital status,
passive smoking,
contraceptive drugs



November 2012
5-616
Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order of
appearance in
the text)
Bloom et al.
(2010)












Bloom et al.
(2011b)








Study
Location,
and
Years Outcome
California Oocyte
2007-2008 maturity,
oocyte
fertilization










California Embryo cell
2007-2008 number,
embryo
fragmentation
score









Study Population
Women who were part of
the Study of Metals and
Assisted Reproductive
Technologies (SMART):
women referred to the
Center for Reproductive
Health of UCSF for
infertility treatment and
their first IVF procedure
N=15





Women who were part of
the Study of Metals and
Assisted Reproductive
Technologies (SMART)
(women referred to the
Center for Reproductive
Health of UCSF for
infertility treatment and
their first IVF procedure)
and who generated
embryos
N=24


Mean Pb
Methodological Pb (SD)
Details Biomarker in ug/dL
Longitudinal Blood Pb at 0.82 (0.32)
cohort using the time of
multivariable log- oocyte
binomial retrieval
regression










Longitudinal Blood Pb at 0.83 (0.30)
cohort using the time of
logistic regression oocyte
retrieval







Adjusted
Effect
Estimates
RR per 1 ug/dL
Oocyte
maturity
(determined by
Metaphase II
arrest):

0.54 (0.31 , 0.93)
0.25 (0.03,
2.50)*
Oocyte
fertilization:
0.97 (0.66, 1 .43)
1 .09 (0.72,
1 .65)*
OR per 1 ug/dL*
Embryo cell
number:
0.25 (0.07, 0.86)

Embryo
fragmentation
score:
1 .71 (0.45, 6.56)




Potential Confounders
Adjusted for in Analysis
Age, cigarette smoking,
race/ethnicity

*Also, controlled for Cd.










Age, race/ethnicity, cigarette
smoking, urine creatinine

Additionally included for
embryo cell number: day of
embryo transfer

*Also controlling for Hg and
Cd

November 2012
5-617
Draft - Do Not Cite or Quote

-------
Reference
(Presented in Study.
order of Location,
appearance in ar|d
the text) Years Outcome
Al-Saleh et al. Riyadh, Achieving
(2008a) Saudi pregnancy
Arabia and/or
2002-2003 fertilization






















Study Population
Women aged 1 9-50
undergoing IVF

N.
.
pregnancy
_or\o
-ZUo
N:
No pregnancy
=321
N:
fertility
=556
N:
No fertility
=63












Mean Pb
Methodological Pb (SD)
Details Biomarker in ug/dL
Longitudinal Blood Pb Blood Pb:
cohort using Follicular 3-34 <2-24)
logistic regression fluid pb B|ood pb
levels
>10 ug/dL:
1.7%


Follicular
fluid:
0.68(1.82)

















Adjusted
Effect
Estimates
OR (95% Cl)
(unit not given,
assume results
are per 1 ug/dL)
Pregnancy
Blood Pb:
0.55
(0.23, 1.31)
Follicular fluid
Pb:
1.36
(0.91 , 2.02)


Fertilization
Blood Pb:
0.30
(0.08, 1 .03)
Follicular fluid
Pb:
1.45
(0.69, 3.02)
Note: In a
reduced
adjusted model
for fertilization,
the OR for blood
Pb was 0.38
(0.14,0.99)
Potential Confounders
Adjusted for in Analysis
Age, husband's age, BMI,
location and duration at that
location, previous location and
duration at that location, age
at first menses, number
of days of menstrual cycle,
education, work status,
husband's education, family
income, husband's smoking
status, blood and follicular Cd
and Hg, follicular cotinine

Also included for pregnancy
as outcome: coffee
consumption, tea
consumption, caffeinated soft
drink consumption,











November 2012
5-618
Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order of
appearance in
the text)
Study
Location,
and
Years      Outcome
Study Population
Methodological
Details
            Mean Pb
Pb          (SD)
Biomarker   in ug/dL
Adjusted
Effect
Estimates
Potential Confounders
Adjusted for in Analysis
Silberstein et
al. (2006)







Providence, Achieving
Rl pregnancy
NS






Women undergoing IVF
at the study hospital

N.
.
pregnancy
=4
N:
No oreanancv
Longitudinal Follicular
cohort study using fluid Pb
Mann-Whitney
U-test





Not given
quantitatively
From a figure
in the paper:

Median Pb in
follicular fluid
of pregnant
women: -1.3
P-value for
difference in
medians by
Mann-Whitney
U test: 0.0059


*Note, This
study only
None








                                       =5
                                                                                         Median Pb in
                                                                                         follicular fluid
                                                                                         of non-
                                                                                         pregnant
                                                                                         women:  -2.2
                                                                                       included 9
                                                                                       women
November 2012
                              5-619
                             Draft - Do Not Cite or Quote

-------
 1                  A prospective cohort study enrolled women who previously participated in a study offish
 2                  consumption for a length of up to 12 menstrual cycles and investigated the relationship
 3                  between blood Pb levels at baseline and having a positive pregnancy test at some point
 4                  during the next 12 menstrual cycles (Bloom etal. 201 la). No association was observed
 5                  between blood Pb and achieving pregnancy.

 6                  Among women aged 23-44 years, a difference in blood Pb was reported between women
 7                  who were seeking help at a fertility clinic after one year of trying to conceive and women
 8                  who had previously delivered an infant and were identified from medical records of a
 9                  postpartum care unit at a medical center (Chang et al.. 2006). Higher odds of infertility
10                  were observed when comparing women with blood Pb levels >2.5 (ig/dL to those with
11                  blood Pb levels < 2.5 (ig/dL although this study is limited by its case-control design.

12                  Epidemiologic studies have also examined women having difficulty conceiving by
13                  performing studies among patients of fertility clinics or undergoing in vitro fertilization
14                  (IVF). The Study of Metals and Assisted Reproductive Technologies (SMART) enrolled
15                  women undergoing their first round of IVF and investigated multiple steps before
16                  pregnancy as the outcomes (Bloom et al., 20 lib; Bloom etal.. 2010). Higher blood Pb
17                  levels were associated with lower oocyte maturity although the lack of power made
18                  interpretation of models controlling for Cd difficult. No association was observed
19                  between blood Pb and oocyte fertilization (Bloom et al.. 2010). In the examination of
20                  markers of IVF success, inconsistent results were observed. Embryo cell number was
21                  lower in association with higher blood Pb levels but no association was observed for
22                  embryo fragmentation score (Bloom etal.. 20 lib). Another study examining  fertility
23                  reported on women in  Saudi Arabia aged 19-50 years who were undergoing IVF
24                  treatment (Al-Saleh et al., 2008a). Women were categorized as having achieved a
25                  pregnancy versus not having achieved a pregnancy and achieved fertilization versus not
26                  achieving fertilization. The majority of women had follicular Pb levels that were below
27                  the limit of detection, whereas less than 2% of women had blood Pb levels below the
28                  limit of detection. In addition,  less than 2% of women had blood Pb levels that were
29                  above 10 (ig/dL. Follicular Pb  levels were not correlated with the blood Pb. No
30                  association was observed between blood or follicular Pb and pregnancy outcomes in
31                  either crude or adjusted models. An association was not detected between follicular Pb
32                  and fertilization, but higher blood Pb was associated with lower rates of fertilization.
33                  Finally, a study that included nine women undergoing IVF treatment in Rhode Island
34                  (Silberstein et al., 2006) found that median follicular Pb levels in women who achieved
35                  pregnancy were lower than the follicular Pb levels among nonpregnant women.
      November 2012                            5-620                 Draft - Do Not Cite or Quote

-------
 1                   Overall, these epidemiologic studies examine a variety of fertility-related endpoints and
 2                   although some studies demonstrate an association between higher Pb levels and
 3                   fertility/pregnancy, as a whole the results are inconsistent across studies. One limitation
 4                   present in most of these studies is that the participants are women who are seeking help
 5                   for fertility problems. The participants are not samples of the general population and
 6                   therefore cannot be generalized to all women of childbearing age. This may also have
 7                   introduced substantial selection bias into the study.

 8                   Animal toxicology studies following female fertility looked at various outcomes. Several
 9                   studies observed a decrease in litter size when females were exposed to Pb before mating
10                   or during pregnancy (Dumitrescu et al.. 2008c: lavicoli et al.. 2006a: Teijon et al.. 2006).
11                   Pups in a study by Teijon et al. (2006) receiving 400 ppm Pb acetate in drinking water
12                   had blood Pb of 97 (ig Pb/dL blood at 1 week post-weaning and 18.2 (ig Pb/dL blood at
13                   2 week post-weaning. Dumitrescu et al. (2008c) observed a modification in sex ratio of
14                   pups born to dams exposed to Pb before mating and during the entirety  of pregnancy. As
15                   the dose of Pb increased, the number of females per litter also increased (i.e., 1  male to
16                   0.8 female in non-Pb exposed group; 1 male to 0.66 female in 50 ppb Pb acetate group; 1
17                   male to 2.25 females in 100 ppb group; and 1 male to  2.5 females in 150 ppb group).
18                   These results are not consistent with earlier results of Ronis et al. (1998b). who did not
19                   observe differences in sex ratio dams and offspring were exposed only during pregnancy.
20                   Thus, Pb exposure in animal studies during or before pregnancy have shown effects on
21                   litter size and mixed effects on sex ratio.

22                   Nandi et al. (2010) demonstrated a concentration-dependent decline in viability rate,
23                   maturation, fertilization, and cleavage rates of buffalo oocytes cultured  in medium
24                   containing 1-10 (ig/mL Pb acetate (24 hour culture). Karaca and §im§ek (2007) observed
25                   an increase in the number of mast cells in ovary tissue after Pb exposure (2,000 (ig/mL in
26                   drinking water for 6 weeks prior to estrous monitoring then for 1 additional month during
27                   which estrous cyclicity was monitored) suggesting that Pb may stimulate an
28                   inflammatory  response in the ovaries which may contribute to Pb-induced female
29                   infertility.

30                   In contrast, Nampoothiri and Gupta (2008) did not observe any statistically significant
31                   change in fertility rate or litter size in female rats subcutaneously administered  Pb
32                   (0.05 mg/kg body weight daily before mating and during pregnancy) with a resulting
33                   blood Pb of 2.49 (ig/mL. Although reproductive performance was not affected in this
34                   study, the authors did report an alteration in implantation enzymes. Cathepsin-D activity
35                   decreased and alkaline phosphatase activity increased after Pb exposure.
      November 2012                             5-621                  Draft - Do Not Cite or Quote

-------
 1                  In summary, recent epidemiologic and toxicological studies on the effect of Pb on
 2                  fertility outcomes have generated inconsistent results. Most of the epidemiologic studies
 3                  are limited by their small sample sizes and selection bias and lack of generalizability due
 4                  to a focus on women seeking help for infertility. Most of the studies control for multiple
 5                  potential confounders, such as smoking status and age. The studies from the toxicological
 6                  literature show that Pb exposure to females affects litter size (decreased litter size), sex
 7                  ratio (ratio of male to female offspring in a litter) and ovarian viability, albeit often at
 8                  higher dose of Pb. However, the bulk of the evidence including the current and historical
 9                  Pb literature (U.S. EPA. 2006b) indicate that  increased Pb exposure may decrease female
10                  fertility.
                    5.8.5.3      Ovaries, Embryo Development, Placental function, and
                                 Spontaneous Abortions

11                  The 2006 Pb AQCD included studies of Pb exposure among men and women and their
12                  associations with spontaneous abortions. The 2006 Pb AQCD concluded that overall
13                  there was little evidence to support an association between Pb exposure among women
14                  and spontaneous abortion (U.S. EPA. 2006b). Most of the studies examined in the
15                  2006 Pb AQCD assigned exposure based on living near a smelter or working in
16                  occupations that often result in Pb exposure and the results of these studies were
17                  inconsistent. Little evidence was available in the 2006 Pb AQCD to suggest an
18                  association with paternal Pb levels, and no recent studies have been performed to
19                  examine paternal Pb levels and spontaneous abortion. Since the 2006 Pb AQCD, multiple
20                  epidemiologic studies have been published that examine Pb levels in women and their
21                  possible association with spontaneous abortion. Table 5-47 provides information on these
22                  longitudinal and cross-sectional studies. Additionally, toxicological studies have studied
23                  the effects of Pb on fetal loss and the contribution of the ovaries and placenta to fetal loss.
      November 2012                            5-622                  Draft - Do Not Cite or Quote

-------
Table 5-47   Summary of recent epidemiologic studies of associations between Pb levels and
             spontaneous abortions.
Reference
(Presented in
order of
appearance in the
text)
Vigeh et al.
(2010)






Study
Location
Tehran,
Iran
2006-2008





Outcome
Pregnancy
ended before
20 weeks of
gestation




Study population
Women who were non-
smokers, non-obese, had
no chronic health
conditions, had their last
menstrual period less than
1 2 weeks prior, and were
pregnant with a singleton
infant
N:


Methodological
Details
Longitudinal
cohort study using
t-test and logistic
regression




Pb
Biomarker
Maternal blood
Pb
during weeks
8-1 2 of
pregnancy



Mean Pb
(SD)
in ug/dL
3.8 (2.0)
Spontaneous
abortion'
3.51 (1.42)
Non-
spontaneous
abortion'
3.83 (1 .99)



Adjusted Effect
Estimates
T-test for difference in
mean values: 0.41
OR:
0.331 (95% Cl: 0.011,
10.096) for an
increase in log-
transformed blood Pb
(units not given,
assume 1 ug/dL)
Potential
Confounders
Adjusted for
in Analysis
Age, parity,
hematocrit,
passive
cigarette
smoking
exposure


                                 a 1 spontaneous abortions
                                 =15
                                 N:
                                 No spontaneous abortions
                                 =336
Yin et a I. (2008) Shanxi Anembryonic Women age 25-35 yr old
Province, pregnancy and at 8-1 2 weeks of
China gestation at study entry;
2004-2006 cases were anembryonic
pregnancies and controls
were normal pregnancies
that ended in a live birth
between 37-42 weeks
N:
Cases
=40
N:
Controls
=40
Case-control Maternal blood
study using t-test Pb after
miscarriage for
cases and at
study
enrollment for
controls







Cases:
5.3 (95% Cl:
5.2, 5.9)
Controls:
4.5 (95% Cl:
37, 5.0)








Comparisons None
between log-
transformed blood Pb
levels of cases and
controls performed via
Student's t-test had a
p-value of 0.03







November 2012
5-623
Draft - Do Not Cite or Quote

-------
Reference
(Presented in
order of
appearance in the
text)
Lamadrid-
Figueroa et al.
(2007)













Gundacker et al.
(2010)









Study
Location Outcome
Mexico Previous
City, miscarriage
Mexico
1997-1999,
2001 -2004













Vienna, Previous
Austria miscarriage
2005









Study population
Women who had a
previous pregnancy and
were currently pregnant
with gestational age of
£ 14 weeks

N:
> 1 previous miscarriages
=71
N:
No previous miscarriages
=136








Women recruited during the
second trimester of
pregnancy

N:
a 1 previous miscarriages
=8
N:
No previous miscarriages
=22


Methodological
Details
Cross-sectional
study using
Poisson
regression













Cross-sectional
study using non-
parametric tests









Pb
Biomarker
Maternal and
umbilical cord
blood Pb,
maternal bone
Pb













Whole
placentas
shortly after
hirf h
DIIIM







Mean Pb
(SD)
in ug/dL
Overall:
Blood Pb: 6.2
(4.5)
Plasma Pb:
0.014(0.013)
Cases:
Blood Pb: 5.8
(3.4)
Plasma Pb:
0.014(0.013)
Controls:
Blood Pb: 6.5
(4.9)
Plasma Pb:
0.013(0.013)





Median
(IQR):
25.8(21.0,
36.8)








Adjusted Effect
Estimates
Categorized Plasma
Blood Pb ratio:
Isttertile:
1 .00 (Ref)
2nd tertile:
1.16(p-value0.61)
3rd tertile:
1.90(p-value0.015)

IRR (95%CI) Per 1
SD increase: Plasma
Pb
1.12(p-value0.22)
Blood Pb
0.93 (p-value 0.56)
Plasma/Blood Pb ratio
1.1 8 (p-value 0.02)
Patella Pb
1.1 5 (p-value 0.39)
Tibia Pb
1 .07 (p-value 0.56)
Median Placenta Pb:
Women who had not
previously miscarried:
27 ug/kg
Women who had
previously miscarried:
39 ug/kg
(p-value for
difference: 0.039)


Potential
Confounders
Adjusted for
in Analysis
Age, education














N/A








November 2012
5-624
Draft - Do Not Cite or Quote

-------
 1                  A longitudinal study examining spontaneous abortions occurring early in the pregnancy
 2                  was conducted in Iran (Vigeh et al.. 2010). Mean blood Pb concentrations, measured at
 3                  8-12 weeks of pregnancy, were similar in women who did and did not have spontaneous
 4                  abortions. Higher blood Pb levels were not associated with greater odds of spontaneous
 5                  abortions before 20 weeks of pregnancy. Yin et al. (2008) performed a study in the
 6                  Shanxi Province of China to examine if plasma Pb levels were associated with
 7                  anembryonic pregnancies (spontaneous abortions during the first trimester, which
 8                  account for 15% of all spontaneous abortions). Women were enrolled at  8-12 weeks of
 9                  gestation. Women who delivered a term pregnancy had mean plasma Pb  levels that were
10                  lower than those of women who had an anembryonic pregnancy (plasma Pb measured at
11                  the time of miscarriage for cases and at 8-12 weeks for controls). Of note, among cases
12                  plasma Pb level was inversely correlated with folate and vitamin B12, but this correlation
13                  was not observed among those who delivered at term; no models examining plasma Pb
14                  levels adjusted for nutrient status. A study in Mexico City examined a group of pregnant
15                  women (maximum gestational period at enrollment was 14 weeks) who had previously
16                  been pregnant and either given birth or had a spontaneous abortion (Lamadrid-Figueroa
17                  et al.. 2007). Women in the highest tertile of plasma/blood Pb ratio had higher rates of
18                  previous spontaneous abortions than did women in the lowest tertile. The authors state
19                  that the plasma/whole blood ratio represents the bioavailability of Pb, which is capable of
20                  crossing the placental barrier for a given blood concentration. No association was
21                  observed when examining the relationship between Pb and spontaneous abortions using
22                  whole blood, plasma, or bone Pb alone. Similarly, a study of placental Pb levels among
23                  pregnant women in Austria observed higher placenta Pb levels among women who had
24                  miscarried a previous pregnancy compared to women who had not miscarried a previous
25                  pregnancy (Gundacker et al.. 2010).  It is important to note that the number of women
26                  included in the study was small (only 8 women reported previously having a miscarriage)

27                  In toxicological studies, isolated embryo cultures are often used to understand the
28                  mechanisms responsible for aberrant embryo development as it may contribute to
29                  teratogenesis, fetal loss or negative postnatal pup outcomes.  Nandi et al.  (2010)
30                  demonstrated an exposure concentration-dependent decline in embryo development of
31                  fertilized buffalo oocytes cultured for 24 hours in medium containing 0.05-10 (ig/mL
32                  Pb acetate as evidenced by reduced morula/blastocyst yield and increased four-to eight-
33                  cell arrest, embryo degeneration, and asynchronous division. This study provides
34                  evidence of the negative effect of Pb on embryo development and contributes
35                  mechanistic understanding to Pb-dependent pregnancy loss.

36                  A possible explanation for reduced fertility and impaired female reproductive success  as
37                  a result of Pb exposure is changes in morphology or histology in female  sex organs and
38                  the placenta (Dumitrescu et al.. 2007; U.S. EPA. 2006b). Wang et al. (2009e) observed
      November 2012                             5-625                  Draft - Do Not Cite or Quote

-------
 1                  that elevated maternal blood Pb (0.6-1.74 (JVI, -12.4-36.0 (ig/dL) compared to control
 2                  (0.04 (JVI, -0.83 (ig/dL) were associated with decreased fetal body weight, pup body
 3                  length, and placental weight in Wistar rats. The authors reported that placentae from
 4                  Pb-exposed groups showed concentration-dependent increasing pathology of
 5                  cytoarchitecture and cytoplasmic organelles. The authors also reported a positive
 6                  expression of NF-KB, a transcription factor that controls the expression of genes involved
 7                  in immune responses, apoptosis, and cell cycle, in the cytotrophoblasts, decidual cells,
 8                  and small vascular endothelial cells in rat placenta under a low-level Pb exposure
 9                  condition which correlated with low blood Pb levels.

10                  Pb-exposed (273 mg/L or 819 mg/L in drinking water, 500 or 1,500 ppm Pb acetate,
11                  respectively) male rats  from Anjum et al.  (2010) that had an exposure concentration-
12                  dependent decreases in serum testosterone, decreased male reproductive organ weight
13                  and decreased sperm were mated to untreated females. These untreated dams bred to the
14                  Pb exposed males had male related exposure concentration-dependent decreased
15                  implantation rate and higher pre- and post-implantation loss, indicating paternally
16                  mediated fetal loss. The magnitude of these  effects in dams was dependent on the
17                  concentration of Pb exposure in their male mating partners.

18                  As observed in sperm cells, Pb stimulates changes in antioxidant enzyme activity in rat
19                  ovaries indicating that oxidative stress may be a contributing factor in Pb-induced ovarian
20                  dysfunction. Nampoothiri et al. (2007) observed a reduction in SOD activity and an
21                  increase  in CAT activity along with a decrease in glutathione content and an increase in
22                  lipid peroxidation in rat granulosa cells  after 15 days of Pb treatment (subcutaneously
23                  administered Pb (0.05 mg/kg body weight daily before mating and during pregnancy)
24                  with a resulting blood Pb of 2.49 (ig/mL).

25                  Previous studies demonstrated that Pb accumulates in the ovaries and causes histological
26                  changes, thus contributing to Pb-induced effects on female fertility (U.S. EPA. 2006b). In
27                  support of historical studies, recent studies demonstrate Pb-induced histological changes
28                  in ovarian cells of pigs (Kolesarova et al.. 2010) and rats (Nampoothiri et al.. 2007;
29                  Nampoothiri and Gupta, 2006). Kolesarova et al. (2010) observed a reduction of the
30                  monolayer of granulosa cells after Pb addition (0.5 mg/mL, 18 hours culture).
31                  Nampoothiri and Gupta (2006) reported that Pb exposure caused a decrease in cholesterol
32                  and total phospholipid content in the membranes of granulosa cells which resulted  in
33                  increased membrane fluidity (subcutaneously administered Pb, 0.05 mg/kg body weight
34                  daily before mating and during pregnancy with a resulting blood Pb of 2.49 (ig/mL).

35                  Overall,  the recent studies support the conclusions of the 2006 Pb AQCD (U.S. EPA.
36                  2006b) that there is mixed evidence among epidemiologic studies to suggest an
37                  association between Pb and spontaneous abortions. It is important to note that studies of


      November 2012                             5-626                  Draft - Do Not Cite or Quote

-------
 1                  spontaneous abortions are difficult to conduct. The majority of spontaneous abortions are
 2                  during the first trimester, which makes them difficult to capture. Women may miscarry
 3                  before being enrolled in a study and many women may not have known they were
 4                  pregnant when they miscarried. This limits the ability to detect subtle effects, especially
 5                  if higher Pb levels do lead to increased risk of early spontaneous abortions. In addition,
 6                  some studies are limited by their retrospective examination of current Pb levels in
 7                  relation to previous miscarriages. Sample size is another limitation of the available
 8                  epidemiologic studies. The epidemiologic studies also had little control for potential
 9                  confounding factors, with some studies including no potential confounders in their
10                  analyses. Toxicological data provide mechanistic understanding of the contribution of Pb
11                  exposure to spontaneous abortions.  These laboratory data show that Pb exposure
12                  impaired placental function, induced oxidative stress and histological changes in the
13                  ovaries, and affected embryo development. The toxicological and epidemiologic data
14                  provide inconsistent findings for the role of Pb in spontaneous abortions.
                    5.8.5.4     Effects on Breast Milk

15                  Experiments in laboratory animals have shown that dietary manipulation of maternal
16                  fatty acid (FA) levels in diet can worsen Pb-related behavioral effects of offspring after
17                  lactational Pb exposure (Lim et al., 2005). To determine if components of dam milk
18                  contributed to this change, dam milk fatty acids were altered via diet. Diets deficient in
19                  n-3 fatty acids can lead to a deficiency of DHA, which is essential for proper nervous
20                  system development. Lim et al. (2005) found that dam Pb exposure (Long-Evans rats,
21                  2,000 ppm Pb acetate trihydrate/BW) during lactation (PNDO-PND21)  led to a decrement
22                  in non-essential fatty acids in the maternal organs at PND25 (mean [SD] blood Pb levels
23                  in dams: 308 [56] (ig/dL). In animals with a diet deficient in n-3 FAs, there was a Pb-diet
24                  interaction with a specific size PUFA (i.e., a 20-carbon n-6 PUFA). In general, Pb
25                  exposure caused a decrement in shorter chain monounsaturated and saturated FAs in
26                  maternal organs.

27                  Dietary supplementation with calcium  can be an especially important contributor to Pb
28                  mobilization during periods of high calcium demand including pregnancy/lactation. For
29                  example, mothers with elevated blood  Pb levels given calcium phosphate and ascorbic
30                  acid supplementation during lactation had a 90% decrease in placental Pb content and a
31                  15% decrease in the concentration of Pb in breast milk (Altmann et al.. 1981) versus the
32                  control group that did not receive dietary treatment. Another study (Gulson et al., 2004a)
33                  has shown that calcium supplementation during the lactation is less beneficial in
34                  modulating maternal blood Pb levels (mean blood Pb at first sampling was 2.4 (ig/dL);
35                  the Gulson cohort (Gulson et al.. 2004a) was limited by power (n=10 women). In a


      November 2012                            5-627                  Draft - Do Not Cite or Quote

-------
 1                  cohort of women from Mexico City, daily calcium supplementation during lactation
 2                  reduced maternal blood Pb by 15-20% and Pb in breast milk by 5-10% (Ettinger et al..
 3                  2004a). Another study by the same investigators showed that using calcium supplements
 4                  daily during pregnancy also reduced blood Pb levels during pregnancy (Ettinger etal..
 5                  2009) with the effect strongest in women with higher biomarkers of Pb exposure
 6                  (elevated baseline bone Pb or >5 (ig/dL blood Pb) or in women with higher Pb exposure
 7                  (self-reported use of Pb-glazed ceramics). Thus, dietary modulation with calcium
 8                  supplementation during pregnancy and lactation may decrease the amount of Pb to which
 9                  the developing fetus of infant is exposed. The evidence for this seems especially strong
10                  for protection during pregnancy and more mixed for protective effects of calcium during
11                  lactation.
                    5.8.5.5     Summary of Effects on Female Reproductive Function

12                  In summary, Pb exposure was found to affect female reproductive function as
13                  demonstrated by both epidemiologic and toxicological studies. Some evidence is also
14                  available regarding blood Pb levels and altered hormone levels in adults, but varied
15                  among studies. The differences may have been due to the different hormones examined
16                  and the different timing in the menstrual and life cycles of the women. Although studies
17                  reported inconsistent findings for the association between Pb and fertility, there is some
18                  evidence of a potential relationship. Adjustment for potential confounders varies from
19                  study to study, with some potentially important confounders, such as BMI, not included
20                  in all studies. Also, many epidemiologic studies are limited by small samples sizes and
21                  are generally of women attending infertility clinics, which presents the possibility of
22                  selection bias and lack of generalizability. Toxicological studies found effects on female
23                  reproductive function after prenatal or early postnatal exposures. Further coherence for
24                  these findings in laboratory animal models is found in with findings in the ecological
25                  literature for the effects of Pb exposure on reduced fecundity in terrestrial and aquatic
26                  animal species (Sections 7.4.5.2. 7.3.4.2. 7.3.4.3. and 7.4.5.3). Although epidemiologic
27                  and toxicological studies provide information on different exposure periods, both types of
28                  studies support the conclusion that Pb affects at least some aspects of female reproductive
29                  function.
      November 2012                            5-628                  Draft - Do Not Cite or Quote

-------
      5.8.6        Summary and Causal Determination

 1                  Many epidemiologic and toxicological studies of the effects of Pb on reproductive and
 2                  developmental outcomes have been conducted since the 2006 Pb AQCD. The evaluation
 3                  of causal relationships with Pb exposure focuses on four areas: developmental effects,
 4                  birth outcomes, reproductive function among males, and reproductive function among
 5                  females. The sections that follow describe the evaluation of evidence for these outcomes
 6                  with respect to causal relationships with Pb exposure using the framework described in
 7                  Table II of the Preamble. The application of the key supporting evidence to the causal
 8                  framework is summarized in Table 5-48.
                    5.8.6.1      Effects on Development

 9                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported Pb-associated effects on development in
10                  toxicological studies. Multiple recent epidemiologic studies of Pb and puberty have
11                  shown associations between concurrent blood Pb levels and delayed pubertal onset for
12                  girls and boys. Delayed puberty has been linked to decreased peak bone mass and
13                  increased risk of osteoporotic fractures (Gilsanz etal., 2011; Naves et al., 2005). In cross-
14                  sectional epidemiologic studies of girls (ages 6-18 years) with mean and/or median
15                  concurrent blood Pb levels less than 5 (ig/dL consistent associations with delayed
16                  pubertal onset (measured by age at menarche, pubic hair development, and breast
17                  development) were observed. In boys (ages 8-15 years), fewer epidemiologic studies
18                  were conducted but associations were observed, including associations among boys in a
19                  longitudinal study. These associations are consistently observed in populations with
20                  concurrent blood Pb levels <10(ig/dL. Potential confounders considered in the
21                  epidemiologic studies of both boys and girls that performed regression analyses varied.
22                  Most studies controlled for age and BMI. Other variables, such as measures of diet, SES,
23                  and race/ethnicity, were included in some of the studies. Adjustment for nutritional
24                  factors was done less often and this could be an important confounder. A study using a
25                  cohort of girls from the NHANES analysis controlled for various dietary factors as well
26                  as other potential confounders and reported an association between increased concurrent
27                  blood Pb levels and delayed pubertal onset (Selevan et al.. 2003). A limitation across
28                  most of the epidemiologic studies of blood Pb levels and  delayed puberty is their cross-
29                  sectional design, which does not allow for an understanding of temporality. There is
30                  uncertainty with regard to the exposure frequency, timing, duration, and level that
31                  contributed to the associations observed in these studies.

32                  Recent toxicological studies show that pubertal onset is one of the more sensitive markers
33                  of Pb exposure with effects observed after gestational exposures leading to blood Pb
      November 2012                            5-629                 Draft - Do Not Cite or Quote

-------
 1                  levels in the female pup of 3.5-13 (ig/dL (lavicoli et al.. 2006a; lavicoli et al. 2004).
 2                  Toxicological studies have reported delayed male sexual maturity as measured with sex
 3                  organ weight, among other outcomes, seeing significant decrements at blood Pb levels of
 4                  34 (ig/dL (Sokol et al.  1985). Thus, data from the toxicological literature and from
 5                  epidemiologic findings demonstrate that puberty onset in both males and females is
 6                  delayed with Pb exposure.

 7                  Findings from epidemiologic studies of postnatal growth are inconsistent and findings
 8                  from the toxicological literature are mixed with recent growth findings showing adult
 9                  onset obesity. Toxicological studies demonstrated that the effects of Pb exposure during
10                  early development include impairment of retinal development and alterations in the
11                  developing hematopoietic and hepatic systems. Affected developmental outcomes with
12                  Pb exposure also included effects on the eyes and teeth.

13                  The collective body of evidence integrated across epidemiologic and toxicological
14                  studies, based on the findings of delayed pubertal onset among males and females, is
15                  sufficient to conclude that there is a causal relationship between Pb exposure and
16                  developmental effects.
                    5.8.6.2     Effects on Birth Outcomes

17                  Overall, results of pregnancy outcomes were similar to those of the 2006 Pb AQCD;
18                  (U.S. EPA. 2006b) inconsistent evidence of a relationship with Pb was available for
19                  preterm birth. The 2006 Pb AQCD included a few studies that reported potential
20                  associations between Pb and neural tube defects, but the recent epidemiologic studies
21                  found no association. Some associations were observed between Pb and low birth weight
22                  when epidemiologic studies used measures of postpartum maternal bone Pb or air
23                  exposures. The associations were less consistent when using maternal blood Pb measured
24                  during pregnancy or at delivery or umbilical cord and placenta Pb (maternal blood Pb or
25                  umbilical cord and placenta Pb were the biomarkers most commonly used in studies of
26                  low birth weight) but some associations between increased Pb levels and decreased low
27                  birth weight/fetal growth were observed.  The  effects of Pb exposure during gestation in
28                  animal toxicological studies included mixed findings with some studies showing
29                  reduction in litter size, implantation, and birth weight, and some showing no effect.
30                  Based on the mix of inconsistent results of studies on various birth outcomes but some
31                  associations observed in well-conducted epidemiologic studies of preterm birth and low
32                  birth weight/fetal growth, the evidence is suggestive of a relationship between Pb
33                  exposure and birth outcomes.
      November 2012                            5-630                  Draft - Do Not Cite or Quote

-------
                    5.8.6.3     Effects on Male Reproductive Function

 1                  Toxicological evidence and supporting epidemiologic evidence indicate that a causal
 2                  relationship exists between Pb exposure and effects on male reproductive function. Key
 3                  evidence is provided by toxicological studies in rodents, non-human primates, and rabbits
 4                  showing detrimental effects on semen quality, sperm and fecundity/fertility with
 5                  supporting evidence in epidemiologic studies of associations between Pb exposure and
 6                  detrimental effects on sperm. This is consistent with studies reported in the
 7                  2006 Pb AQCD (U.S. EPA. 2006b).

 8                  Toxicological studies with relevant Pb exposure routes reported effects on rodent sperm
 9                  quality and sperm production rate [(Sokol and Berman. 1991; Sokol etal.. 1985). blood
10                  Pb level 34-37 (ig/dL], sperm DNA damage lYNava-Hernandez et al.. 2009). blood Pb
11                  levels 19 and 22 jig/dL], and histological or ultrastructural damage to the male
12                  reproductive organs in studies from rodents [(El Shafai et al.. 2011). blood Pb level
13                  5.1 (ig/dL] and non-human primates [(Cullen etal.. 1993). blood Pb level 43 (ig/dL].
14                  These effects were found in animals exposed to Pb during peripuberty or adults for 1
15                  week to 3 months. The toxicological studies reported an association between Pb exposure
16                  and decreases in reproductive organ weight and organ histological changes in the testes
17                  and germ cells. Subfecundity (decreased number of pups born/litter) was reported in
18                  unexposed females mated to Pb exposed males. Also, sperm from Pb-exposed rats (blood
19                  Pb level: 33 to 46 ug/dL) used for in vitro fertilization of eggs harvested from unexposed
20                  females yielded lower rates of fertilization.  (Sokol etal..  1994). Supporting evidence was
21                  provided by decrements in sperm quality from rabbits administered Pb subcutaneously
22                  (blood Pb levels of 25 (ig/dL) (Moorman et al.. 1998).

23                  The detrimental effects of Pb on sperm were observed in epidemiologic studies with
24                  concurrent blood Pb levels of 25 (ig/dL and greater among men occupationally exposed
25                  (Hsu et  al.. 2009b: Kasperczyk et al.. 2008; Naha and Manna. 2007; Nahaand
26                  Chowdhury. 2006). The epidemiologic studies were limited due to these high exposure
27                  levels among the occupational cohorts and the lack of consideration for potential
28                  confounding factors, including other occupational exposures. Studies among men with
29                  lower Pb levels were limited to infertility clinic studies, which may be a biased sample
30                  and lack generalizability. However, a well-conducted epidemiologic study that enrolled
31                  men going to a clinic for either infertility issues or to make  a semen donation and
32                  controlled for other metals as well as smoking reported a positive association with
33                  various  detrimental effects in sperm (Telisman et al.. 2007). The median concurrent
34                  blood Pb levels in this study were 4.92  (ig/dL (range:  1.13-14.91). A similar study
35                  (Meeker et al.. 2008) also reported possible associations between concurrent blood Pb
      November 2012                            5-631                 Draft - Do Not Cite or Quote

-------
 1                  and various semen parameters, but the results were extremely imprecise, making it
 2                  difficult to draw conclusions.

 3                  Similar to the 2006 Pb AQCD (U.S. EPA. 2006b). recent epidemiologic and toxicological
 4                  studies reported inconsistent results regarding hormone aberrations associated with Pb
 5                  exposure. Due to the complexity of the reproductive system, uncertainty exists as to
 6                  whether Pb exerts its toxic effects on the reproductive system by affecting the
 7                  responsiveness of the hypothalamic-pituitary-gonad axis, by suppressing circulating
 8                  hormone levels or by some other pathway. Mixed findings were also apparent among
 9                  epidemiologic studies of fertility among men.

10                  More recent toxicological studies suggest that oxidative stress is a major contributor to
11                  the effects of Pb exposure on the male reproductive system, providing mode of action
12                  support. The effects of ROS may involve interference with cellular defense  systems
13                  leading to increased lipid peroxidation and free radical attack on lipids, proteins, and
14                  DNA. Several recent studies showed that Pb induced an increased generation of ROS
15                  within the male sex organs, and germ cell injury, as evidenced by aberrant germ cell
16                  structure and function. Co-administration of Pb with various antioxidant compounds
17                  either eliminated Pb-induced injury or greatly attenuated its effects. In addition,  many
18                  studies that observed increased oxidative stress also observed increased apoptosis which
19                  is likely a critical underlying mechanism in Pb-induced germ cell DNA damage  and
20                  dysfunction.

21                  Based on the consistency and coherence of findings for the detrimental effects of Pb
22                  exposure on sperm and semen in the toxicological literature, the support from
23                  epidemiologic studies, and biological plausibility provided by mode of action evidence,
24                  the evidence is sufficient to conclude that there is a causal relationship between Pb
25                  exposures and male reproductive function.
                    5.8.6.4     Effects on Female Reproductive Function

26                  Epidemiologic and toxicological studies of reproductive function among females
27                  investigated whether Pb biomarker levels were associated with hormone levels, fertility,
28                  estrus cycle changes, and morphology or histology of female reproductive organs
29                  including the placenta. Toxicological studies reported in the 2006 Pb AQCD (U.S. EPA.
30                  2006b) reported associations between Pb exposure and female reproductive function,
31                  although little evidence was provided by epidemiologic studies. Some epidemiologic
32                  studies have shown associations with concurrent blood Pb levels and altered hormone
33                  levels in adults, but varied among studies,  likely due to the different hormones examined
34                  and the different timing in the menstrual and life cycles. There is some evidence of a

      November 2012                            5-632                 Draft - Do Not Cite or Quote

-------
 1

 2

 3

 4

 5

 6

 7

 8

 9

10

11

12

13
potential relationship between Pb exposure and female fertility, but findings are mixed.

The majority of the epidemiologic studies are cross-sectional, and adjustment for

potential confounders varies from study to study, with some potentially important

confounders, such as BMI, not included in all studies. Also, most of the studies have

small samples sizes and are generally of women attending infertility clinics.

Toxicological study design often employs prenatal or early postnatal Pb exposures with

Pb contributing to placental pathology and inflammation, decreased ovarian antioxidant

capacity, altered ovarian steriodogenesis and aberrant gestational hormone levels.

Although epidemiologic and toxicological studies provide information on different

exposure periods, both types of studies support the conclusion that Pb possibly affects at

least some aspects of female reproductive function. Overall, the relationship observed

with female reproductive outcomes is sufficient to conclude that there is a suggestive

relationship between Pb exposure and female reproductive function.
      Table 5-48   Summary of evidence supporting reproductive and developmental
                     causal determinations.
      Attribute in Causal
      Framework3
     Key Supporting Evidence
Recent References
Pb Exposure or
Blood Pb Levels
Associated with Effects0
      Effects on Development - Causal
      Delayed Puberty Onset

       Consistent          Consistent evidence in multiple large
       associations        cross-sectional epidemiologic studies for
       between higher      females and males plus a longitudinal
       blood Pb levels in    study for males. Most of these studies
       high-quality         have large sample sizes and controlled for
       epidemiologic       potential confounding by covariates such
       studies            as age, BMI, and education/SES.
                                       Tomoum et al. (2010).
                                       Mauser et al. (2008),
                                       Williams et al. (2010).
                                       Denham et al .(2005).
                                       Naickeretal. (2010).
                                       Wu et al. (2003b).
                                       Gollenberg et al. (2010)
                                       Sections
                                       5.8.1.1 and 5.8.1.2
                      Concurrent blood Pb levels:
                      <10 ug/dL

Epidemiologic
evidence supported
by consistent
toxicological findings
with relevant Pb
exposure
Evidence clearly
describes Mode of
Action.
A large study using females aged 8-18
years from the NHANES III study also
controlled for various dietary factors and
reported associations between blood Pb
levels and delayed puberty onset
Consistent toxicological evidence from
multiple laboratories of delayed male and
female puberty onset with Pb exposure via
diet or oral gavage in rodents
Toxicological evidence describes HPG axis
dysfunction providing mechanism of action
support for delayed puberty findings. MOA
further supported by IGF-1 changes
contributing to Pb-induced delay in puberty
onset.
Selevan et al. (2003)
Dumitrescu et al. (2008c).
lavicoli et al. (2006a).
Pine et al. (2006)
Sections
5.8.1.1 and 5.8. 1.2

Section 5.8.1.1


Blood Pb level after dietary
exposure from gestation to
estrus:
3.5-13ug/dL

      November 2012
                            5-633
            Draft - Do Not Cite or Quote

-------
Attribute in Causal
Framework3
  Key Supporting Evidence
 Recent References
Pb Exposure or
Blood Pb Levels
Associated with Effects0
Postnatal Growth
  Available
  toxicological studies
  reported mixed
  findings of effects of
  Pb on postnatal
  growth
 There are mixed findings in the
 toxicological literature on Pb exposure and
 postnatal growth with some studies
 showing stunted growth in  animals
 exposed to Pb and some showing no
 effects.
 Recent toxicological evidence of effect of
 Pb on postnatal growth: obesity in adult
 male offspring after gestational +
 lactational Pb exposure
                                                                Section 5.8.1.3
  Available
  epidemiologic
  findings of
  associations
  between higher
  blood Pb levels and
  postnatal growth are
  inconsistent
 Multiple studies, mostly cross-sectional, for
 children of varying ages have reported
 inconsistent results for the association
 between blood Pb levels and various
 measures of growth
                                                                Section 5.8.1.3
Impaired Organ Systems
Consistent
toxicological
findings of effects
on sensory organ
systems, bone,
teeth, and Gl
system but not
always at relevant
Pb exposure levels.
Relevant gestational and
exposure of rats resulted
lactational Pb Fox et al. (2008)
in retinal ERG
aberrations and increased retinal cell layer
thickness.











Blood Pb level after
gestational-lactational
exposure:
10-12 ug/dL





Effects on Birth Outcomes - Suggestive
Inconsistent findings
in epidemiologic
studies of various
birth outcomes
Inconsistent findings for studies for birth
defects, preterm birth, and low birth
weight/fetal growth.

A few well-conducted epidemiologic studies
of preterm birth and low birth weight/fetal
growth using measures of maternal blood
Pb at the time of pregnancy reported
associations.
See Section 5.8.3 (and all
subsections): 5.8.3.1.
5.8.3.2. 5.8.3.3. 5.8.3.4

Jelliffe-Pawlowski et al.
(2006).
Vigehetal.(2011).
Zhu et al. (2010).
Chenetal.  (2006a).
Gundacker et al. (2010)
                                                                                          Maternal pregnancy blood
                                                                                          Pb levels:
                                                                                          >10 ug/dL
Inconsistent findings
in toxicological
literature for birth
outcomes
The toxicological literature reported mixed
findings with some studies showing smaller
litter size (fewer pups born) or decreased
birth weight with Pb exposure and some
studies showing no effect.
See Section 5.8.3.4 and
5.8.3.1
November 2012
                             5-634
               Draft - Do Not Cite or Quote

-------
Attribute in Causal
Framework3
 Key Supporting Evidence
 Recent References
 Pb Exposure or
 Blood Pb Levels
 Associated with Effects0
Effects on Male Reproductive Function - Causal
High-quality and
consistent findings in
toxicological studies of
detrimental effects of
Pb exposure on
sperm or semen in
multiple species.
Decreased sperm counts, decreased
sperm production rate, dose-dependent
suppression of spermatiogenesis in
rodents with drinking water Pb exposure.
All results:
Section 5.8.4.1
Sokoletal. (1985)

Sokol and Berman (1991)
Blood Pb level after adult
drinking water exposure for
30 days: 34 ug/dL

Blood Pb level after
peripubertal or adult drinking
water exposure for 30 days:
35 and 37 ug/dL.
                       Ultrastructural damage and histological to
                       non-human primates testis and
                       seminiferous tubules.
                       Histologic damage to rodent seminiferous
                       tubules including spermatids and
                       developing sperm.

                       Ultrastructural abnormalities to rat
                       spermatogenesis.


                       Direct effects on rodent sperm DMA after
                       drinking water Pb exposure.


                       Sperm from Pb exposed rats used for in
                       vitro fertilization of eggs harvested from
                       unexposed females yielded lower rates of
                       fertilization.

                       Semen and  sperm quality in rabbits with
                       subcutaneous Pb treatment; Ultrastructural
                       damage to spermatids with  i.p. injection of
                       Pb.

                       Findings of detrimental effects of Pb
                       exposure on sperm from multiple species
                       (Deer, Asian earthworm, rainbow trout,
                       marine worm, H. elegans,Fathead
                       minnow)
                                         Singh et al. (1993a)
                                         Foster etal. (1998)
                                         El Shafai et al. (2011)
                                         Murthv etal.(1995)
                                         Nava-Hernandez et al.
                                         (2009)
                                         Sokol et al.(1994)
                                         Moorman etal.(1998).
                                         See Ecological Effects;
                                         Sections 7.4.12.1 and
                                         7.4.21.1)
                          Maximum blood Pb levels
                          after daily oral Pb exposure
                          (gelatin capsule) during
                          infancy, post infancy, or over
                          a lifetime (up to  10yr):
                          32 to 36 ug/dL

                          Blood Pb level after adult
                          exposure (oral gavage) for 3
                          months: 5.31 ug/dL

                          Blood Pb level after i.p.
                          injection for 16 days:
                          7.4 ug/dL

                          Blood Pb level after adult
                          exposure for 13  weeks:
                          19 and 22 ug/dL

                          Blood Pb level after adult
                          exposure for 14-60 days:
                          33-46 ug/dL


                          Blood Pb level after adult
                          exposure for 15  weeks:
                          16-24ug/dL
Toxicological evidence
is supported by
consistent findings in
epidemiologic studies
of associations
between higher blood
Pb levels and
decrements in sperm
count and quality in
occupational cohorts
Consistent evidence from studies of
occupational cohorts with high blood Pb
levels. Results from occupational cohorts
may have been confounded by other
workplace exposures, which were not
adjusted for in the epidemiologic studies.
Potential confounding by smoking was
considered in one study.

Results less consistent at lower blood Pb
level. A well-conducted epidemiologic
study at an infertility clinic reported
associations between detrimental effects
in sperm and blood Pb levels after
controlling for smoking and other metal
exposure. A similar study also reported
some elevated effect estimates but the
results were too imprecise to draw
definitive conclusions.
Naha and Manna (2007).
Naha and Chowdhury
(2006). Hsu et al.
(2009b). Kasperczyk et al.
(2008)
Telisman etal. (2007),
Meeker etal. (2008)
Concurrent blood Pb levels:
> 25 ug/dL
Concurrent blood Pb level:
<10ug/dL
November 2012
                            5-635
               Draft - Do Not Cite or Quote

-------
Attribute in Causal
Framework3
  Key Supporting Evidence
 Recent References
 Pb Exposure or
 Blood Pb Levels
 Associated with Effects0
Evidence describes
mode of action for
effects on sperm.
Inconsistent findings
of associations
between Pb levels
and hormone levels in
epidemiologic studies;
few studies available
Lack of large, well-
conducted
epidemiologic studies
examining
associations between
Pb levels and fertility
Limited toxicological
findings of Pb
exposure inducing
effects on fertility
Consistent MOA evidence in reproductive
organs of Pb-exposed male animals of
increased apoptosis, decreased
antioxidant activity (SOD and CAT), and
increased oxidative stress (MDA).
There are a small number of studies
examining hormone levels and the results
are inconsistent.
The few epidemiologic studies examining
this outcome generally have small
samples sizes and are drawn from men
attending infertility clinics.
Paternal Pb exposure resulted in less
successful copulation, fewer implantations,
and longer periods of time copulating for
successful matings. Unexposed females
with Pb-exposed male partners did not
have fewer pregnancies, but did produce
smaller litters.
Sections 5. 8. 4.1 and
5.8.4.2
Telisman et al. (2007).
Naha and Manna (2007).
Hsiehetal. (2009a),
Meeker etal. (2010),
Mendiola et al. (2011)
Section 5.8.4.2
Kizileretal. (2007).
Bloom etal. (201 1b).
Bloom etal. (2010).
Gracia et al. (2005)
Section 5.8.4.3
Anjum et al. (2010) 45-day exposure of adult
Sainatha et al. (2011) male rats to 500 or
Pace et al. (2005) 1 ,500 ppm Pb acetate
Section 5.8.4.3 exposure in drinking water,
followed by behavioral
mating studies with
unexposed females.
Effects on Female Reproductive Function - Suggestive
Epidemiologic studies
of Pb levels and
hormones
demonstrate
associations but are
inconsistent overall
Evidence in some high-quality cross-
sectional epidemiologic studies
demonstrates associations with hormone
levels but results are mixed and vary by
hormone examined and timing in  a
woman's menstrual and life cycles.  In
addition, the potential confounders vary
from study to study, with some potentially
important confounders, such as BMI, not
included in all studies.
Jackson et al. (2011).
Pollack etal. (2011).
Chang et al. (2006).
Krieg (2007)
Section 5.8.5.1
Concurrent mean blood
Pb levels:
<5 ug/dL
Lack of large, well-
conducted
epidemiologic studies
examining
associations between
Pb levels and fertility
Epidemiologic studies of this association
are limited by the small sample sizes
included in those studies. In addition, most
of the study populations were drawn from
women undergoing IVF and/or attending
infertility clinics.
                                                               Section 5.8.5.2
Toxicological studies
of Pb and effects on
female reproduction
demonstrate effects in
some studies.
Evidence in the toxicological literature of
Pb contributing to placental pathology and
inflammation, decreased ovarian
antioxidant capacity, altered ovarian
steriodogenesis and aberrant gestational
hormone levels.
Section 5.8.5.1 and
5.8.5.3
"Described in detail in Table II of the Preamble.
bDescribes the key evidence and references contributing the most heavily to causal determination. References to earlier sections
indicate where full body of evidence is described.
""Describes the blood Pb level in humans with which evidence is substantiated or the blood Pb levels or Pb exposure
concentrations in animals relevant to humans.
November 2012
                             5-636
               Draft - Do Not Cite or Quote

-------
      5.9       Effects on  Other Organ Systems

 1                  The 2006 Pb AQCD described limited evidence for the effects of Pb exposure on various
 2                  organ systems including the liver, GI tract, endocrine system, bone and teeth, eyes, and
 3                  respiratory tract. These lines of evidence largely are supported by recent toxicological
 4                  and epidemiologic studies, although the collective evidence remains relatively limited in
 5                  terms of the quantity and design of studies and/or the populations examined.
      5.9.1        Effects on the Hepatic System

 6                  Hepatotoxic effects of Pb exposure indicated in various animal models and/or human
 7                  populations include alterations in hepatic metabolism, hepatic cell proliferation, changes
 8                  in cholesterol metabolism, as well as oxidative stress-related injury.
                    5.9.1.1     Summary of Key Findings of the Effects on the Hepatic
                                System (2006 Pb AQCD)

 9                  The 2006 Pb AQCD (U.S. EPA. 2006b) stated that the experimental animal database
10                  indicated hepatotoxic effects, including liver hyperplasia, at very high dose Pb exposures.
11                  Other effects noted in the liver following exposure to Pb included altered cholesterol
12                  synthesis, DNA synthesis, and glucose-6-phosphotase dehydrogenase (G6DP) activity.
13                  The 2006 Pb AQCD reported that cytochrome (CYP) P450 levels decreased following
14                  single doses of Pb nitrate. Induced and constitutive expression of microsomal CYP1A1
15                  and CYP1A2 was inhibited by Pb exposure. Inhibition of these (Phase I) xenobiotic
16                  metabolizing enzymes was accompanied by an increase in Phase II enzymes following
17                  exposure to Pb  nitrate and other Pb compounds, suggesting that Pb is capable of inducing
18                  a biochemical phenotype similar to hepatic nodules. Studies related to Pb-induced hepatic
19                  hyperplasia suggested alterations in the gluconeogenic mechanism, DNA
20                  hypomethylation, changes in proto-oncogene expression, as well as cholesterol synthesis.
21                  Cholesterol metabolism changes following exposure to Pb were reportedly mediated by
22                  induction of several enzymes related to cholesterol metabolism as well as a decrease in
23                  the cholesterol  catabolizing enzyme, 7 a-hydroxylase. Tumor necrosis factor alpha
24                  (TNF-a) was reported to be one of the major mitogenic signals that mediated Pb nitrate-
25                  induced hepatic hyperplasia, based on findings showing that inhibitors blocking TNF-a
26                  activity also blocked Pb-induced hyperplasia. Other Pb-related effects presented in the
27                  2006 Pb AQCD included liver cell apoptosis mediated by Kupffer cell derived signals
28                  and Pb-induced oxidative stress in vitro cell cultures. The 2006 Pb AQCD further
29                  suggested that alterations in liver heme metabolism may involve changes in


      November 2012                            5-637                 Draft - Do Not Cite or Quote

-------
 1                   5-aminolevulinic acid dehydrogenase (ALAD) activity, porphyrin metabolism, transferrin
 2                   gene expression and changes in iron metabolism.

 3                   With regard to human studies, the 2006 Pb AQCD stated that increases in serum liver
 4                   enzymes suggest that Pb exposure results in nonspecific liver injury in occupationally-
 5                   exposed adults. However, studies did not adjust for potential confounding factors,
 6                   including other occupational exposures or establish explicit associations between Pb
 7                   exposure and hepatic injury (i.e., observation of histopathological effects). In addition,
 8                   similar to effects were noted in animal studies, and decreased cytochrome P450 activity
 9                   was associated with higher blood Pb levels in a few studies of children or adults (drawn
10                   from the general population). The 2006 Pb AQCD reported that hepatic effects in humans
11                   were associated only with high blood Pb levels, i.e., >30 (ig/dL.
                     5.9.1.2      Recent Epidemiologic Studies

12                   A few epidemiologic studies examined antioxidant status and oxidative stress effects, as
13                   measured by liver biochemical parameters, associated with occupational exposure to Pb.
14                   However, all of these occupationally-exposed cohorts represented populations highly
15                   exposed to Pb, with mean or median blood Pb levels ranging from 29 to 53 (ig/dL.
16                   Although the hepatic effects observed within these cohorts may not be  generalizable to
17                   the general population as a whole, they are useful in demonstrating consistent effects on a
18                   number of liver outcomes, including altered liver function (i.e., changes in the level of
19                   liver function enzymes), oxidative stress, and antioxidant status (Can et al.. 2008; Khan et
20                   al., 2008; Patil et al., 2007). However, these studies were cross-sectional in design with
21                   concurrent blood Pb measurement. Thus, there is uncertainty regarding the directionality
22                   of effects and the magnitude, timing, frequency, and duration of Pb exposure that
23                   contributed to  the observed associations. Further, analyses did not consider potential
24                   confounding by factors such as age, diet, BMI, smoking, or other occupational exposures.

25                   In spray painters from Kolhapur City in western Maharashtra, India, exposed to Pb for
26                   >6 hours/day for 2 to 20 years examined by Patil et al. (2007). mean concurrent (SD)
27                   blood Pb levels in 30 workers (mean [SD]: 22.32 [8.87]  (ig/dL) were significantly higher
28                   (p <0.001, t-test) than those in the 35  concurrent controls (mean [SD]:  12.52
29                   [4.08]  (ig/dL), who had no history of Pb exposure and lived in rural areas. Levels of liver
30                   function enzymes, including the two serum transaminase enzymes SGOT (also known as
31                   AST; serum glutamic oxaloacetic transaminase/aspartate aminotransferase) and SGPT
32                   (also known as ALT; serum glutamic pyruvic transaminase/alanine aminotransferase),
33                   were increased in spray painters compared to those in controls, whereas total serum
34                   protein levels were decreased compared to controls (p <0.01, t-test).  In another
      November 2012                             5-638                  Draft - Do Not Cite or Quote

-------
 1                  occupational study, Conterato et al. (In Press) investigated liver function parameters in
 2                  automotive painters exposed to Pb in Brazil. Mean (SD) concurrent blood Pb levels were
 3                  5.4 (0.4) (ig/dL in the 50 exposed painters and 1.5 (0.1) (ig/dL in the 36 unexposed
 4                  controls. The mean (SD) duration of exposure to Pb in painters was 133.9 (14.5) months.
 5                  In exposed workers, the levels of AST, but not y-glutamyltransferase, were increased
 6                  approximately 2-fold compared to levels in controls (p <0.05). The activity of AST was
 7                  positively correlated with blood Pb levels (r = 0.26, p <0.05). The authors suggested that
 8                  confounding exposures to toxic constituents of the paints regularly used by painters, and
 9                  not Pb, may be the etiological cause of decrements in AST function as these effects were
10                  not also seen in battery workers with much higher blood Pb  levels (49.8 (ig/dL)
11                  (Conterato et al.. In Press).  Co-exposure to other environmental contaminants may also
12                  explain the effects that were previously reported in occupationally-exposed spray-
13                  painters in Patil et al. (2007).
                    5.9.1.3      Recent lexicological Studies

                    Hepatic Metabolism

14                  As stated in the 2006 Pb AQCD (U.S. EPA. 2006b). acute (e.g., single dose) treatment of
15                  rodents with Pb nitrate and other Pb compounds was found to result in a decrease in
16                  Phase I enzymes and a simultaneous increase in Phase II enzymes. The conclusions
17                  presented in the 2006 Pb AQCD were also reviewed by Mudipalli (2007).

18                  Recent studies found changes in biochemical parameters, suggestive of liver damage, in
19                  animals exposed to Pb; however, the relevance to humans is uncertain because of the
20                  high blood Pb levels used in animal studies and the exposure routes of Pb administration.
21                  Undernourished male Wistar rats (fed low-protein diet without mineral supplements)
22                  exposed to 500 ppm Pb acetate in drinking water over a 10 month period had decreases in
23                  serum protein and albumin levels as well as  increases in AST, ALT, serum alkaline
24                  phosphatase (ALP), and gamma glutamyl transpeptidase (GGT) levels (Herman et al..
25                  2009). In Pb-treated animals, the blood Pb levels steadily increased throughout the initial
26                  portion of the study period, reaching a maximum of approximately 30 (ig/dL after
27                  2 months.  After this time, blood Pb levels rapidly increased to approximately 110 (ig/dL
28                  by six months time, and remained at this level until the termination of exposure at
29                  10 months. Similar biochemical changes were not observed in animals treated with
30                  Pb acetate maintained on protein-adequate, mineral rich diet.

31                  Similarly, mice gavaged with Pb nitrate (50  mg/kg for 40 days) also demonstrated
32                  increased activities of AST, ALT, ALP, and acid phosphatase (ACP) compared to
      November 2012                            5-639                  Draft - Do Not Cite or Quote

-------
 1                  controls (Sharma et al., 2010a). Upadhyay et al. (2009) reported that treatment of
 2                  Sprague-Dawley rats with Pb acetate (35 mg/kg via i.p. injection for 3 days, blood Pb not
 3                  reported) significantly increased the activities of ALT, AST, serum ALP, and acid
 4                  phosphatase over those in controls but decreased liver ALP activity. Concomitant
 5                  treatment with Zn and varying levels of vitamin C were observed to ameliorate the toxic
 6                  effects of Pb. The serum activities of glutamic pyruvic transaminase (GPT) and lactate
 7                  dehydrogenase (LDH) were similarly significantly increased over those in controls in
 8                  mice subcutaneously injected with Pb acetate (50 mg/kg daily for 15 days, blood Pb not
 9                  reported) (Wang etal. 2010g). Swarup et al. (2007) investigated serum biochemical
10                  changes in cows living in Pb-contaminated environments. Serum levels of ALT, AST,
11                  ALP, total protein, albumin, globulin, and A/G ratio were significantly altered in cows
12                  living near Pb-Zn smelters (mean [SD] blood Pb level: 86 [6] (ig/dL) compared to control
13                  cows (mean [SD] blood Pb level: 7 [1]  (ig/dL). Significant positive correlations were
14                  found between blood Pb level and ALT and AST, whereas a negative correlation was
15                  observed between blood Pb level and total lipids, protein, and albumin.

16                  Pillai et al. (2009) investigated hepatic Phase I and II enzymes in male and female rats
17                  born to dams that were treated with Pb acetate (50 (ig/kg, via s.c. injection daily
18                  throughout gestation and continuing until PND21). Thus, the offspring of treated dams
19                  were exposed to Pb via placental and lactational transfer. The female and male pups were
20                  then allowed to reach sexual maturity (PND55-PND56) to assess continuing exposure to
21                  bioaccumulated Pb. The activities of hepatic Phase I enzymes NADPH- and NADH-
22                  cytochrome  c reductase were significantly  reduced in Pb-exposed male and female rats
23                  on PND56 (blood Pb not reported), compared to controls. In rats treated with 25 (ig/kg Pb
24                  and Cd, the effect on Phase I enzymes was increased. Pb treatment additionally decreased
25                  the activities of Phase  II enzymes uridine diphosphate-glucoronyl transferase and GST in
26                  males and females, but no effect was observed on GGT or 17|3-hydroxysteroid
27                  oxidoreductase. Additionally, no effect was observed on serum glutamate pyruvate
28                  dehydrogenase or ALP activities in Pb-treated males or females. Histological
29                  observations demonstrated fatty degeneration of the liver, vacuolization, and pycnotic
30                  nuclei, indicating general hepatotoxicity following Pb treatment in both male and female
31                  rats.

32                  In a similar study, Teijon et al. (2006) exposed Wistar rats to Pb acetate (200 or 400 ppm
33                  drinking water) throughout gestation, lactation, and 3 months postweaning, or only
34                  1 month postweaning. In the animals exposed continuously throughout gestation and
35                  lactation, the concentrations of Pb in the liver were elevated in the 200- and 400-ppm
36                  groups 1 and 3 months postweaning. Liver concentrations of Pb were greater in the
37                  200 ppm animals compared to the 400 ppm animals at one month postweaning (mean
38                  [SE]: 1.19 [0.30] (ig Pb/g tissue versus 0.76 [0.06] ng Pb/g tissue, respectively), but were
      November 2012                            5-640                 Draft - Do Not Cite or Quote

-------
 1                  similar between the 2 dosing regimens (200 ppm versus 400 ppm) at 3 months
 2                  postweaning (mean [SE]: 0.54 [0.06] versus 0.55  [0.07] \ig Pb/g tissue, respectively).
 3                  ALP activity was increased at 2 weeks postweaning in animals continuously exposed to
 4                  Pb throughout gestation and lactation, whereas ALT activity was decreased only at 2 and
 5                  3 months postweaning. In animals exposed for 1 month postweaning alone, only serum
 6                  ALP activity was significantly increased, although not in a concentration-dependent
 7                  manner. ALT and AST activities did not show significant changes.

 8                  Cheng et al. (2006) studied the mechanism of Pb effects on bacterial lipopolysaccharide
 9                  (LPS)-induced TNF-a expression. A/J mice were  injected with Pb acetate (100 (imol/kg
10                  via i.p.), with or without LPS (5 mg/kg). Pb alone did not affect AST or ALT activity or
11                  the level of TNF-a in the serum of the mice. In comparison, treatment of mice with low
12                  doses of Pb and LPS together caused a statistically significant increase in TNF-a
13                  induction as well as enhanced liver injury, suggesting that Pb potentiated LPS-induced
14                  inflammation. In a complementary in vitro experiment, co-exposure of Pb and LPS
15                  stimulated the phosphorylation of p42/44 mitogen-activated protein kinase (MAPK) and
16                  increased TNF-a expression in mouse whole blood cells, peritoneal macrophages, and
17                  RAW264.7 cells (a macrophage cell line). These results indicated that Pb increased LPS-
18                  induced TNF-a levels via the protein kinase C (PKC)/MAPK pathway in
19                  monocytes/macrophages rather than hepatocytes.  Similarly, Pb chloride potentiated
20                  bovine serum albumin (BSA)-induced inflammation in the livers  of mice subcutaneously
21                  injected with Pb (Saet al.. 2012).


                    Lipid Metabolism

22                  Several recent toxicological studies indicated Pb-induced impaired lipid metabolism, as
23                  evidenced by increases in liver cholesterol. There was some evidence in animals exposed
24                  to Pb in diet, albeit at relatively high exposure concentrations or measured blood Pb.
25                  Ademuyiwa et al. (2009) reported that male albino Sprague Dawley rats exposed to 200,
26                  300 and 400 ppm Pb acetate in drinking water had mean (SD) blood Pb levels of 40.63
27                  (9.21), 61.44 (4.63), and 39.00 (7.90) (ig/dL, respectively. Animals exposed to 200 ppm
28                  Pb had mean (SD) liver Pb concentrations of 10.04 (1.14) (ig/g, compared to 3.24
29                  (1-19) (ig/g and 2.41 (0.31) (ig/g in animals exposed to 300 or 400 ppm Pb, respectively.
30                  Animals exposed to Pb exhibited increased hepatic cholesterogenesis at all doses tested
31                  compared to controls. Additionally, a decrease in  triglyceride levels was observed at 300
32                  and 400 ppm Pb; a decrease in phospholipid levels was observed  at 400 ppm Pb. The
33                  authors also reported positive correlations between tissue cholesterol and phospholipids
34                  and Pb accumulation in liver across all doses. In contrast, the association between tissue
35                  triglyceride levels and Pb accumulation was negative. In related studies, Khotimchenko
36                  and Kolenchenko (2007) reported that adult male  albino rats treated with Pb acetate

      November 2012                            5-641                  Draft - Do  Not Cite or Quote

-------
 1                  (100 mg/kg for 14 days, blood Pb not reported) exhibited disorders in lipid metabolism
 2                  that were accompanied by increased levels of total cholesterol and triglycerides in the
 3                  liver tissue. Sharma et al. (2010a) reported increased liver cholesterol in mice gavaged
 4                  with Pb nitrate, 50 mg/kg for 40 days. Pillai et al. (2009) observed decreases in total liver
 5                  cholesterol in PND56 male and female rats that had been treated with Pb acetate (via s.c.
 6                  injection, 50 (ig/kg, continuously throughout gestation and lactation). These results
 7                  suggest that Pb induction of cholesterogenesis  and phospholipidosis in the liver may
 8                  cause subtle effects  at the cellular level that may lead to hepatotoxicity.

 9                  Kojima and Degawa (2006) examined sex-related differences in Pb-induced gene
10                  expression of a rate  limiting hepatic cholesterol biosynthesis enzyme, 3-hydroxy-3-
11                  methylglutaryl-CoA reductase (HMGR) and its transcription factor, sterol regulatory
12                  element binding protein-2 (SREBP-2). Male and female  Sprague Dawley rats were
13                  injected with Pb nitrate (100 (imol/kg body weight, intravenously, blood Pb not reported).
14                  SREBP-2 expression was significantly increased in males and females with the increase
15                  occurring earlier in male rats (6-12 hours, compared to 24-36 hours in females). In
16                  contrast, expression of HMGR was significantly increased in both Pb-exposed males and
17                  females at earlier time frames and greater range of onset (3-48 hours in males; 12-48
18                  hours in females) compared to that of SREBP-2. Significant increases in total liver
19                  cholesterol were also observed in Pb-exposed males and females at 3-48 and 24-48 hours,
20                  respectively. These results suggest that the SREBP-2 and HMGR gene expressions and
21                  increase in total cholesterol levels in the liver in response to Pb exposure occur earlier in
22                  males compared to females and also suggest that the HMGR gene expression and
23                  increase in total cholesterol levels in the liver occur before an increase in the SREBP-2
24                  gene expression in each sex.


                    Hepatic Oxidative Stress

25                  A number of studies demonstrated increased hepatic oxidative stress as a result of
26                  exposure to various Pb compounds, demonstrated by increases in reactive  oxygen species
27                  (ROS) or decreases  in antioxidant levels or enzyme activity. ROS can potentially result in
28                  damage to hepatic function and structure.  Several of these observations were made in
29                  animals exposed to Pb in drinking water that produced blood Pb levels relevant to
30                  humans. In a study examining the effects of Pb exposure to fetuses, Masso et al. (2007)
31                  exposed pregnant Wistar rats to 300 ppm Pb in drinking water from GD1 to parturition,
32                  or to weaning. Blood Pb levels were higher at parturition (mean [SD]: 31.5 [0.80]  (ig/dL)
33                  than at weaning (mean [SD]: 22.8 [0.50] (ig/dL). Pups exhibited liver damage that was
34                  accompanied by an increased  production of thiobarbituric acid-reactive species (TEARS,
35                  an indicator of lipid peroxidation) and increased  CAT activity compared to controls.  In
36                  addition, increased ALP and acid phosphatase  activity was observed. Uzbekov et al.

      November 2012                             5-642                  Draft - Do Not Cite or Quote

-------
 1                  (2007) showed differential effects by duration of maternal exposure before mating.
 2                  Female Wistar rats were exposed to 0.3 and 3.0 ppm Pb nitrate in drinking water, for
 3                  1 month or 5 months prior to pregnancy, and also continuing during pregnancy; then the
 4                  livers from both groups of GD20 fetuses were examined for hepatic SOD activity. The
 5                  pregnant control rats had a mean (SD) blood Pb level of 16.1 (0.63) (ig/dL, whereas the
 6                  dams exposed to 0.3 and 3.0 ppm Pb had mean blood Pb levels of 20.4 (ig/dL and
 7                  24.4 (ig/dL, respectively. In the GD20 fetuses from dams exposed for 1 month prior to
 8                  pregnancy, a concentration-dependent increase in liver SOD activity was observed,
 9                  whereas SOD  activity was decreased in the GD20 fetuses from dams exposed for
10                  5 months prior to pregnancy. The increase in SOD activity in the livers of fetuses from
11                  dams exposed to 0.3 or 3.0 ppm Pb nitrate for one month suggests an initial activation of
12                  SOD in response to increased free radical production, while the decrease in SOD
13                  production in fetal livers from dams exposed to the same concentrations for 5 months
14                  suggests that longer durations of Pb exposure impairs the antioxidant defense mechanism.

15                  Increased oxidative stress also was found in animals with postnatal Pb exposure  in
16                  drinking water. Jurczuk et al. (2007) reported that male Wistar rats treated with 500 ppm
17                  Pb in drinking water (blood Pb not reported) exhibited decreases in liver vitamin E and
18                  GSH levels along with an increase in lipid peroxidation. The correlation between vitamin
19                  E and lipid peroxidation suggested that vitamin E is involved in the mechanism of
20                  peroxidative action of Pb in the liver. In a study examining the role of low molecular
21                  weight thiols on peroxidative mechanisms, Jurczuk et al. (2006) found that male  Wistar
22                  rats treated with 500 ppm Pb acetate in drinking water exhibited a decrease in blood
23                  ALAD as well as decreases in GSH and nonprotein sulfhydryl levels in the liver.
24                  Metallothionein levels were also reported to be higher in the liver following exposure to
25                  Pb. Yu et al. (2008) reported concentration-dependent increases in lipid peroxide levels
26                  and decreases  in GSH levels and CAT, SOD and GPx activities in livers from castrated
27                  male pigs that received a diet mixed with 0, 5, 10, or 20 mg/kg Pb nitrate exposure,
28                  during ages 55-100 days. The level of hepatic CuZnSOD mRNA was also reduced in
29                  Pb-treated animals. The study authors suggested that this decrease in SOD mRNA
30                  expression and activity of antioxidant enzymes may lead to a reduction in free radical
31                  scavenging capability and increased lipid peroxidation.

32                  Studies administering Pb by bolus doses had similar findings. Adegbesan and Adenuga
33                  (2007) reported that lipid peroxidation was increased and SOD activity was decreased in
34                  protein undernourished male Wistar rats compared to well-fed rats, and that these effects
35                  were further exacerbated in protein undernourished rats injected with Pb nitrate
36                  (100 (imol/kg, blood Pb not reported). Protein undernourishment also decreased  GSH
37                  levels and CAT activity compared to normal diet; however, co-treatment with Pb
38                  mitigated the severity of these effects. GSH levels and CAT activity were still lower in
      November 2012                            5-643                 Draft - Do Not Cite or Quote

-------
 1                   undernourished rats with Pb exposure compared to well-fed rats, but greater than
 2                   undernourished rats with no Pb exposure. The results indicated Pb treatment exacerbated
 3                   the effects of malnutrition on liver lipid peroxidation and altered the involvement of free
 4                   radicals. Male Charles-Foster rats treated with Pb acetate (0.025 mg/kg via i.p. injection,
 5                   blood Pb not reported) also exhibited statistically significant increases in lipid
 6                   peroxidation levels and decreases in SOD, CAT, and glucose-6-phosphatase
 7                   dehydrogenase levels in liver mitochondrial and postmitochondrial fractions (Pandya et
 8                   al.. 2010). Statistically nonsignificant decreases were observed in GSH levels, and in GPx
 9                   and GR activities in Pb-treated animals. In mice gavaged with Pb nitrate (50 mg/kg for
10                   40 days), lipid peroxidation was increased, and SOD, CAT, and GSH were decreased
11                   compared to controls (Sharma et al.. 2010a). Additionally, Pb nitrate treatment resulted in
12                   histopathological changes in the structure of the liver: hepatocytes were damaged and
13                   were marked by cytoplasmic vacuolization and pycnotic nuclei.

14                   Khotimchenko and Kolinchenko (2007) also reported an increase in lipid peroxidation
15                   and development of hepatitis in male albino rat liver parenchyma following intragastric
16                   treatment with Pb acetate (100 mg/kg for 14 days). Lipid peroxidation was demonstrated
17                   by increases in malondialdehyde (MDA) levels along with decreases in GSH and thiol
18                   groups; indicating injury in the liver antioxidant system. Levels of hepatic lipid
19                   peroxidation were observed to be significantly increased in rats treated with Pb acetate
20                   (35 mg/kg via i.p. injection daily for 3 days, blood Pb not reported), whereas hepatic
21                   GSH was significantly decreased (Upadhyay et al.. 2009). A study examining male and
22                   female rat pups that were continuously exposed to Pb during gestation and lactation
23                   (pregnant dams were injected [s.c.] with Pb  acetate, 50 (ig/kg per day [GDO to PND21],
24                   blood Pb not reported), did not find effects on GSH or MDA levels at PND56 (Filial et
25                   al., 2009). In vitro exposure of cells from a hepatic human embryonic epithelial cell line
26                   (WRL-68) to 5 (iM Pb acetate for 30  days resulted in increased production of ROS
27                   throughout the incubation period (Hernandez-Franco et al., 2011). Concurrent with this
28                   increase in ROS generation, the activities of SOD and the levels of membrane lipid
29                   peroxidative damage increased throughout the first 24 days of exposure but returned to
30                   normal levels by day 30.


                     Hepatic Apoptosis

31                   Fan et al. (2009b) reported that a single i.v. injection (tail vein) of Pb nitrate
32                   (200  (imol/kg in 0.5 mL) in rats resulted in an increase in the percentage of apoptotic
33                   hepatocytes (mean: 2.5 [SD: 1.4]% of total hepatocytes) compared with controls (mean:
34                   0.31  [SD: 0.31]%). Expression of ferritin light-chain (FTL) protein also increased (mean
35                   [SD]: 3.5 [1.0]-fold increase) over that in controls. Immunohistochemical analysis
36                   revealed that hepatocytes around the central vein were heavily stained by anti-FTL

      November 2012                             5-644                  Draft - Do Not Cite or Quote

-------
 1                  antibodies, as were nonparenchymal cells identified as Kupffer cells. The authors
 2                  hypothesized that the expression of FTL in Kupffer cells may have resulted from the
 3                  phagocytosis of apoptotic cells. Treatment of rats with clofibrate, a lipid lowering agent,
 4                  did not increase FTL expression in Kupffer cells and induced hepatocellular proliferation
 5                  butnotapoptosis.
                    5.9.1.4     Summary of Effects on the Hepatic System

 6                  While explicit associations between hepatic injury (i.e., histopathological effects) and Pb
 7                  exposure have not established, evidence from epidemiologic and toxicological studies
 8                  have indicated that exposure to Pb can result in altered liver function and hepatic
 9                  oxidative stress. A few studies have reported associations of higher blood Pb levels with
10                  decreased cytochrome P450 enzymes (Phase I xenobiotic metabolism) in children and
11                  nonoccupationally-exposed adults. However, most evidence indicates decreases in serum
12                  protein and albumin levels and increased AST, ALT, ALP, and GGT activities (indicators
13                  of decreased liver function), increased oxidative stress, and decreased antioxidant status
14                  in Pb-exposed workers with blood Pb levels >29 (ig/dL (Can et al., 2008; Khan et al.,
15                  2008; Patil et al., 2007; Conterato et al.. In Press). The implications of the epidemiologic
16                  evidence is limited because of its cross-sectional study design nature, the high blood Pb
17                  levels examined, and lack of consideration for potential confounding by factors such as
18                  age, diet, BMI, smoking, or other occupational exposures.

19                  Similar changes in liver function enzymes have been found in mature animals exposed to
20                  high levels of Pb during adulthood (Sharma et al., 2010a; Wang et al., 2010; Herman et
21                  al.. 2009; Cheng et al.. 2006). and animals exposed during gestation and lactation (Pillai
22                  et al.. 2009; Teijon et al., 2006). Pb exposure has been shown to impair lipid metabolism
23                  in animals, as evidenced by increased hepatic cholesterogenesis, and in altered
24                  triglyceride and phospholipid levels (Ademuyiwa et al., 2009; Khotimchenko and
25                  Kolenchenko. 2007). Multiple studies in humans and animals have observed
26                  Pb-associated hepatic oxidative stress, generally indicated by an increase in lipid
27                  peroxidation along with a decrease in GSH levels and CAT, SOD, and GPx activities
28                  (Pandvaetal. 2010; Sharma et al.. 2010a; Khan et al.. 2008; Yu et al.. 2008; Adegbesan
29                  and Adenuga. 2007; Jurczuk et al.. 2007; Khotimchenko and Kolenchenko. 2007;
30                  Jurczuk et al.. 2006). Indices of increased oxidative stress were also observed in the livers
31                  of fetuses exposed to Pb throughout gestation (Masso et al.. 2007). The relevance of the
32                  toxicological evidence is uncertain as many studies administered Pb as bolus doses.
33                  Because of the insufficient quality of studies, the evidence is inadequate to determine if
34                  there is a causal relationship between Pb exposure and hepatic effects.
      November 2012                            5-645                  Draft - Do Not Cite or Quote

-------
      5.9.2        Effects on the Gastrointestinal System

 1                  Gastrointestinal effects examined in relation to Pb exposure include abdominal pain,
 2                  constipation, and internal paralysis in humans and degeneration of the intestinal epithelial
 3                  mucosa and a decrease in duodenal motility in animals.
                    5.9.2.1      Summary of Key Findings on the Effects on the
                                 Gastrointestinal System (2006  Pb AQCD)

 4                  The 2006 Pb AQCD (U.S. EPA. 2006b) stated that a number of factors influence the
 5                  gastrointestinal absorption of Pb; including the chemical and physical form of Pb, the age
 6                  at Pb intake, as well as various nutritional factors. Rats exposed to Pb acetate in drinking
 7                  water had degeneration of the intestinal epithelial mucosa, potentially leading to
 8                  malabsorption of nutrients. In suckling rat pups, increased casein micelles incidences
 9                  were reported as a result of Pb present in bovine and rat milk and in infant milk formula.
10                  Pb ingestion through water was more toxic compared to Pb ingestion via milk. Pb
11                  ingested in milk was reported to be taken up by the ileal tissue, whereas Pb administered
12                  intragastrically as a soluble salt was primarily accumulated in the duodenum irrespective
13                  of vehicle used for administration. Decreases in duodenal motility and the amplitude of
14                  contractility in the intestinal tract were observed in rats following Pb exposure.
15                  Nutritional studies examining different dietary levels of Pb, Ca2+, and vitamin D in rats
16                  indicated competition in absorption between Pb and calcium. Dietary  supplementation
17                  with vitamin D led to an increase in intestinal absorption of Pb and calcium.  In instances
18                  where severe calcium deficiency was noted, ingestion of Pb caused a clear decrease in
19                  1,25-dihydroxy vitamin D (1,25-(OH)2D3) levels. Overall, the 2006 Pb AQCD stated
20                  that studies in rat intestine have shown that the largest amount of Pb absorption occurs in
21                  the  duodenum with the mechanisms of absorption involving active transport and
22                  diffusion via the  intestinal epithelial cells. Absorption has been reported to occur, through
23                  both saturable  and nonsaturable pathways, based on results from various animal studies.
24                  The 2006 Pb AQCD reported evidence that symptoms associated with gastrointestinal
25                  colic (abdominal pain, constipation, intestinal paralysis) were more prevalent in
26                  occupationally-exposed adults with blood Pb levels > 50 (ig/dL.
                    5.9.2.2     Recent Epidemiologic Studies

27                  Consistent with previous findings, Kuruvilla et al. (2006) reported gastrointestinal effects
28                  including stomach pain and gastritis along with other non-gastrointestinal effects in 53
29                  Pb-exposed painters (mean [SD] blood Pb: 8.04 [5.04] (ig/dL) in India compared with 50
      November 2012                            5-646                 Draft - Do Not Cite or Quote

-------
 1                   controls (mean [SD] blood Pb level: 5.76 [4.45] (ig/dL) matched by sex, age, education,
 2                   income, smoking, and alcohol consumption. Prevalence of symptoms in painters did not
 3                   differ from that in battery workers with higher blood Pb levels (mean [SD] blood Pb level
 4                   of 42.40 ([25.53] (ig/dL). Despite the consistent evidence among occupational studies,
 5                   the implications of gastrointestinal findings in Pb-exposed workers are limited by the
 6                   cross-sectional study designs, high blood Pb levels associated with effects (mostly
 7                   > 50 (ig/dL), and limited consideration for potential confounding by factors such as age,
 8                   smoking, alcohol use, nutrition, or other occupational exposures.
                     5.9.2.3      Recent lexicological Studies

 9                   A few recent studies pertaining to gastrointestinal effects of Pb exposure were identified
10                   that provide evidence for additional mechanisms underlying gastrointestinal damage and
11                   impaired function. Santos et al. (2006) examined the impact of Pb exposure on
12                   nonadrenergic noncholinergic (NANC) relaxations in rat gastric fundus. Male Wistar rats
13                   treated with 80 ppm Pb acetate via drinking water for 15, 30, and 120 days (blood Pb not
14                   reported) exhibited a significant difference in NANC relaxations in the gastric fundus
15                   following electrical field stimulus. While frequency-dependent relaxations were observed
16                   in all groups, including the control group, the relaxations were significantly inhibited in
17                   rats treated with Pb acetate for all three durations. When gastric fundus strips from rats
18                   were incubated with L-nitroarginine, a nitric oxide  (NO) synthase inhibiter, no additional
19                   inhibition in relaxations was observed. In contrast,  incubation with sodium nitroprusside
20                   and 8-Br-GMPc (a cyclic guanosine monophosphate [cGMP] analog), resulted in a
21                   concentration-dependent relaxation in strips in the control group and in the group
22                   exposed to Pb acetate for 120 days. The results suggested that long-term exposure to Pb
23                   causes inhibition in NANC relaxation probably due to the modulated release of NO from
24                   the NANC nerves or due to interaction with the intracellular transducer mechanism in the
25                   rat gastric fundus.

26                   In another study examining Pb-induced oxidative stress in the gastric mucosa, Olaleye et
27                   al.  (2007) treated Albino Wistar rats with 100 or 5,000 ppm Pb acetate in drinking water
28                   for 15 weeks (blood Pb not reported). Exposure to Pb acetate caused a significant
29                   increase in gastric mucosal damage caused by pretreatment with acidified ethanol. While
30                   the basal gastric acid secretory rate was not altered, stomach response to histamine was
31                   significantly higher in animals treated with Pb acetate compared to that in the controls.
32                   Additionally, there was a significant increase in gastric lipid peroxidation at both the  100
33                   and 5,000 ppm dose levels. In contrast, CAT, and SOD activities and nitrite levels were
34                   significantly decreased in the gastric mucosa. The results indicated that Pb-induced
35                   gastric damage may be mediated via increases in oxidative stress.


      November 2012                             5-647                 Draft - Do Not Cite or Quote

-------
                    5.9.2.4     Summary of Gastrointestinal Effects

 1                  Relatively few human studies have been conducted on the gastrointestinal toxicity of Pb.
 2                  The evidence points to more prevalent symptoms, such as stomach pain, gastritis,
 3                  constipation, and intestinal paralysis, in Pb-exposed workers (Kuruvilla et al.. 2006).
 4                  However, the implications of gastrointestinal findings in Pb-exposed workers are limited
 5                  by the cross-sectional study designs, high blood Pb levels associated with effects (mostly
 6                  > 40 (ig/dL), and limited consideration of potential confounding by factors such as age,
 7                  smoking, alcohol use, nutrition, or other occupational exposures. Toxicological evidence
 8                  indicates that Pb is absorbed primarily in the duodenum by active transport and diffusion,
 9                  although variability is observed by Pb compound, age of intake, and nutritional factors.
10                  There is some coherence between the evidence in Pb-exposed workers and observations
11                  in animals that Pb induces damage to the intestinal mucosal epithelium, decreases
12                  duodenum contractility and motility, reduces absorption of Ca2+, inhibits NANC
13                  relaxations in the gastric fundus, and induces oxidative stress (lipid peroxidation,
14                  decreased SOD and CAT) in the gastric mucosa (Olaleve et al., 2007; Santos et al.,
15                  2006). The observation of oxidative stress was accompanied by gastric mucosal damage.
16                  Because of the insufficient quantity and quality of studies, the evidence is inadequate to
17                  determine if there is a causal relationship between Pb exposure and gastrointestinal
18                  effects.
      5.9.3       Effects on the Endocrine System

19                  A summary of key findings pertaining to reproductive hormones in males and females is
20                  presented in the section on Reproductive and Developmental Effects (Sections 5.8.1 and
21                  5.8.2). Collective epidemiologic and toxicological evidence is inconsistent in
22                  demonstrating the effects of Pb exposure on male and female sex hormone levels. Other
23                  endocrine processes that are most commonly found to be impacted by Pb exposure
24                  include changes in thyroid hormones, including thyroid stimulating hormone (TSH),
25                  triiodothyronine (T3), and thyroxine (T4). A few studies have examined calcium and
26                  cortisol.
                    5.9.3.1     Summary of Key Findings of the Effects on the Endocrine
                                System (2006 Pb AQCD)

27                  The 2006 Pb AQCD (U.S. EPA. 2006b) reported that endocrine processes impacted by
28                  occupational Pb exposure include thyroid hormone levels, changes in male sex hormone
29                  levels, as well as changes in the production of vitamin D (1,25-(OH)2D3). However,
      November 2012                            5-648                 Draft - Do Not Cite or Quote

-------
 1                  these effects were observed only with blood Pb levels exceeding 30-40 (ig/dL and in
 2                  studies with little or no consideration for potential confounding by factors such as sex,
 3                  SES, nutritional status, BMI, smoking, comorbid conditions, and other occupational
 4                  exposures. In addition, alterations in calcitropic hormones were found in children with
 5                  blood Pb levels ranging from 10-120 (ig/dL and in an opposite direction than that in
 6                  Pb-exposed workers.
                    5.9.3.2     Recent Epidemiologic Studies

 7                  Recent epidemiologic studies have reported associations between indicators of exposure
 8                  to Pb and thyroid hormone levels, in populations of children and in adults with and
 9                  without occupational Pb exposure; although results have not been consistent for a
10                  particular hormone. Further, the implications of these findings are limited because of the
11                  use of cross-sectional study design, lack of rigorous statistical analysis, and limited
12                  consideration for potential confounding factors. Inconsistent associations were found in a
13                  study that considered potential confounding factors. Abdelouahab et al. (2008) examined
14                  a Canadian population characterized by high consumption of freshwater fish
15                  contaminated with Pb and other environmental pollutants. The median concurrent blood
16                  Pb level was 3.1 (ig/dL for men and 1.7 (ig/dL for women. The median blood Pb level for
17                  women was lower than the limit of detection (2.1 (ig/dL), resulting in measurement error
18                  of blood Pb level and greater uncertainty in the results. In an analysis stratified by sex,
19                  TSH levels were negatively correlated with blood Pb in women with adjustment for age,
20                  smoking status, estro-progestative intake, total plasma lipids, and Se. No associations
21                  with T3 and T4 levels were found in women. TSH, T3 and T4 levels were not correlated
22                  with blood Pb level in males, after adjustment for the  same covariates (excluding
23                  hormone intake) plus pesticide exposure, corticoid medication, concurrent alcohol
24                  consumption, and occupational exposure to metals. Overall, the inconsistent associations
25                  and potential influence of other exposures did not strongly demonstrate an effect of Pb
26                  exposure.

27                  Studies with less rigorous methods also did not clearly indicate an association between
28                  blood Pb level and a particular thyroid hormone.  In a Kosovo, Yugoslavia population,
29                  higher pregnancy blood Pb levels were associated with lower pregnancy free T4 level
30                  among the  156 women living in a highly exposed town with a smelter and battery plant,
31                  but not among the  153 women living in a relatively unexposed nearby town (Lamb et al..
32                  2008). The mid-pregnancy blood Pb levels were highly  elevated in the industrial town
33                  compared to the unexposed town (mean [SD]: 20.56 [7.38] versus 5.60 [1.99]  (ig/dL). In
34                  24 newborns delivered in Tokyo, Japan, neither TSH nor free T4 (sampled 4-6 days
35                  postpartum) was correlated  with cord blood Pb level (mean:  0.67 (ig/dL) (lijima et al..


      November 2012                            5-649                 Draft - Do Not Cite or Quote

-------
 1                   2007). Neither of these studies considered potential confounding. Croes et al. (2009)
 2                   examined the hormone levels in 1,679 adolescents residing in nine study areas in
 3                   Belgium with varying exposures to multiple industrial pollutants including Pb. The
 4                   median concurrent blood Pb level of the participants from the nine different regions
 5                   ranged from 1.6 to 2.8 (ig/dL. Analyses only indicated differences in free T3, at the
 6                   region or neighborhood level with adjustment for age, sex, recent disease, and BMI. No
 7                   direct associations with blood Pb level were analyzed, thus the results could be attributed
 8                   to other factors that varied by location.

 9                   Contrasting results were found for free T4 in Pb-exposed workers. Dundar et al. (2006)
10                   examined associations between blood Pb levels and thyroid function in 42 male
11                   adolescent auto repair apprentice workers, with no history of prior disease, exposed long
12                   term to Pb (in the auto repair apprenticeship at least 1 year). Mean blood Pb level was
13                   higher in the auto repair workers compared to the 55 healthy unexposed control subjects
14                   (mean [SD]: 7.3 [2.92] versus 2.08 [1.24] (ig/dL). Free T4 levels were  significantly lower
15                   in the auto workers compared to the control group, which had no abnormal free T4 levels
16                   reported. In contrast, free T3 and TSH levels were comparable between auto workers and
17                   controls. Blood Pb level was negatively correlated with free T4 levels.  In contrast,
18                   another study (Pekcici et al.. 2010) found higher free T4 and TSH in adult auto mechanic
19                   or battery factory workers who were highly exposed to Pb (mean blood Pb: 71.1 (ig/dL)
20                   compared to controls (mean blood Pb level: 0.2 (ig/dL). Free T3 levels were similar
21                   between the two groups. The results from this study are likely not generalizable to the
22                   general public due to the high blood Pb levels of the exposed workers.

23                   Previous findings for blood Pb-associated changes in serum vitamin D (1,25-(OH)2D3)
24                   in children were mixed. A recent study in New Jersey examined winter (December to
25                   March) and summer (July to September) seasonal changes in the associations between
26                   blood Pb level and serum 1,25-(OH)2D3 status, in 142 young, U.S. urban African-
27                   American or Hispanic children (ages 1-8 years, grouped by age [1-3 year-olds and 4-8
28                   year-olds] and race/ethnicity) using a repeated measures design (Kemp et al.. 2007). The
29                   percentage of 1 -3 year-old African-American children (n = 49) with blood Pb levels
30                   > 10 (ig/dL increased from  12.2% in winter to 22.5% in summer. This  large seasonal
31                   increase in blood Pb levels in these 1-3 year-old children was not accompanied by a
32                   significant increase in serum 1,25-(OH)2D3 concentrations. There was also a larger
33                   seasonal increase in blood Pb levels in 1-3 year-old children from both races combined
34                   (n = 78) (mean [SE]: 4.94 [0.45]  (ig/dL winter, 6.54 [0.82] (ig/dL summer) than in 4-8
35                   year-old children from both races combined (n = 64) (mean [SE]: 3.68  [0.31] (ig/dL
36                   winter, 4.16 [0.36] (ig/dL summer). However, no difference in seasonal 1,25-(OH)2D3
37                   was observed in the 1-3 year-old children from both races combined. A larger winter to
38                   summer increase in blood Pb level was correlated with a larger seasonal increase in
      November 2012                             5-650                  Draft - Do Not Cite or Quote

-------
 1                   serum 1,25-(OH)2D3 in the 4-8 year-old children from both races combined and in the
 2                   4-8 year-old African American children (n = 42). In the 4-8 year-old children from both
 3                   races combined, there was a winter to summer increase in 1,25-(OH)2D3 (mean [SE]:
 4                   25.3 [1.2] (ig/L in winter versus 33.8 [1.1] (ig/L in summer.), which may account forthe
 5                   results in this older age group. Based on these results, the study authors concluded that
 6                   higher summertime increase in serum 1,25-(OH)2D3 levels in children between 4 and 8
 7                   years is most likely due to increased sunlight-induced vitamin D synthesis and may be a
 8                   contributing factor to seasonal changes in blood Pb levels via changes in gastrointestinal
 9                   absorption or release of Pb from bone.

10                   HPA function was examined in a prospective analysis of associations between prenatal
11                   maternal blood Pb levels (cord blood collected at delivery) or postnatal blood Pb levels
12                   (at mean [SD] age: 2.62 [1.2] years, data obtained from family physicians or state
13                   records), and saliva cortisol levels during a stress protocol in the children at age 9.5 years
14                   (Gump et al.. 2008). For prenatal blood Pb, the children were divided into the following
15                   quartiles: < 1, 1.1-1.4, 1.5-1.9, and 2.0-6.3 (ig/dL. For postnatal blood Pb, the quartiles
16                   were: 1.5-2.8, 2.9-4.1, 4.2-5.4, and 5.5-13.1 (ig/dL. With adjustment for potential
17                   confounding (by SES-related factors, HOME score, pregnancy health, maternal substance
18                   abuse), blood Pb level was not associated with initial salivary cortisol levels. However,
19                   following an acute stressor, which comprised submerging the dominant arm for a minute
20                   in a gallon of one part ice to one part water, increasing prenatal and postnatal blood Pb
21                   levels were associated with statistically significant increases in salivary cortisol
22                   responses. Children in the 2nd, 3rd, and 4th prenatal blood Pb quartiles and in the 4th
23                   postnatal quartile had increased salivary cortisol responses compared to children in the
24                   1st quartile. When blood Pb was treated as a continuous variable, regression analysis
25                   showed that both prenatal and postnatal blood Pb levels were associated salivary cortisol
26                   reactivity. While associations were found in children with blood Pb levels below
27                   10 (ig/dL, they could have been attributed to higher earlier childhood blood Pb  levels of
28                   these children who were born in 1980-1990s.
                     5.9.3.3      Recent lexicological Studies

29                   Pb-associated changes in thyroid hormones also were found in animal studies. In a study
30                   examining the effects of Pb and Cd in adult cows reared in a polluted environment in
31                   India, Swarup et al. (2007) found significantly higher mean plasma T3 and T4 levels in
32                   cows living near Pb/Zn smelters (mean [SD] blood Pb: 86 [6] (ig/dL) and near closed
33                   Pb/operational Zn smelters (mean [SD] blood Pb: 51 [9] (ig/dL) when compared to cows
34                   in unpolluted areas (mean [SD] blood Pb: 7 [1] (ig/dL). Regression analyses of the 269
35                   cows showed a significant positive correlation between blood Pb levels and plasma T3


      November 2012                             5-651                  Draft - Do Not Cite or Quote

-------
 1                   and T4 levels, whereas the correlation between blood Pb levels and plasma cortisol was
 2                   not statistically significant. Mean plasma estradiol level was significantly higher in cows
 3                   near closed Pb/operational Zn smelters compared to the control group of cows. Because
 4                   of the Pb-Zn co-exposure, the effects cannot be attributed specifically to Pb.

 5                   Biswas and Ghosh (2006) investigated the effect of Pb treatment on adrenal and male
 6                   gonadal functions in Wistar rats treated with Pb acetate (8.0 mg/kg via i.p. injection for
 7                   21 days, blood Pb not reported). Pb treatment significantly increased adrenal
 8                   steroidogenic enzyme activity and serum corticosterone levels. Accessory sex organ
 9                   (prostate and seminal vesicle) weights were decreased in Pb-treated animals, whereas
10                   adrenal weights were increased. These effects were accompanied by a decrease in
11                   spermatogenesis and serum concentrations of testosterone, FSH, and LH and by an
12                   increase in the percent of spermatid degeneration. Supplementation with testosterone
13                   during the last 14 days of Pb treatment was observed to ameliorate these effects.
                     5.9.3.4      Summary of Endocrine Effects

14                   Collective epidemiologic and toxicological evidence is inconsistent in demonstrating the
15                   effects of Pb exposure on male and female sex hormone levels (Sections 5.8.1 and 5.8.2)
16                   and vitamin D levels. Several epidemiologic studies have reported associations between
17                   indicators of Pb exposure and thyroid hormone levels in populations of children and
18                   adults without (Lamb et al.. 2008) and with occupational Pb exposure (Dundar et al.
19                   2006). although results have not been consistent for a particular hormone. Further, the
20                   implications of these findings are limited because of the cross-sectional study design,
21                   high blood Pb levels associated with effects (>30 (ig/dL), lack of rigorous statistical
22                   analysis, and limited consideration for potential confounding factors. Blood Pb level was
23                   positively correlated with plasma T3 and T4 levels in adult cows living near Pb-Zn
24                   smelters; however, the effects could not be attributed specifically to Pb exposure (Swarup
25                   et al.. 2007).

26                   In a prospective study of children in New York, who were challenged with an acute
27                   stressor, higher cord blood levels (as a reflection of prenatal maternal Pb blood level), or
28                   2-year-old blood Pb levels, were associated with significant higher salivary cortisol in
29                   response to a stress challenge at age 9 years (Gump et al.. 2008). While these associations
30                   were found with blood Pb levels <10 (ig/dL, they could have been attributed to higher
31                   earlier childhood blood Pb levels of these children who were born in the 1980s and
32                   1990s. Biswas and Ghosh (2006) found a Pb-induced increase in corticosterone in rats,
33                   albeit by i.p. Pb treatment. Cortisol and corticosterone are the major glucocorticoids in
34                   humans and rodent, respectively.
      November 2012                             5-652                  Draft - Do Not Cite or Quote

-------
 1                  In conclusion, epidemiologic and toxicological evidence indicates Pb-associated
 2                  endocrine effects such as thyroid hormones, cortisol, and vitamin D, although results are
 3                  not consistent. Because of the lack of insufficient quantity and quality of studies, the
 4                  evidence is inadequate to determine if there is a causal relationship between Pb exposure
 5                  and endocrine effects related to thyroid hormones, cortisol, and vitamin D.
      5.9.4        Effects on Bone and Teeth

 6                  Primary effects on bone associated with Pb exposure or biomarker levels have included
 7                  an increase in osteoporosis, increased frequencies of falls and fractures, changes in bone
 8                  cell function as a result of replacement of bone calcium with Pb, and depression in early
 9                  bone growth. Other effects include tooth loss and periodontitis. Mechanistic evidence
10                  from toxicological studies includes effects on cell proliferation, procollagen type I
11                  production, intracellular protein, and osteocalcin in human dental pulp cell cultures.
                    5.9.4.1      Summary of Key Findings of the Effects on Bone and
                                 Teeth (2006 Pb AQCD)

12                  The 2006 Pb AQCD reported many effects on bone and some in teeth in animals
13                  following Pb exposure. Exposure of animals to Pb during gestation and the immediate
14                  postnatal period was reported to significantly depress early bone growth with the effects
15                  showing concentration-dependent trends. In mature animals, long-term Pb exposure (up
16                  to one year), along with poor nutrition (low calcium) reduced bone growth as well as
17                  bone density. Systemic effects of Pb exposure included disruption in bone mineralization
18                  during growth, alteration in bone cell differentiation and function due to alterations in
19                  plasma levels of growth hormones and calcitropic hormones such as 1,25-[OH]2D3 and
20                  impact on Ca2+- binding proteins and increases in Ca2+ and phosphorus concentrations in
21                  the bloodstream. Bone cell cultures exposed to Pb had altered vitamin D-stimulated
22                  production of osteocalcin accompanied by inhibited secretion of bone-related proteins
23                  such as osteonectin and collagen. In addition, Pb exposure caused  suppression in bone
24                  cell proliferation most likely due to interference from factors such as growth hormone
25                  (GH), epidermal growth factor (EGF), transforming growth factor-beta 1 (TGF-|31), and
26                  parathyroid hormone-related protein (PTHrP).

27                  As  in bone, Pb exposure was found to easily substitute for Ca2+ in the teeth and was taken
28                  up and incorporated into developing teeth in experimental animals. Since teeth do not
29                  undergo remodeling like bone does during growth, most of the Pb  in the teeth remains in
30                  a state of permanent storage. High dose Pb exposure to animals (30 mg/kg body weight)
      November 2012                            5-653                  Draft - Do Not Cite or Quote

-------
 1                  was found to induce the formation of a "Pb line" that is visible in both the enamel and
 2                  dentin and is localized in areas of recently formed tooth structure. Areas of mineralization
 3                  were easily evident in the enamel and the dentin within these "Pb lines." Pb has also been
 4                  shown to decrease cell proliferation, procollagen type I production, intracellular protein,
 5                  and osteocalcin in human dental pulp cell cultures. Adult rats exposed to Pb have
 6                  exhibited an inhibition of the posteruptive enamel proteinases, delayed teeth eruption
 7                  times, as well as a decrease in microhardness of surface enamel. Pb was reported to be
 8                  widely dispersed and incorporated into developing apatite crystal during enamel
 9                  formation process; however, post formation, Pb was reported to be capable of entering
10                  and concentrating in specific enamel areas which were Ca2+-deficient. The
11                  2006 Pb AQCD (U.S. EPA. 2006b) also reported that a number of animal studies and a
12                  few epidemiologic studies each suggested that Pb is a caries-promoting element. The
13                  strongest epidemiologic evidence comprised associations between concurrent blood Pb
14                  level and dental caries in an NHANES analysis of thousands of children that adjusted for
15                  age, sex, race/ethnicity, poverty to income ratio, exposure to cigarette smoke, geographic
16                  region, head of household education, carbohydrate and calcium intake, and frequency of
17                  dental visits. Other effects found in humans included bone disease (e.g., Paget's disease);
18                  however, the evidence was provided by occupational or case-control studies.
                    5.9.4.2      Recent lexicological and Epidemiologic Studies

19                  Consistent with evidence reported in the 2006 Pb AQCD, recent studies have found
20                  associations between Pb exposure or biomarker levels and effects in bones of humans and
21                  animals. The association between blood Pb levels and lower bone mineral density was
22                  examined in several epidemiologic studies. Prospective evidence was provided by Khalil
23                  et al. (2008) in 533 older women aged 65-87 years with a mean (SD) blood Pb of 5.3
24                  (2.3) (ig/dL. Bone mineral density was measured in 1986-1988 (calcaneus), again in
25                  1988-1990 (total hip and femoral neck), and again in 1993-1994 (calcaneus, total hip and
26                  femoral neck), while blood Pb levels were measured in between the 2nd and last bone
27                  analyses (during 1990-1991; and categorized as low [n = 122], medium [n = 332], and
28                  high [n = 79] Pb blood levels [range: 1-21 ng/dL]). Information on falls and fractures was
29                  collected every 4 months, starting after the initial enrollment (1986-1988) and continuing
30                  for more than 10 years. The bone mineral density at the last measurement (1993-1994)
31                  was 7% lower in the total hip (p <0.02) and 5% lower in the femoral neck (p <0.03) in the
32                  high blood Pb group (> 8 (ig/dL) compared to the low blood Pb group (< 3 (ig/dL). A
33                  concentration-dependent relationship was found for total hip and femoral neck bone
34                  mineral density across the three blood Pb level groups. In addition, total hip, femoral
35                  neck, and calcaneus bone loss was observed to be greater in the medium (blood Pb:
      November 2012                            5-654                 Draft - Do Not Cite or Quote

-------
 1                  4-7 (ig/dL) and high Pb groups compared to the low Pb group, with a statistically
 2                  significant trend found for calcaneus bone loss. Compared to the low blood Pb level
 3                  group, women in the high blood Pb level group had an increased risk of non-spine
 4                  fracture (10.5 year interview follow-up), and women with medium or high blood Pb
 5                  levels had a higher risk of falls (4 year follow-up) with adjustment for age, clinic, BMI,
 6                  weight change between visits, smoking, chair stands (were able to stand up five times
 7                  from a chair, without using the arms of the chair), fracture history, estrogen use, and
 8                  baseline (1986-1988) bone mineral density. Nutritional factors were not considered. The
 9                  increased risk of lower bone density and falls leading to osteoporosis-related fractures
10                  associated with blood Pb levels >4 (ig/dL are likely influenced by higher past Pb
11                  exposures of these women.

12                  Supporting evidence was provided by cross-sectional epidemiologic studies, although the
13                  direction of the association and the magnitude, timing, frequency, and duration of Pb
14                  exposure that contributed to the observed associations are uncertain. Further, most studies
15                  did not consider potential confounding by nutritional factors. A large NHANES II
16                  analysis of 8,654 adults > 50 years of age (Campbell and Auinger. 2007). which was
17                  stratified by non-Hispanic white men (mean blood Pb: 4.9 [range: 0.7 to 48.1] (ig/dL),
18                  non-Hispanic white women (mean blood Pb: 3.6 [range: 0.7 to 28.7] (ig/dL), African-
19                  American men (mean blood Pb: 7.7 [range: 0.7 to 52.9] (ig/dL), and African-American
20                  women (mean blood Pb: 4.5  [range: 0.7 to 23.3] (ig/dL). In analyses of covariance that
21                  considered potential confounding (by age, race, sex, BMI, menopausal status, tobacco
22                  use, alcohol use, physical activity, Ca2+ intake, chronic medical conditions, certain
23                  medication use, and SES), non-Hispanic white men (n = 1,693, p <0.05) and women
24                  (n = 1,754, p <0.10) in the highest tertile of concurrent blood Pb level had lower mean
25                  total hip bone mineral density than non-Hispanic white  men and women in the lowest
26                  tertile of blood Pb levels (actual concentration not reported). Smaller differences were
27                  observed in African-American men and women (possibly due to the smaller sample sizes
28                  (n = 613, and 629, respectively). No association was observed between blood Pb levels
29                  and osteoporotic fractures in any sex or race/ethnicity group.

30                  Similar observations were made by Sun et al. (2008a) in 155 males and 37 females in
31                  China who were occupationally-exposed to Pb (mean blood Pb: 20.22 and 15.5 (ig/dL,
32                  respectively). In analyses (including all workers, plus 36 male and 21 female unexposed
33                  controls stratified into groups according to blood Pb and urinary Pb levels), groups with
34                  urinary Pb levels > 5 (ig/g creatinine had lower (p <0.01) bone mineral density compared
35                  to groups with lower urinary Pb in each sex. Prevalence of osteoporosis increased with
36                  increasing blood Pb in a linear manner. In contrast, a significant difference was observed
37                  between blood Pb level and bone mineral density, but only in men with blood Pb levels
38                  >30 (ig/dL. Prevalence of osteoporosis increased significantly with increasing blood Pb
      November 2012                            5-655                  Draft - Do Not Cite or Quote

-------
 1                  in a linear manner. Results were not adjusted for potential confounding factors, including
 2                  other occupational exposures.

 3                  Cross-sectional epidemiologic studies also found associations between concurrent blood
 4                  Pb level and biological markers of bone turnover. Among 329 male (mean age: 65 years,
 5                  median blood Pb level: 2.2 (ig/dL) and 342 female (mean age: 62 years, median blood Pb
 6                  level: 1.9 (ig/dL) adults in North Carolina, Nelson et al. (2011) found in women that
 7                  higher blood Pb level was associated with higher uN-telopeptide cross-linked collagen
 8                  type I (uNTX-I, a marker of bone resorption/turnover) and uCTX-II (a marker associated
 9                  with the progression of radiographic knee and hip osteoarthritis) after adjusting for age,
10                  BMI, race, and smoking status. In adjusted analyses of men, higher blood Pb level was
11                  associated with higher uCTX-II, COMP, and C2C:CPII ratio (an indication of the balance
12                  between cartilage collagen degradation and synthesis).  In women, a weaker association
13                  was found for COMP, a cartilage biomarker related to osteroarthritis. The results
14                  indicated that blood Pb level is associated with bone  turnover and mineralized cartilage
15                  turnover in women, and with non-mineralized cartilage turnover in men.

16                  Similarly, Machida et al. (2009) investigated bone matrix turnover in Japanese women
17                  farmers, and how it is related to age-related menopause status and blood Pb level.
18                  Perimenopausal women (n = 319 [age range: 49 to 55 years]) had higher geometric mean
19                  blood Pb level (2.0 (ig/dL) than the other 3 groups did: premenopausal women (n = 261
20                  [age range: 35  to 48 years], blood Pb level: 1.6 (ig/dL), younger postmenopausal women
21                  (n = 397 [age range: 56 to 65 years], blood Pb level:  1.8 (ig/dL), or older postmenopausal
22                  women (n = 248 [age range: 66 to 75 years], blood Pb level: 1.7 (ig/dL). In a model that
23                  simultaneously included bone-mineral density, NTx, osteocalcin, and age, higher blood
24                  Pb levels were positively associated with bone mineral density, NTx, and osteocalcin
25                  (all p <0.01). In perimenopausal women, higher blood Pb level was predicted most
26                  strongly by higher osteocalcin levels. Age was positively associated with higher blood Pb
27                  levels in perimenopausal women only. Associations also were reported for bone-specific
28                  ALP in unadjusted analyses.

29                  To characterize mechanisms underlying the effects of Pb on bone, Jang et al. (2008)
30                  studied the effect of Pb exposure on Ca2+-release activated Ca2+-influx (CRACI) using
31                  cultures of human fetal osteoblast-like hFOB 1.19 cells (OLCs) in vitro. When cells were
32                  incubated with  1,000 or 3,000 (iM Pb in the culture medium, a concentration-dependent
33                  decrease on CRACI was observed, as was a concentration-dependent increase in the
34                  influx of Pb into human OLC. These results suggest that Pb inhibits the  measurable
35                  influx of Ca2+ upon re-addition of Ca2+, which in turn, results in an influx of Pb into the
36                  OLCs.
      November 2012                            5-656                 Draft - Do Not Cite or Quote

-------
 1                  Studies have found inconsistent associations between higher blood Pb level and reduced
 2                  growth in children; however, Zuscik et al. (2007) hypothesized that Pb may alter growth
 3                  by altering chondrogenic commitment of mesenchymal cells and by affecting various
 4                  signaling pathways. Exposure of stage El 1.5 murine limb bud mesenchymal cells
 5                  (MSCs) to 1 (iM Pb in vitro caused increased basal and TGF-(3/BMP induction of
 6                  chondrogenesis, which was accompanied by nodule formation and upregulation of Sox-9,
 7                  type 2 collagen, and aggrecan, which are all key markers of chondrogenesis. Enhanced
 8                  chondrogenesis during induced ectopic bone formation also was found in mice that had
 9                  been pre-exposed to Pb acetate for six weeks via drinking water (55 or 233 ppm,
10                  [previously shown to correspond to 14 or 40 (ig/dL blood Pb level, respectively]). MSCs
11                  exposed to Pb in vitro exhibited an increase in TGF-(3, but BMP-2 signaling was
12                  inhibited.  Pb also induced NF-KB and inhibited AP-1 signaling. These results suggested
13                  that the chondrogenesis induced by Pb exposure most likely involved modulation and
14                  integration of multiple signaling pathways including TGF-(3, BMP, AP-1, and NF-KB.

15                  Effects of Pb exposure on teeth were examined in a few recent cross-sectional
16                  epidemiologic studies. A subset of the U.S. NHANES III (1988-1994) population was
17                  selected for a large periodontitis versus Pb blood level study of both men (n = 2,500) and
18                  women (n = 2,399), 30-55 years-old, that considered potential confounding by a large set
19                  of factors, including nutritional status (Saraiva et al., 2007). Compared to individuals
20                  with a concurrent blood Pb level of <3 (ig/dL, the prevalence ratios of periodontitis were
21                  1.70 (95% CI: 1.02, 2.85) formen with concurrent blood Pb level of >7 (ig/dL and 3.80
22                  (95% CI:  1.66,  8.73) for women with concurrent blood Pb level >7 (ig/dL. These results
23                  were adjusted for age, NHANES III phase, cotinine levels, poverty to income ratio,
24                  race/ethnicity, education, bone mineral density, diabetes, calcium intake, dental visits,
25                  and menopause status in women.

26                  Arora et al. (2009) examined the association between blood and bone Pb level and the
27                  loss of natural teeth, in 333 men (age range: 50 to 94 years) from a subset of the Veterans
28                  Affairs Normative Aging Study (NAS). Tooth loss  was ascertained as the number of teeth
29                  present during a dental assessment, and was categorized into three groups: 0 missing
30                  teeth (n = 44), 1-8 missing teeth (n = 164), or > 9 missing teeth (n = 125). Men with > 9
31                  teeth missing had significantly higher tibia and patella Pb concentrations (measured
32                  within 3 years of dental assessment) compared to those with no tooth loss.  Men with the
33                  highest tibia Pb concentrations (>23 (ig/g) had higher odds of tooth loss (OR: 3.03 [95%
34                  CI: 1.60, 5.75]) compared to men with tibia Pb levels < 15 (ig/g. Men with the highest
35                  patella Pb levels (>36 (ig/g) also had higher odds of tooth loss (> 9 missing teeth versus
36                  0-8 missing teeth; or > 1 missing teeth versus 0 missing teeth:  OR: 2.41 [95% CI: 1.30,
37                  4.49]) compared to men with patella Pb levels < 22.0 (ig/g. Men with tibia Pb levels
38                  16-23  (ig/g, and men with patella Pb levels 23-36 (ig/g also had elevated odds of tooth
      November 2012                            5-657                  Draft - Do Not Cite or Quote

-------
 1                   loss. Results were adjusted for age, education, smoking status, pack-years of smoking,
 2                   and diabetes, but nutritional factors were not considered. Tooth loss was not associated
 3                   with higher blood Pb levels [also measured within 3 years of dental assessment (Hu et al.,
 4                   1996b)1. indicating that long-term cumulative exposure to Pb is associated with increased
 5                   odds of tooth loss. However, because the timing of tooth loss was not ascertained and
 6                   bone Pb levels may represent exposures after tooth loss occurred, the directionality of
 7                   effects is uncertain.
                     5.9.4.3      Summary of Effects on Bone and Teeth

 8                   A few studies have indicated associations of Pb exposure or Pb biomarker levels with
 9                   bone disease (e.g., Paget's disease); however, the implications are limited by examination
10                   of Pb-exposed workers or individuals with bone disease (i.e., case-control). Numerous
11                   epidemiologic studies indicated an association between higher Pb biomarker levels and
12                   lower bone density in adults. Prospective evidence was provided by a study of elderly
13                   women (65-87 years-old), in which higher blood Pb levels were associated with lower
14                   bone density measured after 2-4 years and greater risk of falls and osteoporosis-related
15                   fractures (Khalil et al., 2008). Cross-sectional epidemiologic associations between higher
16                   blood Pb levels and lower bone mineral density were found in adults without (Campbell
17                   and Auinger. 2007) and with occupational Pb exposure (Sun et al., 2008a). Cross-
18                   sectional studies also indicated associations between higher blood Pb levels and higher
19                   markers of bone turnover in elderly populations (Nelson et al., 2011; Machidaet al.,
20                   2009). In the cross-sectional epidemiologic evidence, it is difficult to determine whether
21                   an increase in blood  Pb level results from lower bone density or from higher bone
22                   turnover, and whether these effects lead to a greater release of Pb from bone into the
23                   bloodstream. Except for Sun et al. (2008a), studies adjusted for several potentially
24                   important confounding factors, including age, BMI, and smoking. However, studies did
25                   not consider nutritional status, which could affect the release of Pb from bone to blood.
26                   To support the direction and independent effects of Pb  on bone, toxicological studies
27                   have found Pb-induced (gestational and postnatal) decreases in bone growth in juvenile
28                   animals. Further, these toxicological studies have characterized potential modes of action,
29                   by showing Pb-induced decreases in bone mineralization and bone cell differentiation,
30                   inhibition of CRACI, and alterations in signaling pathways involved in skeletal
31                   development (Jang et al., 2008; Zuscik  et al., 2007).

32                   Epidemiologic studies have found associations between blood Pb levels  and effects on
33                   teeth. Large NHANES analyses adjusted for several potentially important confounding
34                   factors (including age, SES-related factors, and nutritional factors), and found
35                   associations between concurrent blood  Pb level and dental caries in children (Moss et al..
      November 2012                             5-658                  Draft - Do Not Cite or Quote

-------
 1                   1999) and periodontitis in adults (Saraiva et al., 2007). Higher patella and tibia Pb levels
 2                  were associated with tooth loss in NAS men (Arora et al.. 2009). The results for blood Pb
 3                  and bone Pb levels in adults indicate that long-term, cumulative exposure to Pb exposure
 4                  is associated with negative effect on teeth. This epidemiologic evidence was based on
 5                  cross-sectional study design analyses, which precludes conclusions about the
 6                  directionality of effects. However, these findings are supported by toxicological evidence
 7                  in animals for Pb-induced increases in Pb uptake into teeth; and decreases in cell
 8                  proliferation, procollagen type I production, intracellular protein, and osteocalcin in cells
 9                  exposed to Pb in vitro.

10                  The small body of epidemiologic evidence showing associations between Pb biomarker
11                  levels and various bone and teeth effects (after adjusting for potential confounding by
12                  age, SES-related factors, and nutritional factors), plus the supporting toxicological
13                  evidence, is sufficient to conclude that there is a likely causal relationship between Pb
14                  exposure and effects on bone and teeth.
      5.9.5        Effects on Ocular Health

15                  Ocular effects most commonly associated with exposure to Pb include formation of
16                  cataracts, impaired vision, edema and retinal stippling.
                    5.9.5.1      Summary of Key Findings of the Effects on Ocular Health
                                 (2006 Pb AQCD)

17                  The 2006 Pb AQCD stated that various changes in the visual system were observed with
18                  Pb poisoning including retinal stippling and edema, cataracts, ocular muscle paralysis and
19                  impaired vision. Maternal prenatal blood Pb levels in the range of 10.5 to 32.5 ug/dL
20                  were associated with supernormal retinal ERGs in children at age 5-7 years. Cataracts
21                  were noted in middle-aged men with tibia bone Pb levels of 31-126 ug/g.
                    5.9.5.2     Recent Toxicological and Epidemiologic Studies

22                  The recent cross-sectional epidemiologic studies of ocular effects in adults did not
23                  produce clear evidence, and each was limited by the lack of rigorous statistical analysis
24                  and lack of consideration for potential confounding. Erie et al. (2009) measured Pb and
25                  Cd in retinal tissue from 36 eye donors with age-related macular degeneration (cases) and
26                  25 normal control donors. Pb, but not Cd, concentration was significantly elevated in the
27                  neural retina tissue of the 36 donors with macular degeneration (72 eyes; median [IQR]:

      November 2012                            5-659                 Draft - Do Not Cite or Quote

-------
 1                   12.0 [8-18] ng/g Pb) versus normal control donors (50 eyes; median [IQR]: 8.0 [0-11]
 2                   ng/g Pb). Neither of these heavy metals were significantly elevated in the retinal pigment
 3                   epithelium (RPE)/choroid complex in donors with macular degeneration over normal
 4                   controls. Mosad et al. (2010) compared Pb, Cd, vitamin C, vitamin E, and beta carotene
 5                   blood levels between 45 middle-aged male smokers and nonsmokers with cataracts.
 6                   Blood Pb levels were elevated (p <0.0001) in 15 light (mean [SD]: 14.5 [0.41] (ig/dL), 15
 7                   moderate (14.5 [0.41] (ig/dL),  and 15 heavy smokers (18.7 [1.24] (ig/dL) compared to  15
 8                   nonsmokers (12.2 [0.21] (ig/dL). Similar associations were observed for Cd blood levels
 9                   and lens concentrations. There was no direct analysis of the association between Pb blood
10                   level or lens Pb concentration and the severity of cataracts.

11                   Recent animal studies have observed Pb-induced retinal progenitor cell proliferation and
12                   neurogenesis (Section 5.3.7.3). An in vitro study found increased opacity of rat (age
13                   4-6 weeks) lens exposed to 1 (iM Pb nitrate with or without secondary oxidative
14                   challenge after 5-8 days but not after 3 days (Neal et al.. 2010b). Thus, short-term Pb
15                   exposure did not induce osmotic swelling or lens shrinkage. With a 5-day exposure, 30%
16                   of the Pb-exposed lenses displayed "definite cataracts" compared to only 2.5% of control
17                   lenses. By culture day 8, 100% of the exposed lenses were described either as clearly
18                   opaque or definite cataracts, while only 7% of control lenses displayed these
19                   characteristics, indicating that prolonged exposure of lenses to Pb induced an accelerated
20                   formation of opacity/cataract compared to unexposed lenses. Pb-exposed lenses cleared
21                   the media of hydrogen peroxide more rapidly than did control lenses, potentially due to
22                   increased CAT activity. Exposure to hydrogen peroxide resulted in total (100%) opacity
23                   in Pb-exposed lenses at culture day 7, compared to less than 20% in control cells.
24                   Exposure to Pb additionally altered epithelial nutrient transport and lens histology,
25                   relative to that in controls.

26                   In summary, prospective epidemiologic evidence indicates associations between prenatal
27                   maternal blood Pb levels of 10.5-32.5 (ig/dL and supernormal retinal ERGs in children
28                   ages 5-7 years after adjusting for age, sex, and head circumference. However, the
29                   relevance of supernormal ERGs is uncertain. Evidence in adults for associations between
30                   eye tissue Pb levels and macular degeneration (Mosad et al.. 2010) and cataracts in adults
31                   (Erie et al.. 2009) is limited by weak statistical methods and lack of consideration for
32                   potential confounding to warrant conclusions. Toxicology studies have reported
33                   Pb-induced retinal progenitor cell proliferation, retinal ERGs, and lens opacity
34                   (Section 5.3.7.3). Because of the insufficient quantity and quality of studies in the
35                   cumulative body of evidence, the evidence is inadequate to determine a causal
36                   relationship between Pb exposure and ocular effects.
      November 2012                             5-660                 Draft - Do Not Cite or Quote

-------
      5.9.6        Effects on the Respiratory System

 1                  Blood Pb level has been associated with asthma and allergy in children in prospective and
 2                  cross-sectional epidemiologic studies (Section 5.6.5.2). As described in Section 5.2.4.
 3                  Pb exposure has been shown to induce the generation of ROS. ROS are implicated in
 4                  mediating increases in bronchial responsiveness and activating neural reflexes, leading to
 5                  decrements in lung function. Collectively, studies investigating these airway responses in
 6                  asthma-free populations are limited in number, lack rigorous statistical analysis, and
 7                  collectively do not provide strong evidence of an association with blood Pb level
 8                  (Section  5.6.5.3). Collectively, panel and time-series epidemiologic studies demonstrate
 9                  associations between short-term increases in ambient air Pb (measured in PM2 5 or PMi0
10                  air samples), and decreases in lung function and increases in respiratory symptoms, and
11                  asthma hospitalizations in children but not adults (Section 5.6.5.3). Toxicological studies
12                  have found pulmonary inflammation induced by concentrated ambient air particles
13                  (CAPs) in which Pb was one of the numerous components (Wei et al.. 2011; Duvall et al..
14                  2008; Godleski et al.. 2002; Saldiva et al.. 2002). Despite this evidence for respiratory
15                  effects related to air-Pb concentrations, the limitations of air-Pb studies, including the
16                  limited data on the size distribution of Pb-PM (Section 3.5.3). the uncertain relationships
17                  of Pb-PMio and Pb-PM2 5 with blood Pb levels, and the lack of adjustment for other
18                  correlated PM chemical components preclude firm conclusions about air Pb-associated
19                  respiratory effects. Because of the insufficient quantity and quality of studies in the
20                  cumulative body of evidence, the evidence is inadequate to determine a causal
21                  relationship between Pb exposure and respiratory effects in populations without asthma.
      5.10      Cancer

22                  Previous AQCDs have demonstrated that Pb is a well-established animal carcinogen.
23                  Oral Pb acetate exposure to male and female rodents has consistently been shown to be a
24                  kidney carcinogen in multiple separate studies, inducing adenocarcinomas and adenomas
25                  after chronic exposure. Developmental Pb acetate exposure also induced kidney tumors
26                  in offspring whose dams received Pb acetate in drinking water during pregnancy and
27                  lactation. Gliomas of the brain have also been reported after oral Pb exposure. These
28                  rodent toxicological studies have been conducted at high doses of Pb and have shown that
29                  Pb is an animal carcinogen. Because of this strong body of historical data, the
30                  2006 Pb AQCD states, "limited tumorigenesis studies have been conducted in animal
31                  models and the focus has been more on the mechanism of neoplasia... and possible
32                  immunomodulatory effects of Pb in the promotion of cancer." More recent studies have
      November 2012                            5-661                 Draft - Do Not Cite or Quote

-------
 1                   focused on administration of Pb with known carcinogens or modifiers such that lifestage,
 2                   diet, and mechanism of action can be belter understood.

 3                   The previous epidemiologic studies included in the 2006 Pb AQCD (U.S. EPA. 2006b)
 4                   "provide [d] only very limited evidence suggestive of Pb exposure associations with
 5                   carcinogenic or genotoxic effects in humans," and the studies were summarized as
 6                   follows:
                        "The epidemiologic data ... suggest a relationship between Pb exposure and cancers of the
                        lung and the stomach...  Studies of genotoxicity consistently link Pb-exposed populations
                        with DNA damage and micronuclei formation, although less consistently with
                        chromosomal aberrations."
 7                   The International Agency  for Research on Cancer (IARC) classified inorganic Pb
 8                   compounds as probable human carcinogens (Group 2A of IARC classification) based on
 9                   sufficient evidence in animal studies (evidence in human studies was  limited), and
10                   organic Pb compounds as  not classifiable (Group 3 of IARC classifications) (IARC.
11                   2006a; Rousseau et al.. 2005). Additionally, the National Toxicology Program (NTP)
12                   listed Pb and Pb compounds as "reasonably anticipated to be human carcinogens" (NTP.
13                   2011). The typical cancer  bioassays employed by IARC or NTP as evidence of
14                   Pb-induced carcinogenicity used rodents that were continuously exposed to Pb acetate in
15                   chow or drinking water for 18 months to two years in duration. These two year cancer
16                   bioassays and the doses administered are typical of cancer bioassays used with other
17                   chemicals.

18                   In the following sections, recent epidemiologic and toxicological studies published since
19                   the 2006 Pb AQCD, regarding Pb and cancer mortality and incidence are examined. In
20                   addition, recent studies of Pb exposure associated with DNA and cellular damage, as well
21                   as epigenetic effects, are summarized. When the information is available, the form of the
22                   Pb compound under study (e.g., inorganic, organic) is indicated. In epidemiologic
23                   studies, various biological indicators of Pb exposure  are used  including Pb measured in
24                   blood and bone. The biological indicators of Pb associated with cancer-related endpoints
25                   are considered in drawing conclusions about potentially important levels and timing of Pb
26                   exposure. Bone Pb is indicative of cumulative Pb exposure. Blood Pb can represent more
27                   recent exposure, but because it can also represent remobilized Pb occurring during times
28                   of bone remodeling, blood Pb level may also be an indicator of long-term Pb exposure in
29                   adults. Toxicological studies only report exposure by blood Pb or exposure dose. More
30                   detailed discussion of these measures is given in Section 4.3.5. Details of the recent
31                   epidemiologic and toxicological studies follow.
      November 2012                             5-662                  Draft - Do Not Cite or Quote

-------
     5.10.1       Cancer Incidence and Mortality

1                  Recent studies have included epidemiologic evaluations of the associations between Pb
2                  exposure and both specific cancers (such as lung cancer and brain cancer), and overall
3                  cancer (cancer of any type). Table 5-49 provides an overview of the study characteristics
4                  and results for the epidemiologic studies that reported effect estimates. This section also
5                  evaluates toxicological evidence on the potential carcinogenicity of Pb.
     November 2012                            5-663                 Draft - Do Not Cite or Quote

-------
Table 5-49 Summary of recent epidemiologic studies of cancer incidence and overall cancer mortality.
Reference
(In order of Study
appearance in text.) Location

Cancer Study
Outcome Population

Methodological Measure of
Details Pb Exposure

Mean Pb
(SD)

Adjusted Effect
Estimates
Potential
confounders
adjusted for in
analysis
Overall Cancer Mortality:
Menke et al. (2006) Multiple U.S.
locations













Schoberet al. Multiple U.S.
(2006) locations






Overall NHANES III
cancer cohort with
mortality Blood Pb
measures in
1 988-1 994
At least 1 2 years
of follow-up
Blood Pb
<10 ug/dL
N=1 3,946
N for cancer
mortality = 411




Overall NHANES III
cancer cohort
mortality At |east 40 years
of age
N= 9,757
N for cancer
mortality = 543





Cohort study Blood Pb at
using Cox baseline
regression and
other techniques











Cohort study Blood Pb at
using Cox baseline
proportional
hazard regression
analysis and other
techniques






2.58 ug/dL
(geometric
mean)


Tertile 1 :
<1.93 ug/dL


Tertile 2:
1. 94-3.62 ug/dL


Tertile 3:
> 3.63 ug/dL


Blood
Pb<5 ug/dL:
67.7%
Blood Pb
5-9 ug/dL:
26.0%
Blood
Pb>=10 ug/dL:
6.3%




HR (95% Cl):




Tertile 1 :
1.00



Tertile 2:
0 72
(95% Cl: 0.46, 1.12)

Tertile 3:
1.10
(95% Cl: 0.82, 1.47)
RR (95% Cl):
Blood Pb<5 ug/dL:
1.00
Blood Pb 5-9 ug/dL:
1.44 (95% Cl: 1.12,
1.86)
Blood Pb> 10 ug/dL:
1.69 (95% Cl: 1.14,
2.52)

Note: Modification by
age assessed and
associations varied
Age, race-
ethnicity, sex,
diabetes mellitus,
body mass index,
current or former
smoking, alcohol
consumption,
physical activity,
low income, CRP,
total cholesterol,
high school
education, urban
residence,
postmenopausal
status,
hypertension, and
level of kidney
function
Sex, race /
ethnicity,
education, and
smoking status
Age used as time-
scale in models
Additional
covariates
considered but not
included: Census
region and urban
status of
residence, alcohol
intake
                                                                                                         slightly
November 2012
5-664
Draft - Do Not Cite or Quote

-------

Reference
(In order of Study Cancer
appearance in text.) Location Outcome
Weisskopf et al. Boston, MA Overall
(2009) area cancer
mortality

















Khalil et al. (2009b) Baltimore, MD, Overall
and cancer
Monongahela mortality
Valley, PA








Study
Population
Normative Aging
Study (MAS)
Included men
only (mostly
white)
Mean follow-up
period for this
study: 8.9 yr
Blood Pb
measures
available
N=1038

N Tor cancer
mortality=85
Bone Pb
measures
available
N=727
N for cancer
mortality=57



Subgroup of the
Study of
Osteoporotic
Fractures cohort
Included white
women aged
65-87;
1 2 yr (+/- 3 yr)
follow-up
N=533
N for cancer
mortality=38


Methodological Measure of
Details Pb Exposure
Cohort study Blood Pb at
using baseline,
Cox proportional Patella Pb at
hazards baseline
















Cohort study Blood Pb at
using baseline
Cox proportional
hazards
regression
analysis and other
techniques







Mean Pb
(SD)
Blood Pb:
5.6 ug/dL (3.4)
Tertile 1 of
Blood Pb:
<4 ug/dL
Tertile 2 of
Blood Pb:
4-6 ug/dL
Tertile 3 of
Blood Pb:
>6 ug/dL



Tertile 1 of
patella Pb:
<22ug/g
Tertile 2 of
patella Pb:
22-35ug/g
Tertile 3 of
patella Pb:
>35 ug/g
Blood Pb Level
5.3 (2.3)ug/dL








Adjusted Effect
Estimates
HR (95% Cl):


Blood Pb Tertile 1:
1.00
Blood Pb Tertile 2:
1.03(95% Cl: 0.42,
2.55)
Blood Pb Tertile 3:
0.53 (95% Cl: 0.20,
1.39)

Patella Pb Tertile 1:
1.00
Patella Pb Tertile 2:
0.82 (95% Cl: 0.26,
2.59)
Patella Pb Tertile 3:
0.32 (95% Cl: 0.08,
1 .35)


HR (95% Cl):
Blood Pb<8 ug/dL:
1.00
Blood Pb> 8 ug/dL:
1.64(95% Cl: 0.73,
3.71)






Potential
confounders
adjusted for in
analysis
Age, smoking, and
education
Additional
covariates
considered but not
included: alcohol
intake, physical
activity, body
mass index, total
cholesterol, serum
high-density
lipoprotein,
diabetes mellitus,
race, and
hypertension









Age, clinic, BMI,
education,
smoking, alcohol
intake, estrogen
use, hypertension,
walking for
exercise,
diabetes, and total
hip BMD




November 2012
5-665
Draft - Do Not Cite or Quote

-------
Reference
(In order of
appearance in text.)
Study
Location
Cancer
Outcome
Study
Population
Methodological
Details
Measure of
Pb Exposure
Mean Pb
(SD)
Adjusted Effect
Estimates
Potential
confounders
adjusted for in
analysis
Overall Cancer Incidence:
Absalon and Slesak
(2010)
Obhodaset al.
(2007)
Mendy et al.
(2012)
Silesia
province,
Poland
Island of Krk,
Croatia
Multiple U.S.
locations
Overall
cancer
incidence
Incidence
rates for
neoplasms
Incidence of
cancer or
"malignancy
of any kind"
Children living in
this province at
least five years
N = not specified
Individuals living
in the Island of
Krk from
1997-2001
N= 1,940
2007-2008
NHANES cohort
- at least 20
years of age
N= 1,857
Ecologic analysis
using correlations.
Cross-sectional
study using
correlations and
linear regression
Cross-sectional
study using
logistic regression
Pb-related air
pollution
measures
Soil and
vegetation
samples,
household
potable water
samples,
children's hair
samples
Concurrently
measured
creatinine-
adjusted urinary
Pb
NA
NA
Geometric
mean for
creatinine-
adjusted urinary
Pb marker:
0.59 ug/g (95%
Cl: 0.57, 0.61)
Reported
correlations between
changes in Pband
cancer incidence -
no/low correlations
observed (correlation
coefficients between
-0.3 and 0.2)
No association
observed between
Pb in the samples
and incidence of
neoplasm (numerical
results not provided)
OR (95% Cl):
Greater than log-
transformed mean
creatinine-adjusted
urinary Pb level
compared to less
than log-transformed
mean creatinine-
adjusted urinary Pb
level: 0.76 (0.44,
1.33)
None specified
Examined
correlation by sex,
no difference
reported
None specified
Age, sex,
race/ethnicity,
education level,
ratio family
income to poverty,
alcohol
consumption,
cigarette smoking,
and other heavy
metals
November 2012
5-666
Draft - Do Not Cite or Quote

-------

Reference
(In order of Study
appearance in text.) Location


Cancer Study
Outcome Population


Methodological
Details


Measure of
Pb Exposure


Mean Pb
(SD)


Adjusted Effect
Estimates
Potential
confounders
adjusted for in
analysis
Lung Cancer:
Lundstrom et al. Sweden
(2006)












Jones et al. (2007) Humberside,
U.K.













Lung cancer Male Pb smelter
(incidence workers first
and mortality) employed for
a 3 months
between 1 928
and 1979
Followed up for
mortality from
1 955 -1 987
N=187
N lung
cancer=46








Lung cancer Male tin smelter
mortality employees
N=1,462












Nested case-
referent study
using conditional
logistic regression












Cohort study
using Poisson
regression












Median peak
blood Pb level


Median number
of years with at
least one blood
sample
obtained

Median
cumulative
blood Pb index
(sum of annual
blood Pb Level)





Personnel
record cards
and air
sampling
conducted from
1972-1991
Three exposure
scenarios
determined for
working lifetime
cumulative
exposure - all
have similar
medians of
approximately
2 mg/m3-yr
Median peak
blood Pb Level:
cases
49.7 ug/dL,
controls
55.9 ug/dL

Median number
of years with at
least one blood
sample
obtained:
cases 4.5 yr,
controls 6.0 yr
Median
cumulative
blood Pb index:
cases
186 ug/dL,
controls
246 ug/dL



~2.0 mg/m3-yr














OR (95% Cl):
Median peak blood
Pb Level:
1 .00 (0.71 , 1 .42)


Median number of
years with at least
one blood sample
obtained:
0.98(0.96, 1.01) per
10ug/dL

Median cumulative
Blood Pb index:
1.00(0.98, 1.01) per
10 ug/dL

Note: similar results
were observed when
restricted to smokers
only
RR for Pb exposure
weighted age and
time since exposure
(90% Cl): 1.54(1.14,
2.08)

Note: Similar results
for other exposure
determination
scenarios.






Matched by age
Adjusted for
smoking and As
exposure












Not specified














November 2012
5-667
Draft - Do Not Cite or Quote

-------
Reference
(In order of
appearance in text.)
Study
Location
Cancer
Outcome
Study
Population
Methodological
Details
Measure of
Pb Exposure
Mean Pb
(SD)
                                                                                                                      Adjusted Effect
                                                                                                                      Estimates
Potential
confounders
adjusted for in
analysis
Rousseau et al.
(2007)
                    Montreal,
                    Canada
                Lung cancer
                incidence
              Men aged 35-79
              N population
              based controls=
              ranged from 271
              to 471
              depending on
              exposure of
              interest
              N controls with
              other cancers=
              ranged from 737
              to 1203
              depending on
              exposure of
              interest
              N lung
              cancer= ranged
              from 433 to 751
              depending on
              exposure of
              interest
                 Population-based    Interview of job
                 case-control study    history and
                 using               exposure matrix
                 unconditional
                 logistic regression
                                   Ever exposed
                                   to:
                                   Organic Pb
                                   3.0%
                                   Inorganic Pb
                                   17.0%
                                   Pb in gasoline
                                   emissions
                                   38.6%
                                 OR (95% Cl):
                                 Organic Pb exposure
                                 compared to no
                                 exposure:
                                 Lung
                                 1.3 (95% 01:0.5,3.1)
                                 Inorganic Pb
                                 exposure compared
                                 to no exposure:
                                 Lung
                                 1.1 (95%CI:0.7, 1.7)
                                 Pb in gasoline
                                 emissions exposure
                                 compared to no
                                 exposure:
                                 Lung
                                 0.8 (95% Cl: 0.6, 1.1)


                                 Note: results are for
                                 comparisons using
                                 population-based
                                 controls; results for
                                 controls with other
                                 types of cancers
                                 were similar
                                     Age, family
                                     income, cultural
                                     origin, proxy
                                     status, ever
                                     exposure to
                                     asbestos, silica,
                                     As, Cd, and
                                     chromium (VI)
November 2012
                             5-668
                                       Draft - Do Not Cite or Quote

-------
Reference
(In order of
appearance in text.)
Study
Location
Cancer
Outcome
Study
Population
Methodological
Details
Measure of
Pb Exposure
Mean Pb
(SD)
Adjusted Effect
Estimates
Potential
confounders
adjusted for in
analysis
Brain Cancer:
van Wijngaarden
and Dosemeci
(2006)
Multiple U.S
locations
Brain cancer
mortality
National
Longitudinal
Mortality Study -
included
individuals with
occupational
information
-included follow-
up from
1970-1989
N= 317,968
Cohort study
using proportional
hazards, Poisson
regression
techniques, and
standardized
mortality ratios
(SMR)
Interview about
current or most
recent job within
the past 5 years
and a job
exposure matrix
                                                                                                           NA
HR (95% Cl):
Any Pb exposure
compared to no
exposure 1.56
(95% Cl: 1.00, 2.43)
RR (95% Cl) from
Poisson regression:
1.42
(0.91, 2.20)
SMR (95% Cl):
Not exposed: 0.87
(0.70, 1.06)
Any exposure:
1.11
(0.74, 1.59)
Gender, age, race,
living in an urban
area, marital
status and
educational level
Additional
covariate
considered but not
included: Family
income (not used
due to  large %
missing; additional
analysis including
it gave similar
results)
                                                                                                                            Note: Effect
                                                                                                                            estimates were
                                                                                                                            greatest among
                                                                                                                            those with high
                                                                                                                            probabilities of
                                                                                                                            exposure and
                                                                                                                            medium/high
                                                                                                                            exposure intensity
November 2012
                               5-669
                                         Draft - Do Not Cite or Quote

-------
Reference
(In order of
appearance in text.)
Study
Location
Cancer
Outcome
Study
Population
Methodological
Details
Measure of
Pb Exposure
Mean Pb
(SD)
                                                                                                                     Adjusted Effect
                                                                                                                     Estimates
Potential
confounders
adjusted for in
analysis
Rajaraman et al.
(2006)
                    Phoenix, AZ,
                    Boston, MA,
                    and Pittsburgh,
                    PA
                Brain cancer
                incidence
              NCI Brain Tumor
              Study
              - included
              individuals >=18
              yr diagnosed
              with brain cancer
              less than 8 week
              before
              hospitalization;
              frequency-
              matched controls
              were individuals
              admitted to the
              same hospitals
              for non-
              neoplastic
              conditions
              N controls
              =799
              N glioma
              = 489
              N meningioma
              =197
                 Case-control
                 study using
                 unconditional
                 logistic regression
                  Interviews of
                  lifetime work
                  history and
                  exposure
                  databases
                                                                                                      NA
                                 OR (95% Cl):
                                 Meningioma:
                                 Ever exposure to Pb
                                 0.8(0.5, 1.3)
                                 Glioma:
                                 Ever exposure to Pb
                                 0.8(0.6, 1.1)
                                                                                                                      Note: positive
                                                                                                                      associations
                                                                                                                      between Pb
                                                                                                                      exposure and
                                                                                                                      meningioma
                                                                                                                      incidence was
                                                                                                                      observed among
                                                                                                                      individuals with
                                                                                                                      ALAD2 genotypes,
                                                                                                                      but not individuals
                                                                                                                      with ALAD1
                                                                                                                      genotypes; these
                                                                                                                      associations were
                                                                                                                      not observed for
                                                                                                                      glioma incidence
                                     Age, sex, race /
                                     ethnicity, hospital,
                                     and residential
                                     proximity to
                                     hospital
November 2012
                             5-670
                                       Draft - Do Not Cite or Quote

-------
Reference
(In order of Study Cancer Study Methodological Measure of
appearance in text.) Location Outcome Population Details Pb Exposure
Bhatti et al. (2009) Phoenix, AZ, Brain cancer NCI Brain Tumor Case-control Interviews of
Boston, MA, incidence Study study using lifetime work
and Pittsburgh, included non unconditional history and
PA Hispanic whites logistic regression exposure
> .{jj yr databases
diagnosed with
brain cancer less
than 8 week
before
hospitalization;
frequency-
matched controls
were individuals
admitted to the
same hospitals
for non-
neoplastic
conditions
N controls
=494
N glioma
= 362
N meningioma
=134





Mean Pb
(SD)



Glioma:
70.5 ug/m -yr
(193.8 ug/m i -y)



Glioblastoma
multiform:
97.5 ug/m3-yr
(233.9 ug/m -y)







Meningioma:
101.1 ug/m3-yr
(408.7 ug/m3-y)

Controls'
69.7 ug/m3-yr
(248.8 ug/rrvSO
Potential
confounders
Adjusted Effect adjusted for in
Estimates analysis
OR (95% Cl) per Age, sex, hospital,
100 ug/m3y increase and residential
in cumulative Pb proximity to the
exposure hospital
Glioma'
1.0(0.9, 1.1)





Glioblastoma
multiform:
1.0(0.9, 1.1)






Meningioma: 1.1
On *i o\
.0, 1.2)

Note: modification by
SNPs was conducted
and associations
varied by SNP
November 2012
5-671
Draft - Do Not Cite or Quote

-------

Reference
(In order of Study
appearance in text.) Location


Cancer Study
Outcome Population


Methodological
Details


Measure of Mean Pb
Pb Exposure (SD)


Adjusted Effect
Estimates
Potential
confounders
adjusted for in
analysis
Breast Cancer:
Pan et al Canada
.(2011)



















Breast cancer National
incidence Enhanced
Cancer
Surveillance
System
(NECSS)-
population-
based sample of
cancer cases
and controls with
information
collected from
1 994-1 997
N controls
=2467

N cases=
2343





Population-based
case-control study
using
unconditional
logistic regression


















Self-reported NA
previous
addresses and
their proximity
to Pb smelters
(determined
using
Environmental
Quality
Database
[EQDB])












OR (95% Cl):
Residing
>3.2 km from Pb
smelter or no nearby
smelter
1.00


Residing
0.8-3.2 km from Pb
smelter:
0.41
(0.11, 1.51)


Residing
<0.8 km from Pb
smelter:
0.61
(0.11 , 3.42)



Age, province of
residence,
education,
smoking pack
years, alcohol
consumption,
body mass index,
recreational
physical activity,
number of live
births, age at
menarche,
menopausal
status, total
energy intake, and
employment in the
industry under
consideration
Additional
covariate
considered but not
included: Family
income
Other Cancers:
Rousseau et al. Montreal,
(2007) Canada












Various Men aged 35-79
cancer N p0pu|ati0n
incidences baseS contro|s=
ranged from 271
to 471
depending on
the cancer and
exposure of
interest
N controls with
other cancers=
ranged from 697
to 2,250
depending on
exposure of
interest
N
cancer= ranged
from 60 to 442
depending on
the cancer and
exposure of
interest
Population-based
case-control study
using
unconditional
logistic regression












Interview of job Ever exposed
history and to:
exposure matrix organic Pb
3.0%
Inorganic Pb
1 7.0%
Pb in gasoline
emissions
38.6%











OR (95% Cl):
Never exposed is
referent group
Organic Pb:
Esophageal
1.7(0.5,6.4)
Stomach
3.0(1.2,7.3)
Colon
1.5(0.7,3.6)
Rectum
3.0(1.2,7.5)
Pancreas
0.9(0.1,5.2)
Prostate
1.9(0.8,4.6)
Bladder
1.7(0.7,4.2)


Kidney
Age, family
income, cultural
origin or
birthplace, and
proxy status; all
models except
those for
melanoma and
non-Hodgkin's
lymphoma were
adjusted for
smoking











November 2012
5-672
Draft - Do Not Cite or Quote

-------
Reference
(In order of
appearance in text.)
Study
Location
Cancer
Outcome
Study
Population
Methodological
Details
Measure of
Pb Exposure
Mean Pb
(SD)
Adjusted Effect
Estimates
Potential
confounders
adjusted for in
analysis
                                                                                                                  2.3(0.8,6.7)
                                                                                                                  Non-Hodgkin's
                                                                                                                  lymphoma
                                                                                                                  0.4(0.1,2.2)

                                                                                                                  Inorganic Pb:
                                                                                                                  Esophageal
                                                                                                                  0.6(0.3,1.2)
                                                                                                                  Stomach
                                                                                                                  0.9(0.6, 1.5)
                                                                                                                  Colon
                                                                                                                  0.8(0.5,1.1)
                                                                                                                  Rectum
                                                                                                                  0.8(0.5, 1.3)
                                                                                                                  Pancreas
                                                                                                                  0.9(0.4,1.8)
                                                                                                                  Prostate
                                                                                                                  1.1 (0.7, 1.6)
                                                                                                                  Bladder
                                                                                                                  1.1 (0.7,1.5)
                                                                                                                  Kidney
                                                                                                                  1.0(0.6, 1.7)
                                                                                                                  Melanoma
                                                                                                                  0.4(0.2,1.0)
                                                                                                                  Non-Hodgkin's
                                                                                                                  lymphoma
                                                                                                                  0.7(0.4, 1.2)

                                                                                                                  Pb in gasoline
                                                                                                                  emissions:
                                                                                                                  Esophageal
                                                                                                                  0.6(0.4,1.1)
                                                                                                                  Stomach
                                                                                                                  1.0(0.7, 1.4)
                                                                                                                  Colon
                                                                                                                  0.8(0.6,1.1)
                                                                                                                  Rectum
                                                                                                                  1.0(0.7, 1.4)
                                                                                                                  Pancreas
                                                                                                                  0.9(0.5,1.4)
                                                                                                                  Prostate
                                                                                                                  0.9(0.7, 1.2)
                                                                                                                  Bladder
                                                                                                                  0.8(0.6, 1.1)
                                                                                                                  Kidney
                                                                                                                  1.0(0.7,1.5)
                                                                                                                  Melanoma
November 2012
                            5-673
                                      Draft - Do Not Cite or Quote

-------
Reference
(In order of
appearance in text.)
Study
Location
Cancer Study
Outcome Population
Methodological Measure of Mean Pb
Details Pb Exposure (SD)
Adjusted Effect
Estimates
Potential
confounders
adjusted for in
analysis
0.8(0.5,1.4)
Non-Hodgkin's
lymphoma
0.7(0.5,1.0)
                                                                                                                        Note: results are for
                                                                                                                        comparisons using
                                                                                                                        population-based
                                                                                                                        controls; results for
                                                                                                                        controls with other
                                                                                                                        types of cancers
                                                                                                                        were similar except
                                                                                                                        no association was
                                                                                                                        present between
                                                                                                                        organic Pb and rectal
                                                                                                                        cancer
November 2012
5-674
Draft - Do Not Cite or Quote

-------
Reference
(In order of Study Cancer
appearance in text.) Location Outcome
Santibanez et al. Valencia and Esophageal
(2008) Alicante, Spain cancer
incidence


































Study Methodological Measure of Mean Pb
Population Details Pb Exposure (SD)
PANESOES Case-control Interviews to NA
study study using determine
included 30-80 unconditional occupational
vr old men logistic regression history and a
hospitalized in job exposure
any of the matrlx
participating
study hospitals
N controls=285

N cancer=185
(147 squamous
cell, 38 adeno.
carcinoma)



























Potential
confounders
Adjusted Effect adjusted for in
Estimates analysis
OR (95% Cl): Age, hospital
All esophageal location,
cancers: educational level,
Unexposed: 1.00 smoking and
Low workplace alcohol use
Pb exposure
(< 4.9 ug/dL):
0.79 (0.43, 1 .46)
High workplace
Pb exposure
(>4.9 ug/dL):
1.69(0.57,5.03)

Esophageal
squamous cell
carcinoma:
Unexposed: 1.00
Low workplace
Pb exposure
(< 4.9 ug/dL):
0.70 (0.34, 1 .43)
High workplace Pb
exposure (>4.9
ug/dL):
0.91 (0.22, 3.75)
Adenocarcinoma:
Unexposed: 1.00
Low workplace
Pb exposure
(< 4.9 ug/dL):
0.95 (0.32, 2.82)
High workplace
Pb exposure
(>4.9 ug/dL):
5.30(1.39,20.22)
*Note: associations
not changed or
slightly increased
when restricted to
occupational
exposures > 15yr
November 2012
5-675
Draft - Do Not Cite or Quote

-------
                    5.10.1.1    Overall Cancer Mortality

 1                  Several recent cohort studies examined the association between Pb levels and cancer
 2                  mortality, including multiple analyses of the NHANES III population. In one NHANES
 3                  III analysis, the cohort of 13,946 (N for cancer mortality=411) was followed for 12 years
 4                  and individuals with blood Pb levels greater than 10 (ig/dL were excluded from the study
 5                  (mean baseline blood Pb level was 2.58 (ig/dL). No association was observed between
 6                  blood Pb and cancer mortality (HR of highest tertile [> 3.63 (ig/dL] compared to lowest
 7                  tertile [<1.93 (ig/dL]: 1.10 [95% CI: 0.82, 1.47]) (Menke et al.. 2006). Another analysis
 8                  of the NHANES III population, which was restricted to individuals 40 years and older at
 9                  the time of blood Pb collection and included 9,757 (N for cancer mortality=543)
10                  individuals with all blood Pb levels (including those greater than 10 (ig/dL), reported
11                  associations between blood Pb and  cancer mortality (Schober et al.. 2006). In this study,
12                  median follow-up time was 8.6 years. The RRs were 1.69 (95% CI: 1.14, 2.52) for
13                  individuals with blood Pb levels of at least 10 (ig/dL and 1.44 (95% CI:  1.12, 1.86) for
14                  blood Pb levels of 5-9 (ig/dL compared to individuals with blood Pb levels less than
15                  5 (ig/dL. When stratified by age, point estimates comparing blood Pb levels of 5-9 versus
16                  less than 5 (ig/dL were similar across all age groups but only statistically significant
17                  among 75-84 year olds. The risks of mortality associated with blood Pb levels > 10 (ig/dL
18                  in the groups aged 40-74 years and  85 years and older were elevated.

19                  A study of men (primarily white) from the greater Boston, MA area enrolled in the
20                  Normative Aging Study (NAS) found no association between blood or bone Pb and
21                  cancer mortality in adjusted analyses (N=l,038, N for cancer mortality=85 when using
22                  blood Pb measures; N=727, N for cancer mortality=57 when using bone Pb measures). At
23                  baseline, the mean (SD) blood Pb level for this population was 5.6 (3.4)  (ig/dL and blood
24                  Pb was poorly correlated with measured bone Pb (Weisskopf et al.. 2009). As part of the
25                  Study of Osteoporotic Fractures , 533 white women aged 65-87 (N for cancer
26                  mortality=38) were included in a sub-study of blood Pb level and cancer mortality and
27                  were followed for approximately 12 years (Khalil et al.. 2009b). The mean (SD) blood Pb
28                  level at baseline was 5.3 (2.3) (ig/dL and no association was detected between blood Pb
29                  and cancer mortality in the study population.

30                  Overall, epidemiologic studies of blood Pb levels and cancer mortality reported
31                  inconsistent results. An epidemiologic study using NHANES III data demonstrated the
32                  strongest association between blood Pb and increased cancer mortality; however, other
33                  studies reported weak or no associations. These cohort studies were well-conducted
34                  longitudinal studies with control for potential confounders, such as age, smoking, and
      November 2012                            5-676                  Draft - Do Not Cite or Quote

-------
 1                  education (see list of the potential confounders addressed in each study in Table 5-49).
 2                  One limitation is that the studies in populations other than NHANES cohorts each had a
 3                  small number of cancer mortality cases.
                    5.10.1.2    Overall Cancer Incidence

 4                  Studies of overall cancer incidence have also been performed (Table 5-49). An ecologic
 5                  analysis compared Pb-related air pollution over 5 year increments from 1990 to 2005
 6                  with incidence rates of cancer during this time period (cancer sites not specified) among
 7                  children (N not specified) (Absalon and Slesak. 2010). The highest Pb air pollution levels
 8                  were measured in 1990 when over 50% of the study area exceeded the limit of
 9                  1 (ig/m2-year. No correlation was observed both overall and in sex-specific analyses.
10                  Another study (N= 1,940) examined correlations between Pb concentrations in soil, water,
11                  vegetation, and hair samples with incidence of neoplasms (Obhodas et al.. 2007). The Pb
12                  concentrations were not correlated with incidence of neoplasms. A recent study using the
13                  2007-2008 NHANES cohort reported no association between higher creatinine-adjusted
14                  urine Pb levels and  elevated odds of having ever had cancer or a malignancy (N=l,857)
15                  (Mendy et al.. 2012). The timing of cancer diagnosis in relation to the urine sample
16                  collection was not identified.

17                  Overall, epidemiologic studies reported no positive associations between various
18                  measures of Pb exposure and overall cancer incidence. These studies are limited by their
19                  ecologic and cross-sectional designs. Absalon and Slesak (2010) and Obhodas (2007) did
20                  not collect biological measurements, and no control for potential confounding was
21                  mentioned.
                    5.10.1.3    Lung Cancer

22                  Most of the recent evidence regarding lung cancer incidence is provided by a few studies
23                  of occupationally-exposed adults. These are described in Table 5-49. Some studies in the
24                  2006 Pb AQCD (U.S. EPA. 2006b) reported associations between Pb exposure and lung
25                  cancer in occupational cohorts, although the studies were limited due to possible
26                  confounding by smoking or other workplace exposures. In a more recently published
27                  study of smelter workers (N= 187, N for lung cancer=46), no association was observed
28                  between several metrics of Pb exposure (peak blood Pb values, number of years Pb
29                  samples were obtained, and cumulative blood Pb index) and lung cancer incidence and
30                  mortality combined (Lundstrom et al.. 2006). The median follow-up in the study was
31                  about 30 years, and the median peak blood Pb values  during employment were
      November 2012                             5-677                  Draft - Do Not Cite or Quote

-------
 1                  49.7 (ig/dL for lung cancer cases and 55.9 (ig/dL for controls. In a study of 1,462 tin
 2                  smelter workers, no association was observed between Pb exposure and lung cancer
 3                  mortality in unweighted analyses, but when the analyses were weighted by age and time
 4                  since exposure, positive associations were apparent (Jones et al.. 2007). In this study,
 5                  cumulative Pb exposure was calculated by combining historical air sampling data and
 6                  personnel record cards, which specified work histories. The median cumulative Pb
 7                  exposure was estimated to be approximately y 2 mg/m3«yr. It is important to note that the
 8                  tin smelter workers were exposed to other metals as well,  such as As and antimony and
 9                  the study did not specify if additional potential confounders were evaluated (beyond the
10                  weighting for age and time since exposure). A population-based case-control study
11                  performed among men  in Montreal, Canada in the 1980s assessed occupational Pb
12                  exposure via interviews regarding job histories and determined the likely Pb exposures
13                  associated with the job  activities (Rousseau et al.. 2007). No association was apparent
14                  between organic Pb, inorganic Pb, or Pb from gasoline emissions and lung cancer (N
15                  ranged from 271 to 1,203 depending on the exposure of interest).

16                  Studies were also conducted that compared lung tissue Pb measurements for individuals
17                  with lung cancer to those without lung cancer. The controls for these studies were
18                  individuals with metastases in the lung from other primary cancers (De Palmaet al..
19                  2008) and individuals with non-cancerous lung diseases (De Palma et al., 2008; Kuo et
20                  al.. 2006). Limitations in these studies include their cross-sectional; design, the
21                  measurement of Pb in cancerous tissue, which may have altered Pb distribution, and the
22                  use of controls with other cancers and lung diseases. Findings are mixed among the
23                  studies. De Palma et al. (2008) reported higher Pb concentrations in the cancerous  and
24                  non-cancerous lung tissue of individuals with non-small cell lung cancer compared to
25                  control groups, although the authors report these results may be confounded by smoking.
26                  Kuo et al. (2006) found no statistical difference in Pb levels for lung tissue of individuals
27                  with lung cancer compared to controls.

28                  Some studies in the 2006 Pb AQCD reported associations between Pb exposure and lung
29                  cancer among occupational cohorts; however, recent epidemiologic studies of lung
30                  cancer reported no associations.  Overall, these recent epidemiologic studies included only
31                  men, limiting the generalizability. The studies by Jones et al. (2007) and Rousseau et al.
32                  (2007) also have the disadvantage of not obtaining actual measures of Pb exposure or
33                  biomarker levels. In addition, these studies, as well as those in the 2006 Pb AQCD, are of
34                  occupational cohorts, and the relationships with Pb exposures may be confounded by
35                  other workplace exposures and covariates that were not considered, such as smoking.
      November 2012                            5-678                  Draft - Do Not Cite or Quote

-------
                    5.10.1.4    Brain Cancer

 1                  A few studies of brain cancer examined the association between cancer and occupational
 2                  Pb exposure using exposures determined via exposure databases and patient interviews
 3                  about past jobs and known exposures (Table 5-49). The National Longitudinal Mortality
 4                  Study, a study that included a national sample of the U.S. population (N=317,968),
 5                  estimated occupational Pb exposure based on current/most recent employment among
 6                  individuals (van Wijngaarden and Dosemeci. 2006). Although not all estimates using
 7                  various statistical techniques and measures of Pb exposure/intensity are statistically
 8                  significant, a pattern of increased associations between Pb exposure and brain cancer
 9                  mortality was observed in the study population. In a case-control study of brain tumors
10                  (N for controls=799, N for glioma=489, n for meningioma =197), glioma was reported to
11                  have no association with any Pb exposure metric.  However, positive associations were
12                  observed between high cumulative occupational Pb exposure and meningioma among
13                  individuals with ALAD2 genotypes (OR 2.4 [95% CI: 0.7, 8.8] comparing individuals
14                  ever exposed to Pb with those not exposed to Pb; OR 12.8  [95% CI: 1.4, 120.8]
15                  comparing individuals with cumulative Pb exposure > 100 (ig-year/m3 to those not
16                  exposed to Pb) (Rajaraman et al.. 2006). This association was not present among
17                  individuals with the ALAD1 genotypes (OR 0.5 [95% CI: 0.3, 1.0] comparing individuals
18                  ever exposed to Pb with those not exposed to Pb; OR 0.7 [95% CI: 0.2, 1.8] comparing
19                  individuals with cumulative Pb exposure > 100 (ig-year/m3 to those not exposed to Pb).
20                  Another study of the association between occupational Pb exposure (measured using self-
21                  reported occupational exposure history) and brain tumors reported none or slight overall
22                  associations with types of brain tumors; however, positive associations were observed
23                  among individuals with certain genetic single nucleotide polymorphisms (SNPs) (N for
24                  controls=494, N for glioma=362, n for meningioma =134)  (Bhatti et al.. 2009). After
25                  control for multiple comparisons, individuals with GPX1 variants (rs 105 0450) had
26                  positive associations between cumulative Pb exposure and glioblastoma multiforme and
27                  meningioma. Individuals without RAC2 variants (rs2239774) showed a positive
28                  association between Pb and glioblastoma multiforme. Also, individuals withoutXDH
29                  variants (rs75 74920) displayed a positive association between Pb and meningioma.

30                  Overall, associations between occupational Pb exposure  and brain cancer incidence and
31                  mortality were found to vary according to several genetic variants.  Studies of the
32                  association between Pb exposure and brain cancer were not reported in the
33                  2006 Pb AQCD. These studies were limited in their methods because they do not have
34                  individual level biological or exposure Pb measurements and the potential for
35                  confounding by other workplace exposures exist. Additional research is needed to
36                  characterize these associations and the modification by various genetic variants.
      November 2012                            5-679                  Draft - Do Not Cite or Quote

-------
                    5.10.1.5    Breast Cancer

 1                  The association between proximity to a Pb smelter and breast cancer was evaluated in a
 2                  non-occupational cohort. A population-based case-control study in Canada (N for
 3                  controls=2,467, N for cases=2,343) examined the proximity to a Pb smelter based on
 4                  residential addresses (Pan etal. 2011) (Table 5-49). No association was reported
 5                  between proximity of a Pb smelter and breast cancer incidence, but the study was limited
 6                  by the small number of women who resided near a Pb smelter (n=13 lived < 3.2 km from
 7                  Pb smelter). No biological samples to determine Pb levels in the body were used in the
 8                  study, nor were Pb biomarker or exposure data were available.

 9                  A few case-control studies examined Pb levels in biological samples among individuals
10                  with and without breast tumor and/or cancer. A study of newly diagnosed breast cancer
11                  patients and controls examined Pb levels in blood and hair samples and reported higher
12                  Pb levels in both for cancer cases, although the difference in the Pb content in hair
13                  samples was not statistically significant (Alatise and Schrauzer. 2010). Siddiqui et  al.
14                  (2006) observed higher blood Pb levels in women with benign and malignant tumors
15                  compared to controls. Additionally, although blood Pb levels were higher among those
16                  with malignant breast tumors compared to those with benign tumors, both had similar
17                  levels of Pb detected in breast tissues. Another study of Pb levels present in breast tissue
18                  reported no statistical difference in Pb levels (Pasha et al., 2008b). However, a study of
19                  breast tissue did observe a statistically significant difference between Pb levels in the
20                  breast tissue of cancer cases and controls (lonescu et al.. 2007). Finally, a study of Pb
21                  levels in urine reported a positive association between urine Pb and breast cancer, but this
22                  association became null when women taking nonsteroidal aromatase inhibitors but not
23                  taking bisphostphonates (a combination responsible for bone loss) were excluded from
24                  the analysis (McElroy et al.. 2008).

25                  The 2006 Pb AQCD (U.S. EPA. 2006b) did not report any studies examining Pb levels
26                  and breast cancer. Overall, recent studies suggest that women with breast cancer may
27                  have higher blood Pb levels than those without breast cancer. However, results are mixed
28                  in studies that compared breast tissue Pb concentrations between breast tumor and  control
29                  samples. These studies are limited by their study design. The samples are taken after
30                  cancer is already present in the cases, thus, the directionality between tissue or blood Pb
31                  levels and cancer development cannot be established. Additionally, the sample sizes are
32                  often small and the studies may be underpowered (most of the studies had less than 25
33                  cases (Alatise and Schrauzer. 2010; lonescu et al.. 2007; Siddiqui et al., 2006). A case-
34                  control study, also limited by its method of exposure measurement, reported no
35                  association between living near a Pb smelter and breast cancer (Pan et al., 2011).
      November 2012                            5-680                  Draft - Do Not Cite or Quote

-------
                    5.10.1.6    Other Cancers

 1                  There have been a few studies of cancer types other than those listed above. The
 2                  2006 Pb AQCD reported evidence of an association between Pb exposure and stomach
 3                  cancer in several occupational cohorts. A study performed among men in Montreal,
 4                  Canada in the 1980s evaluated multiple cancer outcomes and estimated occupational
 5                  exposures to organic Pb, inorganic Pb, and Pb from gasoline emissions via interviews
 6                  regarding job histories and subsequent exposure approximations by chemists and
 7                  hygienists (N for cases and controls varied from 60 to 2,250 based on the cancer type and
 8                  exposure) (Rousseau et al.. 2007). Adults with occupational exposure to organic Pb
 9                  exposure had greater odds of having stomach cancer compared to adults without
10                  occupational exposure to organic Pb. A positive association was also observed for rectal
11                  cancer when population-based controls were used but was null when the control
12                  population was limited to individuals with other types of cancers. No association was
13                  detected for cancers of the esophagus, colon, pancreas, prostate, bladder, kidney,
14                  melanoma, or non-Hodgkin's lymphoma. None of the cancers were associated with
15                  occupational exposure to inorganic Pb. When occupational exposure to Pb in gasoline
16                  emissions was categorized as "unexposed," "nonsubstantial level," or "substantial level,"
17                  a positive association with stomach cancer was observed when cancer controls were used
18                  as the comparison group; however, the association was not present when population
19                  controls were utilized as the control group). Another case-control study using participant
20                  interviews and a job exposure matrix, including only men, (N for controls=285, N for
21                  cancer=185) reported no association between occupational Pb exposure and esophageal
22                  squamous cell carcinomas, but an association was present between high  occupational Pb
23                  exposure and adenocarcinoma of the esophagus (Santibanez et al.. 2008). However,
24                  neither of these studies quantified Pb levels in biological or exposure samples.

25                  Several studies compared Pb levels in blood, tissue, and urine of individuals who have
26                  cancer with Pb-levels in individuals who are cancer-free. Compared to control groups,
27                  higher Pb levels were observed in the blood and bladder tissue of individuals with
28                  bladder cancer (Golabek et al.. 2009). the kidney tissue of individuals with renal cell
29                  carcinoma (with  highest levels among those with the highest stage tumors) (Calvo et al..
30                  2009). the tissue  (but not serum) of individuals with laryngeal cancer (Olszewski et al..
31                  2006). the blood of individuals with gastric cancer (Khorasani et al.. 2008). the plasma
32                  and hair of individuals with gastrointestinal cancer (Pasha etal.. 2010). the blood and hair
33                  of individuals with non-specified types of cancer (Pasha et al.. 2008c: Pasha et al.. 2007).
34                  and the hair of individuals with benign tumors (Pasha et al.. 2008a). No statistical
35                  difference in Pb levels was reported for colon tissue of individuals with colorectal polyps
36                  (Alimonti et al.. 2008) or urine of individuals with bladder cancer (Lin et al.. 2009)
37                  compared to control groups. A study examining Pb levels in kidney tissue reported the

      November 2012                            5-681                 Draft - Do Not Cite or Quote

-------
 1                  highest levels of Pb in normal kidney tissue samples that were adjacent to neoplastic
 2                  tumors. The Pb levels reported in the kidney tissue of neoplastic tumors were elevated
 3                  compared to those detected in corpses without neoplastic tumors of the kidney (Cerulli et
 4                  al.. 2006). All of these comparison studies are limited by the inability to determine
 5                  temporality as Pb  biomarkers were measured after the cancer diagnosis; the level of Pb
 6                  may be due to changes that result from having cancer, not changes that result in cancer.
 7                  Many of these studies attempted to control for this by including only cases who have not
 8                  undergone certain treatments. Additionally, studies are limited by their small sample
 9                  sizes and the selection of the control populations. Control populations are supposed to
10                  represent the general population from which the cases are drawn; some of the control
11                  subjects in these studies are individuals with diseases/conditions warranting tissue
12                  resections, which  are not prevalent in the general population.

13                  In summary, epidemiologic studies examining the potential for associations of Pb
14                  exposure with the incidence of specific cancers reported varying associations with
15                  occupational Pb exposure. Associations were null for occupational Pb exposure and most
16                  cancer sites examined. However, a positive association was observed between
17                  occupational Pb exposure and adenocarcinoma of the esophagus as well as exposure to
18                  occupational organic Pb and stomach cancer, which is supported by evidence of a
19                  relationship between Pb exposure and stomach cancer in occupational cohorts reported in
20                  the 2006 Pb AQCD. Associations between occupational organic Pb exposure and rectal
21                  cancer and occupational exposure to Pb in gasoline emissions and stomach cancer were
22                  inconsistent. These studies of various cancer sites have limited generalizability due to the
23                  study populations comprising only men. In addition, there are no personal biological or
24                  exposure samples used in the epidemiologic analyses and confounding by other
25                  occupational exposures is possible. In other studies, biological samples were used in
26                  biomarker comparisons of cancer and cancer-free individuals but as stated above, these
27                  studies have multiple other limitations.
                    5.10.1.7    Animal Models of Carcinogenicity

28                  Previous AQCDs have established that Pb has been shown to act as a carcinogen in
29                  animal toxicology models, albeit at relatively high concentrations. Chronic oral
30                  Pb acetate exposure to male and female rodents has consistently been shown to be a
31                  kidney carcinogen in multiple separate studies, inducing adenocarcinomas and adenomas
32                  after chronic exposure. Gliomas of the brain have also been reported after oral Pb
33                  exposure. The kidneys are the most common target of Pb-dependent carcinogenicity
34                  (Kasprzak et al.. 1985; Roller etal.. 1985; Azaretal.. 1973; Van Esch and Kroes. 1969)
35                  but the testes, brain, adrenals, prostate, pituitary, and mammary gland have also been


      November 2012                             5-682                 Draft - Do Not Cite or Quote

-------
 1                  affected (IARC. 2006a). The typical cancer bioassays used by IARC or NTP as evidence
 2                  of Pb-induced carcinogenicity were designed using rodents, typically males but
 3                  sometimes animals of both sexes, that were continuously exposed to Pb acetate in chow
 4                  (i.e., 1,000 or 10,000 ppm Pb acetate) or drinking water (i.e.,  26 or 2,600 ppm Pb acetate)
 5                  for 18 months to two years in duration, the typical life span of a rodent (Kasprzak et al.
 6                  1985: Roller etal.. 1985: Azaretal. 1973: Van Esch and Kroes. 1969). These two-year
 7                  cancer bioassays and the doses employed are typical of cancer bioassays employed for
 8                  other chemicals, albeit at doses that are higher than Pb doses  cited in other toxicological
 9                  sections of the ISA. In cancer bioassays, to obtain statistically valid data from small
10                  groups of animals, doses are selected such that any dose-related effects will occur
11                  frequently enough to be detected. The 2006 Pb AQCD (U.S. EPA. 2006b) pointed out
12                  that because Pb is a "well-established animal carcinogen...., focus has been more on the
13                  mechanism of neoplasia and possible immunomodulatory effects of Pb in the promotion
14                  of cancer." This focus continues to date. More recent studies have focused  on
15                  administration of Pb with known carcinogens or antioxidants such that lifestage, diet, and
16                  mode of action can be better understood. Developmental Pb acetate exposure also
17                  induced kidney tumors  in offspring whose dams received Pb  acetate in drinking water
18                  during pregnancy and lactation.

19                  Recognition of the importance of windows of exposure in Pb-induced cancer bioassays is
20                  a focus of more recent studies. In one study, gestational and lactational exposure of
21                  laboratory rodents to inorganic Pb (500, 750 or  1,000 ppm Pb acetate in drinking water)
22                  induced carcinogenicity in adult offspring (Waalkes etal..  1995). Another recent study
23                  considered Pb-induced  carcinogenesis in laboratory animals with early life Pb exposure
24                  (gestation and lactation) in which Tokar et al. (2010) examined tumorigenesis in
25                  homozygous metallothionein I/II knockout mice and their corresponding wild type
26                  controls (groups often mice each). The dams/mothers were exposed by drinking water to
27                  2,000 or 4,000 ppm Pb  acetate during gestation  and lactation  and compared to untreated
28                  controls. Study animals were exposed in utero, through birth  and lactation, and then
29                  postnatally to drinking water until 8 weeks old.  The Pb-exposed metallothionein I/II
30                  knockout mice had increased testicular teratomas and renal and urinary bladder
31                  preneoplasia. The tumor burden of Pb-exposed wild-type mice were not statistically
32                  significantly different than controls. The data suggest that metallothionein can protect
33                  against Pb-induced tumorigenesis. Concerns with the study are  that the doses are at levels
34                  of Pb to which humans would not likely be exposed and there is no metallothionein null
35                  condition in humans, though there is variability in the expression of metallothionein. The
36                  data do not address whether this variability would have any impact on Pb-induced
37                  carcinogenesis in humans. Thus, the animal toxicology data demonstrate that Pb is a
38                  well-established animal carcinogen in studies employing high-dose Pb exposure over a
39                  continuous extended duration of exposure (i.e., 2 years), which  is typical of cancer

      November 2012                            5-683                  Draft - Do Not Cite or Quote

-------
 1                  bioassays. Newer studies are showing early-life maternal Pb exposure can contribute to
 2                  carcinogenicity in offspring and have shown that metallothionein is protective against
 3                  cancer in this pathway.
      5.10.2      Cancer Biomarkers

 4                  A cross-sectional study of men aged 21-40 years without occupational history of
 5                  exposure to metals examined prostate specific antigen (PSA), a biomarker for prostate
 6                  cancer (N=57). Studies of Pb exposure and PSA were not reported in the 2006 Pb AQCD
 7                  (U.S. EPA. 2006b). This recent study reported a positive association between Pb levels
 8                  and PSA levels (measured in the same blood samples) in regression models adjusted for
 9                  the following potential confounders: age, smoking, alcohol consumption, and other
10                  metals (Cd, Zn, Se, and Cu) (Pizent et al., 2009). The median concurrent blood Pb level
11                  was 2.6 (ig/dL (range 1.0-10.8 (ig/dL). The authors note that the study population was
12                  young  and at lower risk of prostate cancer than are older men.
      5.10.3       Modes of Action for Pb-induced Carcinogenicity

13                  The carcinogenic mode of action of Pb is poorly understood. It is unclear whether the
14                  mode of action of Pb is best understood within the framework of multistage
15                  carcinogenesis, genomic instability or epigenetic modification. For example, multistage
16                  carcinogenesis involves a series of cellular and molecular changes that result from the
17                  progressive accumulation of mutations that induce alterations in cancer-related genes. Pb
18                  does not appear to follow this paradigm, and the literature suggests it is weakly
19                  mutagenic. Pb does appear to have some ability to induce DNA damage
20                  (Section 5.10.3.2). However, the ability of Pb to alter gene expression through epigenetic
21                  mechanism (Section 5.10.3.3) and to interact with proteins may be a means by which Pb
22                  induces carcinogenicity. It is known that Pb can  replace Zn in Zn-binding (Zn-finger)
23                  proteins (Section 5.2). which include hormone receptors, cell-cycle regulatory proteins,
24                  the Ah receptor,  estrogen receptor, p53, DNA repair proteins, protamines, and histones.
25                  These Zn-finger  proteins all bind to specific recognition elements in  DNA. Thus, Pb may
26                  act at a post-translational stage to alter protein structure of Zn-finger proteins, which can
27                  in turn alter gene expression, DNA repair and other cellular functions. To recapitulate,
28                  cancer develops  from one or a combination of multiple mechanisms including
29                  modification of DNA via epigenetics or enzyme dysfunction and genetic instability or
30                  mutation. These  modifications then provide the cancer cells with a selective growth
31                  advantage. In this schematic, Pb may contribute  to epigenetic changes and chromosomal
32                  aberrations.

      November 2012                            5-684                 Draft - Do Not Cite or Quote

-------
 1                  The genomic instability paradigm requires a cascade of genome-wide changes caused by
 2                  impaired DNA repair, kinetochore assembly, cellular checkpoints, centrosome
 3                  duplication, microtubule dynamics or a number of cell maintenance processes. These
 4                  processes have been rarely studied for Pb, thus there are few data that suggest Pb may
 5                  interfere with some of these processes. Furthermore, the bulk of the literature in this area
 6                  involves Pb chromate and it is unclear if the effects are due to Pb or chromate. Epigenetic
 7                  modifications may lead to cancer by altering cellular functions without altering the DNA
 8                  sequence. The most commonly studied epigenetic change is methylation alterations. A
 9                  small number of studies show that Pb can induce epigenetic changes (Section 5.10.3.3).
10                  but studies  are still missing to clearly tie these effects to Pb-induced carcinogenesis and
11                  genotoxicity. Thus, either genomic instability or epigenetic modification paradigms or
12                  some combination of the two may underlie Pb-induced carcinogenicity.

13                  Exposure to mixtures can also contribute to understanding of modes of action. No recent
14                  studies of the protective role of Ca2+ or Zn in Pb-induced carcinogenesis or genotoxicity
15                  were found. Pb can displace these and other divalent cations, affecting physiological
16                  processes. There were some data suggesting that metallothionein (Section 5.10.4). which
17                  sequesters Pb and makes it less bioavailable, protects rodents from Pb-induced cancers.
18                  Boron, melatonin, N-acetylcysteine, turmeric and myrrh protected cells  against
19                  Pb-induced genotoxicity (Section 5.10.3.2) and affected antioxidant status, especially the
20                  glutathione pathway. There were some data suggesting that Pb mimics or antagonizes the
21                  essential micronutrient Se in rodents. These data are discussed in more detail elsewhere
22                  (Section 5.10.4) and point to the relevance of mixtures in assessing toxicity.
                    5.10.3.1    Neoplastic Transformation Studies, Human Cell Cultures

23                  Carcinogenesis can be measured in cell culture systems through neoplastic transformation
24                  models that monitor change by following morphological transformation of cells,
25                  i.e., formation of a focus (or foci) of cell growth. Xie et al. (2007) treated BEP2D cells
26                  (human papilloma virus- immortalized human bronchial cells) with 0, 1,  5, or 10 (ig/cm2
27                  PbCrO4 for 120 hours. PbCrO4 induced foci formation in a concentration-dependent
28                  manner. Xie et al. (2008) treated BJhTERT cells (hTERT-immortalized human skin
29                  fibroblasts) and ATLD-2 cells (hTERT-immortalized human skin fibroblasts deficient in
30                  Mrel 1) with 0, 0.1, 0.5, and 1 (ig/cm2 PbCrO4 for 120 hours. PbCrO4 induced foci
31                  formation in a concentration-dependent manner in the Mre 11  deficient cells. Mre 11 was
32                  required to prevent PbCrO4-induced neoplastic transformation.
      November 2012                            5-685                 Draft - Do Not Cite or Quote

-------
                    Immune Modulation of Tumorigenesis by Pb

 1                  As described in the 2006 Pb AQCD (U.S. EPA. 2006b). Pb-induced immunotoxicity can
 2                  contribute to increased risk of cancer, primarily as the result of suppressed Thl responses
 3                  and misregulated inflammation. First, Pb-induced misregulation of inflammation
 4                  involving innate immune cells has been shown to result in chronic insult to tissues. These
 5                  insults, excessive lipid and DNA oxidation production by overproduction of ROS and
 6                  weakened antioxidant defenses, can increase the likelihood of mutagenesis, cellular
 7                  instability, and tumor cell formation. For example, results from Xu et al. (2008) support
 8                  the association with Pb exposure and DNA damage, and investigators concluded that it is
 9                  a possible route to increased Pb-induced tumorigenesis. The second component of
10                  increased risk of cancer involves Pb-induced suppression of Thl-dependent anti-tumor
11                  immunity as acquired immunity shifts statistically significantly toward Th2 responses.
12                  With cytotoxic T lymphocytes and other cell-mediated defenses dramatically lessened,
13                  the capacity to resist cancer may be compromised.
                    5.10.3.2    DNA and Cellular Damage

14                  Multiple studies have been performed examining the relationship between Pb and DNA
15                  and cellular damage. Details of the recent epidemiologic and toxicological studies follow.


                    Epidemiologic Evidence for DNA and Cellular Damage

16                  Multiple studies examined the relationship between Pb and sister chromatid exchange
17                  (SCE). SCEs are exchanges of homologous DNA material between chromatids on a
18                  chromosome and are a test for mutagenicity or DNA damage. A study of male policemen
19                  reported mean blood Pb levels for the study population of 43.5 (ig/dL (Wiwanitkit et al.,
20                  2008). In analyses dichotomized as high or low blood Pb levels (cut-off at 49.7 (ig/dL),
21                  the higher blood Pb group was observed to have higher mean SCE. Another study of
22                  adult males compared the SCE of storage battery manufacturing workers (mean blood Pb
23                  levels of 40.14 (ig/dL) and office workers (mean blood Pb levels of 9.11 (ig/dL) (Duydu
24                  et al.. 2005). The exposed workers had higher SCE levels and also a greater number of
25                  cells  in which the SCEs per cell were higher than the 95th percentile of the population.
26                  Finally, a study of children aged 5-14 years old (mean [SD] blood Pb levels of 7.69
27                  [4.29] (ig/dL) reported no correlation between blood Pb levels and SCE (Mielzvnska et
28                  al.. 2006). However, the study did report a positive association between blood Pb and
29                  micronuclei (MN) levels.
     November 2012                            5-686                  Draft - Do Not Cite or Quote

-------
 1                  Other studies of DNA damage have reported mixed results. A study of children ages 6-11
 2                  years old and environmentally-exposed to Pb reported no association between blood Pb
 3                  and baseline DNA damage or repair ability after a peroxide challenge (children attending
 4                  a school far from a Pb smelter: median blood Pb level 4.6 (ig/dL; children attending a
 5                  school near a Pb smelter: median blood Pb level 28.6 (ig/dL) (Mendez-Gomez et al.
 6                  2008). Another study included adult participants aged 50-65 years and reported an
 7                  association between blood Pb and carcinoembryonic antigen (CEA) but not with DNA-
 8                  strand breaks, MN frequency, or oxidative DNA damage (median blood Pb level of the
 9                  study population: 3.92 (ig/dL) (De Coster et al., 2008). A study conducted among
10                  workers exposed to Pb (mean blood Pb level: 30.3 (ig/dL) and unexposed controls (mean
11                  blood Pb level: 3.2 (ig/dL) reported greater cytogenetic damage (measured by MN
12                  frequency), chromosomal aberrations, and DNA damage in the Pb-exposed group
13                  (although this was not statistically significant in linear regression models controlling for
14                  age) (Grover et al., 2010). A study of painters in India, where Pb concentrations in paint
15                  are high, reported a mean (SD) blood Pb level of 21.56 (6.43) (ig/dL among painters who
16                  reported painting houses for 8-9 hours/day for 5-10 years (Khan et al., 2010b); the mean
17                  (SD) blood Pb level was 2.84 (0.96) (ig/dL for healthy workers who had not been
18                  occupationally exposed  to Pb. Cytogenetic damage was higher among the painters
19                  compared to the healthy controls. Another study compared the blood Pb of metal workers
20                  and office workers and reported higher blood Pb levels (both current and 2 year average)
21                  among the metal workers (blood Pb level > 20 (ig/dL) compared to the office workers
22                  (blood Pb level <10 (ig/dL) (Olewinska et al., 2010). Overall, the workers had increased
23                  DNA strand breaks versus the office workers (this held true at various blood Pb levels).
24                  Finally, a study of Pb battery workers with symptoms of Pb toxicity and a group of
25                  controls were examined (Shaik and Jamil. 2009). Higher chromosomal aberrations, MN
26                  frequency, and DNA damage were reported for the battery workers as compared to the
27                  controls.  These workplace studies are limited by the lack of consideration for potential
28                  confounding factors, including other occupational exposures.


                    lexicological Evidence for DNA and Cellular Damage

                        Sister Chromatic/ Exchanges
29                  Pb has been  shown to induce SCEs both in vivo and in vitro. Tapisso et al. (2009).
30                  considered SCEs  in adult Algerian mice (groups of six mice each) that were treated by
31                  i.p. injection with 5 or 10 doses of 0.46 mg/kg Pb acetate. The SCE in bone marrow were
32                  elevated after Pb exposure alone and increased with time. Co-exposure with Cd or Zn
33                  further increased  SCE levels.
      November 2012                             5-687                 Draft - Do Not Cite or Quote

-------
 1                   SCE was also followed in cultured human cells. Ustundag and Duydu (2007) considered
 2                   the ability of N-acetylcysteine and melatonin to reduce Pb nitrate-induced SCE in a
 3                   single human donor. Cells were treated with 0, 1, 5, 10, or 50 (iM Pb nitrate. SCE
 4                   statistically significantly increased at every Pb concentration in a concentration
 5                   dependent manner. Both 1 and 2 mM N-acetylcysteine and melatonin were able to
 6                   statistically significantly reduce SCE levels in Pb-exposed cells. In another study, Turkez
 7                   et al. (2011) considered the ability of boron compounds, essential micronutrients, to
 8                   prevent Pb chloride-induced SCE in human lymphocytes. Cells were obtained from 4
 9                   non-smoking donors. Both 3 and 5 ppm Pb chloride induced a statistically significant
10                   increase in SCE levels over controls. Boron was able to statistically significantly
11                   diminish these levels. For both studies, exposure times were not provided, and the full
12                   interpretation of these data is limited by the limited number of donors and the absence of
13                   an exposure time for the SCE assay.

                         Micronuclei Formation
14                   The 2006 Pb AQCD stated "studies of genotoxicity consistently find associations of Pb
15                   exposure with DNA damage and MN formation" and recent studies continue to report
16                   these associations. Alghazal et al. (2008b) considered the ability of Pb acetate trihydrate
17                   to induce MN in bone marrow of adult Wistar rats. Animals were given a daily dose of
18                   100 mg/L in their drinking water for 125 days. The mean number of MN in male and
19                   female rats was statistically significantly higher in Pb-exposed animals than in unexposed
20                   controls. Tapisso et al. (2009) considered Pb-induced MN in rodents. Algerian mice were
21                   treated by i.p. injection with 5 or 10 doses of 0.46 mg/kg Pb acetate and compared to
22                   untreated controls. The MN in bone marrow were elevated after Pb exposure and
23                   increased with time

24                   MN formation has also been followed  in cultured human cells. Ustundag and Duydu
25                   (2007) considered the ability of N-acetylcysteine and melatonin to reduce Pb nitrate-
26                   induced MN in a single human donor.  Cells were treated with 0,  1,5, 10, or 50 (iM
27                   Pb nitrate. MN formation statistically significantly increased at the two highest Pb
28                   concentrations in a concentration-dependent manner. Both 1 and 2 mM N-acetylcysteine
29                   and melatonin were not able to statistically significantly reduce MN  levels. In another
30                   study, Turkez et al.  considered the ability of boron compounds to prevent Pb chloride-
31                   induced MN in human lymphocytes. Cells were obtained from 4 non-smoking donors.
32                   Both 3 and 5 ppm Pb chloride induced a statistically significant increase in MN levels
33                   over controls. Boron induced a statistically significant attenuation of these Pb-induced
34                   levels. For both studies, exposure times were not provided, and the full interpretation of
35                   these data is limited by the limited number of donors and the absence of an exposure time
36                   for the MN assay. Gastaldo et al. (2007) evaluated the ability of Pb to induce MN.
      November 2012                             5-688                  Draft - Do Not Cite or Quote

-------
 1                  Human endothelial HMEC cell line was treated with 1-1,000 (iM Pb nitrate for 24 hours.
 2                  MN increased in a statistically significant, concentration-dependent manner.

                        Hypoxanthine-guanine phosphoribosyltransferase Mutations
 3                  The potential mutagenicity of Pb in human or animal cells has been evaluated by
 4                  monitoring mutations at the hypoxanthine-guanine phosphoribosytransferase (HPRT)
 5                  locus. Li et al. (2008a) evaluated Pb acetate-induced HPRT in the non-small-cell lung
 6                  carcinoma tumor cell line, CL3, and in normal human diploid fibroblasts (specific tissue
 7                  source not reported). All cells were exposed to 0, 100, 300 or 500 (iM Pb acetate for 24
 8                  hours in serum-free medium ± a 1-hour pretreatment with a MKK1/2 inhibitor or a
 9                  PKC-alpha inhibitor. Pb alone did not induce HPRT mutations.  Inhibiting the ERK
10                  pathway via either inhibitor statistically significantly increased Pb-induced mutagenesis.
11                  Wang et al. (2008c_), investigated Pb acetate -induced HPRT mutations in CL3 cells. All
12                  cells were exposed to 0, 100, 300 or 500 (iM Pb acetate for 24 hours in serum-free
13                  medium ± a 1-hour pretreatment with a PKC-alpha inhibitor or siRNA for PKC-alpha. Pb
14                  alone  did not induce HPRT mutations.  Inhibiting PKC-alpha via either inhibitor
15                  statistically significantly increased Pb-induced mutagenesis. McNeill et al. (2007)
16                  examined Pb acetate induced HPRT mutations in Chinese hamster ovary AA8 cells and
17                  AA8 cells overexpressing human Apel. Cells were treated with 5 (iM Pb acetate for 6
18                  hours. No increases in HPRT mutations were observed after Pb  exposure in either cell
19                  line but with specific pathway perturbations (PKC-alpha or ERK), Pb was able to induce
20                  HPRT mutations.

                        Chromosomal Aberrations
21                  Chromosomal aberrations, an indicator of cancer risk, were followed in Pb-exposed
22                  rodents (El-Ashmawv et al..  2006). Dietary exposure to Pb acetate administered as a
23                  single dose of 5,000 ppm w/w to adult  male Swiss albino mice caused statistically
24                  significant increased levels of chromosomal aberrations in the Pb treatment alone group,
25                  particularly with respect to fragments, deletions, ring chromosomes, gaps, and end-to-end
26                  associations. In addition, the authors found turmeric and myrrh powders were protective.
27                  Concerns with the study include the use of only a single dose of Pb acetate along with the
28                  high levels of unusual aberrations such as ring chromosomes and end-to-end associations.
29                  Typically, these aberrations are rare after metal exposure, but were the most commonly
30                  observed aberration in this study raising questions about the quality of the metaphase
31                  preparations. An additional concern was that only 50 metaphases per dose were analyzed
32                  instead of the more common 100 metaphases per dose. The authors did not explain why
33                  their spectrum of aberrations was so different, why they only used one dose, or analyzed
34                  fewer metaphases per dose.
      November 2012                             5-689                  Draft - Do Not Cite or Quote

-------
 1                  Multiple studies considered the ability of Pb to induce chromosomal aberrations in
 2                  cultured human cells. The ability of Pb nitrate to induce chromosomal aberrations was
 3                  examined in primary human peripheral blood lymphocytes obtained from healthy,
 4                  nonsmoking donors (Pasha Shaik et al.. 2006). Cells were treated with 0, 1.2 or 2 mM
 5                  Pb nitrate for 2 hours. No increase in chromosomal aberrations was reported. Some
 6                  aneuploidy was observed. Concerns with the study are that only a 2-hour exposure was
 7                  used, which may not be long enough  for DNA damage to be expressed as a chromosomal
 8                  aberration.  It also appears from the data presentation that only three subjects were used;
 9                  one for a control, one for the low dose and one for the high dose. Experiments were not
10                  repeated, thus given the small number of subjects, this study may not have had sufficient
11                  power to detect any effects. Holmes et al. (2QQ6a), treated WHTBF-6 cells (hTERT-
12                  immortalized human lung cells) with 0, 0.1, 0.5, or 1 (ig/cm2 Pb chromate for 24-120
13                  hours or with 0, 0.1, 0.5, 1, 5 or 10 (ig/cm2 Pb oxide for 24 or 120 hours. Pb chromate
14                  induced statistically significant, concentration-dependent increases in centrosome
15                  abnormalities and aneuploidy. Wise et al. (2006a) treated BEP2D cells with 0, 0.5, 1, 5,
16                  or 10 (ig/cm2 Pb chromate for 24 hours. Pb chromate induced statistically significant
17                  concentration-dependent increases in chromosomal aberrations. Holmes et al. (2006b),
18                  treated WHTBF-6 cells with 0,  0.1, 0.5, or 1 (ig/cm2 Pb chromate for 24-72 hours.
19                  Pb chromate induced statistically significant, concentration-dependent increases in
20                  chromosomal aberrations. The effects of the chromate anion cannot be ruled out as
21                  causative in inducing these chromosomal aberrations. Wise et al. (2006b), treated
22                  WHTBF-6  cells with 0, 0.1, 0.5, or 1  (ig/cm2 Pb chromate for 24-120 hours. Pb chromate
23                  induced statistically significant, concentration-dependent increases in spindle assembly
24                  and checkpoint disruption, effects of mitosis and aneuploidy. By contrast, chromate-free
25                  Pb oxide did not induce centrosome amplification. The effects were likely attributable to
26                  the chromate anion. Xie et al. (2007)  treated BEP2D cells with 0, 1, 5, or 10 (ig/cm2
27                  Pb chromate for 24 hours. Pb chromate induced statistically significant,  concentration-
28                  dependent increases in chromosomal  aberrations and aneuploidy. Wise et al. (2010)
29                  treated WHTBF-6 cells with 0,  0.1, 0.5, or 1 (ig/cm2 Pb chromate for 24 hours in a study
30                  comparing  4 chromate compounds. Pb chromate induced statistically significant,
31                  concentration-dependent increases in chromosomal aberrations.

32                  Multiple investigators considered the ability of Pb chromate to induce chromosome
33                  aberrations in rodent cell cultures. Grlickova-Duzevik et al. (2006) treated Chinese
34                  hamster ovary (CHO) cells with 0, 0.1, 0.5, or 1  (ig/cm2 Pb chromate for 24 hours.
35                  Specific CHO lines used included AA8 (wildtype) EM9 (XRCC1-deficient), and H9T3
36                  (EM9 complemented with human XRCC1 gene). Pb chromate induced statistically
37                  significant, concentration-dependent increases in chromosomal aberrations that were
38                  statistically significantly increased by XRCC1 deficiency. Nestmann and Zhang (2007)
39                  treated Chinese hamster ovary cells (clone WB(L)) with 0, 0.1, 0.5, 1, 5, or 10 (ig/cm2

      November 2012                            5-690                  Draft - Do Not Cite or Quote

-------
 1                  Pb chromate (as pigment yellow) for 18 hours. No increases in chromosomal aberrations
 2                  were observed. Savery et al. (2007) treated CHO cells with 0, 0.1, 0.5, 1, or 5 (ig/cm2
 3                  Pb chromate for 24 hours. Specific CHO lines used included AA8 (wildtype), KO40
 4                  (,Fa«cg-deficient), and 40BP6 (.Fawcg-complemented). The Fancg gene plays an
 5                  important role in cellular resistance to DNA interstrand crosslinks, protecting against
 6                  genetic instability. Pb chromate induced statistically significant, concentration-dependent
 7                  increases in chromosomal aberrations that were increased by ^a«cg-deficiency. Camyre
 8                  et al. (2007) treated CHO cells with  0, 0.1, 0.5,  1, 5, or 10 (ig/cm2 Pb chromate for 24
 9                  hours. Specific CHO lines used included CHO-K1 (parental), xrs-6 (Ku80 deficient), and
10                  2E (xrs-6 complemented with Ku80 gene). Pb chromate induced statistically significant,
11                  concentration-dependent increases in chromosomal aberrations that were not affected by
12                  Ku80 deficiency. Ku80 is a gene involved in nonhomologous end-joining repair and its
13                  absence can contribute to genetic instability. Stackpole et al.  (2007) treated CHO and
14                  Chinese hamster lung (CHL) cells with  0, 0.1, 0.5, or 1 (ig/cm2 Pb chromate for 24 hours.
15                  Specific CHO lines used included AA8  (wildtype), irs 1SF (XRCC3-deficient), and
16                  ISFwtS (XRCC3 complemented). XRCC3 is DNA repair enzyme involved in
17                  homologous recombination. CHL lines used included V79 (wildtype), irs3 (Rad51C
18                  deficient) and irs3#6 (Rad51C complemented).  Rad51C is a gene that encodes strand-
19                  transfer proteins that are thought to be involved in recombinational repair of damaged
20                  DNA and in meiotic recombination. Pb  chromate induced statistically significant,
21                  concentration-dependent increases in chromosomal aberrations that were statistically
22                  significantly increased by both XRCC3  and Rad51C deficiency.

23                  Multiple studies considered the ability of Pb chromate to induce chromosome aberrations
24                  in marine mammal cell cultures. Li Chen et al. (2009)  treated primary North Atlantic
25                  right whale lung and skin fibroblasts with 0, 0.5, 1.0, 2.0, and 4.0 (ig/cm2 Pb chromate for
26                  24 hours. Wise et al. (2009) treated primary Steller sea lion lung fibroblasts with 0, 0.1,
27                  0.5, 1 and 5 (ig/cm2 Pb chromate for 24 hours. Wise et al. (2011) treated primary sperm
28                  whale skin fibroblasts with 0, 0.5, 1,3,5, and 10 (ig/cm2 Pb chromate for 24 hours.  In all
29                  three studies, Pb chromate induced statistically significant, concentration-dependent
30                  increases in chromosomal aberrations.

31                  In summary, exposure  of various cell models and an in vivo model to Pb (acetate,
32                  chromate, or nitrate) induced significant increases in chromosomal aberration that often
33                  responded in a concentration dependent manner. The use of various cell lines deficient in
34                  specific DNA repair enzymes helped to elucidate which pathways may be most sensitive
35                  to Pb-dependent chromosomal aberration. However, a number of studies used
36                  Pb chromate exposures and the effects of the chromate anion cannot be ruled out as
37                  causative in inducing these chromosomal aberrations.
      November 2012                            5-691                  Draft - Do Not Cite or Quote

-------
                        COMETAssay
 I                  Multiple studies considered the ability of Pb to induce DNA single strand breaks in
 2                  laboratory animals and human and animal cells using the comet assays. The COMET
 3                  assay measures DNA damage assessed by single cell electrophoresis of a lysed cell and
 4                  measurement of the fragmented DNA or tail length. Xu et al. (2008) examined DNA
 5                  damage in male ICR mice treated with Pb acetate. Animals (5 per group) were given
 6                  Pb acetate by gavage at doses of 0, 10, 50, or 100 mg/kg body weight every other day for
 7                  4 weeks. Pb exposure statistically significantly increased both tail length and tail moment
 8                  in a dose-dependent manner. Nava-Hernandez et al. (2009) considered the ability of
 9                  Pb acetate to induce DNA damage in primary spermatocyte DNA of male Wistar rats.
10                  Animals (3 per group) were treated for 13 weeks with 0, 250, or 500 mg/L Pb in their
11                  drinking water. There was statistically significantly less DNA damage in the controls
12                  compared to the two treatment groups. Narayana and Al-Bader (2011) examined DNA
13                  damage in liver tissue of adult male Wistar rats exposed to Pb nitrate. Animals (8 per
14                  group) were treated for 60 days with doses of 0, 5,000, or 10,000 ppm Pb nitrate in their
15                  drinking water. There were no statistical differences between treated groups and controls.
16                  Drosophila melanogaster larvae (72 hours old) exposed to Pb nitrate (2,000, 4,000, and
17                  8,000 (iM in culture media for 24 hours) yielded haemocytes that tested positive in the
18                  comet assay; Pb chloride (8,000 (iM) did not cause DNA damage with the comet assay
19                  (Carmona et al.. 2011).

20                  Other studies used the COMET assay in cultured human cells. Pasha Shaik et al. (2006)
21                  treated primary human peripheral blood lymphocytes obtained from healthy, nonsmoking
22                  donors with 0, 2.1, 2.4, 2.7, 3.0, 3.3* 103 (JVI Pb nitrate for 2 hours and found dose-
23                  dependent increases in  Comet tail length. Concerns with the study are that apparently no
24                  negative control was used. It also appears from the data presentation that only five
25                  subjects were used; one for each dose. Experiments were not repeated. Thus, given the
26                  small number of subjects and the absence of a negative control, this study may only be
27                  detecting background levels of DNA damage. Xie et al. (2008) treated BJhTERT cells
28                  (hTERT-immortalized human skin fibroblasts) and ATLD-2 cells (hTERT-immortalized
29                  human skin fibroblasts deficient in Mrel 1) with 0, 0.1, 0.5, and  1 (ig/cm2 Pb chromate  for
30                  24 hours. Mrel 1 is a component of the MRN complex and plays a role in telomere
31                  maintenance and double-strand break repair. Pb chromate induced a concentration-
32                  dependent increase in DNA double strand breaks measured by the comet assay.
33                  Pb chromate exposure and the effects of the chromate anion cannot be ruled out as
34                  causative in inducing these aberrations. In another study,  Pb nitrate exposure (30 (ig/mL)
35                  induced statistically significant increased DNA damage in human liver HepG2 cells that
36                  was attenuated with co-exposure with the antioxidant NAC (500 \\M) (Yedjou et al..
37                  2010).
      November 2012                            5-692                 Draft - Do Not Cite or Quote

-------
 1                  Other studies used the comet assay to examine Pb-induced DNA single strand breaks in
 2                  rodent cell cultures. Xu et al. (2006). treated PC12 cells with 0, 0.1, 1 or 10 (iM
 3                  Pb acetate. Both tail length and tail moment statistically significantly increased in a
 4                  concentration-dependent manner. Kermani et al. (2008) exposed mouse bone marrow-
 5                  mesenchymal stem cells to 60 (iM Pb acetate for 48 hours. There was an increase in
 6                  several comet assay measurements including tail length.

 7                  The COMET assay showed multiple positive findings after Pb exposure in rodents, flies,
 8                  primary human cells,  and cell lines. In vivo studies with rodents exposed to Pb acetate
 9                  yielded significant increases in tail length and moment via COMET assays in separate
10                  studies that used lymphocytes and sperm. In drosophila, Pb nitrate but not Pb chloride
11                  produced significant increases with the COMET assay. Human cell culture from primary
12                  cells (lymphocytes) and from cell lines (fibroblasts and liver) produced positive COMET
13                  assays with separate Pb nitrate and Pb chromate exposures. Thus, the COMET assay
14                  showed multiple positive findings of DNA damage after in vitro and in vivo Pb exposure.

                        Other Indicators of DNA Damage
15                  Other studies considered the ability of Pb to induce DNA double strand breaks by
16                  measuring gamma-H2A.X foci formation in cultured human cells. Xie et al. (2008)
17                  treated BJhTERT cells (hTERT-immortalized human skin fibroblasts) and ATLD-2 cells
18                  (hTERT-immortalized human skin fibroblasts deficient in Mrel 1) with 0, 0.1, 0.5, and
19                  1 (ig/cm2 Pb chromate for 24 hours. Pb chromate induced a concentration-dependent
20                  increase in DNA double strand breaks measured by gamma-H2A.X foci formation.
21                  Pb chromate exposure and the effects of the chromate anion cannot be ruled out as
22                  contributory. Gastaldo et al. (2007) evaluated the ability of Pb to induce DNA double
23                  strand breaks with both gamma-H2A.X foci formation and pulse-field gel electrophoresis
24                  in cultured human cells. The human  endothelial HMEC cell line was treated with 1 to
25                  1,000 (iM Pb nitrate for 24 hours. DNA double strand breaks increased in a
26                  concentration-dependent manner. Wise et al. (2010) treated WHTBF-6 cells with 0, 0.1,
27                  0.5, or 1 (ig/cm2 Pb chromate for 24  hours in a study comparing four chromate
28                  compounds.  Pb chromate induced statistically significant, concentration-dependent
29                  increases in DNA double strand breaks measured by gamma-H2A.X foci formation, at a
30                  similar level to the three other compounds. A few studies demonstrated the ability  of Pb
31                  to destabilize DNA by forming DNA-histone cross links, which can lead to histone
32                  aggregation (Rabbani-Chadegani et al.. 2011; Rabbani-Chadegani et al.. 2009). In
33                  extracts of rat liver, Pb nitrate (<300 (iM) was shown to react with chromatin components
34                  and induce chromatin aggregation via histone-DNA cross links.

35                  Genotoxicity testing of Drosophila melanogaster larvae (72 hours old) using the Wing
36                  Spot test showed that neither Pb chloride nor Pb nitrate (at concentrations of 2,000, 4,000


      November 2012                            5-693                  Draft - Do Not Cite or Quote

-------
 1                  and 8,000 (iM in culture media with exposure until pupation) was able to induce
 2                  significant increases in the frequency of wing spots (Carmona et al.. 2011). The wing
 3                  spot test can detect mitotic recombination and multiple mutational events such as point
 4                  mutations, deletions, and certain types of chromosome aberrations (Graf and Wurgler.
 5                  1986). Further, wing spot assays employing Pb co-exposure with gamma radiation
 6                  showed no effect of Pb on gamma radiation induced spotting frequency.

 7                  Multiple studies examined the effects of Pb on DNA repair. Most were conducted in
 8                  cultured cells, and one was done in an animal model. El-Ghor et al. (2011) followed
 9                  microsatellite instability (MSI) in Pb acetate trihydrate exposed adult male rats. MSI
10                  reflects impaired DNA mismatch repair and contributes to an increased risk of cancer.
11                  DNA from leukocytes of adult male albino rats exposed to  Pb acetate (acute: single oral
12                  dose of 467 mg/kg BW or sub-chronic: 47 mg/kg BW six days/week for 4 week)  showed
13                  increased MSI at three microsatellite loci (D6mit3, D9mit2, and DISMghl). This study is
14                  limited by its small sample size (n=2 to 3 rodents per treatment group). Li et al. (2008a)
15                  evaluated Pb acetate-induced effects on nucleotide excision repair efficiency in CL3
16                  cells. All cells were exposed to 0,  100, 300 or 500 (iM Pb acetate for 24 hours in serum-
17                  free medium. Pb increased nucleotide excision repair efficiency. Gastaldo et al. (2007)
18                  evaluated the ability of Pb to affect DNA repair in cultured human cells. The human
19                  endothelial HMEC cell line was treated with 100 (iM Pb nitrate for 24 hours. Pb inhibited
20                  non-homologous end joining repair, over activated MRE11-dependent repair, and
21                  increased Rad51 -related repair. Xie et al. (2008) treated BJhTERT cells (hTERT-
22                  immortalized human skin fibroblasts) and ATLD-2 cells (hTERT-immortalized human
23                  skin fibroblasts deficient in Mrel 1) with 0, 0.1,  0.5, and  1 (ig/cm2 Pb chromate for 24 or
24                  120 hours. Mrel 1 was required to prevent Pb chromate-induced DNA double strand
25                  breaks. In this finding, Pb chromate exposure  and the effects of the chromate anion
26                  cannot be ruled out as causative. McNeill et al, (2007) considered Pb acetate effects on
27                  Apel. Chinese hamster ovary cells (AA8) were  treated with 0, 0.5, 5, 50, or 500 (iM
28                  Pb acetate and then whole cell extracts were used to determine AP site incision activity.
29                  The data show that Pb reduced AP endonuclease function. Finally, studies considered
30                  Pb-induced cellular proliferation in laboratory animals. Kermani et al. (2008) exposed
31                  mouse bone marrow-mesenchymal stem cells to 0-100 (iM Pb acetate for 48 hours. As
32                  measured by the MTT assay, Pb decreased cell proliferation at all concentrations tested.
33                  An earlier study in rats showed Pb nitrate-induced increased proliferation of liver cells
34                  after a partial hepatectomy, with more prominent effects found in males than females
35                  (sexual dimorphism) (Tessitore et al., 1995). Recent studies showed similar trends in
36                  males. Fortoul et al. (2005) exposed adult male CD1 mice (24 animals per group) to
37                  1* 104 (iM Pb acetate, 0.006 M Cd chloride or a  mixture of the two chemicals for  1 h
38                  twice a week for 4 weeks by inhalation. Electron microscopy indicated Pb-induced
39                  cellular proliferation in the lungs.

      November 2012                            5-694                 Draft - Do Not Cite or Quote

-------
                    5.10.3.3    Epigenetics

 1                  Air pollution exposure is being linked increasingly with epigenetic changes in humans
 2                  and toxicological models (Pavanello et al., 2010; Baccarelli and Bollati. 2009; Tarantini
 3                  et al.. 2009; Bollati et al.. 2007). Epigenetic changes are changes in DNA expression that
 4                  occur without actual changes in the DNA sequence, and these changes may be heritable.
 5                  Epigenetic changes are mediated by histone modification, DNA methylation, miRNA
 6                  changes, or pathways that affect these three mediators. Differential epigenetic
 7                  modification has the possibility to contribute to disease. Epigenetic studies have been
 8                  conducted to examine the associations between Pb biomarker levels and global DNA
 9                  methylation markers [Alu and long interspersed nuclear element-1 (LINE-1)] in humans
10                  (Wright etal.. 2010; Pilsner et al.. 2009). Wright et al. (2010) examined men from the
11                  Normative Aging Study (N=517) with mean (SD) Pb levels of 20.5(14.8)g/g in tibia, 27.4
12                  (19.7)g/g in patella, and 4.1 (2.4) (ig/dL in blood. In both crude and adjusted analyses,
13                  patella Pb levels were inversely associated with LINE-1 methylation but not with Alu.
14                  The adjusted models all  included age, BMI, percent lymphocytes, with some adjusted
15                  models also controlling for education, smoking, and blood Pb levels. In examination of
16                  the relationship between patella Pb and LINE-1 more closely, a non-linear trend was
17                  observed with a smaller magnitude of effect estimated for higher patella Pb (> 40 (ig/g).
18                  No associations were observed for tibia or blood Pb and either LINE-1 or Alu. Another
19                  study included maternal-infant pairs from the Early Life Exposures in Mexico to
20                  Environmental  Toxicants study (N=103) and measured LINE-1 and Alu methylation in
21                  umbilical cord blood samples (Pilsner et al., 2009). In unadjusted models, maternal tibia
22                  Pb levels one month postpartum (mean [SD]: 10.5 [8.4] (ig/g) were inversely associated
23                  with Alu methylation in the cord blood. Maternal patella Pb levels one month postpartum
24                  (mean [SD]:  12.9 [14.3] (ig/g) were inversely associated with LINE-1 methylation. The
25                  associations persisted  in models adjusted for maternal age, maternal education, infant sex,
26                  smoking during pregnancy, and umbilical cord blood Pb levels (the results were no
27                  longer statistically significant when umbilical cord blood was removed from the model).
28                  No association was detected between umbilical cord Pb levels and the DNA methylation
29                  markers. Overall, the studies consistently demonstrate an association between higher
30                  patella Pb levels and lower LINE-1 methylation. Lower DNA methylation is associated
31                  with increased gene expression; however, the link between global DNA methylation and
32                  risk of disease,  has not been established.

33                  Toxicological studies have examined Pb-induced epigenetic changes and gene
34                  expression, DNA repair, and mitogenesis. Glahn et al., (2008) performed a gene array
35                  study in primary normal human bronchial epithelial cells from four donors after in vitro
36                  treatment of the cells with 55 (ig/dL Pb chloride, 15 (ig/L Cd sulfate, 25 (ig/L Co chloride
37                  or all three combined for 72 hours. The authors describe a pattern of RNA expression

      November 2012                            5-695                  Draft - Do Not Cite or Quote

-------
 1                  changes indicating "... coordinated stress-response and cell-survival signaling,
 2                  deregulation of cell proliferation, increased steroid metabolism, and increased expression
 3                  of xenobiotic metabolizing enzymes." These are all known targets of possible epigenetic
 4                  changes, but attributing the results to epigenetic changes is complicated. In a recent
 5                  publication (Li etal.. 2011). exposure of HepG2 cells to a high dose of Pb (100 (iM
 6                  Pb acetate) resulted in ALAD gene promoter hypermethylation and decreased ALAD
 7                  transcription. This was in agreement with findings in battery plant workers who showed
 8                  ALAD hypermethylation (versus non-occupationally exposed controls) and an
 9                  association of this hypermethylation with elevated risk of Pb poisoning (Li et al.. 2011).
10                  These latter results have implications for Pb toxicokinetics or disposition of Pb as
11                  modified by ALAD.
      5.10.4       Effects of Pb within Mixtures

12                  Several studies considered the impact of Pb as part of a mixture on mixtures genotoxicity
13                  and mutagenesis. Mendez-Gomez et al. (2008) evaluated 65 children in Mexico with high
14                  environmental exposures to both As and Pb. DNA damage and decreased DNA repair
15                  were seen using the comet assay and other assays but did not correlate with urinary As or
16                  blood Pb levels. Tapisso et al. (2009) examined Pb alone, Pb plus Zn and Pb plus Cd-
17                  induced MN in rodents. Algerian mice (groups of six mice each) were treated i.p. with 5
18                  or 10 doses of 0.46 mg/kg Pb acetate and compared to untreated controls. The MN in
19                  bone marrow were elevated after Pb treatment alone and increased with time.
20                  Co-exposure with Cd or Zn did not further increase MN levels but did increase SCE
21                  levels. Glahn et al. (2008) performed a gene array study in primary normal human
22                  bronchial epithelial cells from four donors treated with 55 (ig/dL Pb chloride, 15 (ig/L Cd
23                  sulfate, 25 (ig/L Co chloride or all three  combined for 72 hours. There was a clear
24                  interaction of all three metals impacting  RNA expression.

25                  Studies in the 2006 Pb AQCD (U.S. EPA. 2006b) found a protective role for calcium in
26                  genotoxic and mutagenic assays with Pb co-exposure. No recent studies of the protective
27                  role of calcium in Pb-induced carcinogenesis or genotoxicity were found. There were
28                  some data suggesting that boron,  melatonin, N-acetylcysteine, turmeric and myrrh protect
29                  cells against Pb-induced genotoxicity (Section 5.10.3.2).

30                  A recent study details Pb and Se interactions in virus-dependent carcinogenesis in
31                  laboratory animals. Schrauzer (2008) considered the impact of Se on carcinogenesis by
32                  studying four groups of weanling virgin  female C3H/St mice infected with murine
33                  mammary tumor virus (groups of 20-30 mice), which induces mammary tumor
34                  formation. One set of two groups were fed a diet containing 0.15 ppm  Se and then were
      November 2012                            5-696                 Draft - Do Not Cite or Quote

-------
 1                  exposed via drinking water to acetic acid (control group) or 0.5 ppm Pb acetate. The
 2                  second set of two groups were fed a diet containing 0.65 ppm Se and then similarly
 3                  exposed to acetic acid or 0.5 ppm Pb acetate. The study was primarily focused on the
 4                  general effects of a low Se diet. The data suggest that Se is anticarcinogenic as in the
 5                  groups without Pb exposure, the animals exposed to the higher Se levels had fewer
 6                  mammary tumors and these tumors had a delayed onset of appearance. Pb exposure with
 7                  low Se caused the same delayed onset as did the higher dose of Se and also caused some
 8                  reduction in the tumor frequency. Pb exposure with higher Se increased the tumor
 9                  frequency and the onset of the tumors. Pb also induced weight loss at 14 months in both
10                  exposed groups. The data suggest that there may be interactions of Pb and Se, but they
11                  suggest that Pb mimics or antagonizes Se. They do not suggest that Se is protective of
12                  Pb-induced toxicity or carcinogenesis.

13                  In summary, the new data on Pb exposure as part of a mixture  is derived from studies
14                  designed with co-exposure to metals or antioxidants. Children  in Mexico with
15                  co-exposure to high levels of Pb and As showed elevated DNA damage and impaired
16                  DNA repair. Pb and Cd co-exposure in mice elevated SCE levels but did not further
17                  exacerbate MN levels above Pb exposure alone. Primary lung cells exposed to a metals
18                  mixture showed an interaction at the mRNA level among the three metals tested. In other
19                  genotoxicity assays, various antioxidants (melatonin, NAC, turmeric and myrrh) and
20                  metals (boron) were protective against Pb-induced genotoxicity. In an animal model of
21                  breast cancer, Se modified the onset and multiplicity of murine mammary tumor virus-
22                  induced tumorogenicity in Pb-exposed animals. These data show that co-exposure of Pb
23                  with antioxidants or metals, modifies the effect of Pb on DNA damage, DNA repair,
24                  mutagenicity, genotoxicity, or tumorogenicity.
      5.10.5      Summary and Causal Determination

25                  Toxicological and epidemiologic studies of the association between Pb exposure and
26                  cancer and cancer-related outcomes have been reviewed in the preceding sections.
27                  Evaluation of the relationship between Pb exposure and cancer with respect to causality
28                  was based on evidence for tumor incidence in experimental animals, associations of Pb
29                  exposure with cancer incidence and mortality in humans, and evidence describing
30                  potential modes of action including mutagenesis, clastogenesis, and epigenetic changes.
31                  The application of the key supporting evidence from these studies to the causal
32                  framework is summarized in Table 5-50 and the following text.
      November 2012                             5-697                 Draft - Do Not Cite or Quote

-------
 1                  The toxicological literature provides consistent evidence of the carcinogenic potential of
 2                  Pb and possible contributing modes of action, including genotoxic, mutagenic and
 3                  epigenetic effects. In laboratory studies, chronic Pb exposure for 18 months or two years
 4                  to high concentrations such as 10,000 ppm Pb acetate in diet or 2,600 ppm Pb acetate in
 5                  drinking water has been demonstrated to be an animal carcinogen. Chronic Pb exposure
 6                  to male and female rodents has consistently induced kidney and brain carcinogenesis in
 7                  multiple separate studies, inducing various tumors, (i.e., adenocarcinomas, adenomas,
 8                  and gliomas. Pb has also been shown to cause mammary gland, prostate, adrenal, and
 9                  testicular tumors in animals. Developmental Pb acetate exposure also induced tumors in
10                  offspring whose dams received Pb acetate in drinking water during pregnancy and
11                  lactation. Multiple toxicological studies showed neoplastic transformation in cultured
12                  cells providing an additional potential mode of action, but most used Pb chromate, and it
13                  is possible that the chromate ion contributed to these findings. The toxicological and
14                  epidemiologic literature provides evidence for potential carcinogenic modes of action
15                  from genotoxic, mutagenic and epigenetic assays. Multiple longitudinal epidemiologic
16                  studies have been performed examining the association between cancer incidence and
17                  mortality and Pb exposures, estimated with biological measures and exposure databases.
18                  Mixed results have been reported for cancer mortality studies; a large NHANES
19                  epidemiologic study demonstrated a positive association between blood Pb and cancer
20                  mortality with median 8.6 years of follow up on subjects (Schober et al., 2006). but the
21                  other studies reported null results (Khalil et al.. 2009b: Weisskopf et al.. 2009; Menke et
22                  al., 2006). These were well-conducted epidemiologic studies with control for important
23                  potential confounders such as age, smoking, and education. Although the 2006 Pb AQCD
24                  (U.S. EPA. 2006b) reported some studies that found an association between Pb exposure
25                  indicators and lung cancer, recent studies mostly included occupationally-exposed adults
26                  and observed no associations. Most studies  of Pb and brain cancer were null among the
27                  overall study population, but positive associations were observed among individuals with
28                  certain genetic variants. However, the studies of Pb and brain cancer were all limited by
29                  the use of occupational cohorts and interviews instead of biological measurements to
30                  represent Pb exposure,  and by possible confounding by several factors, including other
31                  workplace exposures. A limited amount of research has been performed on other types of
32                  cancer. The 2006 Pb AQCD reported evidence that suggested an association between Pb
33                  exposure and stomach cancer, but in a recent study of stomach cancer the results were
34                  inconsistent, reporting a positive association between organic Pb exposure and stomach
35                  cancer but null findings for exposure to inorganic Pb or Pb from gasoline emissions and
36                  stomach cancer.

37                  Among epidemiologic studies, high Pb levels (over 40 (ig/dL in adults) were associated
38                  with SCEs among adults. This association was not observed among children (mean blood
39                  Pb 7.69 (ig/dL). Other epidemiologic studies of DNA damage reported inconsistent

      November 2012                            5-698                 Draft - Do Not Cite or Quote

-------
 1                   results. Consistent with previous toxicological findings, Pb does appear to have genotoxic
 2                   activity in animal and in vitro models, inducing SCE, MN and DNA strand breaks. The
 3                   majority of the chromosomal aberration studies with Pb-induced significant finding used
 4                   Pb chromate exposure and the aberrations are likely due to the chromate. Pb does not
 5                   appear to be  very mutagenic as the HPRT assays were typically negative unless a cell
 6                   signaling pathway was disturbed.

 7                   Mechanistic  understanding of the carcinogenicity of Pb in toxicological models is
 8                   expanding with work on the antioxidant Se and metallothionein, a protein that binds Pb
 9                   and reduces its bioavailability. Low Se diet affects tumorigenesis and tumor multiplicity
10                   with Pb exposure. Metallothionein has been shown to be protective against the effect of
11                   Pb on carcinogenicity. Pb is clastogenic and mutagenic in some but not all models.
12                   Clastogenicity and mutagenicity may be possible mechanisms contributing to cancer but
13                   are not absolutely associated with the induction of cancer. Because Pb has a higher
14                   atomic weight than does Zn, Pb replaces Zn at many Zn binding sites or Zn finger
15                   proteins. This substitution has the potential to induce effects that can indirectly contribute
16                   to carcinogenicity via interactions with hormone receptors, cell-cycle regulatory proteins,
17                   tumor suppressor genes like p53, DNA repair enzymes, histones, etc. These indirect
18                   effects may act at a post-translational level to negatively alter protein structure and DNA
19                   repair.

20                   Epigenetic changes associated with Pb exposure or biological markers, particularly,
21                   methylation and effects on DNA repair, are beginning to appear in the literature.
22                   Epigenetic modifications may contribute to carcinogenicity by altering DNA repair or
23                   changing the expression of a tumor suppressor gene or oncogene. A small number of
24                   epidemiologic studies examining Pb  and global epigenetic changes demonstrated an
25                   inverse association between bone Pb and LINE-1 or Alu methylation. Lower DNA
26                   methylation is associated with increased gene expression, but epigenetic contributions to
27                   cancer are not yet fully characterized in this emerging  area of research. Toxicological
28                   studies show that Pb can activate or interfere with a number of signaling and repair
29                   pathways, though it is unclear whether these are due to epigenetic responses or
30                   genotoxicity. Thus, an underlying mechanism is still uncertain, but likely involves
31                   genomic instability, epigenetic modifications, or both.

32                   In conclusion, the toxicological literature provides the strong evidence  for long-term
33                   exposure (i.e., 18 months or 2 years) to high concentrations of Pb (> 2,600 ppm) and
34                   cancer. The consistent evidence indicating Pb-induced carcinogenicity  in animal models
35                   is substantiated by the mode of action findings from multiple high-quality toxicological
36                   studies in animal and in vitro models from different laboratories. This is substantiated  by
37                   the findings of other agencies including IARC, which has classified inorganic Pb
      November 2012                             5-699                  Draft - Do Not Cite or Quote

-------
 1                  compounds as a probable human carcinogen and the National Toxicology Program,
 2                  which has listed Pb and Pb compounds as "reasonably anticipated to be human
 3                  carcinogens." Strong evidence from animal toxicological studies demonstrates an
 4                  association between Pb and cancer, genotoxicity, mutagenicity or epigenetic
 5                  modification. Carcinogenicity in animal toxicology studies with relevant routes of Pb
 6                  exposure has been reported in the kidneys, testes, brain, adrenals, prostate, pituitary, and
 7                  mammary gland, albeit at high doses of Pb. Epidemiologic studies of cancer incidence
 8                  and mortality reported inconsistent results; one strong epidemiologic study demonstrated
 9                  an association between blood Pb and increased cancer mortality, but the other studies
10                  reported weak or no associations. In the 2006 Pb AQCD, indicators of Pb exposure were
11                  found to be associated with stomach cancer, and a recent study on stomach cancer and
12                  occupational Pb exposure, reported mixed findings depending on the type of Pb exposure
13                  (organic Pb, inorganic Pb, or Pb from gasoline emissions). Similarly, some studies in the
14                  2006 Pb AQCD reported associations between Pb exposure indicators and lung cancer.
15                  Recent epidemiologic studies of lung cancer focused on occupational exposures and
16                  reported inconsistent associations. The majority of epidemiologic studies of brain cancer
17                  had null results overall, but positive associations between Pb exposure indicators and
18                  brain cancer were observed among individuals with certain genotypes. Overall, the
19                  consistent and strong body of evidence from toxicological studies but inconsistent
20                  epidemiologic evidence is sufficient to conclude that a causal relationship is likely to
21                  exist between Pb exposure and cancer.
      November 2012                             5-700                  Draft - Do Not Cite or Quote

-------
Table 5-50     Summary of evidence supporting cancer and genotoxicity
                  causal determinations.
Attribute in
Causal
Framework3
Key Supporting Evidence13
Pb Exposure or
Blood Levels
Associated
References13 with Effects
Cancer - Likely Causal
Consistent         Consistent findings across multiple
toxicological        toxicology studies using 18 month or
results of tumors    two year cancer bioassays in rats
in laboratory        wherein rodents are fed chow or
animals with        received drinking water enriched with
chronic Pb         Pb acetate, showing tumor
exposure          development.
                                     Azaretal. (1973).
                                     Kasprzak et al. (1985).
                                     Kolleretal. (1985),
                                     Van Esch and Kroes (1969)
                                   Chronic 10,000 ppm
                                   Pb acetate diet or
                                   2,600 ppm drinking water
                                   Pb acetate, no blood Pb
                                   measurement available.
Toxicological
studies of early
life Pb exposure
induced tumor
formation in
adulthood
Gestational and lactational Pb exposure
induced carcinogenicity in adult
offspring
Waalkesetal., (1995)
500, 750 and 1,000 ppm
Pb in drinking water, no
blood Pb measurement
available.
Limited and        Epidemiologic studies of overall cancer
inconsistent        mortality have inconsistent findings.
epidemiologic      These are high-quality, longitudinal
studies            studies with control for confounders,
                  such as age, smoking, and education.
                  Epidemiologic studies of specific cancer
                  sites were limited. Recent studies were
                  not consistent with previous findings of
                  possible associations for lung and
                  stomach cancers reported in the
                  2006 PbAQCD. Many of the
                  epidemiologic studies examining
                  specific cancer sites were case-control
                  studies and not all included potentially
                  important confounders, such as
                  smoking.
                                     Menke et al. (2006)
                                     Schoberetal. (2006)
                                     Weisskopfetal. (2009),
                                     Khalil (2009b)
                                     Overall Cancer Mortality:
                                     See Section 5.10.1.1
                                     Specific Cancer: See Sections:
                                     5.10.1.3 (Lung), 5.10.1.4 (Brain),
                                     5.10.1.5 (Breast): and
                                     Section 5.10.1.6 (Other cancers).
Evidence clearly
supports mode of
action

  Mutagenic,
  carcinogenic
  and genotoxic
  assays provide
  consistent
  support to the
  MOA

  Clastogenic
  assays provide
  inconsistent
  support to the
  MOA

  Epigenetic
  evidence
  provides
  support to the
  MOA
Consistent evidence of toxicological
findings of mutagenicity,
carcinogenicity, and genotoxicity has
been reported by multiple laboratories
in animals and in vitro models using
multiple assays (MN, SCE, COMET).
Toxicological studies of the clastogenic
effects of Pb often employ
Pb chromate. The effect of the
chromate ion in contributing to the
clastogenic effects cannot be ruled out.

Bone Pb levels were inversely
associated with LINE-1 methylation in
adult men.
Maternal pregnancy bone Pb levels
were inversely associated with Alu and
LINE-1 methylation in child cord blood.
Occupational battery workers had
ALAD hypermethylation compared with
controls; cell culture study of high dose
Pb exposure caused ALAD
hypermethylation.
Epidemiology evidence of DMA and
cellular damage:
See Section 5.10.3.2
Toxicology evidence of DMA and
cellular damage:
See Section 5.10.3.2
See Section 5.10.3.2 (Toxicological
Evidence for DMA and Cellular
Damage)
                                                       See Section 5.10.3.3
Described in detail in Table II of the Preamble.
bDescribes the key evidence and references contributing the most heavily to causal determination. References to earlier sections
indicate where full body of evidence is described.
November 2012
                                5-701
                      Draft - Do Not Cite or Quote

-------
References for Chapter 5
     Abam. E; Okediran. BS: Odukoya. OO: Adamson. I; Ademuyiwa. O. (2008). Reversal of ionoregulatory
        disruptions in occupational lead exposure by vitamin C. Environ Toxicol Pharmacol 26: 297-304.
        http://dx.doi.0rg/10.1016/i.etap.2008.05.008

     Abdel Moneim. AE; Dkhil. MA; Al-Quraishy. S. (2011 a). Effects of flaxseed oil on lead acetate-induced
        neurotoxicity in rats. Biol Trace Elem Res 144: 904-913. http://dx.doi.org/10.1007/sl2011-011-9055-4

     Abdel Moneim. AE; Dkhil. MA; Al-Quraishy. S. (20 lib). The protective effect of flaxseed oil on lead
        acetate-induced renal toxicity in rats. J Hazard Mater 194: 250-255.
        http://dx.doi.0rg/10.1016/i.jhazmat.2011.07.097

     Abdelouahab. N; Mergler. D; Takser. L; Vanier. C: St-Jean. M; Baldwin. M; Spear. PA; Chan. HM.
        (2008). Gender differences in the effects of organochlorines, mercury, and lead on thyroid hormone
        levels in lakeside communities of Quebec (Canada). Environ Res 107: 380-392.
        http://dx.doi.0rg/10.1016/i.envres.2008.01.006

     Aboul-Ela. EL (2002). The protective effect of calcium against genotoxicity of lead acetate administration
        on bone marrow and spermatocyte cells of mice in vivo. Mutat Res Genet Toxicol Environ Mutagen
        516: 1-9. http://dx.doi.org/10.1016/81383-5718(01)00332-1

     Absalon. D; Slesak. B. (2010). The effects of changes in cadmium and lead air pollution on cancer
        incidence in children. Sci Total Environ 408: 4420-4428.
        http://dx.doi.0rg/10.1016/i.scitotenv.2010.06.030

     Adegbesan. BO; Adenuga. GA. (2007). Effect of lead exposure on liver lipid peroxidative and antioxidant
        defense systems of protein-undernourished rats. Biol Trace Elem Res  116: 219-225.
        http://dx.doi.org/10.1007/BF02685932

     Ademuyiwa. O: Agarwal. R; Chandra. R; Behari. JR. (2009). Lead-induced phospholipidosis and
        cholesterogenesis in rat tissues. Chem Biol Interact 179: 314-320.
        http://dx.doi.org/10.1016/i.cbi.20Q8.10.057

     Ademuyiwa. O: Ugbaja. RN; Idumebor. F; Adebawo. O. (2005a). Plasma lipid profiles and risk of
        cardiovascular disease in occupational lead exposure in Abeokuta, Nigeria. Lipids Health Dis 4:19.
        http://dx.doi.org/10.1186/1476-511X-4-19

     Ademuyiwa. O: Ugbaja. RN; Ojo. DA; Owoigbe. AO; Adeokun. SE. (2005b). Reversal of aminolevulinic
        acid dehydratase (ALAD) inhibition and reduction of erythrocyte protoporphyrin levels by Vitamin C
        in occupational lead exposure in Abeokuta, Nigeria. Environ Toxicol Pharmacol 20: 404-411.
        http://dx.doi.0rg/10.1016/i.etap.2005.04.002

     Adonaylo. VN; Oteiza. PI. (1999).  Pb2+ promotes lipid oxidation and alterations in membrane physical
        properties. Toxicology 132: 19-32. http://dx.doi.org/10.1016/S0300-483X(98)00134-6

     Afeiche. M; Peterson. KE; Sanchez. BN;  Cantonwine. D: Lamadrid-Figueroa. H; Schnaas. L; Ettinger.
        AS: Hernandez-Avila. M; Hu. H; Tellez-Rojo. MM. (2011). Prenatal lead exposure and weight of 0 to
        5 year-old children in Mexico City. Environ Health Perspect 119: 1436-1441.
        http://dx.doi.org/10.1289/ehp.1003184

     Agarwal. S: Roy. S: Ray. A; Mazumder. S: Bhattacharya. S. (2009). Arsenic trioxide and lead acetate
        induce apoptosis in adult rat hepatic stem cells. Cell Biol Toxicol 25: 403-413.
        http://dx.doi.org/10.1007/sl0565-008-9094-6

     Aguilar. RP; Genta. S: Sanchez. S. (2008). Renal gangliosides are involved in lead intoxication. J Appl
        Toxicol 28: 122-131. http://dx.doi.org/10.1002/iat.1256
November 2012                             5-702                  Draft - Do Not Cite or Quote

-------
     Ahamed. M; Fareed. M; Kumar. A; Siddiqui. WA; Siddiqui. MKJ. (2008). Oxidative stress and
        neurological disorders in relation to blood lead levels in children. Redox Rep 13: 117-122.
        http://dx.doi.org/10.1179/135100008X259213

     Ahamed. M; Mehrotra. PK; Kumar. P; Siddiqui. MKJ. (2009). Placenta! lead-induced oxidative stress and
        preterm delivery. Environ Toxicol Pharmacol 27: 70-74. http://dx.doi.Org/10.1016/i.etap.2008.08.013

     Ahamed. M; Siddiqui. MK. (2007). Low level lead exposure and oxidative stress: Current opinions
        [Review]. Clin Chim Acta 383: 57-64. http://dx.doi.Org/10.1016/i.cca.2007.04.024

     Ahamed. M; Singh. S: Behari. JR; Kumar. A; Siddiqui. MK. (2007). Interaction of lead with some
        essential trace metals in the blood of anemic children from Lucknow, India. Clin Chim Acta 377:
        92-97. http://dx.doi.0rg/10.1016/j.cca.2006.08.032

     Ahamed. M; Verma. S: Kumar. A; Siddiqui. MKJ. (2005). Environmental exposure to lead and its
        correlation with biochemical indices in children. Sci Total Environ 346: 48-55.
        http://dx.doi.0rg/10.1016/i.scitotenv.2004.12.019

     Ahamed. M; Verma. S: Kumar. A; Siddiqui. MKJ. (2006). Delta-aminolevulinic acid dehydratase
        inhibition and oxidative stress in relation to blood lead among urban adolescents. Hum Exp Toxicol
        25: 547-553. http://dx.doi.org/10.1191/0960327106het657oa

     Aisemberg. J; Nahabedian. DE; Wider. EA; Verrengia Guerrero. NR. (2005). Comparative study on two
        freshwater invertebrates for monitoring environmental lead exposure. Toxicology 210: 45-53.
        http://dx.doi.0rg/10.1016/i.tox.2005.01.005

     Akel. A; Wagner. CA; Kovacikova. J: Kasinathan. RS: Kiedaisch. V; Koka. S: Alper. SL; Bernhardt. I;
        Wieder. T; Huber. SM; Lang. F. (2007). Enhanced suicidal death of erythrocytes from gene-targeted
        mice lacking the C1-/HCO3- exchanger AE1. Am J Physiol Cell Physiol 292: C1759-C1767.
        http://dx.doi.org/10.1152/aipcell.00158.2006

     Akesson. A; Lundh. T; Vahter. M; Bjellerup. P; Lidfeldt. J: Nerbrand. C: Goran. S: Stromberg. U;
        Skerfving. S. (2005). Tubular and glomerular kidney effects in Swedish women with low
        environmental cadmium exposure. Environ Health Perspect 113: 1627-1631.
        httrj://dx.doi.org/10.1289/ehp.8Q33

     Al-Saleh. I; Coskun. S: Mashhour. A; Shinwari. N; El-Doush. I; Billedo. G: Jaroudi. K; Al-Shahrani. A;
        Al-Kabra. M;  El Din Mohamed. G. (2008a). Exposure to heavy metals (lead, cadmium and mercury)
        and its effect on the outcome of in-vitro fertilization treatment. Int J Hyg Environ Health 211:
        560-579. http://dx.doi.0rg/10.1016/i.iiheh.2007.09.005

     Al-Saleh. I; Shinwari. N; Nester. M; Mashhour. A; Moncari. L; Mohamed. GED; Rabah. A. (2008b).
        Longitudinal study of prenatal and postnatal lead exposure and early cognitive development in Al-
        Kharj, Saudi Arabia: A preliminary results of cord blood lead levels. JTrop Pediatr 54: 300-307.
        http://dx.doi.org/10.1093/tropej/fmn019

     Alasia. DP; Emem-Chioma. PC: Wokoma. FS. (2010). Occupational and environmental lead exposure in
        Port Harcourt, Nigeria: Analysis of its association with renal function indices. Niger J Med 19:
        407-414.

     Alatise. OI; Schrauzer. GN. (2010). Lead exposure: A contributing cause of the current breast cancer
        epidemic in Nigerian women. Biol Trace Elem Res 136:  127-139.
        http://dx.doi.org/10.1007/sl2011-010-8608-2

     Alber. SA; Strupp. BJ. (1996). An in-depth analysis of lead effects in a delayed spatial alternation task:
        assessment of mnemonic effects, side bias, and proactive interference. Neurotoxicol Teratol 18: 3-15.
        http://dx.doi.org/10.1016/0892-0362(95)02026-8

     Alexander. KR; Fishman. GA. (1984). Supernormal scotopic ERG in cone dystrophy. Br J Ophthalmol
        68: 69-78.

     Alfano. DP; Petit. TL. (1981). Behavioral effects of postnatal lead exposure: possible relationship to
        hippocampal dysfunction.  Behav Neural Biol 32: 319-333.
November 2012                             5-703                   Draft - Do Not Cite or Quote

-------
     Alghazal. MA; Lenartova. V: Holovska. K; Sobekova. A; Falls. M; Legatfa. J. (2008a). Activities of
        antioxidant and detoxifying enzymes in rats after lead exposure. Acta Vet (Beogr) 77: 347-354.
        http://dx.doi.org/10.2754/avb200877030347

     Alghazal. MA; Sutiakova. I; Kovalkovicova. N; Legath. J; Palis. M; Pistl. J: Sabo. R; Benova. K; Sabova.
        L; Vaczi. P. (2008b). Induction of micronuclei in rat bone marrow after chronic exposure to lead
        acetate trihydrate. Toxicol Ind Health 24: 587-593. http://dx.doi.org/10.1177/0748233708100089

     Aligne. CA; Auinger. P; Byrd. RS: Weitzman. M. (2000). Risk factors for pediatric asthma: Contributions
        of poverty, race, and urban residence. Am J Respir Crit Care Med 162:  873-877.

     Alimonti. A; Bocca. B; Lamazza. A; Forte. G: Rahimi. S: Mattei. D; Fiori. E; lacomino. M; Schillaci. A;
        De Masi. E; Pino. A. (2008). A study on metals content in patients with colorectal polyps. J Toxicol
        Environ Health A 71: 342-347. http://dx.doi.org/10.1080/15287390701839133

     Alinovi. R;  Scotti. E; Andreoli. R; De Palma. G: Goldoni. M; Apostoli. P; Mutti. A. (2005).
        [Neuroendocrine and renal effects of inorganic lead]. G Ital Med Lav Ergon 27: 33-38.

     Allouche. L; Hamadouche. M; Touabti. A. (2009). Chronic effects  of low lead levels on sperm quality,
        gonadotropins and testosterone in albino rats. Exp Toxicol Pathol 61: 503-510.
        http://dx.doi.0rg/10.1016/i.etp.2008.12.003

     Alomran. AH; Shleamoon. MN. (1988). The influence of chronic lead exposure on lymphocyte
        proliferative response and immunoglobulin levels in storage battery workers. J Biol Sci 19: 575-585.

     Altmann. L; Weinsberg. F; Sveinsson. K; Lilienthal. H; Wiegand. H; Winneke. G. (1993). Impairment of
        long-term potentiation and learning following chronic lead exposure. Toxicol Lett 66: 105-112.
        http://dx.doi.org/10.1016/0378-4274(93)90085-C

     Altmann. P; Maruna. RF; Maruna. H; Michalica. W: Wagner. G. (1981). [Lead detoxication effect of a
        combined calcium phosphate and ascorbic acid therapy in pregnant women with increased lead burden
        (author's transl)]. Wiener medizinische Wochenschrift 131: 311-314.

     Alvarez. J: Garcia-Sancho. J: Herreros. B. (1986). Inhibition of Ca2+-dependent K+ channels by lead in
        one-step inside-out vesicles from human red cell membranes. Biochim Biophys Acta 857: 291-294.
        http://dx.doi.org/10.1016/0005-2736(86)90359-7

     Amici. A; Emanuelli. M; Magni. G: Raffaelli. N; Ruggieri. S.  (1997). Pyrimidine nucleotidases from
        human erythrocyte possess phosphotransferase activities specific for pyrimidine nucleotides. FEES
        Lett 419: 263-267. http://dx.doi.org/10.1016/S0014-5793(97)01464-6

     Andersen. ZJ; Wahlin. P; Raaschou-Nielsen. O; Scheike. T; Loft. S. (2007). Ambient particle source
        apportionment and daily hospital admissions among children and elderly in Copenhagen. J Expo Sci
        Environ Epidemiol 17: 625-636. http://dx.doi.org/10.1038/sj.jes.7500546

     Andreae. MO. (1983). Biotransformation of arsenic  in the marine environment. In WH Lederer; RJ
        Fensterheim (Eds.), Arsenic: Industrial, biomedical and environmental perspectives: Proceedings of
        the Arsenic Symposium (pp. 378-391). New York, NY: Van Nostrand Reinhold.

     Anetor. JI; Adeniyi. FAA.  (1998). Decreased immune status in Nigerian workers occupationally exposed
        to lead. Afr J Med Med Sci 28: 169-172.

     Anjum. RM; Sainath. SB; Suneetha. Y; Reddv. SP (2010). Lead acetate induced reproductive and
        paternal mediated developmental toxicity in rats. Ecotoxicol Environ Saf 74: 793-799.
        http://dx.doi.0rg/10.1016/i.ecoenv.2010.10.044

     Annesi-Maesano. I; Pollitt R: King. G: Bousquet. J: Hellier. G: Sahuquillo. J: Huel. G. (2003). In utero
        exposure to lead and cord blood total IgE: Is there a connection? Allergy 58: 589-594.
        http://dx.doi.0rg/10.1034/i.1398-9995.2003.00111.x

     Antonowicz. J: Andrzejak. R; Smolik. R. (1990). Influence of heavy metal mixtures on erythrocyte
        metabolism. Int Arch Occup Environ Health 62:  195-198. http://dx.doi.org/10.1007/BF00379431
November 2012                              5-704                   Draft - Do Not Cite or Quote

-------
     Antonyuk. SV: Strange. RW: Marklund. SL; Hasnain. SS. (2009). The structure of human extracellular
        copper-zinc superoxide dismutase at 1.7 A resolution: Insights into heparin and collagen binding. J
        Mol Biol 338: 310-326. http://dx.doi.Org/10.1016/j.jmb.2009.03.026

     Anwar. WA; Kamal. AAM. (1988). Cytogenetic effects in a group of traffic policemen in Cairo. Mutat
        Res Lett 208: 225-231. http://dx.doi.org/10.1016/0165-7992(88)90065-6

     Apostoli. P: Huard. C: Chaumontet C: Mattel. P; Alleskx L; Mazzoleni. G. (2000). Effects of four
        inorganic lead compounds on the proliferation and junctional coupling of cultured REL liver cells. Am
        JIndMed 38: 340-348.

     Apostoli. P: Romeo. L; De Matteis. MC: Menegazzi. M; Faggionato. G: Vettore. L. (1988). Effects of
        lead on red blood cell membrane proteins. Int Arch Occup Environ Health 61: 71-75.
        http://dx.doi.org/10.1007/BF00381610

     Aquilonius. SM; Hartvig. P. (1986). A Swedish county with unexpectedly high utilization of anti-
        parkinsonian drugs. ActaNeurol Scand 74: 379-382.
        http://dx.doi.org/10.1111/i. 1600-0404.1986.tb03529.x

     Ari. E; Kava. Y; Demir. H; Asicioglu. E; Keskin. S. (2011). The Correlation of Serum Trace  Elements and
        Heavy Metals with Carotid Artery Atherosclerosis in Maintenance Hemodialysis Patients. Biol Trace
        Elem Res 144: 351-359. http://dx.doi.org/10.1007/sl2011-011-9103-0

     Ariza. ME; Bijur. GN; Williams. MV. (1998). Lead and mercury mutagenesis: Role of H2O2, superoxide
        dismutase, and xanthine oxidase. Environ Mol Mutagen 31: 352-361.
        http://dx.doi.org/10.1002/(SICI)1098-2280(1998)31:4<352::AID-EM8>3.0.CO:2-K

     Ariza. ME; Williams. MV. (1996). Mutagenesis of AS52 cells by low concentrations of lead(II) and
        mercury(II). Environ Mol Mutagen 27:  30-33.
        httrj://dx.doi.org/10.1002/(SICI)1098-2280(1996)27:l<:30::AID-EM4>:3.0.CO:2-I

     Ariza. ME; Williams. MV. (1999). Lead and mercury mutagenesis: Type of mutation dependent upon
        metal concentration. J Biochem Mol Toxicol 13: 107-112.
        http://dx.doi.org/10.1002/(SICI)1099-0461(1999)13:2<107::AID-JBT6>3.0.CO:2-0

     Arnold. WJ: Kellev. WN. (1978). Adenine phosphoribosyltransferase. Methods Enzymol 51: 568-574.
        http://dx.doi.org/10.1016/80076-6879(78)51079-3

     Arora. M; Chan. SW: Ryan. CG: Kennedy. BJ; Walker. DM. (2005). Spatial distribution of lead in enamel
        and coronal dentine of wistar rats. Biol Trace Elem Res 105: 159-170.
        http://dx.doi.Org/10.1385/BTER:105:l-3:159

     Arora. M; Weuve. J: Weisskopf. MG: Sparrow. D; Nie. H; Garcia. RI; Hu. H. (2009). Cumulative lead
        exposure and tooth loss in men: The normative aging study. Environ Health Perspect 117: 1531-1534.
        http://dx.doi.org/10.1289/ehp.0900739

     Arrieta. O: Palencia. G: Garcia-Arenas. G: Morales-Espinosa. D; Hernandez-Pedro. N; Sotelo. J. (2005).
        Prolonged exposure to lead lowers the threshold of pentylenetetrazole-induced seizures in rats.
        Epilepsia46: 1599-1602. http://dx.doi.org/10.1111/i. 1528-1167.2005.00267.x

     Astrin. KH; Bishop. DF; Wetmur. JG: Kaul. B; Davidow. B; Desnick. RJ. (1987). 5-aminolevulinic acid
        dehydratase isozymes and lead toxicity. Ann N Y Acad  Sci 514: 23-29.
        http://dx.doi.org/10.1111/i. 1749-6632.1987.tb48757.x

     Atabek. ME; Kurtoglu. S: Pirgon. O; Uzum. K; Saraymen. R. (2007). Relation of in utero lead exposure
        with insulin-like growth factor-I levels and neonatal anthropometric parameters. Int J Hyg Environ
        Health 210: 91-95. httrj://dx.doi.org/10.1016/i.iiheh.2006.06.007

     Atchison. WD. (2003). Effects of toxic environmental contaminants on voltage-gated calcium channel
        function: From past to present. J Bioenerg Biomembr 35: 507-532.
        http://dx.doi.Org/10.1023/B:JOBB.0000008023.11211.13
November 2012                              5-705                   Draft - Do Not Cite or Quote

-------
     ATS (American Thoracic Society). (2000). What constitutes an adverse health effect of air pollution? This
        official statement of the American Thoracic Society was adopted by the ATS Board of Directors, July
        1999. Am J Respir Crit Care Med 161: 665-673.

     Attri. J; Dhawan. V; Mahmood. S: Pandhi. P; Parwana. HK; Nath. R. (2003). Effect of vitamin C
        supplementation on oxidative DNA damage in an experimental model of lead-induced hypertension.
        Ann Nutr Metab 47: 294-301. http://dx.doi.org/10.1159/000072402

     Audesirk. G: Audesirk. T. (1991). Effects of inorganic lead on voltage-sensitive calcium channels in
        N1E-115 neuroblastoma cells. Neurotoxicology 12: 519-528.

     Audesirk. G: Audesirk. T. (1993). The effects of inorganic lead on voltage-sensitive calcium channels
        differ among cell types and among channel subtypes. Neurotoxicology 142: 259-266.

     Azar. A; Trochimowicz. HJ; Maxfield. ME. (1973). Review of lead studies in animals carried out at
        Haskell Laboratory: Two year feeding study and response to hemorrhage study. In D Earth; A Berlin;
        R Engel; P Recht; J Smeets (Eds.), Environmental health aspects of lead: Proceedings  of an
        international symposium (pp. 199-210). Luxembourg: Commission of the European Communities.

     Baccarelli. A; Bollati. V. (2009). Epigenetics and environmental chemicals [Review]. Curr Opin Pediatr
        21: 243-251. http://dx.doi.org/10.1097/MOP.Ob013e32832925cc

     Badavi. M; Mehrgerdi. FZ; Sarkaki. A; Naseri. MK; Dianat. M. (2008). Effect of grape seed extract on
        lead induced hypertension and heart rate in rat. Pak J Biol Sci 11: 882-887.

     Bagchi. D; Preuss. HG. (2005). Effects of acute and chronic oral exposure of lead on  blood pressure and
        bone mineral density in rats. J Inorg Biochem 99: 1155-1164.
        http://dx.doi.0rg/10.1016/i.jinorgbio.2005.02.011

     Baghurst. PA; McMichael. AJ; Wigg. NR; Vimpani. GV; Robertson. EF; Roberts. RJ; Tong. SL. (1992).
        Environmental exposure to lead and children's  intelligence at the age of seven years: The Port Pirie
        cohort study. N Engl J Med 327:  1279-1284. http://dx.doi.org/10.1056/NEJM199210293271805

     Baker. EL: Feldman. RG: White. RA: Harlev. JP: Niles. CA: Dinse. GE: Berkev. CS.  (1984).
        Occupational lead neurotoxicity: A behavioural and electrophysiological evaluation: Study design and
        year one results. Occup Environ Med 41: 352-361. http://dx.doi.Org/10.1136/oem.41.3.352

     Baker. EL; Landrigan. PJ; Barbour. AG; Cox. DH; Folland. PS; Ligo. RN; Throckmorton. J. (1979).
        Occupational lead poisoning in the United States: Clinical and biochemical findings related to blood
        lead levels.  Occup Environ Med 36: 314-322. http://dx.doi.Org/10.1136/oem.36.4.314

     Baker. EL; White. RF; Pothier. LJ; Berkev. CS; Dinse. GE; Travers. PH; Harlev. JP; Feldman. RG.
        (1985). Occupational lead neurotoxicity: Improvement in behavioural effects after reduction of
        exposure. Br J Ind Med 42: 507-516. http://dx.doi.Org/10.1136/oem.42.8.507

     Balbuena. P; Li. W; Ehrich. M. (2011). Assessments of tight junction proteins occludin, claudin 5 and
        scaffold proteins ZO1 and ZO2 in endothelial cells of the rat blood-brain barrier: cellular responses to
        neurotoxicants malathion and lead acetate. Neurotoxicology 32: 58-67.
        http://dx.doi.0rg/10.1016/i.neuro.2010.10.004

     Balbuena. P; Li. W; Magnin-Bissel. G; Meldrum. B; Ehrich. M. (2010). Comparison  of Two Blood-Brain
        Barrier in Vitro Systems: Cytotoxicity and Transfer Assessments of Malathion/oxon and Lead Acetate.
        Toxicol Sci 114: 260-271.  http://dx.doi.org/10.1093/toxsci/kfqOO 1

     Ball. GV; Sorensen. LB. (1969). Pathogenesis of hyperuricemia in saturnine gout. N Engl J Med 280:
        1199-1202.  http://dx.doi.org/10.1056/neiml96905292802203

     Bandeen-Roche. K; Glass. TA; Bolla. KI; Todd. AC; Schwartz. BS. (2009). Cumulative lead dose and
        cognitive function in older adults. Epidemiology 20: 831-839.
        http://dx.doi.org/10.1097/EDE.Ob013e3181b5flOO
November 2012                             5-706                  Draft - Do Not Cite or Quote

-------
     Bao. OS: Lu. CY; Song. H; Wang. M; Ling. WH; Chen. WQ: Deng. XQ; Hao. YT; Rao. SO. (2009).
        Behavioural development of school-aged children who live around a multi-metal sulphide mine in
        Guangdong province, China: Across-sectional study. BMC Public Health 9: 217.
        http://dx.doi.org/10.1186/1471-2458-9-217

     Baranowska-Bosiacka. I; D^browska-Bouta. B; Struzynska. L. (2011 a). Regional changes in purines and
        selected purinergic receptors in immature rat brain exposed to lead. Toxicology 279: 100-107.
        htrp://dx.doi.org/10.1016/i.tox.2010.09.016

     Baranowska-Bosiacka. I; Dziedziejko. V; Safranow. K; Gutowska. I; Marchlewicz. M; Dolegowska. B;
        Rac. ME; Wiszniewska. B; Chlubek. D. (2009).  Inhibition of erythrocyte phosphoribosyltransferases
        (APRT and HPRT) by Pb2+: A potential mechanism of lead toxicity. Toxicology 259: 77-83.
        htrp://dx.doi.org/10.1016/i.tox.2009.02.005

     Baranowska-Bosiacka. I; Gutowska. I; Marchetti. C: Rutkowska. M; Marchlewicz. M; Kolasa. A;
        Prokopowicz. A; Wiernicki. I; Piotrowska. K; Baskiewicz. M; Safranow. K; Wiszniewska. B;
        Chlubek. D. (20 lib). Altered energy status of primary cerebellar granule neuronal cultures from rats
        exposed to lead in the pre- and neonatal period. Toxicology 280: 24-32.
        http://dx.doi.0rg/10.1016/i.tox.2010.ll.004

     Baranowska-Bosiacka. I; Hlynczak. AJ.  (2003). The effect of lead ions on the energy metabolism of
        human erythrocytes in vitro. Comp Biochem Physiol C Toxicol Pharmacol 134: 403-416.
        http://dx.doi.org/10.1016/S1532-0456(03)00008-5

     Baranowska-Bosiacka. I; Hlynczak. AJ.  (2004). Effect of lead ions on rat erythrocyte purine content. Biol
        Trace Elem Res 100: 259-273. http://dx.doi.Org/10.1385/BTER:100:3:259

     Barbeito. AG: Martinez-Palma. L; Vargas. MR; Pehar. M; Mafiav. N; Beckman. JS: Barbeito. L; Cassina.
        P (2010). Lead exposure stimulates VEGF expression in the spinal cord and extends survival in a
        mouse model of ALS. Neurobiol Dis 37: 574-580. http://dx.doi.Org/10.1016/i.nbd.2009.ll.007

     Barbosa. F. Jr; Sandrim. VC: Uzuelli.  JA; Gerlach. RF; Tanus-Santos. JE.  (2006c). eNOS genotype-
        dependent correlation between whole blood lead and plasma nitric oxide products concentrations.
        Nitric Oxide 14: 58-64. http://dx.doi.Org/10.1016/i.niox.2005.09.007

     Barker. DJ; Osmond. C:  Golding. J; Kuh. D; Wadsworth. ME. (1989). Growth in utero, blood pressure in
        childhood and adult life, and mortality from cardiovascular disease. BMJ 298:  564-567.

     Barker. DJP; Bagbv. SP  (2005). Developmental antecedents of cardiovascular disease: a historical
        perspective. J Am Soc Nephrol 16: 2537-2544. http://dx.doi.org/10.1681/ASN.2005020160

     Baronas. ET; Lee. JW: Alden. C: Hsieh. FY. (2007). Biomarkers to monitor drug-induced
        phospholipidosis. Toxicol Appl Pharmacol 218:  72-78. http://dx.doi.Org/10.1016/i.taap.2006.10.015

     Basha. MR; Murali. M; Siddiqi. HK; Ghosal. K; Siddiqi. OK; Lashuel. HA; Ge. YW: Lahiri. DK; Zawia.
        NH. (2005). Lead (Pb) exposure and its effect on APP proteolysis and A beta aggregation. FASEB J
        19: 2083-2084. http://dx.doi.org/10.1096/fi.05-4375fje

     Basha. MR; Wei. W: Brydie. M; Razmiafshari. M; Zawia. NH. (2003). Lead-induced developmental
        perturbations in hippocampal Spl DNA-binding are prevented by zinc supplementation: In vivo
        evidence for Pb and Zn competition.  Int J Dev Neurosci 21: 1-12.
        htrp://dx.doi.org/10.1016/S0736-5748(02)00137-5

     Batalha.  J. R.: Saldiva P. H; Clarke. RW: Coull. BA; Stearns. RC: Lawrence. J: Murthy. GG: Koutrakis.
        P; Godleski. JJ. (2002). Concentrated ambient air particles induce vasoconstriction of small
        pulmonary arteries in rats. Environ Health Perspect 110: 1191-1197.

     Batuman. V. (1993). Lead nephropathy, gout, and hypertension [Review]. Am J Med Sci 305: 241-247.

     Batuman. V; Landv. E; Maesaka. JK; Wedeen. RP (1983). Contribution of lead to hypertension with renal
        impairment. N Engl J Med 309: 17-21. http://dx.doi.org/10.1056/NEJM198307073090104

     Bayer. SA. (1989). Cellular aspects of brain development [Review]. Neurotoxicology 10: 307-320.
November 2012                             5-707                  Draft - Do Not Cite or Quote

-------
     Baykov. B; Gunova. M; Stovanov. M; Nevchev. H; Stefanova. T; Nicolova. N. (1996). Designing an
        artificial ecological mesocosm for the study of Cd and Pb impact on the immune system of
        experimental animals. Toxicol Lett 89: 5-10.

     Bear. MF; Malenka. RC. (1994).  Synaptic plasticity: LTP and LTD. Curr Opin Neurobiol 4: 389-399.
        http://dx.doi.org/10.1016/0959-4388(94)90101-5

     Beaudin. SA; Stangle. DE; Smith. PR; Levitsky. DA; Strupp. BJ. (2007). Succimer chelation normalizes
        reactivity to reward omission  and errors in lead-exposed rats. Neurotoxicol Teratol 29: 188-202.
        http://dx.doi.0rg/10.1016/i.ntt.2006.ll.004

     Bechara. EJH; Medeiros. MHG: Monteiro. HP; Hermes-Lima. M; Pereira.  B; Demasi. M; Costa. CA;
        Abdalla. DSP: Onuki. J; Wendel. CMA; Mascio. PP. (1993). A free radical hypothesis of lead
        poisoning and inborn porphyrias associated with 5-aminolevulinic acid  overload. Quimica Nova 16:
        385-392.

     Behl. M; Zhang. Y; Zheng. W. (2009). Involvement of insulin-degrading enzyme in the clearance of beta-
        amyloid at the blood-CSF barrier: Consequences of lead exposure.  Cerebrospinal Fluid Research 6:
        11. http://dx.doi.org/10.1186/1743-8454-6-ll

     Bell. ML; Ebisu. K; Peng. RD; Samet. JM; Dominici. F. (2009). Hospital admissions and chemical
        composition of fine particle air pollution. Am J Respir Crit Care Med 179: 1115-1120.
        http://dx.doi.org/10.1164/rccm.200808-1240OC

     Belles-Isles. M; Ayotte. P; Dewailly. E; Weber. JP; Roy. R.  (2002). Cord blood lymphocyte functions in
        newborns from a remote maritime population exposed to organochlorines and methylmercury. J
        Toxicol Environ Health A 65:  165-182. http://dx.doi.org/10.1080/152873902753396794

     Bellinger. D; Hu. H; Titlebaum. L; Needleman. HL. (1994a). Attentional correlates of dentin and bone
        lead levels in adolescents. Arch Environ Health 49: 98-105.
        http://dx.doi.org/10.1080/00039896.1994.9937461

     Bellinger. D; Leviton. A; Allred. E; Rabinowitz. M. (1994b). Pre- and postnatal lead exposure and
        behavior problems in school-aged children. Environ  Res 66: 12-30.
        http://dx.doi.org/10.1006/enrs.1994.1041

     Bellinger. D; Leviton. A; Sloman. J. (1990). Antecedents and correlates of improved cognitive
        performance in children exposed in utero to low levels of lead. Environ Health Perspect 89: 5-11.

     Bellinger. D; Leviton. A; Waternaux. C: Needleman. H; Rabinowitz. M. (1987). Longitudinal  analyses of
        prenatal and postnatal lead exposure and early  cognitive development. N Engl J Med 316:  1037-1043.
        http://dx.doi.org/10.1056/NEJM198704233161701

     Bellinger. D; Needleman. HL. (2003). Intellectual impairment and blood lead levels [Letter]. N Engl J
        Med 349: 500. http://dx.doi.org/10.1056/NEJM200307313490515

     Bellinger. D; Sloman. J: Leviton. A; Rabinowitz. M; Needleman. HL; Waternaux. C. (1991). Low-level
        lead exposure and children's cognitive function in the preschool years. Pediatrics 87: 219-227.

     Bellinger. D; Stiles. KM; Needleman. HL. (1992). Low-level lead exposure, intelligence and academic
        achievement: A long-term follow-up study. Pediatrics 90: 855-861.

     Bellinger. DC. (2004a). Assessing environmental neurotoxicant exposures and child neurobehavior:
        Confounded by confounding? Epidemiology 15: 383-384.
        http://dx.doi.org/10.1097/01.ede.0000129525.15064.a4

     Bellinger. DC. (2004b). What is an  adverse effect? A possible resolution of clinical and epidemiological
        perspectives on neurobehavioral toxicity. Environ Res 95: 394-405.
        http://dx.doi.0rg/10.1016/i.envres.2003.07.013

     Bellinger. DC. (2007). Interpretation of small effect sizes in occupational and environmental
        neurotoxicology: Individual versus population risk. Neurotoxicology 28: 245-251.
        http://dx.doi.0rg/10.1016/i.neuro.2006.05.009
November 2012                             5-708                  Draft - Do Not Cite or Quote

-------
     Bellinger. DC. (2008). Email from David Bellinger, Harvard Medical School, to Jee-Young Kim, EPA.
        SUBJECT: Request for additional information on Bellinger and Needleman (2003). (EPA-HQ-
        OAR-2006-0735-5156). U.S. Environmental Protection Agency.
        http://www.regulations.gov/#!documentDetail:D=EPA-HQ-OAR-2006-0735-5156

     Belloni-Olivi. L; Annadata. M; Goldstein. GW: Bressler. JP (1996). Phosphorylation of membrane
        proteins in erythrocytes treated with lead. Biochem J 315: 401-406.

     Bener. A; Almehdi. AM; Alwash. R; Al-Neamy. FRM. (200 la). A pilot survey of blood lead levels in
        various types of workers in the United Arab Emirates. Environ Int 27: 311-314.
        http://dx.doi.org/10.1016/S0160-4120(01)00061-7

     Bener. A; Obineche. E; Gillett. M; Pasha. MAH; Bishawi. B. (200Ib). Association between blood levels
        of lead, blood pressure and risk of diabetes and heart disease in workers. Int Arch Occup Environ
        Health 74: 375-378. http://dx.doi.org/10.1007/s004200100231

     Bennet. C: Bettaiya. R; Rajanna. S: Baker. L; Yallapragada. PR; Brice. JJ: White. SL; Bokara. KK.
        (2007). Region specific increase in the antioxidant enzymes and lipid peroxidation products in the
        brain of rats exposed to lead.  Free Radic Res 41: 267-273.
        http://dx.doi.org/10.1080/10715760600889855

     Berenson. GS. (2002). Childhood risk factors predict adult risk associated with subclinical cardiovascular
        disease: The  Bogalusa Heart Study [Review]. Am J Cardiol 90: 3L-7L.
        http://dx.doi.org/10.1016/S0002-9149(02)02953-3

     Bergdahl. IA; Grubb. A; Schiitz. A; Desnick. RJ; Wetmur. JG: Sassa. S: Skerfving. S. (1997a). Lead
        binding to delta-aminolevulinic acid dehydratase (ALAD) in human erythrocytes. Basic Clin
        Pharmacol Toxicol 81: 153-158. http://dx.doi.Org/10.llll/i.1600-0773.1997.tb02061.x

     Bergdahl. IA; Schutz. A;  Gerhardsson. L; Jensen. A; Skerfving. S. (1997b). Lead concentrations in human
        plasma, urine and whole blood. Scand J Work Environ Health 23: 359-363.

     Bergdahl. IA; Sheveleva. M; Schutz. A; Artamonova. VG: Skerfving. S. (1998). Plasma and blood lead in
        humans: Capacity-limited binding to delta-aminolevulinic acid dehydratase and other lead-binding
        components.  Toxicol Sci 46: 247-253. http://dx.doi.Org/10.1093/toxsci/46.2.247

     Bergeret. A; Pouget. E; Tedone. R: Meygert. T; Cadot. R: Descotes. J. (1990). Neutrophil functions in
        lead-exposed workers. Hum Exp Toxicol 9: 231-233. http://dx.doi.org/10.1177/096032719000900405

     Berglund. AMM; Ingvarsson. PK; Danielsson. H; Nyholm. NEI. (2010). Lead exposure and biological
        effects in pied flycatchers (Ficedula hypoleuca) before and after the closure of a lead mine in northern
        Sweden. Environ Pollut 158:  1368-1375. http://dx.doi.Org/10.1016/j.envpol.2010.01.005

     Berkowitz. Z; Price-Green. P; Bove. FJ; Kave. WE. (2006). Lead exposure and birth outcomes in five
        communities in Shoshone County, Idaho. Int J Hyg Environ Health 209: 123-132.
        http://dx.doi.0rg/10.1016/i.ijheh.2005.ll.001

     Berrahal. AA; Lasram. M; El Elj. N; Kerkeni. A; Gharbi. N; El-Fazaa. S. (2011). Effect of age-dependent
        exposure to lead on hepatotoxicity and nephrotoxicity in male rats. Environ Toxicol 26: 68-78.
        http://dx.doi.org/10.1002/tox.20530

     Berrahal. AA; Nehdi. A; Haijaii.  N; Gharbi. N; El-Fazaa.  S. (2007). Antioxidant enzymes activities and
        bilirubin level in adult rat treated with lead. C R Biol 330: 581-588.
        http://dx.doi.0rg/10.1016/i.crvi.2007.05.007

     Berry. WD. Jr; Moriartv.  CM; Lau. YS. (2002). Lead attenuation of episodic growth hormone secretion in
        male rats. Int J Toxicol 21: 93-98.

     Bhattacharya. A; Shukla. R: Dietrich. K; Bornschein. R; Berger. O. (1995). Effect of early lead exposure
        on children's postural balance. Dev Med Child Neurol 37: 861-878.

     Bhattacharva. A; Shukla. R: Dietrich. KN; Bornschein. RL. (2006). Effect of early lead exposure on the
        maturation of children's postural balance: A longitudinal study. Neurotoxicol Teratol 3: 376-385.
        http://dx.doi.0rg/10.1016/i.ntt.2006.02.003
November 2012                              5-709                  Draft - Do Not Cite or Quote

-------
     Bhatti. P: Stewart. PA; Hutchinson. A; Rothman. N; Linet. MS: Inskip. PD; Rajaraman. P. (2009). Lead
        exposure, polymorphisms in genes related to oxidative stress, and risk of adult brain tumors. Cancer
        Epidemiol Biomarkers Prev 18: 1841-1848. http://dx.doi.org/10.1158/1055-9965.EPI-09-0197

     Bielarczyk. H; Tian. X; Suszkiw. JB. (1996). Cholinergic denervation-like changes in rat hippocampus
        following developmental lead exposure. Brain Res 708: 108-115.
        http://dx.doi.org/10.1016/0006-8993(95)01315-6

     Bihaqi. SW: Huang. H; Wu. J: Zawia. NH. (2011). Infant exposure to lead (Pb) and epigenetic
        modifications in the aging primate brain: Implications for Alzheimer's disease. J Alzheimers Dis 27:
        819-833. http://dx.doi.org/10.3233/JAD-2011-111013

     Bilban. M. (1998). Influence of the work environment in a Pb-Zn mine on the incidence of cytogenetic
        damage in miners. Am J Ind Med 34: 455-463.
        http://dx.doi.org/10.1002/(SICI)1097-0274(199811)34:5<455::AID-AJIM6>3.0.CO:2-P

     Billick. IH; Curran. AS: Shier. DR. (1979). Analysis of pediatric blood lead levels in New York City for
        1970-1976. Environ Health Perspect 31: 183-190.

     Bishayi. B; Sengupta. M.  (2006). Synergism in immunotoxicological effects due to repeated combined
        administration of arsenic and lead in mice. Int Immunopharmacol 6: 454-464.
        http://dx.doi.0rg/10.1016/i.intimp.2005.09.011

     Biswas. NM; Ghosh. PK.  (2006). Protection of adrenal and male gonadal functions by androgen in lead-
        treated rats. Kathmandu Univ Med J 4: 218-221.

     Bitto. E; Bingman. CA; Wesenberg. GE; McCoy. JG: Phillips. GN. Jr. (2006). Structure of pyrimidine 5'-
        nucleotidase type 1: Insight into mechanism of action and inhibition during lead poisoning. J Biol
        Chem 281: 20521-20529. http://dx.doi.org/10.1074/ibc.M602000200

     Blackman. SS. Jr. (1936). Intranuclear inclusion bodies in the kidney and liver caused by lead poisoning.
        Bull Johns Hopkins Hosp 58: 384-402.

     Blaklev. BR; Sisodia. CS: Mukkur. TK.  (1980). The effect of methylmercury, tetraethyl lead, and sodium
        arsenite on the humoral immune response in mice. Toxicol Appl Pharmacol 52: 245-254.
        http://dx.doi.org/10.1016/0041-008X(80)90111-8

     Bleecker. ML; Ford. DP;  Celio. MA; Vaughan. CG: Lindgren. KN. (2007a). Impact of cognitive reserve
        on the relationship of lead exposure and neurobehavioral performance. Neurology 69: 470-476.
        http://dx.doi.org/10.1212/01.wnl.0000266628.43760.8c

     Bleecker. ML; Ford. DP; Vaughan. CG:  Walsh. KS: Lindgren. KN. (2007b). The association of lead
        exposure and motor performance mediated by cerebral white matter change. Neurotoxicology 28:
        318-323. http://dx.doi.0rg/10.1016/i.neuro.2006.04.008

     Bleecker. ML; Lindgren. KN: Ford. DP. (1997). Differential contribution of current and cumulative
        indices of lead dose to neuropsychological performance by  age. Neurology 48: 639-645.

     Bloom. MS: Louis. GM; Sundaram. R; Kostyniak. PJ; Jain. J. (2011 a). Associations between blood
        metals and fecundity among women  residing in New York State. Reprod Toxicol 31: 158-163.
        http://dx.doi.0rg/10.1016/i.reprotox.2010.09.013

     Bloom. MS: Parsons. PJ:  Kim. D;  Steuerwald. AJ; Vaccari. S: Cheng. G: Fujimoto. VY (20lib). Toxic
        trace metals and embryo quality indicators during in vitro fertilization (IVF).  Reprod Toxicol 31:
        164-170. http://dx.doi.0rg/10.1016/i.reprotox.2010.ll.011

     Bloom. MS: Parsons. PJ:  Steuerwald. AJ: Schisterman. EF; Browne. RW: Kim. K; Coccaro. GA; Conti.
        GC: Naravan. N; Fujimoto. VY. (2010).  Toxic trace metals and human oocytes during in vitro
        fertilization (IVF). Reprod Toxicol 29: 298-305. http://dx.doi.Org/10.1016/j.reprotox.2010.01.003

     Boerrigter. G: Burnett. JC. Jr. (2009). Soluble guanylate cyclase:  Not a dull enzyme [Editorial].
        Circulation 119: 2752-2754. http://dx.doi.org/10.1161/CIRCULATIONAHA.109.860288
November 2012                             5-710                  Draft - Do Not Cite or Quote

-------
     Boillee. S: Yamanaka. K; Lobsiger. CS: Copeland. NG: Jenkins. NA; Kassiotis. G: Kollias. G: Cleveland.
        DW. (2006). Onset and progression in inherited ALS determined by motor neurons and microglia.
        Science 312: 1389-1392. http://dx.doi.org/10.1126/science.1123511

     Bokara. KK; Blaylock. I; Denise. SB; Bettaiya. R; Raj anna. S: Yallapragada. PR. (2009). Influence of
        lead acetate on glutathione and its related enzymes in different regions of rat brain. J Appl Toxicol 29:
        452-458. http://dx.doi.org/10.1002/iat.1423

     Bolin. CM; Basha. R; Cox. D; Zawia. NH; Malonev. B; Lahiri. DK; Cardozo-Pelaez. F.  (2006). Exposure
        to lead (Pb) and the developmental origin of oxidative DNA damage in the aging brain. FASEB J 20:
        788-790. http://dx.doi.org/10.1096/fi.05-5091fie

     Bollati. V: Baccarelli. A; Hou. L; Bonzini. M; Fustinoni. S: Cavallo. D; Byun. HM; Jiang. J: Marinelli. B;
        Pesatori. AC: Bertazzi. PA; Yang. AS. (2007). Changes in DNA methylation patterns in subjects
        exposed to low-dose benzene. Cancer Res 67:  876-880.
        http://dx.doi.org/10.1158/0008-5472.CAN-06-2995

     Bonacker. D; Stoiber. T; Bohm. KJ; Prots. I; Wang. M; Unger. E; Thier. R; Bolt. HM; Degen. GH. (2005).
        Genotoxicity of inorganic lead salts and disturbance of microtubule function. Environ Mol Mutagen
        45: 346-353. http://dx.doi.org/10.1002/em.20100

     Borgstahl. GEO; Parge. HE; Hickev. MJ; Beyer. WF. Jr; Hallewell. RA; Tainer. JA. (1992). The structure
        of human mitochondria! manganese superoxide dismutase reveals a novel tetrameric interface of two
        4-helix bundles. Cell 71: 107-118. http://dx.doi.org/10.1016/0092-8674(92)90270-M

     Boscolo. P: Di Gioacchino. M; Sabbioni. E; Benvenuti. F; Conti. P; Reale. M; Bavazzano. P; Giuliano. G.
        (1999). Expression of lymphocyte subpopulations, cytokine serum levels, and blood  and urinary trace
        elements in asymptomatic atopic men exposed to an urban environment. Int Arch Occup Environ
        Health 72: 26-32. http://dx.doi.org/10.1007/s004200050330

     Boscolo. P: Di Gioacchino. M; Sabbioni. E; Di Giacomo. F; Reale. M; Volpe. AR; Di Sciascio. MB;
        Conti. P: Giuliano. G. (2000). Lymphocyte subpopulations,  cytokines and trace elements in
        asymptomatic atopic women exposed to an urban environment. Life Sci 67: 1119-1126.
        http://dx.doi.org/10.1016/S0024-3205(00)00712-8

     Bouchard. MF; Bellinger. DC: Weuve. J: Matthews-Bellinger. J: Gilman. SE; Wright. RO: Schwartz. J:
        Weisskopf. MG. (2009). Blood lead levels and major depressive disorder, panic disorder, and
        generalized anxiety disorder in US young adults. Arch Gen Psychiatry 66: 1313-1319.
        http://dx.doi.org/10. lOOl/archgenpsvchiatry.2009.164

     Boudene. C: Despaux-Pages. N; Comoy. E; Bohuon. C. (1984). Immunological and enzymatic studies of
        erythrocytic delta-aminolevulinate dehydratase: Comparison of results obtained in normal and lead-
        exposed subjects. Int Arch Occup Environ Health 55: 87-96. http://dx.doi.org/10.1007/BF00378071

     Bound. JP; Harvey. PW: Francis. BJ; Awwad. F; Gatrell. AC. (1997). Involvement of deprivation and
        environmental lead in neural tube defects: A matched case-control study. Arch Dis Child 76: 107-112.
        http://dx.doi.0rg/10.1136/adc.76.2.107

     Bouton. CM; Frelin. LP; Forde. CE; Arnold Godwin.  H; Pevsner. J. (2001). Synaptotagmin I is a
        molecular target for lead. JNeurochem 76: 1724-1735.
        http://dx.doi.0rg/10.1046/i.1471-4159.2001.00168.x

     Bowers. TS: Beck. BD. (2006). What is the meaning of non-linear dose-response relations between blood
        lead concentrations  and IQ? Neurotoxicology 27: 520-524.
        http://dx.doi.0rg/10.1016/i.neuro.2006.02.001

     Bragadin. M; Marton. D; Manente. S. (2007). Trialkyllead compounds induce the opening of the MTP
        pore in rat liver mitochondria. J Inorg Biochem 101: 876-878.
        http://dx.doi.0rg/10.1016/i.iinorgbio.2007.01.016

     Bragadin. M; Marton. D; Murgia. M; Rizzoli. V; Scutari. G: Deana. R. (1998). Interactions of tri alky lie ad
        compounds with rat liver mitochondria. J Inorg Biochem 69: 259-262.
        http://dx.doi.org/10.1016/S0162-0134(97)10036-8
November 2012                             5-711                   Draft - Do Not Cite or Quote

-------
     Bratton. GR; Hinev. JK; Dees. WL. (1994). Lead (Pb) alters the norepinephrine-induced secretion of
        luteinizing hormone releasing hormone from the medium eminence of adult male rats in vitro. Life Sci
        55: 563-571. http://dx.doi.org/10.1016/0024-3205(94)00482-X

     Braude. S: Rovston. D; Coe. C: Barnes. PJ. (1984). Histamine increases lung permeability by an H2-
        receptor mechanism. Lancet 2: 372-374. http://dx.doi.org/10.1016/S0140-6736(84)90542-7

     Braun. JM; Froehlich. TE; Daniels. JL; Dietrich. KN; Hornung. R; Auinger. P; Lanphear. BP (2008).
        Association of environmental toxicants and conduct disorder in U.S. children: NHANES 2001-2004.
        Environ Health Perspect 116: 956-962. http://dx.doi.org/10.1289/ehp. 11177

     Braun. JM; Kahn. RS:  Froehlich. T; Auinger. P; Lanphear. BP.  (2006). Exposures to environmental
        toxicants and attention deficit hyperactivity disorder in U.S. children. Environ Health Perspect 114:
        1904-1909. http://dx.doi.org/10.1289/ehp.9478

     Bravo. Y; Quiroz. Y; Ferrebuz. A; Vaziri. ND; Rodriguez-Iturbe. B. (2007). Mycophenolate mofetil
        administration reduces renal inflammation, oxidative stress, and arterial pressure in rats with lead-
        induced hypertension. Am J Physiol Renal Physiol 293: F616-F623.
        http://dx.doi.org/10.1152/ajprenal.00507.2006

     Brender. JD; Suarez. L; Felkner. M; Gilani. Z; Stinchcomb. D;  Moody. K; Henry. J; Hendricks. K. (2006).
        Maternal exposure to arsenic, cadmium, lead, and mercury and neural tube defects in offspring.
        Environ Res  101: 132-139. http://dx.doi.Org/10.1016/j.envres.2005.08.003

     Brockel. BJ; Cory-Slechta. DA. (1998). Lead, attention, and impulsive behavior: Changes in  a fixed-ratio
        waiting-for-reward  paradigm. Pharmacol Biochem Behav 60: 545-552.
        http://dx.doi.org/10.1016/S0091-3057(98)00023-9

     Brockel. BJ: Cory-Slechta. DA. (1999a). The effects of postweaning low-level Pb exposure on sustained
        attention: a study of target densities, stimulus presentation rate, and stimulus predictability.
        Neurotoxicology 20: 921-933.

     Brockel. BJ: Cory-Slechta. DA. (1999b). Lead-induced decrements in waiting behavior: involvement of
        D2-like dopamine receptors. Pharmacol Biochem Behav 63: 423-434.

     Brondum. J. (2007). Environmental exposures and ADHD [Letter]. Environ Health Perspect 115: A398;
        author reply A399. http://dx.doi.org/10.1289/ehp.10274

     Brondum. J. (2011). ADHD, lead, and PCBs: Appropriate comparison studies [Letter]. Environ Health
        Perspect 119: a282-a282. http://dx.doi.org/10.1289/ehp.1103513

     Brubaker. CJ: Dietrich. KN; Lanphear. BP: Cecil. KM. (2010). The influence of age of lead exposure on
        adult gray matter volume. Neurotoxicology 31: 259-266.
        http://dx.doi.0rg/10.1016/i.neuro.2010.03.004

     Brubaker. CJ: Schmithorst. VJ: Haynes. EN; Dietrich. KN; Egelhoff. JC: Lindquist. DM; Lanphear. BP:
        Cecil. KM. (2009).  Altered myelination and axonal integrity in adults with childhood lead exposure: A
        diffusion tensor imaging study. Neurotoxicology 30: 867-875.
        http://dx.doi.0rg/10.1016/i.neuro.2009.07.007

     Bryant-Stephens. T. (2009). Asthma disparities in urban environments. J Allergy  Clin Immunol 123:
        1199-1206. http://dx.doi.0rg/10.1016/j.jaci.2009.04.030

     Buccelletti. E; Gilardi.  E; Scaini. E; Galiuto. L; Persiani. R; Biondi. A; Basile. F; Silveri. NG. (2009).
        Heart rate variability and myocardial infarction: systematic  literature review and metanalysis
        [Review]. Eur Rev Med Pharmacol Sci 13: 299-307.

     Buchanan. LH; Counter. SA; Ortega.  F. (2011). Environmental lead exposure and otoacoustic emissions
        in Andean children. J Toxicol Environ Health A 74: 1280-1293.
        http://dx.doi.org/10.1080/15287394.2011.587106

     Budtz-Jorgensen. E; Keiding. N; Grandjean. P. (2001). Benchmark dose calculation from epidemiological
        data. Biometrics 57: 698-706. http://dx.doi.Org/10.llll/i.0006-341X.2001.00698.x
November 2012                              5-712                  Draft - Do Not Cite or Quote

-------
     Bunn. TL; Dietert. RR; Ladies. GS: Holsapple. MR (200la). Developmental immunotoxicology
        assessment in the rat: Age, gender, and strain comparisons after exposure to Pb. Toxicol Methods 11:
        41-58. http://dx.doi.org/10.1080/105172301300055151

     Bunn. TL; Marsh. JA; Dietert. RR. (2000). Gender differences in developmental immunotoxicity to lead
        in a chicken: Analysis following a single early low-level exposure in ovo. J Toxicol Environ Health A
        61: 677-693. http://dx.doi.org/10.1080/00984100050195152

     Bunn. TL; Parsons. PJ: Kao. E; Dietert. RR. (200Ib). Exposure to lead during critical windows of
        embryonic development: Differential immunotoxic outcome based on stage of exposure and gender.
        Toxicol Sci 64: 57-66. http://dx.doi.Org/10.1093/toxsci/64.l.57

     Bunn. TL; Parsons. PJ: Kao. E; Dietert. RR. (2001c). Gender-based profiles of developmental
        immunotoxicity to lead in the rat: Assessment in juveniles and adults. J Toxicol Environ Health A 64:
        223-240. http://dx.doi.org/10.1080/15287390152543708

     Burger. J: Gochfeld. M. (2005). Effects of lead on learning  in herring gulls: An avian wildlife model for
        neurobehavioral deficits. Neurotoxicology 26: 615-624. http://dx.doi.Org/10.1016/i.neuro.2005.01.005

     Burns. JM; Baghurst PA; Sawyer. MG: McMichael. AJ; Tong. SL. (1999). Lifetime low-level exposure
        to environmental  lead and children's emotional and behavioral development at ages 11-13 years: The
        Port Pirie cohort study. Am J Epidemiol 149: 740-749.

     Bussolaro. D; Neto. FF; Gargioni. R; Fernandes. LC: Randi. MAP; Pelletier. E; Oliveira Ribeiro. CA.
        (2008). The immune response of peritoneal macrophages due to exposure to inorganic lead in the
        house mouse Mus musculus. Toxicol In Vitro 22: 254-260. http://dx.doi.Org/10.1016/i.tiv.2007.09.003

     Cabaravdic. M; Mijanovic. M; Kusturica. J: Cabaravdic. A. (2010). Occupational exposure of workers at
        gas station to inorganic lead. Med Arh 64:  107-109.

     Caffo. B; Chen. S: Stewart. W: Bolla. K; Yousem. D; Davatzikos. C: Schwartz. BS. (2008). Are brain
        volumes based on magnetic resonance imaging mediators of the associations  of cumulative lead dose
        with cognitive function? Am J Epidemiol 167: 429-437. http://dx.doi.org/10.1093/aje/kwm326

     Cai. MY; Arenaz. P. (1998). Antimutagenic effect of crown ethers on heavy metal-induced sister
        chromatid exchanges. Mutagenesis 13: 27-32. http://dx.doi.Org/10.1093/mutage/13.l.27

     Calderon-Salinas. JV: Quintanar-Escorcia. MA; Gonzalez-Martinez. MT; Hernandez-Luna. CE.  (1999a).
        Lead and calcium transport in human erythrocyte. Hum Exp Toxicol 18: 327-332.
        http://dx.doi.org/10.1191/096032799678840138

     Calderon-Salinas. JV: Quintanar-Escorza.  MA; Hernandez-Luna. CE; Gonzalez-Martinez. MT. (1999b).
        Effect of lead on the calcium transport in human erythrocyte. Hum Exp Toxicol 18: 146-153.
        http://dx.doi.org/10.1177/096032719901800303

     Calvo. FB; Santos. D: Rodrigues. CJ: Krug. FJ; Marumo. JT; Schor. N; Bellini. MH. (2009). Variation in
        the distribution of trace elements in renal cell carcinoma. Biol Trace Elem Res 130: 107-113.
        htrp://dx.doi.org/10.1007/sl2011-009-8325-x

     Camoratto. AM; White. LM; Lau. YS: Ware. GO: Berry. WD; Moriartv. CM. (1993). Effect of exposure
        to low level lead on growth and growth hormone release in rats. Toxicology 83: 101-114.
        http://dx.doi.org/10.1016/0300-483X(93)90095-A

     Campagna. D: Huel.  G: Hellier. G: Girard. F;  Sahuquillo. J: Fagot-Campagna. A; Godin. J: Blot. P.
        (2000). Negative relationships between erythrocyte Ca-pump activity and lead levels in mothers and
        newborns. Life Sci 68: 203-215. http://dx.doi.org/10.1016/S0024-3205(00)00928-0

     Campbell. JR; Auinger. P. (2007). The association between blood lead levels and osteoporosis among
        adults-results from the Third National Health and Nutrition Examination Survey (NHANES  III).
        Environ Health Perspect 115: 1018-1022. http://dx.doi.org/10.1289/ehp.9716
November 2012                             5-713                  Draft - Do Not Cite or Quote

-------
     Camyre. E; Wise. SS: Milligan. P; Gordon. N; Goodale. B; Stackpole. M; Patzlaff N; Aboueissa. AM;
        Wise. JP. Sr. (2007). Ku80 deficiency does not affect paniculate chromate-induced chromosome
        damage and cytotoxicity in Chinese hamster ovary cells. Toxicol Sci 97: 348-354.
        http://dx.doi.org/10.1093/toxsci/kfm045

     Can. S: Bagci. C: Ozaslan. M; Bozkurt. AI; Cengiz. B; Cakmak. EA; Kocabas. R; Karadag. E;
        Tarakcioglu. M. (2008). Occupational lead exposure effect on liver functions and biochemical
        parameters. Acta Physiol Hung 95: 395-403. http://dx.doi.Org/10.1556/APhysiol.95.2008.4.6

     Canfield. R. (2008). Email from Dr. Richard  Canfield, Cornell University, to Jee-Young Kim, USEPA
        Subject: Request for additional information on Canfield et al.  (2003). (EPA-HQ-
        OAR-2006-0735-5811). U.S. Environmental Protection Agency.
        http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OAR-2006-0735-5811

     Canfield. RL; Gendle. MH; Cory-Slechta. DA. (2004). Impaired neuropsychological functioning in lead-
        exposed children. Dev Neuropsychol 26:  513-540. http://dx.doi.org/10.1207/sl5326942dn2601 8

     Canfield. RL; Henderson. CR. Jr; Cory-Slechta. DA; Cox. C: Jusko. TA; Lanphear. BP (2003a).
        Intellectual impairment in children with blood lead concentrations below 10 micrograms per deciliter.
        NEnglJMed348: 1517-1526. http://dx.doi.org/10.1056/NEJMoa022848

     Canfield. RL: Kreher. DA; Cornwell. C: Henderson. CR. Jr. (2003b). Low-level lead exposure, executive
        functioning, and learning in early childhood. Child Neuropsychol 9:  35-53.
        http://dx.doi.org/10.1076/chin.9.1.35.14496

     Cantarow. A; Trumper. M. (1944). Lead poisoning. Hagerstown, MD: Williams & Wilkins Co.

     Cantonwine. D; Hu. H; Sanchez. BN; Lamadrid-Figueroa. H; Smith. D; Ettinger. AS: Mercado-Garcia. A;
        Hernandez-Avila. M; Wright. RO: Tellez-Rojo. MM. (2010a). Critical windows of fetal lead exposure:
        Adverse impacts on length of gestation and risk of premature delivery. J Occup Environ Med 52:
        1106-1111. http://dx.doi.org/10.1097/JOM.Ob013e3181f86fee

     Cantonwine. D: Hu. H; Tellez-Rojo. MM; Sanchez. BN; Lamadrid-Figueroa. H; Ettinger. AS: Mercado-
        Garcia. A; Hernandez-Avila. M; Wright. RO. (2010b). HFE gene variants modify the  association
        between maternal lead burden and infant birthweight: A prospective birth cohort study in Mexico City,
        Mexico. Environ Health Global Access Sci  Source 9: 43. http://dx.doi.org/10.1186/1476-069X-9-43

     Cao. XJ; Huang. SH; Wang. M; Chen. JT; Ruan. DY. (2008). S-adenosyl-L-methionine improves
        impaired hippocampal long-term potentiation  and water maze performance induced by developmental
        lead exposure in rats. Eur J Pharmacol 595: 30-34. http://dx.doi.Org/10.1016/i.ejphar.2008.07.061

     Cao. XJ; Wang. M; Chen. WH; Zhu. DM;  She. JO: Ruan. DY. (2009). Effects of chronic  administration of
        melatonin on spatial learning ability and long-term potentiation in lead-exposed and control rats.
        Biomed Environ Sci 22: 70-75. http://dx.doi.org/10.1016/S0895-3988(09)60025-8

     Carafoli. E. (2005). The Theodor Bucher Lecture - Calcium - a universal carrier of biological signals.
        Delivered on 3 July 2003 at the Special FEES Meeting in Brussels. FEES J 272: 1073-1089.
        http://dx.doi.0rg/10.llll/i.1742-4658.2005.04546.x

     Cardenas. A; Roels. H; Bernard. AM; Barbon. R; Buchet. JP: Lauwerys. RR; Rosello. J: Ramis. I; Mutti.
        A; Franchini. I; Pels. LM; Stolte. H; De Broe. ME; Nuyts. GD; Taylor. SA; Price. RG. (1993).
        Markers of early renal changes induced by industrial pollutants: II Application to workers exposed to
        lead. Br J Ind Med 50: 28-36. http://dx.doi.Org/10.1136/oem.50.l.28

     Carey. JB; Allshire. A; Van Pelt. FN. (2006).  Immune modulation by cadmium and lead in the acute
        reporter antigen-popliteal lymph node assay. Toxicol Sci 91: 113-122.
        http://dx.doi.org/10.1093/toxsci/kfi 142

     Carmignani. M; Boscolo. P; Poma. A; Volpe. AR. (1999). Kininergic system and arterial  hypertension
        following chronic exposure to inorganic lead.  Int Immunopharmacol 44: 105-110.
        http://dx.doi.org/10.1016/S0162-3109(99)00115-0
November 2012                             5-714                  Draft - Do Not Cite or Quote

-------
     Carmignani. M; Volpe. AR; Boscolo. P; Qiao. N; MDi. G: Grilli. A; Felaco. M. (2000). Catcholamine and
        nitric oxide systems as targets of chronic lead exposure in inducing selective functional impairment.
        Life Sci 68: 401-415. http://dx.doi.org/10.1016/S0024-3205(00)00954-l

     Carmona. ER; Creus. A; Marcos. R. (2011). Genotoxicity testing of two lead-compounds in somatic cells
        of Drosophila melanogaster. Mutat Res 724: 35-40. http://dx.doi.Org/10.1016/i.mrgentox.2011.05.008

     Carreras. HA; Wannaz. ED; Pignata. ML. (2009). Assessment of human health risk related to metals by
        the use of biomonitors in the province of Cordoba, Argentina. Environ Pollut 157: 117-122.
        http://dx.doi.0rg/10.1016/i.envpol.2008.07.018

     Casado. MF; Cecchini. AL; Simao. ANC: Oliveira. RD; Cecchini. R. (2007). Free radical-mediated pre-
        hemolytic injury in human red blood cells subjected to lead acetate as evaluated by
        chemiluminescence. Food Chem Toxicol 45: 945-952. http://dx.doi.Org/10.1016/j.fct.2006.12.001

     Castranova. V: Bowman. L; Reasor. MJ; Miles. PR. (1980). Effects of heavy metal ions on selected
        oxidative metabolic processes in rat alveolar macrophages. Toxicol Appl Pharmacol 53:  14-23.
        http://dx.doi.org/10.1016/0041-008X(80)90375-0

     Caylak. E; Aytekin. M; Halifeoglu. I. (2008). Antioxidant effects of methionine, alpha-lipoic acid, N-
        acetylcysteine and homocysteine on lead-induced oxidative stress to erythrocytes in rats. Exp Toxicol
        Pathol 60: 289-294. http://dx.doi.Org/10.1016/i.etp.2007.ll.004

     CDC (Centers for Disease Control and Prevention). (2004). The health consequences of smoking: A
        report of the Surgeon General. Washington, DC: U.S. Department of Health and Human Services.
        http ://www. surgeongeneral.gov/librarv/smokingconsequences/

     CDC (Centers for Disease Control and Prevention). (20 lib). State-based prevalence data of ADHD
        diagnosis. Atlanta, GA. Retrieved from http://www.cdc.gov/ncbddd/adhd/prevalence.html

     CDC (Centers for Disease Control and Prevention). (2012). Tested and confirmed elevated blood lead
        levels by state, year and blood lead level group for children <72 months. Atlanta, GA. Retrieved from
        http://www.cdc.gov/nceh/lead/data/national.htm

     Cecil. K; Yuan. W: Holland. S: Wessel. S: Dietrich. K; Ris. D; Lanphear. B. (2005). The influence of
        childhood lead exposure on language function in young adults: An FMRI study [Abstract]. Proc Int
        Soc Magn Reson Med 13: 1441.

     Cecil. KM; Brubaker. CJ: Adler. CM; Dietrich. KN; Altave. M; Egelhoff. JC: Wessel. S: Elangovan. I;
        Hornung. R; Jarvis. K; Lanphear. BP (2008). Decreased brain volume in adults with childhood lead
        exposure. PLoS Med 5: el 12. http://dx.doi.org/10.1371/iournal.pmed.0050112

     Cecil. KM; Dietrich. KN; Altave. M; Egelhoff. JC: Lindquist. DM; Brubaker. CJ: Lanphear. BP. (2011).
        Proton magnetic resonance spectroscopy in adults with childhood lead exposure. Environ Health
        Perspect 119: 403-408. http://dx.doi.org/10.1289/ehp.1002176

     Celik. A; Ogenler. O; Comelekoglu. U. (2005). The evaluation of micronucleus frequency by acridine
        orange fluorescent staining in peripheral blood of rats treated with lead acetate. Mutagenesis 20:
        411-415. http://dx.doi.org/10.1093/mutage/gei055

     Cerulli. N; Campanella. L; Grossi. R: Politi. L; Scandurra. R: Soda. G: Gallo. F; Damiani. S: Alimonti. A;
        Petrucci. F; Caroli. S. (2006). Determination of Cd, Cu, Pb and Zn in neoplastic kidneys and in renal
        tissue of fetuses, newborns and corpses. J Trace Elem Med Biol 20: 171-179.
        http://dx.doi.0rg/10.1016/i.itemb.2006.03.002

     Chai. S: Webb. RC. (1988). Effects of lead on vascular reactivity [Review]. Environ Health Perspect 78:
        85-89.

     Chakrabortv. I; Sharma. A; Talukder. G. (1987). Antagonistic and synergistic effects of lead and selenium
        in Rattus norvegicus. Toxicol Lett 37: 21-26. http://dx.doi.org/10.1016/0378-4274(87)90162-7

     Chancellor. AM; Slatterv. JM; Fraser. H; Warlow. CP  (1993). Risk factors for motor neuron disease: A
        case-control study based on patients from the Scottish Motor Neuron Disease Register. J Neurol
        Neurosurg Psychiatry 56: 1200-1206. http://dx.doi.org/10.1136/innp.56.ll.1200
November 2012                              5-715                  Draft - Do Not Cite or Quote

-------
     Chandramouli. K; Steer. CD; Ellis. M; Emond. AM. (2009). Effects of early childhood lead exposure on
        academic performance and behaviour of school age children. Arch Dis Child 94: 844-848.
        http://dx.doi.org/10.1136/adc.2008.149955

     Chang. HR: Chen. SS: Tsao. DA: Cheng. JT: Ho. CK: Yu. HS. (1997). Change of cardiac beta-
        adrenoceptors in lead-exposed rats. Toxicology 123: 27-32.
        httrj://dx.doi.org/10.1016/S0300-483X(97)00100-5

     Chang. HR; Tsao. DA; Yu. HS: Ho. CK. (2005). The change of "beta"-adrenergic system after cessation
        of lead exposure. Toxicology 207: 73-80. http://dx.doi.Org/10.1016/i.tox.2004.08.018

     Chang. MK; Raggatt.  LJ; Alexander. KA; Kuliwaba. JS: Fazzalari. NL; Schroder. K; Maylin. ER; Ripoll.
        VM; Hume. DA; Pettit AR. (2008a). Osteal tissue macrophages are intercalated throughout human
        and mouse bone lining tissues and regulate osteoblast function in vitro and in vivo. J Immunol 181:
        1232-1244.

     Chang. SH: Cheng. BH: Lee. SL: Chuang. HY: Yang. CY: Sung. FC: Wu. TN. (2006). Low blood lead
        concentration in association with infertility in women. Environ Res 101: 380-386.
        http://dx.doi.0rg/10.1016/i.envres.2005.10.004

     Chang. WC: Chang. CC: Wang. YS: Wang. YS: Weng. WT; Yoshioka. T; Juo. SH. (2011). Involvement of
        the epidermal growth factor receptor in Pb(2+)-induced activation of cPLA(2)/COX-2 genes and
        PGE(2) production in vascular smooth muscle cells. Toxicology 279: 45-53.
        http://dx.doi.0rg/10.1016/i.tox.2010.09.004

     Chang. YF: Teng. HC: Cheng. SY: Wang. CT: Chiou. SH: Kao. LS: Kao. FJ: Chiou. A: Yang. DM.
        (2008b). Orail-STIMl formed store-operated Ca[sub]2+[/sub] channels (SOCs) as the molecular
        components needed for Pb[sub]2+[/sub] entry in living cells. Toxicol Appl Pharmacol 227: 430-439.
        http://dx.doi.0rg/10.1016/i.taap.2007.ll.025

     Chao. SL: Moss. JM; Harry. GJ. (2007). Lead-induced alterations of apoptosis and neurotrophic factor
        mRNA in the developing rat cortex, hippocampus, and cerebellum. J Biochem Mol Toxicol 21:
        265-272. http://dx.doi.org/10.1002/ibt.20191

     Chaurasia. SS: Gupta. P: Maiti. PK; Kar. A. (1998). Possible involvement of lipid peroxidation in the
        inhibition of type I iodothyronine 5'-monodeiodinase activity by lead in chicken  liver. J Appl Toxicol
        18: 299-300. http://dx.doi.org/10.1002/(SICI)1099-1263(199807/08)18:4<299::AID-
        JAT493>3.0.CO:2-J

     Chaurasia. SS: Kar. A. (1997). Protective effects of vitamin E against lead-induced deterioration of
        membrane associated type-I iodothyronine 5'-monodeiodinase (5'D-I) activity in male mice.
        Toxicology 124: 203-209. http://dx.doi.org/10.1016/S0300-483X(97)00155-8

     Chen. A; Cai. B; Dietrich. KN; Radcliffe. J: Rogan. WJ. (2007). Lead exposure, IQ, and behavior in urban
        5- to 7-year-olds: Does lead affect behavior only by lowering IQ? Pediatrics 119: e650-e658.
        http://dx.doi.org/10.1542/peds.2006-1973

     Chen. A; Dietrich. KN; Ware. JH; Radcliffe. J:  Rogan. WJ. (2005). IQ and blood lead from 2 to 7 years of
        age: Are the effects in older children the residual of high blood lead concentrations in 2-year-olds?
        Environ Health Perspect 113:  597-601. http://dx.doi.org/10.1289/ehp.7625

     Chen. B: Lamberts. LV: Behets. GJ: Zhao. TT: Zhou. MX: Liu. G: Hou. XH: Guan. GJ: D'Haese. PC.
        (2009a). Selenium, lead, and cadmium levels in renal failure patients in China. Biol Trace Elem  Res
        131: 1-12. http://dx.doi.org/10.1007/sl2011-009-8340-v

     Chen. HH; Chan. MH. (2002). Developmental lead exposure differentially alters the susceptibility to
        chemoconvulsants in rats. Toxicology 173: 249-257. http://dx.doi.org/10.1016/S0300-483X(02)00032-
        X

     Chen. HI; Chiu. YW: Hsu. YK; Li. WF; Chen. YC: Chuang. HY. (2010a). The association of
        metallothionein-4 gene polymorphism and renal function in long-term lead-exposed workers. Biol
        Trace Elem Res 137: 55-62. http://dx.doi.org/10.1007/sl2011-009-8564-x
November 2012                             5-716                  Draft - Do Not Cite or Quote

-------
     Chen. J; Mercer. G: Rotfa. SR; Abraham. L; Lutz. P; Ercal. N; Neal. RE. (20lib). Sub-chronic lead
        exposure alters kidney proteome profiles. Hum Exp Toxicol 30: 1616-1625.
        http://dx.doi.org/10.1177/0960327110396521

     Chen. PC: Pan. IJ; Wang. JD. (2006a). Parental exposure to lead and small for gestational age births. Am
        J Ind Med 49: 417-422. http://dx.doi.org/10.1002/aiim.20313

     Chen. PS: Tan. AY. (2007). Autonomic nerve activity and atrial fibrillation [Review]. Heart Rhythm 4:
        S61-S64. http://dx.doi.0rg/10.1016/i.hrthm.2006.12.006

     Chen. S: Golemboski. KA; Piepenbrink. M; Dietert. RR. (2004). Developmental immunotoxicity of lead
        in the  rat: Influence of maternal diet. J Toxicol Environ Health A 67: 495-511.
        http://dx.doi.org/10.1080/15287390490276520

     Chen. S: Golemboski. KA; Sanders. FS: Dietert. RR.  (1999). Persistent effect of in utero meso-2,3-
        dimercaptosuccinic acid (DMSA) on immune function and lead-induced immunotoxicity. Toxicology
        132: 67-79. http://dx.doi.org/10.1016/S0300-483X(98)00139-5

     Chen. S: Miller. TE; Golemboski. KA: Dietert. RR. (1997). Suppression of macrophage metabolite
        production by lead glutamate in vitro is reversed by meso-2,3-dimercaptosuccinic acid (DMSA).  In
        Vitro Toxicol 10: 351-358.

     Chen. XK; Yang. Q; Smith. G: Krewski. D; Walker. M; Wen. SW. (2006c). Environmental lead level and
        pregnancy-induced hypertension. Environ Res 100: 424-430.
        http://dx.doi.0rg/10.1016/i.envres.2005.07.006

     Chen. Y; Zhao. JX; Liu. JW: Cui. J: Li.  L; Tian. W. (2008c). Lack of association of delta-aminolevulinic
        acid dehydratase genotype with blood lead levels in environmentally exposed children of Uygur and
        Han populations. Acta Paediatr 97: 1717-1720. http://dx.doi.Org/10.llll/j.1651-2227.2008.01003.x

     Cheng. Y; Schwartz. J: Sparrow. D; Aro. A; Weiss.  ST; Hu. H. (2001). Bone lead and blood lead levels in
        relation to baseline blood pressure and the prospective development of hypertension: The Normative
        Aging Study. Am J Epidemiol 153: 164-171. http://dx.doi.Org/10.1093/aje/153.2.164

     Cheng. Y; Schwartz. J: Vokonas. PS: Weiss. ST; Aro. A; Hu. H. (1998). Electrocardiographic conduction
        disturbances in association with low-level lead exposure (the Normative Aging Study). Am J Cardiol
        82: 594-599. http://dx.doi.org/10.1016/S0002-9149(98)00402-0

     Cheng. YJ; Yang. BC: Hsieh. WC: Huang. BM; Liu. MY. (2002). Enhancement of TNF-alpha expression
        does not trigger apoptosis upon  exposure of glial cells to lead and lipopolysaccharide. Toxicology 178:
        183-191. http://dx.doi.org/10.1016/S0300-483X(02)00225-l

     Cheng. YJ; Yang. BC: Liu. MY.  (2006).  Lead increases lipopolysaccharide-induced liver injury through
        tumor necrosis factor-alpha overexpression by monocytes/macrophages: Role of protein kinase C and
        p42/44 mitogen-activated protein kinase. Environ Health Perspect 114: 507-513.
        http://dx.doi.org/10.1289/ehp.8550

     Chettv. CS: Reddv. GR; Murthy. KS: Johnson. J: Saiwan. K; Desaiah. D. (2001). Perinatal lead exposure
        alters the expression of neuronal nitric oxide synthase in rat brain. Int J Toxicol 20:  113-120.
        http://dx.doi.org/10.1080/109158101317097692

     Chettv. CS: Vemuri. MC: Campbell. K;  Suresh. C. (2005). Lead-induced cell death of human
        neuroblastoma cells involves GSH deprivation.  Cell Mol Biol Lett 10: 413-423.

     Chia. SE: Zhou. H: Tham. MT: Yap. E: Dong. NV: Tu. NH: Chia. KS. (2005). Possible influence of
        "delta"-aminolevulinic acid dehydratase polymorphism and susceptibility to renal toxicity of lead: A
        study of a Vietnamese population. Environ Health Perspect 113: 1313-1317.
        http://dx.doi.org/10.1289/ehp.7904

     Chia. SE; Zhou. HJ; Theng. TM; Yap. E. (2007). Possibilities of newer ALAD polymorphism influencing
        human susceptibility to effects of inorganic lead on the neurobehavioral functions. Neurotoxicology
        28: 312-317. http://dx.doi.Org/10.1016/i.neuro.2006.04.003
November 2012                             5-717                   Draft - Do Not Cite or Quote

-------
     Chia. SE; Zhou. HJ; Yap. E; Tham. MT; Dong. NV; Hong Tu. NT; Chia. KS. (2006). Association of renal
        function and delta-aminolevulinic acid dehydratase polymorphism among Vietnamese and Singapore
        workers exposed to inorganic lead. Occup Environ Med 63:  180-186.
        http://dx.doi.org/10.1136/oem.2005.021154

     Chiodo. LM; Covington. C: Sokol. RJ; Hannigan. JH; Jannise. J: Ager. J; Greenwald. M; Delanev-Black.
        V (2007). Blood lead levels and specific attention effects in young children. Neurotoxicol Teratol 29:
        538-546. http://dx.doi.0rg/10.1016/i.ntt.2007.04.001

     Chiodo. LM; Jacobson. SW: Jacobson. JL. (2004). Neurodevelopmental effects of postnatal lead exposure
        at very low levels. Neurotoxicol Teratol 26: 359-371. http://dx.doi.Org/10.1016/i.ntt.2004.01.010

     Chiu. TY; Teng. HC: Huang. PC: Kao. FJ; Yang. DM. (2009). Dominant role of Orail with STIM1  on the
        cytosolic entry and cytotoxicity of lead ions. Toxicol Sci 110: 353-362.
        http://dx.doi.org/10.1093/toxsci/kfp099

     Cho. SC:  Kim. BN: Hong. YC: Shin. MS: Yoo. HJ: Kim. JW: Bhang. SY: Cho. IH: Kim. HW. (2010).
        Effect of environmental exposure to lead and tobacco smoke on inattentive and hyperactive symptoms
        and neurocognitive performance in children. J Child Psychol Psychiatry 51: 1050-1057.
        http://dx.doi.0rg/10.llll/i.1469-7610.2010.02250.x

     Chobanian. AV; Bakris. GL; Black. HR; Cushman. WC: Green. LA; Izzo. JL. Jr; Jones. DW: Materson.
        BJ;  Oparil. S: Wright. JT. Jr; Roccella. EJ. (2003). Seventh report of the Joint National Committee on
        Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 42:
        1206-1252. http://dx.doi.org/10.1161/01.HYP.0000107251.49515.c2

     Choie. DP; Richter. GW. (1972). Lead poisoning: Rapid formation of intranuclear inclusions. Science
        177: 1194-1195. http://dx.doi.org/10.1126/science.177.4055.1194

     Choie. DP; Puchter. GW: Young. LB. (1975). Biogenesis of intranuclear lead-protein inclusions in mouse
        kidney. Pathol Res Pract 155: 197-203.

     Chou. YH; Woon. PY; Huang. WC: Shiurba. R; Tsai. YT; Wang. YS: Hsieh. TJ; Chang. WC: Chuang.
        HY; Chang. WC.  (2011). Divalent lead cations induce cyclooxygenase-2 gene expression by
        epidermal growth factor receptor/nuclear factor-kappa B signaling in A43 Icarcinoma cells. Toxicol
        Lett 203: 147-153. http://dx.doi.Org/10.1016/i.toxlet.2011.03.017

     Chuang. HY: Kuo. CH: Chiu. YW: Ho. CK: Chen. CJ: Wu. TN. (2007). A case-control study on the
        relationship of hearing function and blood concentrations of lead, manganese, arsenic, and selenium.
        Sci Total Environ 387: 79-85. http://dx.doi.Org/10.1016/i.scitotenv.2007.07.032

     Chuang. HY; Schwartz. J: Gonzales-Cossio. T; Lugo. MC: Palazuelos. E; Aro. A; Hu. H; Hernandez-
        Avila. M. (2001).  Interrelations of lead levels in bone, venous blood,  and umbilical cord blood with
        exogenous lead exposure through maternal plasma lead in peripartum women. Environ Health
        Perspect 109: 527-532. http://dx.doi.org/10.1289/ehp.01109527

     Church. HJ: Day. JP; Braithwaite.  RA; Brown. SS. (1993a). Binding of lead to a metallothionein-like
        protein in human erythrocytes.  J Inorg Biochem 49: 55-68.
        http://dx.doi.org/10.1016/0162-0134(93)80048-E

     Church. HJ: Day. JP: Braithwaite.  RA; Brown. SS. (1993b). The speciation of lead in erythrocytes in
        relation to lead toxicity: Case studies of two lead-exposed workers. Neurotoxicology 14: 359-364.

     Ciubar. R; Ciofrangeamu C: Mitran. V; Cimpean. A; Stanescu. A; lordachescu. D. (2007).  The effect of
        lead ions on human erythrocytes in vitro. Rev Chim 58: 895-897.

     Clark. CS: Bornschein. RL; Succop. P; Que Hee. SS:  Hammond. PB; Peace. B. (1985). Condition and
        type of housing as an indicator of potential environmental lead exposure and pediatric blood lead
        levels. Environ Res 38: 46-53.  http://dx.doi.org/10.1016/0013-9351(85)90071-4
November 2012                             5-718                   Draft - Do Not Cite or Quote

-------
     Glaus Henn. B; Schnaas. L; Ettinger. AS: Schwartz. J; Lamadrid-Figueroa. H; Hernandez-Avila. M;
        Amarasiriwardena. C: Hu. H; Bellinger. DC: Wright. RO: Tellez-Rojo. MM. (2012). Associations of
        early childhood manganese and lead co-exposure with neurodevelopment. Environ Health Perspect
        120:  126-131. http://dx.doi.org/10.1289/ehp.1003300

     Cline. HT; Witte. S: Jones. KW. (1996). Low lead levels stunt neuronal growth in a reversible manner.
        PNAS 93: 9915-9920.

     Coban. TA; Senturk. M; Ciftci. M; Kufrevioglu. OI. (2007). Effects of some metal ions on human
        erythrocyte glutathione reductase: an in vitro study. Protein Pept Lett 14: 1027-1030.
        http://dx.doi.org/10.2174/092986607782541060

     Cocco. P: Fadda. D; Atzeri. S: Avataneo. G: Meloni. M; Flore. C. (2007). Causes of death among lead
        smelters in relation to the glucose-6-phosphate dehydrogenase polymorphism. Occup Environ Med
        64: 414-416. http://dx.doi.org/10.1136/oem.2006.028779

     Cockcroft. DW: Gault. MH. (1976). Prediction of creatinine clearance from serum creatinine. Nephron
        16:31-41.

     Cohen. BS: Li. W: Xiong. JQ; Lippmann. M. (2000). Detecting H+ in ultrafine ambient aerosol using iron
        nano-film detectors and scanning probe microscopy. Appl Occup Environ Hyg  15: 80-89.
        http://dx.doi.org/10.1080/104732200301881

     Cohen. N; Modai. D; Golik. A; Weissgarten. J; Peller. S: Katz. A; Averbukh. Z; Shaked. U. (1989).
        Increased concanavalin A-induced suppressor cell activity in humans with occupational lead exposure.
        Environ Res 48:  1-6. http://dx.doi.org/10.1016/80013-9351(89)80079-9

     Columbano. A; Endoh. T; Denda. A; Noguchi. O; Nakae. D: Hasegawa. K; Ledda-Columbano. GM;
        Zedda. AI; Konishi. Y. (1996). Effects of cell proliferation and cell death (apoptosis and necrosis) on
        the early stages of rat hepatocarcinogenesis. Carcinogenesis 17: 395-400.
        http://dx.doi.0rg/10.1093/carcin/17.3.395

     Columbano. A; Ledda-Columbano. GM; Coni. P; Pani. P. (1987). Failure of mitogen-induced cell
        proliferation to achieve initiation of rat liver carcinogenesis. Carcinogenesis 8: 345-347.
        http://dx.doi.Org/10.1093/carcin/8.2.345

     Columbano. A; Ledda-Columbano. GM: Ennas. MG: Curto. M; Chelo. A; Pani. P. (1990). Cell
        proliferation and promotion of rat liver carcinogenesis: Different effect of hepatic regeneration and
        mitogen induced hyperplasia on the development of enzyme-altered foci.  Carcinogenesis 11:  771-776.
        http://dx.doi.0rg/10.1093/carcin/ll.5.771

     Company. R; Serafim. A; Lopes. B; Cravo. A; Kalman. J: Riba. I; DelValls. TA; Blasco. J: Delgado. J:
        Sarmiento. AM; Nieto. JM; Shepherd. TJ; Nowell. G: Bebianno. MJ.  (2011). Source and impact of
        lead contamination on delta-aminolevulinic acid dehydratase activity in several marine bivalve species
        along the Gulf of Cadiz. AquatToxicol 101: 146-154. http://dx.doi.Org/10.1016/i.aquatox.2010.09.012

     Comper. WD; Russo. LM. (2009). The glomerular filter: An imperfect barrier is required for perfect renal
        function [Review]. Curr Opin Nephrol Hypertens 18: 336-342.
        http://dx.doi.org/10.1097/Mnh.Ob013e32832cb96a

     Coni. P;  Pichiri-Coni. G: Ledda-Columbano. GM: Semple. E; Raialakshmi. S: Rao. PM; Sarma. DSR;
        Columbano. A. (1992). Stimulation of DNA synthesis by rat plasma following in vivo treatment with
        three liver mitogens. Cancer Lett 61: 233-238.  http://dx.doi.org/10.1016/0304-3835(92)90293-5

     Conroy. MC: Adkinson. NF. Jr; Lichtenstein. LM. (1977). Measurement of IgE on human basophils:
        Relation to serum IgE and anti-IgE-induced histamine release. J Immunol 118:  1317-1321.

     Conterato. GMM; Augusti. PR; Somacal. S: Einsfeld. L; Sobieski. R; Torres. JRV: Emanuelli. T. (2007).
        Effect of lead acetate on cytosolic thioredoxin reductase activity and oxidative stress parameters in rat
        kidneys. Basic Clin Pharmacol Toxicol 101: 96-100.
        http://dx.doi.0rg/10.llll/i.1742-7843.2007.00084.x
November 2012                             5-719                  Draft - Do Not Cite or Quote

-------
     Conterato. GMM; Bulcao. RP; Sobieski. R; Moro. AM; Charao. MF; de Freitas. FA; de Almeida. FL;
        Moreira. APL; Roehrs. M; Tonello. R; Batista. BL; Grotto. D; Barbosa. F. Jr; Garcia. SC; Emanuelli.
        T\ (In Press) Blood thioredoxin reductase activity, oxidative stress and hematological parameters in
        painters and battery workers: Relationship with lead and cadmium levels in blood. J Appl Toxicol.
        http://dx.doi.org/10.1002/iat. 1731

     Cook. JA; Hoffmann. EO; Di Luzio. NR. (1975). Influence of lead and cadmium on the susceptibility of
        rats to bacterial challenge. Proc  Soc Exp Biol Med 150: 741-747.

     Coon. S: Stark. A; Peterson. E; Gloi. A; Kortsha. G: Pounds. J: Chettle. D; Gorell. J. (2006). Whole-body
        lifetime occupational lead exposure and risk of Parkinson's disease. Environ Health Perspect 114:
        1872-1876. http://dx.doi.org/10.1289/ehp.9102

     Coonev. G: Bell. A; Stavrou. C. (1991). Low level exposures to lead and neurobehavioural development:
        The Sydney study at seven years. In JG Farmer (Ed.), International conference: Heavy metals in the
        environment, Vol 1; Edinburgh - September 1991 (pp.  16-19). Edinburgh, United Kingdom: CEP
        Consultants, Ltd.

     Coonev. GH; Bell. A; McBride. W: Carter.  C. (1989a). Low-level exposures to lead: the Sydney lead
        study. Dev Med  Child Neural 31: 640-649.

     Coonev. GH; Bell. A; McBride. W: Carter.  C. (1989b). Neurobehavioural consequences of prenatal low
        level exposures to lead. Neurotoxicol Teratol 11: 95-104.

     Cooper.  GP; Manalis. RS. (1984). Interactions of lead and cadmium on acetylcholine release at the frog
        neuromuscular junction. Toxicol Appl Pharmacol 74: 411-416.
        http://dx.doi.org/10.1016/0041-008X(84)90294-l

     Coratelli. P; Giannattasio. M; Lomonte. C:  Marzolla. R: Rana. F; L'Abbate. N. (1988). Enzymuria to
        detect tubular injury in workers  exposed to lead: A 12-month follow-up. In C Bianchi; V Bocci; FA
        Carone; R Rabkin (Eds.), Kidney and proteins in health and disease: Fifth international symposium in
        health and disease; July 1987; Montecatini Terme, Italy (pp. 207-211). Basel,  Switzerland: S. Karger.

     Coria. C; Cabello. A; Tassara. E; Lopez. E; Rosales. H; Perez. M; Zavala. C; Munoz. P; Orellana. G;
        Inostroza. MI; Contreras. L; Kirsten. L.  (2009). [Long term consequences among children exposed to
        lead poisoning].  Rev Med Chil 137: 1037-1044.  http://dx.doi.org/10.4067/S0034-98872009000800006

     Coria. F; Berciano. MT; Berciano. J; Lafarga. M. (1984). Axon membrane remodeling in the lead-induced
        demyelinating neuropathy of the rat. Brain Res 291: 369-372.
        http://dx.doi.org/10.1016/0006-8993(84)91271-x

     Corongiu. FP; Milia. A. (1982). Rise of hepatic glutathione concentration induced in rats by chronic lead
        nitrate treatment: Its role in aflatoxin Bl intoxication. Res Comm Chem Pathol Pharmacol 38: 97-112.

     Cory-Slechta. DA. (1990). Exposure duration modifies the effects of low level lead on fixed-interval
        performance. Neurotoxicology 11: 427-441.

     Cory-Slechta. DA. (1994). Neurotoxicant-induced changes in schedule-controlled behavior. In LW Chang
        (Ed.), Principles of neurotoxicology (pp. 313-344). New York, NY: Marcel Dekker, Inc.

     Cory-Slechta. DA; Pokora. MJ; Widzowski. DV. (1991). Behavioral manifestations of prolonged lead
        exposure initiated at different stages of the life cycle: II. Delayed spatial alternation. Neurotoxicology
        12: 761-776.

     Cory-Slechta. DA; Pokora. MJ; Widzowski. DV. (1992). Postnatal lead exposure induces supersensitivity
        to the stimulus properties of D2D3  agonist. Brain Res 598: 162-172.
        http://dx.doi.org/10.1016/0006-8993(92)90180-H

     Cory-Slechta. DA; Stern. S; Weston. D; Allen. JL; Liu. S. (2010). Enhanced learning deficits in female
        rats following lifetime pb exposure combined with prenatal stress.  Toxicol Sci 117: 427-438.
        http://dx.doi.org/10.1093/toxsci/kfq221
November 2012                              5-720                   Draft - Do Not Cite or Quote

-------
     Costa. LG: Fox. DA. (1983). A selective decrease of cholinergic muscarinic receptors in the visual cortex
        of adult rats following developmental lead exposure. Brain Res 276: 259-266.
        http://dx.doi.org/10.1016/0006-8993(83)90733-3

     Costa. M; Zhitkovich. A; Gargas. M; Paustenbach. D; Finlev. B; Kuykendall. J: Billings. R; Carlson. TJ;
        Wetterhahn. K; Xu. J; Patierno. S: Bogdanffy. M. (1996). Interlaboratory validation of a new assay for
        DNA-protein crosslinks. Mutat Res Genet Toxicol 369: 13-21.
        http://dx.doi.org/10.1016/S0165-1218(96)90043-9

     Counter. SA; Buchanan. LH; Ortega. F. (2008). Zinc protoporphyrin levels, blood lead levels and
        neurocognitive deficits in Andean children with chronic lead exposure. Clin Biochem 41: 41-47.
        http://dx.doi.0rg/10.1016/i.clinbiochem.2007.10.002

     Counter. SA; Buchanan. LH; Ortega. F. (2009). Neurocognitive screening of lead-exposed andean
        adolescents and young adults. J Toxicol Environ Health A 72: 625-632.
        http://dx.doi.org/10.1080/15287390902769410

     Counter. SA; Buchanan. LH: Ortega. F; Rifai. N; Shannon. MW.  (2007). Comparative analysis of zinc
        protoporphyrin and blood lead levels in lead-exposed Andean children. Clin Biochem 40: 787-792.
        http://dx.doi.0rg/10.1016/i.clinbiochem.2007.03.003

     Counter. SA; Buchanan. LH: Ortega. F; van der Velde. J: Borg. E. (2011). Assessment of auditory
        brainstem function in lead-exposed children using stapedius muscle reflexes. J Neurol Sci 306: 29-37.
        http://dx.doi.0rg/10.1016/i.jns.2011.04.003

     Courtois. E; Marques. M; Barrientos. A; Casado. S: Lopez-Farre. A. (2003). Lead-induced
        downregulation of soluble guanylage cyclase in isolated rat aortic segments mediated by reactive
        oxygen species and cyclooxygenase-2. JAm SocNephrol 14: 1464-1470.
        http://dx.doi.org/10.1097/01.ASN.0000064947.14997.69

     Coyle. JT; Tsai.  G. (2004). The NMDA receptor glycine modulatory site: A therapeutic target for
        improving cognition and reducing negative symptoms in schizophrenia [Review].
        Psychopharmacology 174: 32-38. htrp://dx.doi.org/10.1007/s00213-003-1709-2

     Croes. K; Baevens. W: Bruckers. L; Den Hond. E; Koppen. G: Nelen. V; Van de Mieroop. E; Keune. H;
        Dhooge. W:  Schoeters. G: Van Larebeke. N. (2009). Hormone levels and sexual development in
        Flemish adolescents residing in areas differing in pollution pressure. Int J Hyg Environ Health 212:
        612-625. http://dx.doi.0rg/10.1016/i.ijheh.2009.05.002

     Crooks. PR; Ghosh. MC: Haller. RG: Tong. WH; Rouault. TA. (2010). Posttranslational stability of the
        heme biosynthetic enzyme ferrochelatase is  dependent on iron availability and intact iron-sulfur
        cluster assembly machinery. Blood 115: 860-869. http://dx.doi.org/10.1182/blood-2009-09-243105

     Cullen. C: Singh. A; Dykeman. A; Rice. D; Foster. W. (1993). Chronic lead exposure induces
        ultrastructural alterations in the monkey seminal vesicle. J Submicrosc Cytol Pathol 25: 127-135.

     Daggett DA; Oberlev. TD; Nelson. SA; Wright. LS: Kornguth. SE; Siegel. FL. (1998). Effects of lead on
        rat kidney and liver: GST expression and oxidative stress. Toxicology 128:  191-206.
        http://dx.doi.org/10.1016/S0300-483X(98)00080-8

     Danadevi. K; Rozati. R; Saleha Banu. B; Hanumanth Rao. P; Grover. P. (2003). DNA damage in workers
        exposed to lead using comet assay. Toxicology 187: 183-193.
        htrp://dx.doi.org/10.1016/S0300-483X(03)00054-4

     Daniels. SR. (2011). International differences in secular trends in childhood blood pressure: A puzzle to
        be solved [Editorial].  Circulation 124: 378-380.
        http://dx.doi.org/10.1161/CIRCULATIONAHA. 111.042747

     Davenport. A; Murcutt. G: Whiting. S. (2009). Cross-sectional audit of blood lead levels in regular
        outpatient haemodialysis patients dialysing in north London. Nephrology 14: 476-481.
        http://dx.doi.0rg/10.llll/i.1440-1797.2009.01087.x
November 2012                              5-721                   Draft - Do Not Cite or Quote

-------
     David. O: Clark. J; Voeller. K. (1972). Lead and hyperactivity. Lancet 300: 900-903.
        http://dx.doi.org/10.1016/S0140-6736(72)92534-2

     David. OJ: Hoffman. SP; Sverd. J: Clark. J; Voeller. K. (1976). Lead and hyperactivity: Behavorial
        response to chelation: Apilot study. Am J Psychiatry 133: 1155-1158.

     Davidson. R; MacKinnon. JG. (1981). Several tests for model specification in the presence of alternative
        hypotheses. Econometrica49: 781-793.

     de Burbure. C: Buchet. JP; Leroyer. A; Nisse. C: Haguenoer. JM; Mutti. A; Smerhovsky. Z; Cikrt. M;
        Trczinka-Ochocka. M; Razniewska. G: Jakubowski. M; Bernard. A. (2006). Renal and neurologic
        effects of cadmium, lead, mercury, and arsenic in children: Evidence of early effects and multiple
        interactions at environmental exposure levels. Environ Health Perspect 114: 584-590.
        http:/7dx.doi.org/10.1289/ehp.82Q2

     De Coster. S: Koppen. G: Bracke. M; Schroijen. C: Den Hond. E; Nelen. V; Van De Mieroop. E;
        Bruckers. L; Bilau. M; Baevens. W: Schoeters. G: van Larebeke. N. (2008). Pollutant effects on
        genotoxic parameters and tumor-associated protein levels in adults: Across sectional study. Environ
        Health Global Access Sci Source 7: 26. http://dx.doi.org/10.1186/1476-069X-7-26

     De Guise. S: Bernier. J: Lapierre. P; Dufresne. MM; Dubreuil. P; Fornier.  M. (2000). Immune function of
        bovine leukocytes after in vitro exposure to selected heavy metals. Am J Vet Res 61: 339-344.
        http://dx.doi.org/10.2460/aivr.2000.61.339

     De. M; Ghosh. S: Palit. S: Ghosh. A; Talukder. G: Sharma. A. (1995). Clastogenic effects in human
        samples following prolonged exposure in metal industry. Bull Environ Contam Toxicol 54: 357-362.
        http://dx.doi.org/10.1007/BF00195105

     De Palma. G: Goldoni. M; Catalani. S: Carbognani. P; Poli. D; Mozzoni. P; Acampa. O: Internullo. E;
        Rusca. M; Apostoli. P. (2008). Metallic elements in pulmonary biopsies from lung cancer  and control
        subjects. Acta Biomed 79: 43-51.

     de Restrepo. HG: Sicard. D; Torres. MM. (2000). DNA damage and repair in cells of lead exposed
        people. Am J Ind Med 38: 330-334. http://dx.doi.org/10.1002/1097-0274(200009)38:3<330::AID-
        AJIM13>3.0.CO:2-Z

     de Souza Lisboa. SF; Goncalves. G: Komatsu. F; Queiroz. CAS: Almeida. AA; Moreira. EG.  (2005).
        Developmental lead exposure induces depressive-like behavior in female rats. Drug Chem Toxicol 28:
        67-77. http://dx.doi.org/10.1081/dct-200039696

     Dearth. RK; Hinev. JK; Srivastava. V: Burdick. SB; Bratton. GR; Dees. WL. (2002). Effects of lead (Pb)
        exposure during gestation and lactation on female pubertal development in the rat. Reprod Toxicol 16:
        343-352. http://dx.doi.org/10.1016/S0890-6238(02)00037-0

     Degawa. M; Arai. H; Miura. S: Hashimoto. Y. (1993). Preferential inhibitions of hepatic P450IA2
        expression and induction by lead nitrate in the rat. Carcinogenesis 14: 1091-1094.
        http://dx.doi.0rg/10.1093/carcin/14.6.1091

     Dehlink. E; Baker. AH; Yen. E; Nurko. S: Fiebiger. E. (2010). Relationships between levels of serum IgE,
        cell-bound IgE, and IgE-receptors on peripheral blood cells in a pediatric population. PLoS ONE 5:
        e!2204. http://dx.doi.org/10.1371/iournal.pone.0012204

     Del Donno. M; Chetta. A; Foresi. A; Gavaruzzi. G: Ugolotti. G: Olivieri. D. (1997). Lung epithelial
        permeability and bronchial responsiveness in subjects with stable asthma. Chest 111: 1255-1260.
        http://dx.doi.0rg/10.1378/chest.lll.5.1255

     Delville. Y. (1999). Exposure to lead during development alters aggressive behavior in golden hamsters.
        Neurotoxicol Teratol 21: 445-449.

     Dempster. EL; Pidslev. R; Schalkwyk. LC: Owens. S: Georgiades. A; Kane. F; Kalidindi. S: Picchioni. M;
        Kravariti. E; Toulopoulou. T; Murray. RM; Mill. J. (2011). Disease-associated epigenetic changes in
        monozygotic twins discordant for schizophrenia and bipolar disorder. Hum Mol Genet 20: 4786-4796.
        http://dx.doi.org/10.1093/hmg/ddr416
November 2012                              5-722                  Draft - Do Not Cite or Quote

-------
     Den Hond E. D. hooge W. B..nickers L. S..choeters G. N..elen V. van de Mieroop E. K..oppen G. B.jlau
        M. S..chroijen C. K..eune H. B..aevens W. van Larebeke N. (2011). Internal exposure to pollutants and
        sexual maturation in Flemish adolescents. J Expo Sci Environ Epidemiol 21: 224-233.
        http://dx.doi.0rg/10.1038/ies.2010.2

     Deng. H; Callender. R; Schramm. VL; Grubmever. C. (2010). Pyrophosphate activation in hypoxanthine—
        guanine phosphoribosyltransferase with transition state analogue. Biochemistry 49: 2705-2714.
        http://dx.doi.org/10.1021M100012u

     Deng. W: Poretz. RD. (2002). Protein kinase C activation is required for the lead-induced inhibition of
        proliferation and differentiation of cultured oligodendroglial progenitor cells. Brain Res 929: 87-95.
        http://dx.doi.org/10.1016/S0006-8993(01)03385-6

     Denham. M; Schell. LM; Deane. G:  Gallo. MV; Ravenscroft. J: DeCaprio. AP. (2005). Relationship of
        lead, mercury, mirex, dichlorodiphenyldichloroethylene, hexachlorobenzene, and polychlorinated
        biphenyls to timing of menarche  among Akwesasne Mohawk girls. Pediatrics 115: e!27-e!34.
        http://dx.doi.org/10.1542/peds.2004-1161

     Dentener. MA; Greve. JW: Maessen. JG: Buurman. WA. (1989). Role of tumour necrosis factor in the
        enhanced sensitivity of mice to endotoxin after exposure to lead. Immunopharmacol Immunotoxicol
        11: 321-334. http://dx.doi.org/10.3109/08923978909005373

     Despres. C: Beuter. A; Richer. F; Poitras. K; Veilleux. A; Ayotte. P; Dewailly. E; Saint-Amour. D;
        Muckle. G. (2005). Neuromotor functions in Inuit preschool children exposed to Pb, PCBs, and Hg.
        Neurotoxicol Teratol 27: 245-257. http://dx.doi.Org/10.1016/i.ntt.2004.12.001

     Devaraian. P. (2007). Emerging biomarkers of acute kidney injury. In C Ronco; R Bellomo; JA Kellum
        (Eds.), Acute kidney injury (pp. 203-212). Basel, Switzerland: Karger.
        http://dx.doi.org/10.1159/000102085

     Deveci. E. (2006). Ultrastructural effects of lead acetate on brain of rats. Toxicol Ind Health 22: 419-422.
        http://dx.doi.org/10.1177/07482337060220100101

     Devi. CB; Reddv. GH;  Prasanthi. RPJ; Chettv. CS: Reddv. GR. (2005). Developmental lead exposure
        alters mitochondrial monoamine  oxidase and synaptosomal catecholamine levels in rat brain. Int J
        DevNeurosci 23: 375-381. http://dx.doi.org/10.1016/Ujdevneu.2004.ll.003

     Devi. KD; Banu. BS: Grover. P; Jamil. K. (2000). Genotoxic effect of lead nitrate on mice using SCGE
        (comet assay). Toxicology 145: 195-201. http://dx.doi.org/10.1016/S0300-483X(00)00154-2

     Dhir. H; Ghosh. S: Sharma. A; Talukder. G. (1992a). Interaction between two group IV metals-lead and
        zirconium—in bone marrow cells of Mus musculus in vivo. Biometals 5: 81-86.
        http://dx.doi.org/10.1007/BF01062218

     Dhir. H; Roy. AK; Sharma. A. (1993). Relative efficiency of Phyllanthus emblica fruit extract and
        ascorbic acid in modifying lead and aluminium-induced sister-chromatid exchanges in mouse bone
        marrow. Environ Mol Mutagen 21: 229-236. http://dx.doi.org/10.1002/em.2850210305

     Dhir. H; Roy. AK; Sharma. A; Talukder. G. (1990). Modification of clastogenicity of lead and aluminium
        in mouse bone marrow cells by dietary ingestion of Phyllanthus emblica fruit extract. Mutat Res
        Genet Toxicol 241: 305-312. http://dx.doi.org/10.1016/0165-1218(90)90029-2

     Dhir. H; Sharma. A; Talukder. G. (1992b). Modifying effect of iron on lead-induced clastogenicity in
        mouse bone marrow cells. Biol Trace Elem Res 34: 279-286. http://dx.doi.org/10.1007/BF02783683

     Di Lorenzo. L; Silvestroni. A; Martino. MG: Gagliardi. T; Corfiati. M; Soleo. L. (2006). Evaluation of
        peripheral blood neutrophil leucocytes in lead-exposed workers. Int Arch Occup Environ Health 79:
        491-498. http://dx.doi.org/10.1007/s00420-005-0073-4

     Di Lorenzo. L; Vacca. A; Corfiati. M; Lovreglio. P: Soleo. L. (2007). Evaluation of tumor necrosis factor-
        alpha and granulocyte colony-stimulating factor serum levels in lead-exposed smoker workers. Int J
        Immunopathol  Pharmacol 20: 239-247.
November 2012                              5-723                   Draft - Do Not Cite or Quote

-------
     Dietert. RR. (2011). Role of developmental immunotoxicity and immune dysfunction in chronic disease
        and cancer [Review]. Reprod Toxicol 31: 319-326. http://dx.doi.Org/10.1016/j.reprotox.2010.09.006

     Dietert. RR: McCabe. MJ. Jr. (2007). Lead immunotoxicity. In RW Luebke; RV House; I Kimber (Eds.),
        Immunotoxicology and immunopharmacology (3rd ed., pp. 207-223). Boca Raton, FL: CRC Press.

     Dietert. RR: Piepenbrink. MS. (2006). Perinatal immunotoxicity: Why adult exposure assessment fails to
        predict risk [Review]. Environ Health Perspect 114: 477-483. http://dx.doi.org/10.1289/ehp.8566

     Dietrich. KN; Berger. OG: Succop. PA. (1993 a). Lead exposure and the motor developmental status of
        urban six-year-old children in the Cincinnati Prospective Study. Pediatrics 91: 301-307.

     Dietrich. KN; Berger. OG: Succop. PA; Hammond. PB; Bornschein. RL. (1993b). The developmental
        consequences of low to moderate prenatal and postnatal lead exposure: Intellectual attainment in the
        Cincinnati Lead Study Cohort following school entry. Neurotoxicol Teratol 15: 37-44.
        http://dx.doi.org/10.1016/0892-0362(93)90043-N

     Dietrich. KN; Krafft KM; Bier. M; Succop. PA; Berger. O; Bornschein. RL. (1986). Early effects of fetal
        lead exposure: Neurobehavioral findings at 6 months. International Journal for Biosocial Research 8:
        151-168.

     Dietrich. KN; Krafft. KM; Bornschein. RL; Hammond. PB; Berger. O; Succop. PA; Bier. M. (1987a).
        Low-level fetal lead exposure effect on neurobehavioral development in early infancy. Pediatrics 80:
        721-730.

     Dietrich. KN; Krafft. KM; Shukla. R; Bornschein. RL; Succop. PA. (1987b). The neurobehavioral effects
        of early lead exposure. In SR Schroeder (Ed.), Toxic substances and mental retardation:
        Neurobehavioral toxicology  and teratology (pp. 71-95). Washington, DC: American Association on
        Mental Deficiency.

     Dietrich. KN: Ris. MD; Succop. PA; Berger. OG: Bornschein. RL. (2001). Early exposure to lead and
        juvenile delinquency. Neurotoxicol Teratol 23: 511-518.
        http://dx.doi.org/10.1016/S0892-0362(01)00184-2

     Dietrich. KN: Succop. PA; Berger. OG: Hammond. PB: Bornschein. RL. (1991). Lead exposure and the
        cognitive development of urban preschool children: The Cincinnati Lead Study cohort at age 4 years.
        Neurotoxicol Teratol 13: 203-211. http://dx.doi.org/10.1016/0892-0362(91)90012-L

     Dietrich. KN: Succop. PA; Berger. OG: Keith. RW. (1992). Lead exposure and the central auditory
        processing abilities and cognitive development of urban children: The Cincinnati Lead Study cohort at
        age 5 years. Neurotoxicol Teratol 14: 51-56. http://dx.doi.org/10.1016/0892-0362(92)90028-9

     Ding. Y; Gonick. HC: Vaziri. ND. (2000). Lead promotes hydroxyl radical generation and lipid
        peroxidation in cultured aortic endothelial cells. Am J Hypertens 13: 552-555.
        http://dx.doi.org/10.1016/S0895-7061(99)00226-5

     Ding. Y; Gonick. HC: Vaziri. ND; Liang. K; Wei. L. (2001). Lead-induced hypertension. III. Increased
        hydroxyl radical production. Am J Hypertens 14: 169-173.
        http://dx.doi.org/10.1016/S0895-7061(00)01248-6

     Diouf. A; Garcon. G: Diop. Y; Ndiave. B; Thiaw. C: Fall. M; Kane-Barry. O: Ba. D; Haguenoer. JM;
        Shirali. P. (2006). Environmental lead exposure and its relationship to traffic density among
        Senegalese children: A cross-sectional study. Hum Exp Toxicol 25: 637-644.
        http://dx.doi.org/10.1177/0960327106074591

     Dogru. MI; Dogru. AK; Gul. M; Esrefoglu. M; Yurekli. M; Erdogan. S: Ates. B. (2008). The effect of
        adrenomedullin on rats exposed to lead. J Appl Toxicol 28: 140-146.
        http://dx.doi.org/10.1002/iat. 1259

     Dogu. O: Louis. ED; Tamer. L; Unal. O: Yilmaz. A; Kaleagasi. H. (2007). Elevated blood lead
        concentrations in essential tremor: A case-control  study in Mersin, Turkey. Environ Health Perspect
        115: 1564-1568. http://dx.doi.org/10.1289/ehp.10352
November 2012                              5-724                   Draft - Do Not Cite or Quote

-------
     Donald. JM; Cutler. MG: Moore. MR. (1987). Effects of lead in the laboratory mouse. Development and
        social behaviour after lifelong exposure to 12 microM lead in drinking fluid. Neuropharmacology 26:
        391-399. http://dx.doi.org/10.1016/0028-3908(87)90194-8

     Donald. JM; Cutler. MG: Moore. MR; Bardlev. M. (1986). Effects of lead in the laboratory mouse-2:
        Development and social behaviour after lifelong administration of a small dose of lead acetate in
        drinking fluid. Neuropharmacology 25: 151-160.

     Donaldson. WE; Knowles. SO. (1993). Is lead toxicosis a reflection of altered fatty acid composition of
        membranes? [Review]. Comp Biochem Physiol C Toxicol Pharmacol 104: 377-379.
        httrj://dx.doi.org/10.1016/0742-8413(93)90003-4

     Dong. S: Liang. D; An. N; Jia. L; Shan. Y; Chen. C: Sun. K; Niu. F; Li. H; Fu. S. (2009). The role of
        MAPK and FAS death receptor pathways in testicular germ cell apoptosis induced by lead. Acta
        Biochim Biophys Sin 41: 800-807. http://dx.doi.org/10.1093/abbs/gmp069

     Dorman. RV; Freeman. EJ. (2002). Lead-dependent effects on arachidonic acid accumulation and the
        proliferation of vascular smooth muscle. J Biochem Mol Toxicol 16: 245-253.
        http://dx.doi.org/10.1002/ibt.10045

     Dorsev. CD; Lee. BK; Bolla. KI; Weaver. VM; Lee. SS: Lee. GS: Todd. AC: Shi. W: Schwartz. BS.
        (2006). Comparison of patella lead with blood lead and tibia lead and their associations with
        neurobehavioral test scores. J Occup Environ Med 48: 489-496.
        http://dx.doi.org/10.1097/01.jom.0000199678.86629.3b

     Dou. C: Zhang. J. (2011). Effects of lead on neurogenesis during zebrafish embryonic brain development.
        J Hazard Mater 194: 277-282. http://dx.doi.org/10.1016/i.ihazmat.2011.07.106

     Dowd. JB; Aiello. AE. (2009). Socioeconomic differentials in immune response. Epidemiology 20:
        902-908. http://dx.doi.org/10.1097/EDE.Ob013e3181bb5302

     Dowd. TL; Gupta. RK. (1991). 19F-NMR study of the effect of lead on intracellular free calcium in
        human  platelets. Biochim Biophys Acta 1092: 341-346.
        http://dx.doi.org/10.1016/S0167-4889(97)90010-0

     Dowd. TL; Rosen. JF; Gundberg. CM;  Gupta. RK. (1994). The displacement of calcium from osteocalcin
        at submicromolar concentrations of free lead. Biochim Biophys Acta 1226: 131-137.
        http://dx.doi.org/10.1016/0925-4439(94)90020-5

     Dowd. TL: Rosen. JF; Gupta. RK. (1990). 31PNMR and saturation transfer studies of the effect of PB2+
        on cultured osteoblastic bone cells. J Biol Chem 265: 20833-20838.

     Dowd. TL; Rosen. JF: Mints. L; Gundberg. CM. (2001). The effect of Pb2+ on the structure and
        hydroxyapatite binding properties of osteocalcin. Biochim Biophys Acta 1535: 153-163.
        http://dx.doi.org/10.1016/S0925-4439(00)00094-6

     Dumitrescu. E; Trif. A; Argherie. D. (2007). Consequences of chronic exposure of female rats to lead
        acetate  on ovarian and uterine histoarhitectonic. Lucrari Stiintifice, Seria C: Medicina Veterinara LII:
        198-205.

     Dumitrescu. E; Trif. A; Florin. M. (2008a). The consequences of female rats chronic exposure to lead
        acetate  in the biomarkers emphasizing the hormonal disrupting potential of the reproductive function
        for in vivo evaluation. Bull Univ Agric Sci Vet Med  Cluj-Napoca Vet Med 65:  382-387.

     Dumitrescu. E; Trif. A; Muselin. F. (2008b). The impact of lead acetate on some markers of
        developmental toxicity in female rats (body weight, litter size, litter weight until weaning).  Lucrari
        Stiintifice, Seria C: Medicina Veterinara 53: 192-197.

     Dumitrescu. E; Trif. A; Petrovici. S. (2008c). Lead acetate impact on some markers of female
        reproductive system performances (litter size, sex ratio) and physical development (vaginal opening)
        in rats.  Bull Univ Agric Sci Vet Med Cluj-Napoca Vet Med 65: 283-287.
November 2012                             5-725                  Draft - Do Not Cite or Quote

-------
     Dundar. B; Oktem. F; Arslan. MK; Delibas. N; Baykal. B; Arslan. C: Gultepe. M; Ilhan. IE. (2006). The
        effect of long-term low-dose lead exposure on thyroid function in adolescents. Environ Res 101:
        140-145. http://dx.doi.0rg/10.1016/i.envres.2005.10.002

     Dursun. N; Arifoglu. C: Suer. C: Keskinol. L. (2005). Blood pressure relationship to nitric oxide, lipid
        peroxidation, renal function, and renal blood flow in rats exposed to low lead levels. Biol Trace Elem
        Res 104: 141-149. http://dx.doi.Org/10.1385/BTER:104:2:141

     DuVal. G: Fowler. BA. (1989). Preliminary purification and characterization studies of a low molecular
        weight, high affinity cytosolic lead-binding protein in rat brain. Biochem Biophys Res Commun 159:
        177-184. http://dx.doi.org/10.1016/0006-291X(89)92420-0

     Duvall. RM; Norris. GA; Dailev. LA; Burke. JM; McGee. JK; Gilmour. MI; Gordon. T; Devlin. RB.
        (2008). Source apportionment of paniculate matter in the U.S. and associations with lung
        inflammatory markers. Inhal Toxicol 20: 671-683. http://dx.doi.org/10.1080/08958370801935117

     Duvdu. Y; Dur. A; Siizen. HS. (2005). Evaluation of increased proportion of cells with unusually high
        sister chromatid exchange counts as  a cytogenetic biomarker for lead exposure. Biol Trace Elem Res
        104: 121-129. http://dx.doi.Org/10.1385/BTER:104:2:121

     Duvdu. Y; Suzen. HS: Aydin. A; Gander. O; Uysal. H; Isimer. A; Vural. N. (2001). Correlation between
        lead exposure indicators and sister chromatid exchange (SCE) frequencies in lymphocytes from
        inorganic lead exposed workers. Arch Environ Contam Toxicol 41: 241-246.
        http://dx.doi.org/10.1007/s002440010244

     Dyatlov. VA; Lawrence. DA.  (2002). Neonatal lead exposure potentiates sickness behavior induced by
        Listeria monocytogenes infection of mice. Brain Behav Immun 16: 477-492.
        http://dx.doi.org/10.1006/brbi.2001.0641

     Dyatlov. VA; Platoshin. AV; Lawrence. DA; Carpenter. DO. (1998). Lead potentiates cytokine- and
        glutamate-mediated increases in permeability of the blood-brain barrier. Neurotoxicology 19:
        283-291.

     Eder. K; Reichlmayr-Lais. AM; Kirchgessner. M. (1990). Activity of Na-KATPase and Ca-MgATPase in
        red blood cell membranes of lead-depleted rats. J Trace Elem Electrolytes Health Dis 4: 21-24.

     Egle. PM; Shelton. KR. (1986). Chronic lead intoxication causes a brain-specific nuclear protein to
        accumulate in the nuclei of cells lining kidney tubules. J Biol Chem 261: 2294-2298.

     Ehrlich. R; Robins. T; Jordaan. E;  Miller. S: Mbuli. S: Selbv. P; Wynchank. S: Cantrell. A; De Broe. M;
        D'Haese. P; Todd. A; Landrigan. P. (1998). Lead absorption and renal dysfunction in a South African
        battery factory. Occup Environ Med 55: 453-460. http://dx.doi.Org/10.1136/oem.55.7.453

     Ekinci. D; Bevdemir. S: Kufrevioglu. OI. (2007). In vitro inhibitory effects of some heavy metals on
        human erythrocyte carbonic anhydrases. J Enzyme Inhib Med Chem 22: 745-750.
        http://dx.doi.org/10.1080/14756360601176048

     El-Ashmawv.  IM; Ashry. KM; El-Nahas. AF; Salama. OM. (2006). Protection by turmeric and myrrh
        against liver oxidative damage  and genotoxicity induced by lead acetate in mice. Basic Clin
        Pharmacol Toxicol 98: 32-37. http://dx.doi.Org/10.llll/i.1742-7843.2006.pto 228.x

     El-Fawal. HAN; Waterman. SJ: De Feo. A; Shamy. MY.  (1999).  Neuroimmunotoxicology: Humoral
        assessment of neurotoxicity and autoimmune mechanisms [Review]. Environ Health Perspect 107:
        767-775.

     El-Ghor, AA; Noshy, MM; Bid, JI. (2011). Lead acetate and arsenic trioxide induce instability of
        microsatellites at three different fragile sites (6q21, 9q32-9q33 and 15pl4) within the genome of the
        rat. Mutat Res 726:  195-199. http://dx.doi.Org/10.1016/i.mrgentox.2011.10.003

     El-Nekeetv. AA; El-Kadv. AA; Soliman. MS: Hassan. NS: Abdel-Wahhab. MA. (2009). Protective effect
        of Aquilegia vulgaris (L.)  against lead acetate-induced oxidative stress in rats. Food Chem Toxicol 47:
        2209-2215. http://dx.doi.0rg/10.1016/i.fct.2009.06.019
November 2012                              5-726                   Draft - Do Not Cite or Quote

-------
     El-Neweshy. MS: El-Saved. YS. (2011). Influence of vitamin C supplementation on lead-induced
        histopathological alterations in male rats. Exp Toxicol Pathol 63: 221-227.
        http://dx.doi.0rg/10.1016/i.etp.2009.12.003

     El-Sokkary. GH; Abdel-Rahman. GH; Kamel. ES. (2005). Melatonin protects against lead-induced
        hepatic and renal toxicity in male rats. Toxicology 213: 25-33.
        http://dx.doi.0rg/10.1016/i.tox.2005.05.003

     El Shafai. A; Zohdv. N; El Mulla. K; Hassan. M; Morad. N. (2011). Light and electron microscopic study
        of the toxic effect of prolonged lead exposure on the seminiferous tubules of albino rats and the
        possible protective effect of ascorbic acid. Food Chem Toxicol 49: 734-743.
        http://dx.doi.0rg/10.1016/i.fct.2010.ll.033

     Blinder. CG: Friberg. L; Lind. B; Nilsson. B; Svartengren. M; Overmark. I. (1986). Decreased blood lead
        levels in residents of Stockholm for the period 1980-1984. Scand J Work Environ Health 12: 114-120.

     Elmarsafawv. SF; Jain. NB; Schwartz. J; Sparrow. D; Nie. HL;  Hu. H. (2006). Dietary calcium as a
        potential modifier of the relationship of lead burden to blood pressure. Epidemiology 17: 531-537.
        http://dx.doi.org/10.1097/01.ede.0000231285.86968.2b

     Elwood. RK; Kennedy. S: Belzberg. A; Hogg. JC: Pare. PP. (1983). Respiratory mucosal permeability in
        asthma. Am Rev Respir Dis 128: 523-527.

     Emmerson. BT. (1965). The renal excretion of urate in chronic  lead nephropathy. Intern Med J 14:
        295-303.

     Emmerson. BT; Ravenscroft. PJ. (1975). Abnormal renal urate  homeostasis in systemic disorders.
        Nephron 14: 62-80. http://dx.doi.org/10.1159/000180436

     Ercal. N; Treeratphan. P; Hammond. TC: Matthews. RH; Grannemann. NH; Spitz. DR. (1996). In vivo
        indices of oxidative stress in lead-exposed C57BL/6 mice are reduced by treatment with meso-2,3-
        dimercaptosuccinic acid or N-acetylcysteine. Free Radic Biol Med 21:  157-161.
        http://dx.doi.org/10.1016/0891-5849(96)00020-2

     Ergurhan-Ilhan. I; Cadir.  B; Koyuncu-Arslan. M; C. A; Gultepe. FM; Ozkan. G. (2008). Level of
        oxidative stress and damage in erythrocytes in apprentices indirectly exposed to lead. Pediatrics
        International 50: 45-50. http://dx.doi.Org/10.llll/i.1442-200X.2007.02442.x

     Erie. JC: Good. JA; Butz. JA. (2009). Excess lead in the  neural retina in age-related macular
        degeneration. Am J Ophthalmol 148: 890-894. http://dx.doi.Org/10.1016/i.ajo.2009.07.001

     Erjefalt. I; Persson. CGA. (1991). Allergen, bradykinin, and capsaicin increase outward but not inward
        macromolecular permeability of guinea-pig tracheobronchial mucosa. Clin Exp Allergy 21: 217-224.
        http://dx.doi.0rg/10.llll/i.1365-2222.1991.tb00833.x

     Ernhart. CB; Morrow-Tlucak. M; Marler. MR; Wolf. AW. (1987). Low level lead exposure in the prenatal
        and early preschool periods: Early preschool development. Neurotoxicol Teratol 9: 259-270.
        http://dx.doi.org/10.1016/0892-0362(87)90011-0

     Ernhart. CB; Morrow-Tlucak. M; Wolf. AW. (1988). Low level lead exposure and intelligence in the
        preschool years. Sci Total Environ 71: 453-459.

     Ernhart. CB: Morrow-Tlucak. M; Wolf. AW: Super. D; Drotar.  D. (1989). Low level lead exposure in the
        prenatal and early preschool periods: intelligence prior to school entry. Neurotoxicol Teratol 11:
        161-170.

     Ettinger. AS: Lamadrid-Figueroa. H; Tellez-Rojo. MM; Mercado-Garcia. A; Peterson. KE; Schwartz. J:
        Hu. H; Hernandez-Avila. M. (2009). Effect of calcium supplementation on blood lead levels in
        pregnancy: A randomized placebo-controlled trial. Environ Health Perspect 117: 26-31.
        http://dx.doi.org/10.1289/ehp.11868

     Ettinger. AS: Tellez-Rojo. MM; Amarasiriwardena. C: Bellinger. D; Peterson. K; Schwartz. J: Hu. H;
        Hernandez-Avila. M.  (2004a). Effect of breast milk lead on  infant blood lead levels at 1 month of age.
        Environ Health Perspect 112: 1381-1385.
November 2012                              5-727                   Draft - Do Not Cite or Quote

-------
     Bum. KD; Nie. LH; Schwartz. J; Vokonas. PS: Sparrow. D; Hu. H; Weisskopf. MG. (2011). Prospective
        cohort study of lead exposure and electrocardiographic conduction disturbances in the Department of
        Veterans Affairs Normative Aging Study. Environ Health Perspect 119: 490-494.
        http://dx.doi.org/10.1289/ehp.1003279

     Evans. M; Fored. CM; Nise. G:  Bellocco. R; Nyren. O; Blinder. CG. (2010). Occupational lead exposure
        and severe CKD: A population-based case-control and prospective observational cohort study in
        Sweden. Am J Kidney Dis 55: 497-506. http://dx.doi.Org/10.1053/i.aikd.2009.ll.012

     Ewers. U: Stiller-Winkler. R; Idel. H. (1982). Serum immunoglobulin, complement C3, and salivary IgA
        levels in lead workers. Environ Res 29: 351-357. http://dx.doi.org/10.1016/0013-9351(82)90036-6

     Exon. JH; Koller. LD; Kerkvliet. NI. (1979). Lead-cadmium interaction: Effects on viral-induced
        mortality and tissue residues in mice. Arch Environ Health 34: 469-475.

     Factor-Litvak. P;  Graziano. JH;  Kline. JK; Popovac. D; Mehmeti. A; Ahmedi. G: Shrout. P; Murphy. MJ;
        Gashi. E; Haxhiu. R; Raiovic. L; Nenezic. DU: Stein. ZA. (1991). Aprospective study of birthweight
        and length of gestation in a population surrounding a lead smelter in Kosovo, Yugoslavia. Int J
        Epidemiol 20: 722-728.  http://dx.doi.Org/10.1093/iie/20.3.722

     Factor-Litvak. P;  Kline. JK; Popovac. D: Hadzialjevic. S: Lekic. V: Preteni-Rexhepi. E; Capuni-Paracka.
        S:  Slavkovich. V; Graziano.  J. (1996). Blood lead and blood pressure in young children.
        Epidemiology 7: 633-637.

     Factor-Litvak. P;  Slavkovich. V; Liu. X; Popovac. D: Preteni. E; Capuni-Paracka. S: Hadzialjevic. S:
        Lekic. V: Lolacono. N; Kline. J: Graziano. J. (1998). Hyperproduction of erythropoietin in nonanemic
        lead-exposed children. Environ Health Perspect 106: 361-364.

     Factor-Litvak. P; Wasserman. G: Kline. JK: Graziano. J. (1999). The Yugoslavia prospective study of
        environmental lead exposure. Environ Health Perspect 107: 9-15.
        http://dx.doi.org/10.1289/ehp.991079

     Fadrowski. JJ; Navas-Acien. A; Tellez-Plaza. M; Guallar. E; Weaver. VM; Furth. SL. (2010). Blood lead
        level and kidney function in  US adolescents: The Third National Health and Nutrition Examination
        Survey. Arch Intern Med 170: 75-82. http://dx.doi.org/10.1001/archinternmed.2009.417

     Fahmy. MA. (1999). Lead acetate genotoxicity in mice. Cytologia (Tokyo) 64: 357-365.
        http://dx.doi.org/10.1508/cvtologia.64.357

     Faith. RE: Luster. MI; Kimmel.  CA. (1979). Effect of chronic developmental lead exposure on cell-
        mediated immune functions. Clin Exp Immunol 35: 413-420.

     Fan. G: Feng. C: Li. Y; Wang. CH; Yan. J: Li. W: Feng. JG: Shi. XL; Bi. YY (2009a). Selection of
        nutrients for prevention  or amelioration of lead-induced learning and memory impairment in rats. Ann
        Occup Hyg 53: 341-351. http://dx.doi.org/10.1093/annhyg/mep019

     Fan. G: Feng. C: Wu. F: Ye. W:  Lin. F: Wang. C: Yan. J: Zhu. G: Xiao. Y: Bi. Y. (2010). Methionine
        choline reverses lead-induced cognitive and N-methyl-d-aspartate receptor subunit 1 deficits.
        Toxicology 272: 23-31. http://dx.doi.Org/10.1016/i.tox.2010.03.018

     Fan. Y; Yamada. T; Shimizu. T; Nanashima. N; Akita. M; Suto. K; Tsuchida. S. (2009b). Ferritin
        expression in rat hepatocytes and Kupffer cells after lead nitrate treatment. Toxicol Pathol 37:
        209-217.  http://dx.doi.org/10.1177/0192623308328544

     Fang.  F: Kwee. LC: Allen. KD:  Umbach. DM: Ye. W: Watson. M: Keller. J: Oddone. EZ: Sandier. DP:
        Schmidt.  S: Kamel. F. (2010). Association between blood lead and the risk of amyotrophic lateral
        sclerosis. Am J Epidemiol 171: 1126-1133. http://dx.doi.org/10.1093/aje/kwq06

     Farant. JP; Wigfield. DC. (1987). Interaction of divalent metal ions with normal and lead-inhibited human
        erythrocytic porphobilinogen synthase in vitro. Toxicol Appl Pharmacol 89: 9-18.
        http://dx.doi.org/10.1016/0041-008X(87)90171-2

     Farant. JP: Wigfield. DC. (1990). The effects of copper, zinc, mercury, and cadmium on rabbit
        erythrocytic porphobilinogen synthase in vivo. J Anal Toxicol 14: 222-226.
November 2012                              5-728                   Draft - Do Not Cite or Quote

-------
     Faraone. SV: Doyle. AE. (2001). The nature and heritability of attention-deficit/hyperactivity disorder
        [Review]. Child Adolesc Psychiatr ClinNAm 10: 299-316, viii-ix.

     Farmand. F; Ehdaie. A; Roberts. CK; Sindhu. RK. (2005). Lead-induced dysregulation of superoxide
        dismutases, catalase, glutathione peroxidase, and guanylate cyclase. Environ Res 98: 33-39.
        http://dx.doi.0rg/10.1016/i.envres.2004.05.016

     Farrer. DG: Hueber. S: Laiosa. MD; Eckles. KG: McCabe. MJ. Jr. (2008). Reduction of myeloid
        suppressor cell derived nitric oxide provides a mechanistic basis of lead enhancement of alloreactive
        CD4+ T cell proliferation. Toxicol Appl Pharmacol 229: 135-145.
        http://dx.doi.0rg/10.1016/i.taap.2007.12.011

     Farrer. DG: Hueber. SM; McCabe. MJ. Jr. (2005). Lead enhances CD4+ T cell proliferation indirectly by
        targeting antigen presenting cells and modulating antigen-specific interactions. Toxicol Appl
        Pharmacol 207: 125-137. http://dx.doi.Org/10.1016/i.taap.2004.12.017

     Favalli. L; Chiari. MC: Piccinini. F; Rozza. A. (1977). Experimental investigations on the contraction
        induced by lead in arterial smooth muscle. Acta Pharmacol Toxicol 41: 412-420.
        http://dx.doi.org/10.1111/i. 1600-0773.1977.tb03362.x

     Fazli-Tabaei. S: Fahim. M; Khoshbaten. A. (2006). Acute lead exposure and contraction of rat isolated
        aorta induced by Dl-dopaminergic and alpha-adrenergic drugs. Arch Iran Med 9: 119-123.

     Fehlau. R; Grygorczyk. R; Fuhrmann. GF; Schwarz. W. (1989). Modulation of the Ca2+- or Pb2+-
        activated K+-selective channels in human red cells. II. Parallelisms to modulation of the activity of a
        membrane-bound oxidoreductase. Biochim  Biophys Acta 978: 37-42.
        http://dx.doi.org/10.1016/0005-2736(89)90495-l

     Feinberg. AP (2007). Phenotypic plasticity and the epigenetics of human disease [Review]. Nature 447:
        433-440. http://dx.doi.org/10.1038/nature05919

     Pels. LM; Wunsch. M; Baranowski. J; Norska-Borowka. I; Price. RG: Taylor. SA; Patel. S: De Broe. M;
        Elsevier. MM; Lauwerys. R: Roels. H; Bernard. A; Mutti. A; Gelpi.  E; Rosello. J: Stolte. H. (1998).
        Adverse effects of chronic low level lead exposure on kidney function: A risk group study in children.
        Nephrol Dial Transplant 13: 2248-2256. http://dx.doi.Org/10.1093/ndt/13.9.2248

     Ferguson. C: Kern. M; Audesirk. G. (2000). Nanomolar concentrations  of inorganic lead increase Ca2+
        efflux and decrease intracellular free Ca2+ ion concentrations in cultured rat hippocampal neurons by
        a calmodulin-dependent mechanism. Neurotoxicology 21: 365-378.

     Ferguson. MA; Vaidva. VS: Bonventre. JV. (2008). Biomarkers of nephrotoxic acute kidney injury.
        Toxicology 245:  182-193. http://dx.doi.Org/10.1016/i.tox.2007.12.024

     Fergusson. DM; Boden. JM; Horwood. LJ. (2008). Dentine lead levels in childhood and criminal
        behaviour in late adolescence and early adulthood. J Epidemiol Community Health 62: 1045-1050.
        http://dx.doi.org/10.1136/iech.2007.072827

     Fergusson. DM; Fergusson. JE; Horwood. LJ: Kinzett. NG. (1988a). A longitudinal study of dentine lead
        levels, intelligence, school performance and behaviour Part II Dentine lead and cognitive ability. J
        Child Psychol Psychiatry 29: 793-809.

     Fergusson. DM; Fergusson. JE; Horwood. LJ: Kinzett. NG. (1988b). A longitudinal study of dentine lead
        levels, intelligence, school performance and behaviour: Part I. Dentine lead levels and exposure to
        environmental risk factors. J Child Psychol  Psychiatry 29: 781-792.

     Fergusson. DM; Horwood. LJ: Lynskev. MT. (1993). Early dentine lead levels and subsequent cognitive
        and behavioural development. J Child Psychol Psychiatry  34: 215-227.
        http://dx.doi.0rg/10.llll/i.1469-7610.1993.tb00980.x

     Fergusson. DM; Horwood. LJ: Lynskev. MT. (1997). Early dentine lead levels and educational outcomes
        at 18 years. J Child Psychol Psychiatry 38: 471-478.
November 2012                              5-729                  Draft - Do Not Cite or Quote

-------
     Fernandez-Cabezudo. MJ; All. SAE; Ullah. A; Hasan. MY; Kosanovic. M; Fahim. MA; Adem. A; al-
        Ramadi. BK. (2007). Pronounced susceptibility to infection by Salmonella enterica serovar
        Typhimurium in mice chronically exposed to lead correlates with a shift to Th2-type immune
        responses. Toxicol Appl Pharmacol 218: 215-226. http://dx.doi.Org/10.1016/i.taap.2006.ll.018

     Fine. PR; Thomas. CW: Suhs. RH; Cohnberg. RE; Flashner. BA. (1972). Pediatric blood lead levels. A
        study in 14 Illinois cities of intermediate population. JAMA 221: 1475-1479.

     Fiorim. J; Ribeiro Junior. RF; Silveira. EA; Padilha. AS: Vescovi. MV; de Jesus. HC: Stefanon. I;
        Salaices. M; Vassallo. DV. (2011). Low-level lead exposure increases systolic arterial pressure and
        endothelium-derived vasodilator factors in rat aortas. PLoS ONE 6: e!7117.
        http://dx.doi.org/10.1371/iournal.pone.0017117

     Fischbein. A; Tsang. P; Luo. JCJ: Roboz. JP;  Jiang. JD; Bekesi. JG. (1993). Phenotypic aberrations of the
        CD3+ and CD4+ cells and functional impairments of lymphocytes at low-level occupational exposure
        to lead. Clin Immunol Immunopathol 66:  163-168. http://dx.doi.org/10.1006/clin.1993.1020

     Flegal. AR; Smith. DR. (1992). Lead levels in preindustrial humans [Letter]. N Engl J Med 326:
        1293-1294.

     Flohe. SB; Briiggemann. J: Herder. C: Goebel. C: Kolb. H. (2002). Enhanced proinflammatory response
        to endotoxin after priming of macrophages with lead ions.  J Leukoc Biol 71: 417-424.

     Flora. SJS: Saxena. G: Gautam. P; Kaur. P; Gill. KD. (2007). Response of lead-induced oxidative stress
        and alterations in biogenic amines in different rat brain regions to combined administration of DMSA
        and MiADMSA. Chem Biol Interact 170:  209-220. http://dx.doi.Org/10.1016/i.cbi.2007.08.003

     Flora. SJS: Singh. S: Tandon.  SK. (1989). Thiamine and zinc in prevention or therapy of lead
        intoxication. J Int Med Res 17: 68-75.

     Foglieni. C: Fulgenzi. A; Ticozzi. P; Pellegatta. F; Sciorati. C: Belloni. D; Ferrero. E; Ferrero.  ME.
        (2006). Protective effect of EDTA preadministration on renal ischemia. BMC Nephrol 7: 5.
        http://dx.doi.org/10.1186/1471-2369-7-5

     Foller. M; Huber. SM; Lang. F. (2008).  Erythrocyte programmed cell death [Review]. IUBMB Life  60:
        661-668. http://dx.doi.org/10.1002/iub. 106

     Forst. LS: Freels. S: Persky. V. (1997). Occupational lead exposure and hearing loss. J Occup Environ
        Med 39: 658-660.

     Fortier. M; Omara. F; Bernier. J: Brousseau. P: Fournier. M. (2008). Effects of physiological
        concentrations of heavy metals both individually and in mixtures on the viability and function of
        peripheral blood human leukocytes in vitro. J Toxicol Environ Health A 71: 1327-1337.
        http://dx.doi.org/10.1080/15287390802240918

     Fortoul. TI; Saldivar O. L; Espejel-Mavab. G: Bazarro N. P; Mussali-Galante. P: Avila-Casado. M. del C:
        Colin-Barenque. L; Avila-Costa. MR. (2005). Inhalation of cadmium, lead or its mixture effects on the
        bronchiolar structure and its relation with  metal tissue concentrations. Environ Toxicol Pharmacol 19:
        329-334. http://dx.doi.0rg/10.1016/i.etap.2004.08.007

     Fortune. T; Lurie. DI. (2009). Chronic low-level lead exposure affects the monoaminergic system in the
        mouse superior olivary complex. J Comp Neurol 513: 542-558. http://dx.doi.org/10.1002/cne.21978

     Foster. WG. (1992). Reproductive toxicity of chronic lead exposure in the female cynomolgus monkey.
        Reprod Toxicol 6: 123-131.

     Foster. WG: McMahon. A; YoungLai. EV; Hughes. EG: Rice. DC. (1993). Reproductive endocrine
        effects of chronic lead exposure in the male cynomolgus monkey. Reprod Toxicol 7: 203-209.

     Fowler. BA; DuVal. G. (1991). Effects of lead on the kidney: Roles of high-affinity lead-binding proteins.
        Environ Health Perspect 91: 77-80. http://dx.doi.org/10.1289/ehp.919177

     Fowler. BA: Kahng. MW: Smith. PR; Conner. EA; Laughlin. NK. (1993). Implications of lead binding
        proteins for risk assessment of lead exposure. J Expo Sci Environ Epidemiol 3: 441-448.
November 2012                              5-730                   Draft - Do Not Cite or Quote

-------
     Fowler. BA; Kimmel. CA; Woods. JS: McConnell. EE; Grant. LD. (1980). Chronic low-level lead
        toxicity in the rat: III An integrated assessment of long-term toxicity with special reference to the
        kidney. Toxicol Appl Pharmacol 56: 59-77. http://dx.doi.org/10.1016/0041-008X(80)90131-3

     Fox. DA; Campbell. ML; Blocker. YS. (1997). Functional alterations and apoptotic cell death in the retina
        following developmental or adult lead exposure. Neurotoxicology 18: 645-664.

     Fox. DA; Chu. LWF. (1988). Rods are selectively altered by lead: II. Ultrastructure and quantitative
        histology. Exp Eye Res 46: 613-625. http://dx.doi.org/10.1016/S0014-4835(88)80017-4

     Fox. DA; Farber. DB. (1988). Rods are selectively altered by lead: I. Electrophysiology and biochemistry.
        Exp Eye Res 46: 597-611. http://dx.doi.org/10.1016/S0014-4835(88)80016-2

     Fox. DA; Kala. SV: Hamilton. WR; Johnson. JE; O'Callaghan. JA. (2008). Low-level human equivalent
        gestational lead exposure produces supernormal scotopic electroretinograms, increased retinal
        neurogenesis, and decreased retinal dopamine utilization in rats. Environ Health Perspect 116:
        618-625. http://dx.doi.org/10.1289/ehp.11268

     Fox. DA; Katz. LM. (1992). Developmental lead exposure selectively alters the scotopic ERG component
        of dark and light adaptation and increases rod calcium content. Vision Res 32: 249-255.
        http://dx.doi.org/10.1016/0042-6989(92)90134-5

     Fox. DA; Katz. LM; Farber. DB. (1991). Low level developmental lead exposure decreases the
        sensitivity, amplitude and temporal resolution of rods. Neurotoxicology 12:  641-654.

     Fox. DA; Opanashuk. L; Zharkovsky. A; Weiss. B. (2010). Gene-chemical interactions in the developing
        mammalian nervous system: Effects on proliferation, neurogenesis and differentiation.
        Neurotoxicology 31: 589-597. http://dx.doi.Org/10.1016/i.neuro.2010.03.007

     Fox. DA; Sillman. AJ. (1979). Heavy metals affect rod, but not cone, photoreceptors. Science 206:  78-80.
        http://dx.doi.org/10.1126/science.314667

     Fracasso. ME; Perbellini. L; Solda. S: Talamini. G: Franceschetti. P. (2002). Lead induced DNA strand
        breaks in lymphocytes of exposed workers: Role of reactive oxygen species and protein kinase
        C. Mutat Res Genet Toxicol Environ Mutagen 515: 159-169.
        http://dx.doi.org/10.1016/S1383-5718(02)00012-8

     Fraser. S: Muckle. G: Despres. C. (2006). The relationship between lead exposure, motor function and
        behaviour in Inuit preschool children. Neurotoxicol Teratol 28: 18-27.
        http://dx.doi.0rg/10.1016/i.ntt.2005.10.008

     Fried. LF. (2009).  Creatinine and cystatin C: What are the values? [Comment].  Kidney Int 75: 578-580.
        http://dx.doi.org/10.1038/Ki.2008.688

     Froehlich. TE; Lanphear. BP; Auinger. P; Hornung. R; Epstein. JN; Braun. J; Kahn. RS. (2009).
        Association of tobacco and lead exposures with attention-deficit/hyperactivity disorder. Pediatrics
        124: E1054-E1063. http://dx.doi.org/10.1542/peds.2009-0738

     Froehlich. TE; Lanphear. BP; Dietrich. KN; Cory-Slechta. DA; Wang. N; Kahn. RS. (2007). Interactive
        effects of a DRD4 polymorphism, lead and sex on executive functions in children. Biol Psychiatry 62:
        243-249. http://dx.doi.0rg/10.1016/i.biopsvch.2006.09.039

     Fujita. H; Orii. Y; Sano. S. (1981). Evidence of increased synthesis of delta-aminolevulinic acid
        dehydratase in experimental lead-poisoned rats. Biochim Biophys Acta 678: 39-50.
        http://dx.doi.org/10.1016/0304-4165(81)90045-3

     Fujita. H; Sato. K; Sano. S. (1982). Increase in the amount of erythrocyte delta-aminolevulinic acid
        dehydratase in workers with moderate lead exposure. Int Arch Occup Environ Health 50: 287-297.
        http://dx.doi.org/10.1007/BF00378090

     Fujiwara. Y; Kaii. T; Sakurai. S: Sakamoto. M; Kozuka. H. (1997). Inhibitory effect of lead on the  repair
        of wounded monolayers of cultured vascular endothelial cells. Toxicology 117: 193-198.
        http://dx.doi.org/10.1016/S0300-483X(96)03575-5
November 2012                              5-731                   Draft - Do Not Cite or Quote

-------
     Fukumoto. K; Karai. I; Horiguchi. S. (1983). Effect of lead on erythrocyte membranes. Occup Environ
        Med 40: 220-223. http://dx.doi.Org/10.1136/oem.40.2.220

     Fuller. GM; Zhang. Z. (2001). Transcriptional control mechanism of fibrinogen gene expression. Ann N Y
        Acad Sci 936: 469-479. http://dx.doi.Org/10.llll/i.1749-6632.2001.tb03534.x

     Fulton. M; Raab. G: Thomson. G: Laxen. D; Hunter. R; Hepburn. W. (1987). Influence of blood lead on
        the ability and attainment of children in Edinburgh. Lancet 329:  1221-1226.
        http://dx.doi.org/10.1016/S0140-6736(87)92683-3

     Gainer. JH. (1977). Effects of heavy metals and of deficiency of zinc on mortality rates in mice infected
        with encephalomyocarditis virus. Am J Vet Res 38: 869-872.

     Gao. A; Lu. XT; Li. QY; Tian. L. (2010a). Effect of the delta-aminolevulinic acid dehydratase gene
        polymorphism on renal and neurobehavioral function in workers exposed to lead in China. Sci Total
        Environ 408: 4052-4055. http://dx.doi.Org/10.1016/i.scitotenv.2010.04.024

     Gao. D; Kasten-Jolly. J; Lawrence. DA. (2006). The paradoxical effects of lead in interferon-gamma
        knockout BALB/c mice. Toxicol Sci 89: 444-453. http://dx.doi.org/10.1093/toxsci/kfj043

     Gao. D: Mondal. TK; Lawrence. DA. (2007). Lead effects on development and function of bone marrow-
        derived dendritic cells promote Th2 immune responses. Toxicol Appl Pharmacol 222: 69-79.
        http://dx.doi.0rg/10.1016/i.taap.2007.04.001

     Gao. SJ: Jin. YL; Unverzagt. FW: Ma. F; Hall. KS: Murrell. JR; Cheng. YB; Shen. JZ; Ying. B; Ji. RD;
        Matesan. J: Liang. C: Hendrie. HC. (2008). Trace element levels and cognitive function in rural
        elderly Chinese. J Gerontol ABiol Sci Med Sci 63: 635-641.

     Garavan. H; Morgan. RE;  Levitsky. DA; Hermer-Vazquez. L; Strupp. BJ. (2000). Enduring effects of
        early lead exposure: Evidence for a specific deficit in associative ability. Neurotoxicol Teratol 22:
        151-164.

     Garcia-Leston. J: Roma-Torres. J: Vilares. M; Pinto. R; Cunha. LM; Prista. J: Teixeira. JP; Mayan. O;
        Pasaro. E; Mendez. J: Laffon. B. (2011). Biomonitoring of a population of Portuguese workers
        exposed to lead. Mutat Res Genet Toxicol Environ Mutagen 721: 81-88.
        http://dx.doi.0rg/10.1016/i.mrgentox.2011.01.001

     Garcon. G: Leleu. B; Marez. T; Zerimech. F; Jean-Marie. HP; Daniel. FB; Shirali. P. (2007).
        Biomonitoring of the adverse effects induced by the chronic exposure to lead and cadmium on kidney
        function: Usefulness of alpha-glutathione S-transferase. Sci Total Environ 377: 165-172.
        http://dx.doi.0rg/10.1016/i.scitotenv.2007.02.002

     Gargioni. R; Neto. FF; Buchi. DF; Randi. MAP; Franco. CRC: Paludo. KS: Pelletier. E; Ferraro. MVM;
        Cestari. MM; Bussolaro. D; Ribeiro. CAP. (2006). Cell death and DNA damage in peritoneal
        macrophages of mice (Mus musculus) exposed to inorganic lead. Cell Biol Int 30:  615-623.
        http://dx.doi.0rg/10.1016/i.cellbi.2006.03.010

     Garza. A; Vega. R: Soto. E. (2006). Cellular mechanisms of lead neurotoxicity [Review]. Med Sci Monit
        12: RA57-RA65.

     Gastaldo. J: Viau. M; Bencokova. Z; Joubert. A; Charvet AM; Balosso. J: Foray. N. (2007). Lead
        contamination results in late and slowly repairable DNA double-strand breaks and impacts upon the
        ATM-dependent signaling pathways.  Toxicol Lett 173: 201-214.
        http://dx.doi.0rg/10.1016/i.toxlet.2007.08.003

     Gautam. P; Flora. SJ. (2010). Oral supplementation of gossypin during lead exposure  protects alteration
        in heme synthesis pathway and brain  oxidative stress in rats. Nutrition 26: 563-570.
        http://dx.doi.0rg/10.1016/i.jhazmat.2008.12.070

     Gedeon. Y; Ramesh. GT; Wellman. PJ; Jadhav. AL. (2001). Changes in mesocorticolimbic dopamine and
        D1/D2 receptor levels after low level lead exposure: Atime course study. Toxicol Lett 123: 217-226.
        http://dx.doi.org/10.1016/S0378-4274(01)00408-8
November 2012                              5-732                   Draft - Do Not Cite or Quote

-------
     Gent. JF; Koutrakis. P; Belanger. K; Triche. E; Holford. TR; Bracken. MB; Leaderer. BP. (2009).
        Symptoms and medication use in children with asthma and traffic-related sources of fine particle
        pollution. Environ Health Perspect 117: 1168-1174. http://dx.doi.org/10.1289/ehp.0800335

     Gerr. F; Letz. R; Stokes. L; Chettle. D; McNeill. F; Kave. W. (2002). Association between bone lead
        concentration and blood pressure among young  adults. Am J Ind Med 42: 98-106.

     Ghering. AB; Jenkins. LMM; Schenck. BL; Deo. S: Mayer. RA; Pikaart. MJ; Omichinski. JG: Godwin.
        HA. (2005). Spectroscopic and functional determination of the interaction of Pb2+ with GATA
        proteins. JAm Chem Soc 127: 3751-3759. http://dx.doi.org/10.1021/ia0464544

     Ghosh-Narang. J; Jones. TM; Menke. A; Todd. AC: Muntner. P; Batuman. V. (2007). Parathyroid
        hormone status does not influence blood and bone lead levels in dialysis patients. Am J Med Sci 334:
        415-420. http://dx.doi.org/10.1097/MAJ.Ob013e318068b237

     Giddabasappa. A; Hamilton. WR; Chanev. S: Xiao. W: Johnson. JE; Mukherjee. S: Fox. DA. (2011).
        Low-level gestational lead exposure increases retinal progenitor cell proliferation and rod
        photoreceptor and bipolar cell neurogenesis in mice. Environ Health Perspect 119: 71-77.
        http://dx.doi.org/10.1289/ehp.1002524

     Giedd. JN; Lalonde. FM; Celano. MJ: White. SL; Wallace. GL; Lee. NR; Lenroot. RK. (2009).
        Anatomical brain magnetic resonance imaging of typically developing children and adolescents
        [Review]. J Am Acad Child Adolesc Psychiatry 48: 465-470.
        http://dx.doi.org/10.1097/CHI.Ob013e31819f2715

     Gilbert. ME; Kelly. ME; Samsam. TE; Goodman. JH. (2005). Chronic developmental lead exposure
        reduces neurogenesis in adult rat hippocampus but does not impair spatial learning. Toxicol Sci 86:
        365-374. http://dx.doi.org/10.1093/toxsci/kFil56

     Gilbert. ME; Mack. CM; Laslev. SM. (1999). Chronic developmental lead exposure and hippocampal
        long-term potentiation: biphasic dose-response relationship. Neurotoxicology 20: 71-82.

     Gilbert. SG: Rice. DC. (1987). Low-level lifetime lead exposure produces behavioral toxicity (spatial
        discrimination reversal) in adult monkeys. Toxicol Appl Pharmacol 91: 484-490.

     Gilsanz. V; Chalfant. J: Kalkwarf. H; Zemel. B; Lappe. J: Oberfield. S:  Shepherd. J: Wren. T; Winer. K.
        (2011). Age at onset of puberty predicts bone mass in young adulthood. J Pediatr 158: 100-105,
        105.el01-102. http://dx.doi.0rg/10.1016/i.ipeds.2010.06.054

     Giridhar. J: Isom. GE. (1990). Interaction of lead acetate with atrial natriuretic factor in rats. Life Sci 46:
        569-576. http://dx.doi.org/10.1016/0024-3205(90)90124-A

     Gittleman. R; Eskenazi. B. (1983). Lead and hyperactivity revisited: An investigation of non-
        disadvantaged children. Arch Gen Psychiatry 40: 827-833.
        http://dx.doi.org/10.1001/archpsvc.1983.01790070017002

     Glahn. F; Schmidt-Heck. W: Zellmer. S: Guthke. R: Wiese. J: Golka. K; Hergenroder. R; Degen. GH;
        Lehmann. T; Hermes. M; Schormann. W: Brulport. M; Bauer. A; Bedawv. E; Gebhardt. R: Hengstler.
        JG: Foth. H. (2008). Cadmium, cobalt and lead cause stress response, cell cycle deregulation and
        increased steroid as well as xenobiotic metabolism in primary normal human bronchial epithelial cells
        which is coordinated by at least nine transcription factors. Arch Toxicol 82: 513-524.
        http://dx.doi.org/10.1007/s00204-008-0331-9

     Glass. TA; Bandeen-Roche. K; McAtee. M; Bolla. K; Todd. AC:  Schwartz. BS. (2009). Neighborhood
        psychosocial hazards and the association of cumulative lead dose with cognitive function in older
        adults. Am J Epidemiol 169:  683-692. http://dx.doi.org/10.1093/aie/kwn390

     Glenn. BS: Bandeen-Roche. K; Lee. BK; Weaver. VM; Todd. AC: Schwartz. BS. (2006). Changes in
        systolic blood pressure associated with lead in blood and bone. Epidemiology 17: 538-544.
        http://dx.doi.org/10.1097/01.ede.0000231284.19078.4b

     Glymour. M. (2012). Data clarification for Glymour et al 2005. Available online at
        http ://www.regulations. gov/#! documentDetail;D=EPA-HQ-ORD-2011 -0051 -0027
November 2012                             5-733                  Draft - Do Not Cite or Quote

-------
     Glymour. MM; Weuve. J: Berkman. LF; Kawachi. I; Robins. JM. (2005). When is baseline adjustment
        useful in analyses of change? An example with education and cognitive change. Am J Epidemiol 162:
        267-278. http://dx.doi.org/10.1093/aje/kwil87

     Godleski. JJ: Clarke. RW: Coull. BA: Saldiva. PHN: Jiang. NF: Lawrence. J: Koutrakis. P. (2002).
        Composition of inhaled urban air particles determines acute pulmonary responses. Ann Occup Hyg
        46: 419-424. http://dx.doi.org/10.1093/annhyg/46.suppl  1.419

     Goering. PL; Fowler. BA. (1984). Regulation of lead inhibition of delta-aminolevulinic acid dehydratase
        by a low molecular weight, high affinity renal lead-binding protein. J Pharmacol Exp Ther 231:  66-71.
     Goering. PL; Fowler. BA. (1985). Mechanism of renal lead-binding protein reversal of delta-
        aminolevulinic acid dehydratase inhibition by lead. J Pharmacol Exp Ther 234: 365-371.
     Goering. PL; Fowler. BA. (1987a).  Kidney zinc-thionein regulation of delta-aminolevulinic acid
        dehydratase inhibition by lead. Arch Biochem Biophys 253: 48-55.
        http://dx.doi.org/10.1016/0003-9861(87)90635-7

     Goering. PL; Fowler. BA. (1987b).  Regulatory roles of high-affinity metal-binding proteins in mediating
        lead effects on delta-aminolevulinic acid dehydratase. Ann N Y Acad Sci 514: 235-247.
        http://dx.doi.org/10.1111/i. 1749-6632.1987.tb48778.x

     Goering. PL; Mistry. P; Fowler. BA. (1986). A low molecular weight lead-binding protein in brain
        attenuates lead inhibition of delta-aminolevulinic acid dehydratase: Comparison with a renal lead-
        binding protein. J Pharmacol Exp Ther 237: 220-225.
     Golabek. T; Darewicz. B; Borawska. M; Markiewicz. R; Socha.  K; Kudelski. J. (2009). Lead
        concentration in the bladder tissue and blood of patients with bladder cancer. Scand J Urol Nephrol
        43: 467-470. http://dx.doi.org/10.3109/00365590903198991

     Goldstein. GW. (1993). Evidence that lead acts as a calcium substitute in second messenger metabolism
        [Review]. Neurotoxicology 14: 97-101.
     Gollenberg. AL; Hediger. ML; Lee. PA; Himes. JH; Buck Louis. GM. (2010). Association between lead
        and cadmium and reproductive hormones in peripubertal U.S. girls. Environ Health Perspect 118:
        1782-1787. http://dx.doi.org/10.1289/ehp.1001943

     Gomaa. A; Hu. H; Bellinger. D; Schwartz. J: Tsaih. SW: Gonzalez-Cossio. T; Schnaas. L; Peterson. K;
        Aro. A; Hernandez-Avila. M. (2002). Maternal bone lead as an independent risk factor for fetal
        neurotoxicity: A prospective study.  Pediatrics 110: 110-118.
     Gomez-Ramirez. P..; Martinez-Lopez. E..; Maria-Mojica. P..: Leon-Ortega. M..; Garcia-Fernandez. A..
        1 (2011). Blood lead levels and 5-ALAD inhibition in nestlings of Eurasian Eagle Owl (Bubo bubo)
        to assess lead exposure associated to an abandoned mining area. Ecotoxicology 20:  131-138.
        http://dx.doi.org/10.1007/sl0646-010-0563-3

     Gong. Z; Evans. HL. (1997). Effect of chelation with meso-dimercaptosuccinic acid (DMSA) before and
        after the appearance of lead-induced neurotoxicity in the rat. Toxicol Appl Pharmacol 144: 205-214.
        http://dx.doi.org/10.1006/taap.1997.8118

     Gonick. HC: Cohen. AH; Ren. Q; Saldanha. LF; Khalil-Manesh. F; Anzalone. J: Sun. YY. (1996). Effect
        of 2,3-dimercaptosuccinic acid on nephrosclerosis in the Dahl rat I role of reactive oxygen species.
        Kidney Int 50: 1572-1581. http://dx.doi.org/10.1038/ki.1996.473

     Gonick. HC: Ding. Y; Bondv. SC: Ni. Z; Vaziri. ND. (1997). Lead-induced hypertension: Interplay of
        nitric oxide and reactive oxygen species. Hypertension 30: 1487-1492.

     Gonick. HC: Ding. Y; Vaziri. ND. (1998). Effect of low lead exposure on eicosanoid excretion in rats.
        Prostaglandins Other Lipid Mediat  55: 77-82. http://dx.doi.org/10.1016/S0090-6980(98)00010-0
     Gonick. HC: Khalil-Manesh. F; Raghavan. SRV. (1985). Characterization of human erythrocyte lead-
        binding protein. In TD Lekkas (Ed.), Heavy metals in the environment: Athens ~ September 1985 (pp.
        313-316). Edinburgh, U.K.: CEP Consultants.
November 2012                             5-734                   Draft - Do Not Cite or Quote

-------
     Gorell. JM; Johnson. CC: Rybicki. BA; Peterson. EL; Kortsha. GX; Brown. GG: Richardson. RJ. (1997).
        Occupational exposures to metals as risk factors for Parkinson's disease. Neurology 48: 650-658.

     Goswami. K; Gachhui. R; Bandopadhyav. A. (2005). Hepatorenal dysfunctions in lead pollution. J
        Environ Eng Sci 47: 75-80.

     Governa. M; Valentino. M; Visona. I. (1987). In vitro impairment of human granulocyte functions by
        lead. Arch Toxicol 59: 421-425. http://dx.doi.org/10.1007/BF00316208

     Gover. RA. (1968). The renal tubule in lead poisoning: I. mitochondrial swelling and aminoaciduria. Lab
        Invest 19: 71-77.

     Gover. RA; Cherian. MG: Delaquerriere-Richardson. L. (1978). Renal effects of repeated administration
        of calcium disodium ethylenediaminetetraacetate during excessive exposure to lead in rats. J Environ
        Pathol Toxicol Oncol 1: 403-410.

     Gover. RA; Krall. A; Kimball. JP. (1968). The renal tubule in lead poisoning: II. In vitro studies of
        mitochondrial structure and function. Lab Invest 19: 78-83.

     Gover. RA: Leonard. PL; Moore. JF; Rhyne. B; Krigman. MR. (1970a). Lead dosage and the role of the
        intranuclear inclusion body: An experimental study. Arch Environ Health 20: 705-711.

     Gover. RA: May. P; Gates. MM; Krigman. MR. (1970b). Lead and protein content of isolated intranuclear
        inclusion bodies from kidneys of lead-poisoned rats. Lab Invest 22: 245-251.

     Gover. RA; Rhyne. BC. (1973). Pathological effects of lead [Review]. Int Rev Exp Pathol 12: 1-77.

     Gozdzik-Zolnierkiewicz. T; Moszvnski. B. (1969). VIII nerve in experimental lead poisoning. Acta
        Otolaryngol 68: 85-89. http://dx.doi.org/10.3109/00016486909121546

     Grabowska. M; Guminska. M. (1996). The effect of lead on lactate formation, ATP level and membrane
        ATPase activities in human erythrocytes in vitro. Int J Occup Med Environ Health 9: 265-274.

     Gracia. CR; Sammel. MD; Coutifaris. C: Guzick. DS: Barnhart. KT. (2005). Occupational exposures and
        male infertility. Am J Epidemiol 162: 729-733. http://dx.doi.org/10.1093/aie/kwi269

     Graf. U: Wiirgler. FE. (1986). The present status of validation of the wing spot test in Drosophila. Prog
        Clin Biol Res 209B: 391-398.

     Gramigni. E; Tadini-Buoninsegni. F; Bartolommei. G: Santini. G: Chelazzi. G: Moncelli. MR. (2009).
        Inhibitory effect of Pb2+ on the transport cycle of the Na+,K+ATPase. Chem Res Toxicol 22:
        1699-1704. http://dx.doi.org/10.1021/tx9001786

     Grant. LD; Kimmel. CA; West. GL; Martinez-Vargas. CM; Howard. JL.  (1980). Chronic low-level lead
        toxicity in the rat: II. Effects on postnatal physical and behavioral development. Toxicol Appl
        Pharmacol 56: 42-58. http://dx.doi.org/10.1016/0041-008X(80)90130-l

     Graves. AB; Van Duijn. CM; Chandra. V: Fratiglioni. L; Heyman. A; Jorm. AF; Kokmen. E; Kondo. K;
        Mortimer. JA; Rocca. WA; Shalat. SL; Soininen. H; Hofman. A. (1991). Occupational exposures to
        solvents and lead as risk factors for Alzheimer's disease: a collaborative re-analysis of case-control
        studies. Int J Epidemiol 2: S58-S61.

     Graziano. J: Slavkovich. V; Liu X Factor-Litvak. P; Todd. A.  (2004). A prospective study of prenatal and
        childhood lead exposure and erythropoietin production. J Occup Environ Med 46: 924-929.

     Graziano. JH;  Slavkovic. V:  Factorlitvak. P; Popovac. D; Ahmedi. X; Mehmeti. A. (1991). Depressed
        serum erythropoietin in pregnant women with elevated blood lead. Arch Environ Occup Health 46:
        347-350.

     Greene. T; Ernhart. CB. (1993). Dentine lead and intelligence prior to school entry: A statistical
        sensitivity analysis. J Clin Epidemiol 46: 323-339. http://dx.doi.org/10.1016/0895-4356(93)90147-S

     Greene. T; Ernhart. CB; Boyd. TA.  (1992). Contributions of risk factors to elevated blood and dentine
        lead levels in preschool children. Sci Total Environ 115: 239-260.
November 2012                             5-735                   Draft - Do Not Cite or Quote

-------
     Grizzo. LT; Cordellini. S. (2008). Perinatal lead exposure affects nitric oxide and cyclooxygenase
        pathways in aorta of weaned rats. Toxicol Sci 103: 207-214. http://dx.doi.org/10.1093/toxsci/kfnO 18

     Grlickova-Duzevik. E; Wise. SS: Munroe. RC: Thompson. WD; Wise. JP. Sr. (2006). XRCC1 protects
        against paniculate chromate-induced chromosome damage and cytotoxicity in Chinese hamster ovary
        cells. Toxicol Sci 92: 409-415. http://dx.doi.org/!0.1093/toxsci/kfl021

     Grover. P; Rekhadevi. PV: Danadevi. K; Vuyyuri. SB; Mahboob. M; Rahman. MF. (2010). Genotoxicity
        evaluation in workers occupationally exposed to lead. Int J Hyg Environ Health 213: 99-106.
        http://dx.doi.0rg/10.1016/i.ijheh.2010.01.005

     Guilarte. TR; McGlothan. JL. (1998). Hippocampal NMDA receptor mRNA undergoes subunit specific
        changes during developmental lead exposure. Brain Res  790: 98-107.
        http://dx.doi.org/10.1016/S0006-8993(98)00054-7

     Guilarte. TR; Toscano. CD; Mcglothan. JL; Weaver. SA.  (2003). Environmental enrichment reverses
        cognitive and molecular deficits induced by developmental lead exposure. Ann Neurol 53: 50-56.
        http://dx.doi.org/10.1002/ana.10399

     Gulson. B; Mahaffev. KR; Jameson. CW: Patison. N; Law. AJ; Mizon. KJ; Korsch. MJ; Pederson. D.
        (1999). Impact of diet on lead in blood and urine in female adults and relevance to mobilization of
        lead from bone stores. Environ Health Perspect 107: 257-263. http://dx.doi.org/10.1289/ehp.99107257

     Gulson. B; Mizon. KJ: Korsch. MJ: Palmer.  JM; Donnelly. JB. (2003). Mobilization of lead from human
        bone tissue during pregnancy and lactation: A summary of long-term research. Sci Total Environ 303:
        79-104. http://dx.doi.org/10.1016/S0048-9697(02)00355-8

     Gulson. B; Mizon. KJ: Palmer. JM; Korsch.  MJ: Taylor. AJ: Mahaffev. KR. (2004a). Blood lead changes
        during pregnancy and postpartum with calcium supplementation. Environ Health Perspect 112:
        1499-1507. http://dx.doi.org/10.1289/ehp.6548

     Gulson. BL; Cameron. MA; Smith. AJ: Mizon. KJ: Korsch.  MJ: Vimpani. G: McMichael. AJ: Pisaniello.
        D; Jameson. CW: Mahaffev. KR. (1998c). Blood lead urine lead relationships in adults and children.
        Environ Res 2: 152-160. http://dx.doi.org/10.1006/enrs.1997.3810

     Gump. BB; Mackenzie. JA; Bendinskas. K;  Morgan. R; Dumas. AK; Palmer. CD; Parsons. PJ. (2011).
        Low-level Pb and cardiovascular responses to acute stress in children: the role of cardiac autonomic
        regulation. Neurotoxicol Teratol 33: 212-219. http://dx.doi.Org/10.1016/i.ntt.2010.10.001

     Gump. BB; Reihman. J; Stewart. P; Lonky. E; Darvill. T; Matthews. KA. (2007). Blood lead (Pb) levels:
        A potential environmental mechanism explaining the  relation between socioeconomic status and
        cardiovascular reactivity in children. Health Psychol 26:  296-304.
        http://dx.doi.0rg/10.1037/0278-6133.26.3.296

     Gump. BB; Reihman. J: Stewart. P: Lonky. E; Granger. DA; Matthews. KA. (2009). Blood lead (Pb)
        levels: Further evidence for an environmental mechanism explaining the association between
        socioeconomic status and psychophysiological dysregulation in children. Health Psychol 28: 614-620.
        http://dx.doi.org/10.1037/a0015611

     Gump. BB; Stewart. P: Reihman. J: Lonky. E; Darvill. T; Matthews. KA; Parsons. PJ. (2005). Prenatal
        and early childhood blood lead levels and cardiovascular functioning in 9 1/2 year old children.
        Neurotoxicol Teratol 27: 655-665. http://dx.doi.Org/10.1016/i.ntt.2005.04.002

     Gump. BB; Stewart. P: Reihman. J: Lonky. E; Darvill. T; Parsons. PJ: Granger. DA. (2008). Low-level
        prenatal and postnatal blood lead exposure and adrenocortical responses to acute stress in children.
        Environ Health Perspect 116: 249-255. http://dx.doi.org/10.1289/ehp.10391

     Gundacker. C: Frohlich. S: Graf-Rohrmeister. K; Eibenberger. B; Jessenig. V; Gicic. D: Prinz. S:
        Wittmann. KJ: Zeisler. H; Valiant. B; Pollak. A; Husslein. P. (2010). Perinatal lead and mercury
        exposure in Austria. Sci Total Environ 408: 5744-5749.
        http://dx.doi.0rg/10.1016/i.scitotenv.2010.07.079
November 2012                             5-736                   Draft - Do Not Cite or Quote

-------
     Guo. TL; Mudzinski. SP; Lawrence. DA. (1996a). The heavy metal lead modulates the expression of both
        TNF-alpha and TNF-alpha receptors in lipopolysaccharide-activated human peripheral blood
        mononuclear cells. J Leukoc Biol 59: 932-939.

     Guo. TL; Mudzinski. SP: Lawrence. DA. (1996b). Regulation of HLA-DR and invariant chain expression
        by human peripheral blood mononuclear cells with lead, interferon-gamma, or interleukin-4. Cell
        Immunol 171: 1-9. http://dx.doi.org/10.1006/cimm.1996.0166

     Gupta. P: Husain. MM; Shankar. R; Seth. PK; Maheshwari. RK. (2002). Lead exposure enhances virus
        multiplication and pathogenesis in mice. Vet Hum Toxicol 44: 205-210.

     Gustafson. A; Hedner. P;  Schiitz. A; Skerfving. S. (1989). Occupational lead exposure and pituitary
        function. Int Arch Occup Environ Health 61: 277-281. http://dx.doi.org/10.1007/BF00381426

     Gustavsson. P;  Plato. N; Hallqvist. J: Hogstedt. C: Lewne. M; Reuterwall. C: Scheele. P. (2001). A
        population-based case-referent study of myocardial infarction and occupational exposure to motor
        exhaust, other combustion products, organic solvents, lead, and dynamite Stockholm heart
        epidemiology program (SHEEP) study group. Epidemiology 12: 222-228.

     Habermann. E; Crowell. K; Janicki. P. (1983). Lead and other metals can substitute for Ca2+ in
        calmodulin. Arch Toxicol 54: 61-70. http://dx.doi.org/10.1007/BF00277816

     Hage. FG:  Szalai. AJ. (2007). C-reactive protein gene polymorphisms, C-reactive protein blood levels,
        and cardiovascular disease risk [Review]. JAm Coll Cardiol 50: 1115-1122.
        http://dx.doi.0rg/10.1016/i.iacc.2007.06.012

     HaMai. D; Bondv. SC. (2004). Oxidative basis of manganese neurotoxicity [Review]. Ann N Y Acad Sci
        1012: 129-141. http://dx.doi.org/10.1196/annals.1306.010

     Hamed. EA; Meki. AR; Abd El-Mottaleb. NA. (2010). Protective effect of green tea on lead-induced
        oxidative damage in rat's blood and brain tissue homogenates. J Physiol Biochem 66: 143-151.
        http://dx.doi.org/10.1007/sl3105-010-0019-5

     Hanas. JS: Rodgers. JS: Bantle. JA; Cheng. YG. (1999). Lead inhibition of DNA-binding mechanism of
        Cys(2)His(2) zinc finger proteins. Mol Pharmacol 56: 982-988.

     Hansen. JA; Audet. D; Spears. BL; Healy. KA; Brazzle. RE; Hoffman. DJ; Dailev. A; Beyer. WN.
        (201 la). Lead exposure and poisoning of songbirds using the Coeur d'Alene River Basin, Idaho, USA.
        Integr Environ Assess Manag 7: 587-595. http://dx.doi.org/10.1002/ieam.201

     Hara. N; Yamada. K; Terashima. M; Osago. H; Shimoyama. M; Tsuchiya. M. (2003). Molecular
        identification of human glutamine- and ammonia-dependent NAD synthetases: Carbon-nitrogen
        hydrolase domain confers glutamine dependency. J Biol Chem 278: 10914-10921.
        http://dx.doi.org/10.1074/ibc.M209203200

     Haraguchi. T; Ishizu. H; Takehisa. Y; Kawai. K; Yokota. O: Terada. S: Tsuchiya. K; Ikeda. K; Morita. K;
        Horike. T; Kira. S: Kuroda. S. (2001). Lead content of brain tissue in diffuse neurofibrillary tangles
        with calcification (DNTC): the possibility of lead neurotoxicity. Neuroreport 12:  3887-3890.

     Harry. GJ:  Schmitt TJ; Gong. Z; Brown. H; Zawia. N; Evans. HL. (1996). Lead-induced alterations of
        glial fibrillary acidic protein (GFAP) in the developing rat brain. Toxicol Appl Pharmacol 139: 84-93.
        http://dx.doi.org/10.1006/taap. 1996.0145

     Hartwig. A; Schlepegrell. R; Beversmann. D. (1990).  Indirect mechanism of lead-induced genotoxicity in
        cultured mammalian cells. Mutat Res Genet Toxicol 241: 75-82.
        http://dx.doi.org/10.1016/0165-1218(90)90110-N

     Hashemzadeh-Gargari. H; Guilarte. TR. (1999). Divalent cations modulate N-methyl-D-aspartate receptor
        function at the glycine site. J Pharmacol Exp Ther 290: 1356-1362.

     Hashmi. NS: Kachru. DN; Khandelwal. S: Tandon. SK. (1989). Interrelationship between iron deficiency
        and lead intoxication (part 2). Biol Trace Elem Res 22: 299-307.
        http://dx.doi.org/10.1007/BF02916618
November 2012                             5-737                  Draft - Do Not Cite or Quote

-------
     Hauser. R; Sergevev. O: Korrick. S: Lee. MM; Revich. B; Gitin. E; Burns. JS: Williams. PL. (2008).
        Association of blood lead levels with onset of puberty in Russian boys. Environ Health Perspect 116:
        976-980. http://dx.doi.org/10.1289/ehp.10516

     Hausmann. OV; Gentinetta. T; Fux. M; Ducrest. S: Pichler. WJ: Dahinden. CA. (2011). Robust
        expression of CCR3 as a single basophil selection marker in flow cytometry. Allergy 66: 85-91.
        http://dx.doi.0rg/10.llll/i.1398-9995.2010.02431.x

     He. LH; Poblenz. AT; Medrano. CJ: Fox. DA. (2000). Lead and calcium produce rod photoreceptor cell
        apoptosis by opening the mitochondria! permeability transition pore. J Biol Chem 275: 12175-12184.
        http://dx.doi.org/10.1074/ibc.275.16.12175

     He. S: Dong. W: Deng. Q; Weng. S: Sun. W. (2003). Seeing more clearly: Recent advances in
        understanding retinal circuitry [Review].  Science 302: 408-411.
        http://dx.doi.org/10.1126/science.1085457

     He. T; Hirsch. HVB; Ruden. DM; Lnenicka. GA. (2009). Chronic lead exposure alters presynaptic
        calcium regulation and synaptic facilitation in Drosophila larvae. Neurotoxicology 30: 777-784.
        http://dx.doi.0rg/10.1016/i.neuro.2009.08.007

     Hechtenberg. S: Beversmann. D. (1991). Inhibition of sarcoplasmic reticulum Ca2+ATPase activity by
        cadmium, lead and mercury.  Enzyme 45:  109-115.

     Hegazy. RM; Hamdv. R; Kamel. HF. (2011). Modulation of IgE levels in lead exposed children by
        parental cigarette smoking, Qualyobia Governate, Egypt. Int J Pharma Bio Sci 2: 372-385.
     Heier. LS: Lien. IB; Stromseng. AE; Ljones. M; Rosseland. BO; Tollefsen. KE; Salbu. B. (2009).
        Speciation of lead, copper, zinc and antimony in water draining  a shooting range: Time dependant
        metal accumulation and biomarker responses in brown trout (Salmo trutta L.). Sci Total Environ 407:
        4047-4055. http://dx.doi.Org/10.1016/i.scitotenv.2009.03.002

     Heiman. AS: Tonner. LE. (1995). The acute effect of lead acetate on glucocorticoid regulation of tyrosine
        aminotransferase in hepatoma cells. Toxicology 100: 57-68.
        http://dx.doi.org/10.1016/0300-483X(95)03061-J

     Hellstrom. L; Blinder. CG: Dahlberg. B; Lundberg. M; Jarup. L; Persson. B; Axelson. O. (2001).
        Cadmium exposure and end-stage renal disease. Am J Kidney Dis 38: 1001-1008.
        http://dx.doi.org/10.1053/aikd.2001.28589

     Hemdan. NYA; Emmrich. F; Adham. K; Wichmann. G: Lehmann. I; El-Massrv. A; Ghoneim. H;
        Lehmann. J:  Sack. U. (2005). Dose-dependent modulation of the in vitro cytokine production of
        human immune competent cells by lead salts. Toxicol Sci 86:  75-83.
        http://dx.doi.org/10.1093/toxsci/kfil77

     Hemphill. FE; Kaeberle. ML;  Buck. WB. (1971). Lead suppression of mouse resistance to Salmonella
        typhimurium. Science 172: 1031-1032. http://dx.doi.org/10.1126/science. 172.3987.1031

     Hengstler. JG: Bolm-Audorff. U; Faldum. A; Janssen. K; Reifenrath. M; Gotte. W: Jung. D; Maver-
        Popken. O; Fuchs. J: Gebhard. S: Bienfait. HG: Schlink. K; Dietrich. C: Faust. D; Epe. B; Oesch. F.
        (2003). Occupational exposure to heavy metals: DNA damage induction and DNA repair inhibition
        prove co-exposures to cadmium, cobalt and lead as more dangerous than hitherto expected.
        Carcinogenesis 24: 63-73.  http://dx.doi.Org/10.1093/carcin/24.l.63

     Heo. Y; Lee. BK; Ahn. KD; Lawrence. DA. (2004). Serum IgE elevation correlates with blood lead levels
        in battery manufacturing workers. Hum Exp Toxicol 23: 209-213.
        http://dx.doi.org/10.1191/0960327104ht442oa

     Heo. Y; Lee. WT; Lawrence. DA. (1997). In vivo the environmental pollutants lead and mercury induce
        oligoclonal T cell responses skewed toward type-2 reactivities. Cell Immunol 179: 185-195.
        http://dx.doi.org/10.1006/cimm.1997.1160
November 2012                             5-738                  Draft - Do Not Cite or Quote

-------
     Heo. Y; Lee. WT; Lawrence. DA. (1998). Differential effects of lead and cAMP on development and
        activities of Thl- and Th2-lymphocytes. Toxicol Sci 43: 172-185.
        http://dx.doi.0rg/10.1093/toxsci/43.2.172

     Heo. Y; Mondal. TK; Gao. D; Kasten-Jolly. J; Kishikawa. H; Lawrence. DA. (2007). Posttranscriptional
        inhibition of interferon-gamma production by lead. Toxicol Sci 96: 92-100.
        http://dx.doi.org/10.1093/toxsci/kfll82

     Heo. Y; Parsons. PJ: Lawrence. DA. (1996). Lead differentially modifies cytokine production in vitro and
        in vivo. Toxicol Appl Pharmacol 138:  149-157. http://dx.doi.org/10.1006/taap.1996.0108

     Herman. DS: Geraldine. M; T. V. (2009). Influence of minerals on lead-induced alterations in liver
        function in rats exposed to long-term lead exposure. J Hazard Mater 166: 1410-1414.
        http://dx.doi.0rg/10.1016/i.jhazmat.2008.12.070

     Hermes-Lima. M; Pereira. B; Bechara. EJH. (1991). Are free radicals involved in lead poisoning?
        [Review]. Xenobiotica 21: 1085-1090. http://dx.doi.org/10.3109/00498259109039548

     Hernandez-Avila. M; Smith. D; Meneses. F; Sanin. LH; Hu. H. (1998). The influence of bone and blood
        lead on plasma lead levels in environmentally exposed adults. Environ Health Perspect 106: 473-477.
        http://dx.doi.org/10.1289/ehp.98106473

     Hernandez-Franco. P; Silva. M; Valverde. M; Rojas. E. (2011). Induction of oxidative stress by low doses
        of lead in human hepatic  cell line WRL-68. Biometals 24: 951-958.
        http://dx.doi.org/10.1007/sl0534-011-9453-2

     Hernandez-Ochoa. I; Sanchez-Gutierrez. M; Solis-Heredia. MJ; Quintanilla-Vega. B. (2006).
        Spermatozoa nucleus takes up lead during the epididymal maturation altering chromatin condensation.
        Reprod Toxicol 21: 171-178. http://dx.doi.Org/10.1016/i.reprotox.2005.07.015

     Hernandez-Serrato. MI; Fortoul. TI; Rojas-Martinez. R; Mendoza-Alvarado. LR; Canales-Trevino. L;
        Bochichio-Riccardelli. T; Avila-Costa. MR; Olaiz-Fernandez. G. (2006).  Lead blood concentrations
        and renal function evaluation:  Study in an exposed Mexican population. Environ Res 100: 227-231.
        http://dx.doi.0rg/10.1016/i.envres.2005.03.004

     Herold. C. (2010). NMDA and D2-like receptors modulate cognitive flexibility in a color discrimination
        reversal task in pigeons. Behav Neurosci 124: 381-390. http://dx.doi.org/10.1037/a0019504

     Hevdari. A; Norouzzadeh. A; Khoshbaten. A; Asgari. A; Ghasemi. A; Naiafi. S: Badalzadeh. R. (2006).
        Effects of short-term and subchronic lead poisoning on nitric oxide metabolites and vascular
        responsiveness in rat. Toxicol Lett 166: 88-94. http://dx.doi.Org/10.1016/i.toxlet.2006.05.014

     Hilbertz. U; Kramer. U; De Ruiter. N; Baginski. B. (1986). Effects of cadmium and lead on oxidative
        metabolism and phagocytosis by mouse peritoneal macrophages. Toxicology 39: 47-57.

     Hilson. JA; Strupp. BJ. (1997). Analyses of response patterns clarify lead effects in olfactory reversal and
        extradimensional shift tasks: Assessment of inhibitory control, associative ability, and memory. Behav
        Neurosci 111: 532-542.

     Holdstein. Y; Pratt. H; Goldsher. M; Rosen. G: Shenhav. R: Linn. S: Mor. A; Barkai. A. (1986). Auditory
        brainstem evoked potentials in asymptomatic lead-exposed subjects. J Laryngol Otol 100: 1031-1036.

     Hollowav. WR. Jr; Thor. DH. (1987). Low level lead exposure during lactation increases rough and
        tumble play fighting of juvenile rats. Neurotoxicol Teratol 9: 51-57.

     Holmes. AL; Wise. SS: Sandwick. SJ: Lingle. WL; Negron. VC: Thompson. WD; Wise. JP. Sr. (2006a).
        Chronic exposure to lead chromate causes centrosome abnormalities and aneuploidy in human lung
        cells. Cancer Res 66: 4041-4048. http://dx.doi.org/10.1158/0008-5472.can-05-3312

     Holmes. AL; Wise. SS: Sandwick. SJ: Wise. JP. (2006b). The clastogenic effects of chronic exposure to
        paniculate and soluble Cr(VI) in human lung cells. Mutat Res Genet Toxicol Environ Mutagen 610:
        8-13. http://dx.doi.0rg/10.1016/i.mrgentox.2006.06.006
November 2012                              5-739                  Draft - Do Not Cite or Quote

-------
     Holtzman. D; Olson. JE; DeVries. C: Bensch. K. (1987). Lead toxicity in primary cultured cerebral
        astrocytes and cerebellar granular neurons. Toxicol Appl Pharmacol 89: 211-225.
        http://dx.doi.org/10.1016/0041-008X(87)90042-l

     Hon. KL; Ching. GK; Hung. EC: Leung. TF. (2009). Serum lead levels in childhood eczema [Letter]. Clin
        Exp Dermatol 34: e508-e509. http://dx.doi.Org/10.llll/i.1365-2230.2009.03596.x

     Hon. KLE; Wang. SS: Hung. ECW: Lam. HS: Lui. HHK; Chow. CM; Ching. GKW: Fok. TF; Ng. PC:
        Leung. TF. (2010). Serum levels of heavy metals in childhood eczema and skin diseases: Friends or
        foes. Pediatric Allergy and Immunology 21: 831-836.
        http://dx.doi.0rg/10.llll/i.1399-3038.2010.01022.x

     Hong. YC: Hwang. SS: Kim. JH; Lee. KH; Lee. HJ; Lee. KH; Yu. SD; Kim. PS. (2007b). Metals in
        particulate pollutants affect peak expiratory flow of schoolchildren. Environ Health Perspect 115:
        430-434. http://dx.doi.org/10.1289/ehp.9531

     Horiguchi. S: Matsumura. S: Fukumoto. K; Karai. I; Endo. G: Teramoto. K; Shinagawa. K; Kiyota. I;
        Wakitani. F; Takise. S: Kawarava. T. (1991). Erythrocyte deformability in workers exposed to lead.
        Osaka City Med J 37: 149-155.

     Hornung. R. (2008). Email from Richard Hornung, Cincinnati Children's Hospital Medical Center
        (CCHMC) to Jee-Young Kim, USEPA Subject: Request for additional information on Lanphear et al.
        (2005). (EPA-HQ-OAR-2006-0735-5814). U.S. Environmental Protection Agency.
        http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OAR-2006-0735-5814

     Hornung. R; Lanphear. B; Kietrich. K. (2006). Response to: "What is the meaning of non-linear dose-
        response relationships between blood lead concentration and IQ?" [letter] [Letter]. Neurotoxicology
        27: 635.

     Hornung. RW: Lanphear. BP; Dietrich. KN. (2009). Age of greatest susceptibility to childhood lead
        exposure: Anew statistical approach. Environ Health Perspect  117: 1309-1312.
        http://dx.doi.org/10.1289/ehp.0800426

     Hotz. P; Buchet. JP; Bernard. A; Lison. D; Lauwerys. R. (1999). Renal effects of low-level environmental
        cadmium exposure: 5-year follow-up of a subcohort from the Cadmibel study. 354: 1508-1513.
        http://dx.doi.org/10.1016/S0140-6736(99)91145-5

     Howard. JK. (1974). Human erythrocyte glutathione reductase and glucose 6-phosphate dehydrogenase
        activities in normal subjects and in persons exposed to lead. Clin Sci Mol Med 47: 515-520.

     Hsiao. CL; Wu. KH: Wan. KS. (2011). Effects of environmental lead exposure on T-helper cell-specific
        cytokines in children. J Immunotoxicol 8: 284-287. http://dx.doi.org/10.3109/1547691X.2011.592162

     Hsiao. CY; Wu. HDI; Lai. JS: Kuo. HW. (2001). A longitudinal study of the effects of long-term exposure
        to lead among lead battery factory workers in Taiwan (1989-1999). Sci Total Environ 279: 151-158.
        http://dx.doi.org/10.1016/S0048-9697(01)00762-8

     Hsieh. SJ: Chiu. YW: Li. WF; Wu. CH; Chen. HI; Chuang. HY. (2009a). Increased concentrations of
        serum inhibin B among male workers with long-term moderate lead exposure. Sci Total Environ 407:
        2603-2607. http://dx.doi.Org/10.1016/i.scitotenv.2008.12.056

     Hsieh. TJ; Chen. YC: Li. CW: Liu. GC: Chiu. YW: Chuang. HY. (2009b). A proton magnetic resonance
        spectroscopy study of the chronic lead effect on the basal ganglion and frontal and occipital lobes in
        middle-age adults. Environ Health Perspect 117: 941-945. http://dx.doi.org/10.1289/ehp.0800187

     Hsu. CW: Lin. JL: Lin-Tan. DT: Yen. TH: Huang. WH: Ho. TC: Huang. YL: Yeh. LM: Huang. LM.
        (2009a). Association of environmental cadmium exposure with inflammation and malnutrition in
        maintenance haemodialysis patients. Nephrol Dial Transplant 24: 1282-1288.
        http://dx.doi.org/10.1093/ndt/gfn602

     Hsu. JM.  (1981). Lead toxicity as related to glutathione metabolism. JNutr 111: 26-33.
November 2012                             5-740                  Draft - Do Not Cite or Quote

-------
     Hsu. PC: Chang. HY; Quo. YL; Liu. YC: Shih. TS. (2009b). Effect of smoking on blood lead levels in
        workers and role of reactive oxygen species in lead-induced sperm chromatin DNA damage. Fertil
        Steril91: 1096-1103. http://dx.doi.Org/10.1016/i.fertnstert.2008.01.005

     Hsu. PC: Hsu. CC: Liu. MY; Chen. LY; Quo. YL. (1998a). Lead-induced changes in spermatozoa
        function and metabolism. J Toxicol Environ Health A 55: 45-64.
        http://dx.doi.org/10.1080/009841098158610

     Hsu. PC: Liu. MY; Hsu. CC: Chen. LY; Quo. YL. (1998b). Effects of vitamin E and/or C on reactive
        oxygen species-related lead toxicity in the rat sperm. Toxicology 128: 169-179.
        http://dx.doi.org/10.1016/S0300-483X(98)00068-7

     Hu. H. (1991). A 50-year follow-up of childhood plumbism: Hypertension, renal function, and
        hemoglobin levels among survivors. Am J Dis Child 145: 681-687.

     Hu. H; Aro. A; Pavton. M; Korrick. S: Sparrow. (1996a). The relationship of bone and blood lead to
        hypertension: The Normative Aging Study. JAMA 275: 1171-1176.
        http://dx.doi.org/10.1001/iama.1996.03530390037031

     Hu. H; Hernandez-Avila. M. (2002). Lead, bones, women, and pregnancy: The poison within? Am J
        Epidemiol 156: 1088-1091. http://dx.doi.org/10.1093/aje/kwfl64

     Hu. H; Pavton. M; Korrick.  S: Aro. A; Sparrow. D; Weiss. ST; Rotnitzky. A. (1996b). Determinants of
        bone and blood lead levels among community-exposed middle-aged to elderly men: The Normative
        Aging Study. Am J Epidemiol 144: 749-759.

     Hu. H; Tellez-Rojo. MM; Bellinger. D; Smith. D; Ettinger. AS: Lamadrid-Figueroa. H; Schwartz. J:
        Schnaas. L; Mercado-Garca. A; Hernandez-Avila. M. (2006). Fetal lead exposure at each stage of
        pregnancy as a predictor of infant mental development. Environ Health Perspect 114:  1730-1735.
        http://dx.doi.org/10.1289/ehp.9067

     Hu. P; Wang. M; Chen. WH; Liu. J: Chen. L; Yin. ST; Yong. W: Chen. JT; Wang. HL; Ruan. DY
        (2008a). Quercetin relieves chronic lead exposure-induced impairment of synaptic plasticity in rat
        dentate gyrus in vivo. Naunyn Schmiedebergs Arch Pharmacol 378: 43-51.
        http://dx.doi.org/10.1007/s00210-008-0301-z

     Hu. OS: Fu. HJ: Ren. TL: Wang. SY: Zhou. W: Song. H: Han. YF: Dong. SZ. (2008b). Maternal
        low-level lead exposure reduces the expression of PSA-NCAM and the activity of sialyltransferase in
        the hippocampi of neonatal rat pups. Neurotoxicology 29: 675-681.
        http://dx.doi.0rg/10.1016/i.neuro.2008.04.002

     Huang.  BM; Lai. HY; Liu. MY. (2002). Concentration dependency in lead-inhibited steroidogenesis in
        MA-10 mouse Leydig tumor cells. J Toxicol Environ Health A 65: 557-567.
        http://dx.doi.org/10.1080/15287390252808000

     Huang.  BM; Liu. MY. (2004). Inhibitory actions of lead on steroidogenesis in MA-10 mouse Leydig
        tumor cells. Arch Androl 50: 5-9. http://dx.doi.org/10.1080/01485010490250434

     Huang.  H; Bihaqi. SW: Cui. L; Zawia. NH. (2011 a). In vitro Pb exposure disturbs the balance between
        Ap production and elimination: the role of ApPP and neprilysin. Neurotoxicology 32:  300-306.
        http://dx.doi.0rg/10.1016/i.neuro.2011.02.001

     Huang.  J; Wu. J: Li. T; Song. X; Zhang. B; Zhang. P; Zheng. X. (20lib). Effect of exposure to  trace
        elements in the soil on the prevalence of neural tube defects in a high-risk area of China. Biomed
        Environ Sci 24: 94-101. http://dx.doi.org/10.3967/0895-3988.2011.02.002

     Huang.  M; Krepkiy. D; Hu. W: Petering. PH. (2004). Zn-, Cd-, and Pb-transcription factor IIIA:
        Properties, DNA binding, and  comparison with TFIIIA-fmger 3 metal complexes. J Inorg Biochem 98:
        775-785. http://dx.doi.0rg/10.1016/i.iinorgbio.2004.01.014

     Huang.  XP; Feng. ZY; Zhai. WL; Xu. JH. (1988). Chromosomal aberrations and sister chromatid
        exchanges in workers exposed to lead. Biomed Environ Sci 1: 382-387.
November 2012                              5-741                  Draft - Do Not Cite or Quote

-------
     Hudson. CA; Cao. L; Kasten-Jolly. J; Kirkwood. JN; Lawrence. DA. (2003). Susceptibility of lupus-
        prone NZM mouse strains to lead exacerbation of systemic lupus erythematosus symptoms. J Toxicol
        Environ Health A 66: 895-918. http://dx.doi.org/10.1080/15287390306456

     Huel. G: Sahuquillo. J: Debotte. G: Oury. JF; Takser. L. (2008). Hair mercury negatively correlates with
        calcium pump activity in human term newborns and their mothers at delivery. Environ Health Perspect
        116: 263-267. http://dx.doi.org/10.1289/ehp.10381

     Huo. X; Peng. L; Xu. XJ; Zheng. LK; Qiu. B; Qi. ZL; Zhang. B; Han. D; Piao. ZX. (2007). Elevated
        blood lead levels of children in Guiyu, an electronic waste recycling town in China. Environ Health
        Perspect 115: 1113-1117. http://dx.doi.org/10.1289/ehp.9697

     Huseman. CA; Moriartv. CM; Angle. CR. (1987). Childhood lead toxicity and impaired release of
        thyrotropin-stimulating hormone. Environ Res 42: 524-533.
        http://dx.doi.org/10.1016/80013-9351(87)80219-0

     Huseman. CA; Varma. MM; Angle. CR. (1992). Neuroendocrine effects of toxic and low blood lead
        levels in children [Letter]. Pediatrics 90: 186-189.

     Hwang. KY; Lee. BK; Bressler. JP; Bolla. KI; Stewart. WF; Schwartz. BS. (2002). Protein kinase C
        activity and the relations between blood lead and neurobehavioral function in lead workers. Environ
        Health Perspect 110: 133-138. http://dx.doi.org/10.1289/ehp.02110133

     Hwang. YH; Chiang. HY; Yen-Jean. MC: Wang. JD.  (2009). The association between low levels of lead
        in blood and occupational noise-induced hearing loss in steel workers. Sci Total Environ 408: 43-49.
        http://dx.doi.0rg/10.1016/i.scitotenv.2009.09.016

     Hwua. YS: Yang. JL. (1998). Effect of 3-aminotriazole on anchorage independence and mutagenicity in
        cadmium- and lead-treated diploid human fibroblasts. Carcinogenesis 19: 881-888.
        http://dx.doi.0rg/10.1093/carcin/19.5.881

     IARC (International Agency for Research on Cancer). (2006a). Inorganic and organic lead compounds.
        Lyon, France. http://monographs.iarc.fr/ENG/Monographs/vol87/index.php

     lavicoli. I; Carelli. G: III. SE; Castellino. N; Calabrese. EJ. (2004). Effects of low doses of dietary lead on
        puberty  onset in female mice. Reprod Toxicol 19: 35-41.
        http://dx.doi.0rg/10.1016/i.reprotox.2004.06.013

     lavicoli. I; Carelli. G: Stanek. EJ: Castellino. N; Li. Z; Calabrese. EJ. (2006a). Low doses of dietary lead
        are associated with a profound reduction in the time to the onset of puberty in female mice. Reprod
        Toxicol 22: 586-590. http://dx.doi.Org/10.1016/i.reprotox.2006.03.016

     lavicoli. I: Carelli. G: Stanek. EJ. Ill; Castellino. N; Calabrese. EJ. (2006b). Below background levels of
        blood lead impact cytokine levels in male and female mice. Toxicol Appl Pharmacol 210: 94-99.
        http://dx.doi.0rg/10.1016/i.taap.2005.09.016

     lavicoli. I: Sgambato. A; Carelli. G: Ardito. R; Cittadini. A; Castellino. N. (2001). Lead-related effects on
        rat fibroblasts. Mol Cell Biochem 222: 35-40. http://dx.doi.Org/10.1023/A:1017930401906

     Ignasiak. Z; Slawinska. T; Rozek. K; Little. BB; Malina. RM. (2006). Lead and growth status of
        schoolchildren living in the copper basin of south-western Poland: Differential effects on bone growth.
        Ann Hum Biol 33: 401-414. http://dx.doi.org/10.1080/03014460600730752

     lijima. K; Otake. T; Yoshinaga. J: Ikegami. M;  Suzuki. E; Naruse. H; Yamanaka. T; Shibuva. N;
        Yasumizu. T; Kato. N. (2007). Cadmium, lead, and selenium in cord blood and thyroid hormone status
        of newborns. Biol Trace Elem Res 119: 10-18. http://dx.doi.org/10.1007/sl2011-007-0057-l

     Ikebuchi. H; Teshima. R; Suzuki. K; Terao. T; Yamane. Y. (1986). Simultaneous induction of
        Pb-metallothionein-like protein and Zn-thionein in the liver of rats given lead acetate. Biochem J 233:
        541-546.

     Ingelsson. E; Gona. P; Larson. MG: Lloyd-Jones. DM; Kannel. WB; Vasan. RS: Lew. D. (2008). Altered
        blood pressure progression in the community and its relation to clinical events. Arch Intern Med 168:
        1450-1457. http://dx.doi.org/10.1001/archinte.168.13.1450
November 2012                              5-742                  Draft - Do Not Cite or Quote

-------
     Inglis. JA; Henderson. DA; Emmerson. BT. (1978). The pathology and pathogenesis of chronic lead
        nephropathy occurring in Queensland. J Pathol 124: 65-76. http://dx.doi.org/10.1002/path.1711240202

     Institoris. L; Kovacs. D; Kecskemeti-Kovacs. I; Lukacs. A; Szabo. A; Lengyel. Z; Papp. A;
        Nagymaitenyi. L; Desi. I. (2006). Immunotoxicological investigation of subacute combined exposure
        with low doses of PB, HG and CD in rats. Acta Biol Hung 57: 433-439.
        http://dx.doi.Org/10.1556/ABiol.57.2006.4.5

     lonescu. JG: Novotny. J; Stejskal. V: Latsch. A; Blaurock-Busch. E; Eisenmann-Klein. M. (2007). Breast
        tumours strongly accumulate transition metals. Maedica 2: 5-9.

     Iranpour. R; Besharati. AA; Nasseri. F; Hashemipour. M; Balali-Mood. M; Kelishadi. R. (2007).
        Comparison of blood lead levels of mothers and cord blood in intrauterine growth retarded neonates
        and normal term neonates. Saudi Med J 28: 877-880.

     Irgens. A; Kruger. K; Skorve. AH; Irgens. LM. (1998). Reproductive outcome  in offspring of parents
        occupationally exposed to lead in Norway. Am J Ind Med 34: 431-437.

     Iwata. T; Yano. E; Karita. K; Dakeishi. M; Murata. K. (2005). Critical dose of lead affecting postural
        balance in workers. Am J Ind Med 48: 319-325. http://dx.doi.org/10.1002/ajim.20220

     Jabeen. R; Tahir. M; Waqas. S. (2010). Teratogenic effects of lead acetate on kidney. J Ayub Med Coll
        Abbottabad 22: 76-79.

     Jackson. LW: Howards. PP; Wactawski-Wende. J: Schisterman. EF. (2011). The association between
        cadmium, lead and mercury blood levels and reproductive hormones among healthy, premenopausal
        women. Hum  Reprod 26: 2887-2895. http://dx.doi.org/10.1093/humrep/der250

     Jadhav. SH; Sarkar. SN; Ram. GC: Tripathi. HC. (2007). Immunosuppressive effect of subchronic
        exposure to a mixture of eight heavy metals, found as groundwater contaminants in different areas of
        India, through drinking water in male rats. Arch Environ Contam Toxicol 53: 450-458.
        http://dx.doi.org/10.1007/s00244-006-0177-l

     Jagetia. GC: Aruna. R. (1998). Effect of various concentrations of lead nitrate on the induction of
        micronuclei in mouse bone marrow. Mutat Res Genet Toxicol Environ Mutagen 415: 131-137.
        http://dx.doi.org/10.1016/S1383-5718(98)00052-7

     Jain. NB; Potula.  V: Schwartz. J: Vokonas. PS: Sparrow. D; Wright. RO: Nie. H; Hu. H. (2007). Lead
        levels and ischemic heart disease in a prospective study of middle-aged and elderly men: The VA
        Normative Aging Study. Environ Health Perspect 115: 871-875. http://dx.doi.org/10.1289/ehp.9629

     Jamieson. JA; Shuhyta. JN; Taylor. CG. (2007). Lead does not affect transcription of intestinal zinc-
        binding proteins in growing rats. Exp Biol Med 232: 744-753.

     Jamieson. JA: Stringer. DM; Zahradka. P; Taylor. CG. (2008). Dietary zinc attenuates renal lead
        deposition but metallothionein is not directly involved. Biometals 21: 29-40.
        http://dx.doi.org/10.1007/sl0534-007-9090-v

     Jang. HO; Kim. JS: Kwon. WC: Kim. JK; Ko. MS: Kim. DH; Kim. WI; Jeon. YC: Chung. IK; Shin. SH;
        Chung. J; Bae. MK; Yun. I. (2008). The effect of lead on calcium release activated calcium influx in
        primary cultures of human osteoblast-like cells. Arch Pharm Res 31: 188-194.
        http://dx.doi.org/10.1007/sl2272-001-1140-3

     Jang. WH: Lim. KM: Kim. K: Noh. JY: Kang. S: Chang. YK: Chung. JH. (2011). Low level of lead can
        induce phosphatidylserine exposure and erythrophagocytosis: Anew mechanism underlying lead-
        associated anemia. Toxicol Sci 122: 177-184. http://dx.doi.org/10.1093/toxsci/kfr079

     Janjua. NZ; Delzell. E; Larson. RR; Meleth. S: Kristensen. S: Kabagambe. E; Sathiakumar. N. (2009).
        Determinants of low birth weight in urban Pakistan. Public Health Nutr 12: 789-798.
        http://dx.doi.org/10.1017/S1368980008002942

     Javakumar. T; Sridhar. MP; Bharathprasad. TR; Ilavaraia. M; Govindasamy. S: Balasubramanian. MP
        (2009). Experimental studies of Achyranthes aspera (L) preventing nephrotoxicity induced by lead in
        albino rats. J Health Sci 55: 701-708. http://dx.doi.org/10.1248/ihs.55.701
November 2012                             5-743                  Draft - Do Not Cite or Quote

-------
     Jedrychowski. W: Perera. F; Maugeri. U; Miller. RL; Rembiasz. M; Flak. E; Mroz. E; Majewska. R;
        Zembala. M. (2011). Intrauterine exposure to lead may enhance sensitization to common inhalant
        allergens in early childhood: Aprospective prebirth cohort study. Environ Res 111: 119-124.
        http://dx.doi.0rg/10.1016/i.envres.2010.ll.002

     Jedrychowski. W: Perera. FP; Jankowski. J; Mrozek-Budzyn. D; Mroz. E; Flak. E; Edwards. S: Skarupa.
        A; Lisowska-Miszczyk. I. (2009b). Very low prenatal exposure to lead and mental development of
        children in infancy and early childhood: Krakow prospective cohort study. Neuroepidemiology 32:
        270-278. http://dx.doi.org/10.1159/000203075

     Jehan. ZS: Motlag. DB. (1995). Metal induced changes in the erythrocyte membrane of rats. Toxicol Lett
        78: 127-133. http://dx.doi.org/10.1016/0378-4274(94)03245-3

     Jelliffe-Pawlowski. LL; Miles. SO; Courtney. JG: Materna. B; Charlton. V. (2006). Effect of magnitude
        and timing of maternal pregnancy blood lead (Pb) levels on birth outcomes. J Perinatol 26:  154-162.
        http://dx.doi.org/10.1038/si.irj.7211453

     Jett. DA; Kuhlmann. AC: Farmer. SJ: Guilarte. TR. (1997). Age-dependent effects of developmental lead
        exposure on performance in the Morris water maze. Pharmacol Biochem Behav 57: 271-279.
        http://dx.doi.org/10.1016/S0091-3057(96)00350-4

     Jiang. YM; Long. LL; Zhu. XY; Zheng. H; Fu. X; Ou. SY; Wei. PL; Zhou. HL; Zheng. W. (2008).
        Evidence for altered hippocampal volume and brain metabolites in workers occupationally exposed to
        lead: A study by magnetic resonance imaging and 1H magnetic resonance spectroscopy. Toxicol Lett
        181:  118-125. http://dx.doi.0rg/10.1016/j.toxlet.2008.07.009

     Jin. YP;  Liao. YJ; Lu. CW: Li. GX; Yu. F; Zhi. XP; Xu. JH; Liu. SX; Liu. MM; Yang. J. (2006). Health
        effects in children aged 3-6 years induced by environmental lead exposure. Ecotoxicol Environ Saf
        63: 313-317. http://dx.doi.Org/10.1016/i.ecoenv.2005.05.011

     Johnson. BL; Elbatawi. M; Xintaras. C: Baker. EL. Jr; Hanninen. H; Seppalainen. AM. (1987).
        Prevention of neurotoxic illness in working populations. In BL Johnson; M Elbatawi; C Xintaras; EL
        Baker, Jr.; H Hanninen; AM Seppalainen (Eds.). New York: John Willey & Sons.

     Johnson. RJ; Kang. DH; Feig. D; Kivlighn. S; Kanellis. J; Watanabe. S; Turtle. KR; Rodriguez-Iturbe. B;
        Herrera-Acosta. J; Mazzali. M. (2003). Is there a pathogenetic role for uric acid in hypertension and
        cardiovascular and renal disease? Hypertension 41: 1183-1190.
        http://dx.doi.org/10.1161/01.HYP.0000069700.62727.C5

     Johnston. C; Mash. EJ; Miller. N; Ninowski. JE. (2012). Parenting in adults with attention-
        deficit/hyperactivity disorder (ADHD) [Review]. Clin Psychol Rev 32: 215-228.
        http://dx.doi.0rg/10.1016/i.cpr.2012.01.007

     Jomova. K; Jenisova. Z; Feszterova. M; Baros. S; Liska. J; Hudecova. D; Rhodes.  CJ; Valko. M. (2011).
        Arsenic: Toxicity, oxidative stress and human disease [Review]. J Appl Toxicol 31: 95-107.
        http://dx.doi.org/10.1002/iat.1649

     Jomova. K; Valko. M. (2011). Advances in metal-induced oxidative stress and human disease. Toxicology
        283:  65-87. http://dx.doi.Org/10.1016/j.tox.2011.03.001

     Jones. AYM; Lam. PKW; Dean. E. (2006). Respiratory health of bus drivers in Hong Kong. Int Arch
        Occup Environ Health 79: 414-418. http://dx.doi.org/10.1007/s00420-005-0061-8

     Jones. AYM; Lam. PKW; Gohel. MDI. (2008). Respiratory health of road-side vendors in a large
        industrialized city. Environ Sci Pollut Res Int 15: 150-154. http://dx.doi.org/10.1065/espr2006.12.368

     Jones. EA; Wright. JM; Rice. G; Buckley.  BT; Magsumbol. MS; Barr. DB; Williams. BL. (2010). Metal
        exposures in an inner-city neonatal population. Environ Int 36: 649-654.
        http://dx.doi.0rg/10.1016/i.envint.2010.04.007

     Jones. SR; Atkin. P; Holroyd. C; Lutman. E; Batlle. JVI; Wakeford. R; Walker. P. (2007). Lung cancer
        mortality at a U.K. tin smelter. Occup Med (Lond) 57: 238-245.
        http://dx.doi.org/10.1093/occmed/kqll53
November 2012                              5-744                   Draft - Do Not Cite or Quote

-------
     Joseph. CLM; Havstad. S: Ownbv. PR; Peterson. EL; Maliarik. M; McCabe. MJ. Jr; Barone. C: Johnson.
        CC. (2005). Blood lead level and risk of asthma. Environ Health Perspect 113: 900-904.
        http://dx.doi.org/10.1289/ehp.7453

     Jurczuk. M; Brzoska. MM; Moniuszko-Jakoniuk. J. (2007). Hepatic and renal concentrations of vitamins
        E and C in lead- and ethanol-exposed rats. An assessment of their involvement in the mechanisms of
        peroxidative damage.  Food Chem Toxicol 45: 1478-1486. http://dx.doi.Org/10.1016/i.fct.2007.02.007

     Jurczuk. M; Moniuszko-Jakoniuk. J; Brzoska. MM. (2006). Involvement of some low-molecular thiols in
        the peroxidative mechanisms of lead and ethanol action on rat liver and kidney. Toxicology 219:
        11-21. http://dx.doi.0rg/10.1016/i.tox.2005.10.022

     Jusko. TA; Henderson. CR; Lanphear. BP; Cory-Slechta. DA; Parsons. PJ: Canfield. RL. (2008). Blood
        lead concentrations <  10 microg/dL and child intelligence at 6 years of age. Environ Health Perspect
        116: 243-248. http://dx.doi.org/10.1289/ehp.10424

     Kaczynska. K; Walski. M; Szereda-Przestaszewska. M. (2011). Ultrastructural changes in lung tissue after
        acute lead intoxication in the rat. J Electron Microsc (Tokyo) 60: 289-294.
        http://dx.doi.org/10.1093/imicro/dfr035

     Kaji. T; Suzuki. M; Yamamoto. C: Mishima. A; Sakamoto. M; Kozuka. H. (1995). Severe damage of
        cultured vascular endothelial cell monolayer after simultaneous exposure to cadmium and lead. Arch
        Environ Contam Toxicol 28: 168-172. http://dx.doi.org/10.1007/BF00217612

     Kalman. J: Riba. I; Blasco. J: Delvalls. TA. (2008). Is delta-aminolevulinic acid dehydratase activity in
        bivalves from south-west Iberian Peninsula a good biomarker of lead exposure? Mar Environ Res 66:
        38-40. http://dx.doi.0rg/10.1016/j.marenvres.2008.02.016

     Kamal. M; Fathy. MM; Taher. E; Hasan. M; Tolba. M. (2011). Assessment of the role of paraoxonase
        gene polymorphism (Q192R) and paraoxonase activity in the susceptibility to atherosclerosis among
        lead-exposed workers. Ann Saudi Med 31: 481-487. http://dx.doi.org/10.4103/0256-4947.84625

     Kamel. F; Umbach. DM;  Hu. H; Munsat. TL; Shefner. JM; Taylor. JA; Sandier. DP. (2005). Lead
        exposure  as a risk factor for amyotrophic lateral sclerosis. Neurodegener Dis 2: 195-201.
        http://dx.doi.org/10.1159/000089625

     Kamel. F; Umbach. DM;  Munsat. TL; Shefner. JM; Hu. H; Sandier. DP.  (2002). Lead exposure and
        amyotrophic lateral sclerosis.  Epidemiology 13:  311-319.

     Kamel. F; Umbach. DM;  Stallone. L; Richards. M; Hu. H; Sandier. DP. (2008). Association of lead
        exposure  with survival in amyotrophic lateral sclerosis. Environ Health Perspect 116: 943-947.
        http://dx.doi.org/10.1289/ehp.11193

     Kanduc. D; Rossiello. MR; Aresta. A; Cavazza. C: Quagliariello. E; Farber. E. (1991). Transitory DNA
        hypomethylation during liver cell proliferation induced by a single dose of lead nitrate. Arch Biochem
        Biophys 286: 212-216. http://dx.doi.org/10.1016/0003-9861(91)90030-M

     Kannel. WB. (2000a). Elevated systolic blood pressure as a cardiovascular risk factor [Review]. Am J
        Cardiol 85: 251-255. http://dx.doi.org/10.1016/S0002-9149(99)00635-9

     Kannel. WB. (2000b). Risk stratification in hypertension: New insights from the Framingham Study. Am
        JHypertens 13:  3S-10S. http://dx.doi.org/10.1016/S0895-7061(99)00252-6

     Kapuku. GK; Harshfield. GA; Davis. HC: Treiber. FA. (2006). Early markers of cardiovascular disease
        [Review]. Vascul Pharmacol 45: 277-280. http://dx.doi.Org/10.1016/i.vph.2006.08.009

     Karaca. T; Simsek. N. (2007). Effects of spirulina on the number of ovary mast cells in lead-induced
        toxicity in rats. Phytother Res 21: 44-46. http://dx.doi.org/10.1002/ptr.2015

     Karimooy. HN; Mood. MB; Hosseini. M; Shadmanfar. S. (2010). Effects of occupational lead exposure
        on renal and nervous system of workers of traditional tile factories in Mashhad (northeast of Iran).
        Toxicol Ind Health 26: 633-638. http://dx.doi.org/10.1177/0748233710377774
November 2012                              5-745                   Draft - Do Not Cite or Quote

-------
     Karita. K; Yano. E; Dakeishi. M; Iwata. T; Murata. K. (2005). Benchmark dose of lead inducing anemia
        at the workplace. Risk Anal 25: 957-962. http://dx.doi.Org/10.llll/i.1539-6924.2005.00652.x

     Karmaus. W: Brooks. KR; Nebe. T; Witten. J; Obi-Osius. N; Kruse. H. (2005). Immune function
        biomarkers in children exposed to lead and organochlorine compounds: A cross-sectional study.
        Environ Health Global Access Sci Source 4: 5. http://dx.doi.org/10.1186/1476-069X-4-5

     Kasperczyk. A; Kasperczyk. S: Horak. S: Ostalowska. A; Grucka-Mamczar. E; Romuk. E; Olejek. A;
        Birkner. E. (2008). Assessment of semen function and lipid peroxidation among lead exposed men.
        Toxicol Appl Pharmacol 228: 378-384. http://dx.doi.Org/10.1016/i.taap.2007.12.024

     Kasperczyk. S: Kasperczyk. J; Ostalowska. A; Zalejska-Fiolka. J: Wielkoszynski. T; Swietochowska. E;
        Birkner. E. (2009). The role of the antioxidant enzymes in erythrocytes in the development of arterial
        hypertension among humans exposed to lead. Biol Trace Elem Res 130: 95-106.
        http://dx.doi.org/10.1007/sl2011-009-8323-z

     Kasprzak. KS: Hoover. KL; Poirier. LA. (1985). Effects of dietary calcium acetate on lead subacetate
        carcinogenicity in kidneys of male Sprague- Dawley rats. Carcinogenesis 6: 279-282.
        http://dx.doi.Org/10.1093/carcin/6.2.279

     Kasten-Jolly. J: Heo. Y; Lawrence. DA. (2010). Impact of developmental lead exposure on splenic
        factors. Toxicol Appl Pharmacol 247: 105-115. http://dx.doi.Org/10.1016/i.taap.2010.06.003

     Kaufman. AS. (2001). Do low levels of lead produce IQ loss in children? A careful examination of the
        literature. Arch Clin Neuropsychol 16: 303-341.

     Kazi. TG: Jalbani. N; Kazi. N; Jamali. MK; Arain. MB; Afridi. HI; Kandhro. A; Pirzado. Z. (2008).
        Evaluation of toxic metals in blood and urine samples of chronic renal failure patients, before and
        after dialysis. Ren Fail 30: 737-745.  http://dx.doi.org/10.1080/08860220802212999

     Ke. H. (2004). Implications of PDE4 structure on inhibitor selectivity across PDE families. Int J Impot
        Res 16: S24-S27. http://dx.doi.org/10.1038/sj.ijir.3901211

     Kemp. FW: Neti. PVS. V: Howell. RW: Wenger. P; Louria. DB; Bogden. JD. (2007). Elevated blood lead
        concentrations and vitamin D deficiency  in winter and summer in young urban children. Environ
        Health Perspect 115: 630-635. http://dx.doi.org/10.1289/ehp.9389

     Kempe. DA; Lang. PA; Eisele. K; Klarl. BA; Wieder. T;  Huber. SM; Duranton. C: Lang. F. (2005).
        Stimulation of erythrocyte phosphatidylserine exposure by lead ions. Am J Physiol 288: C396-C402.
        http://dx.doi.org/10.1152/ajpcell.00115.2004

     Kempermann. G: Krebs. J: Fabel. K. (2008). The contribution of failing adult hippocampal neurogenesis
        to psychiatric disorders [Review].  Curr Opin Psychiatry 21:  290-295.
        http://dx.doi.org/10.1097/YCO.Ob013e3282fad375

     Kerkvliet. NI; Baecher-Steppan. L. (1982). Immunotoxicology studies on lead: Effects of exposure on
        tumor growth and cell-mediated immunity after syngeneic or allogeneic stimulation.
        Immunopharmacology 4: 213-224. http://dx.doi.org/10.1016/0162-3109(82)90003-0

     Kermani. S: Karbalaie. K; Madani. SH; Jahangirnejad. AA; Eslaminejad. MB; Nasr-Esfahani. MH;
        Baharvand. H. (2008). Effect of lead on proliferation  and neural differentiation of mouse bone
        marrow-mesenchymal stem cells. Toxicol In Vitro 22: 995-1001.
        http://dx.doi.0rg/10.1016/i.tiv.2008.02.009

     Kern. M; Audesirk. G. (2000). Stimulatory and inhibitory effects of inorganic lead on calcineurin.
        Toxicology 150: 171-178. httrj://dx.doi.org/10.1016/S0300-483X(00)00258-4

     Kern. M; Wisniewski. M; Cabell. L; Audesirk. G. (2000). Inorganic lead and calcium interact positively
        in activation of calmodulin. Neurotoxicology 21: 353-363.

     Khairullina. AY; Olshanskava. TV; Filimonenko. DS: Yasinski. VM; Slobozhanina. El; Kozlova. NM.
        (2008). Study of optical and nanostructural metal-induced changes in erythrocyte membranes by
        scattering and atomic force microscopy. Optics and Spectroscopy 105: 154-160.
        http://dx.doi.org/10.1134/S0030400X08070242
November 2012                             5-746                   Draft - Do Not Cite or Quote

-------
     Khalil-Manesh. F; Gonick. HC: Cohen. A; Bergamaschi. E; Mutti. A. (1992a). Experimental model of
        lead nephropathy: II. Effect of removal from lead exposure and chelation treatment with
        dimercaptosuccinic acid (DMSA). Environ Res 58: 35-54.
        http://dx.doi.org/10.1016/80013-9351(05)80203-8

     Khalil-Manesh. F; Gonick. HC: Cohen. AH. (1993a). Experimental model of lead nephropathy III
        Continuous low-level lead administration. Arch Environ Occup Health 48: 271-278.
        httrj://dx.doi.org/10.1080/00039896.1993.9940372

     Khalil-Manesh. F; Gonick. HC: Cohen. AH; Alinovi. R; Bergamaschi. E; Mutti. A; Rosen. VJ. (1992b).
        Experimental model of lead nephropathy. I.  Continuous high-dose lead administration. Kidney Int 41:
        1192-1203. http://dx.doi.org/10.1038/ki.1992.181

     Khalil-Manesh. F; Gonick. HC: Weiler. EWJ: Prins. B; Weber. MA; Purdv. R; Ren. Q. (1994). Effect of
        chelation treatment with dimercaptosuccinic acid (DMSA) on lead-related blood pressure changes.
        Environ Res 65:  86-99. http://dx.doi.org/10.1006/enrs.1994.1023

     Khalil-Manesh. F: Gonick. HC: Weiler. EWJ: Prins. B: Weber. MA: Purdv. RE. (1993b). Lead-induced
        hypertension: Possible role of endothelial factors. Am J Hypertens 6: 723-729.

     Khalil. N. (2010). Re: Blood lead mortality study: Khalil et al., 2009. Khalil,
        N.  http://www.regulations.gov/#!documentDetail:D=EPA-HQ-ORD-2011-0051-0006

     Khalil. N: Caulev. JA: Wilson. JW: Talbott. EO: Morrow. L: Hochberg. MC: Hillier. TA: Muldoon. SB:
        Cummings. SR. (2008). Relationship of blood lead levels to incident nonspine fractures and falls in
        older women: The Study of Osteoporotic Fractures. J Bone Miner Res 23: 1417-1425.
        http://dx.doi.org/10.1359/ibmr.080404

     Khalil. N; Morrow. LA; Needleman. H; Talbott. EO: Wilson. JW: Caulev. JA. (2009a). Association of
        cumulative lead and neurocognitive function in an occupational cohort. Neuropsychology 23: 10-19.
        http://dx.doi.org/10.1037/a0013757

     Khalil. N; Wilson. JW: Talbott. EO: Morrow. LA; Hochberg. MC: Hillier. TA; Muldoon. SB; Cummings.
        SR: Caulev. JA. (2009b). Association of blood lead concentrations with mortality in older women: a
        prospective cohort study. Environ Health Global Access Sci Source 8: 15.
        http://dx.doi.org/10.1186/1476-069x-8-15

     Khan. D; Qawum. S: Saleem. S: Ansari. W: Khan. F. (2010a). Lead exposure and its  adverse health
        effects among occupational workers children. Toxicol Ind Health 26: 497.
        http://dx.doi.org/10.1177/0748233710373085

     Khan. DA; Qawum. S: Saleem. S: Khan. FA. (2008). Lead-induced oxidative stress adversely affects
        health of the occupational workers. Toxicol  Ind Health  24: 611-618.
        http://dx.doi.org/10.1177/0748233708098127

     Khan. MI; Ahmad. I; Mahdi. AA; Akhtar. MJ; Islam. N; Ashquin. M; Venkatesh. T. (201 Ob).  Elevated
        blood lead levels and cytogenetic markers in buccal epithelial cells of painters in India: Genotoxicity
        in painters exposed to lead containing paints. Environ Sci Pollut Res Int 17: 1347-1354.
        http://dx.doi.org/10.1007/sll356-010-0319-x

     Khan. MI; Islam. N; Sahasrabuddhe. AA; Mahdi. AA; Siddiqui. H; Ashquin. M; Ahmad. I. (2011).
        Ubiquitous hazardous metal lead induces TNF-a in human phagocytic THP-1  cells: Primary role of
        ERK 1/2. J Hazard Mater 189: 255-264. http://dx.doi.Org/10.1016/i.ihazmat.2011.02.027

     Kharoubi. O: Slimani. M; Aoues. A; Seddik. L. (2008a). Prophylactic effects of Wormwood on lipid
        peroxidation in an animal model of lead intoxication. Indian J Nephrol 18: 51-57.
        http://dx.doi.org/10.4103/0971-4065.42333

     Kharoubi. O: Slimani. M; Krouf. D; Seddik. L; Aoues. A. (2008b). Role of wormwood (Artemisia
        absinthium) extract on oxidative stress in ameliorating lead induced haematotoxicity. African Journal
        of Traditional, Complementary and Alternative Medicines 5: 263-270.
November 2012                              5-747                  Draft - Do Not Cite or Quote

-------
     Khorasani. G: Shokrzade. M; Salehifar. E; Asadi. M; Shabankhani. B; Rezaeinejad. S. (2008). The
        comparison of lead and zinc plasma levels in gastric cancer patients with healthy volunteers. Res J
        Biol Sci 3: 631-634. http://dx.doi.org/10.3923/rjbsci.2008.631.634

     Khotimchenko. MY; Kolenchenko. EA. (2007). Efficiency of low-esterified pectin in toxic damage to the
        liver inflicted by lead treatment. Bull Exp Biol Med 144: 60-62.
        httrj://dx.doi.org/10.1007/sl0517-007-0254-0

     Kim. D; Lawrence. DA. (2000). Immunotoxic effects of inorganic lead on host resistance of mice with
        different circling behavior preferences. Brain Behav Immun 14: 305-317.
        http://dx.doi.org/10.1006/brbi.2000.0609

     Kim. HS: Lee. SS: Lee. GS: Hwangbo. Y; Ahn. KD; Lee. BK. (2004). The protective effect of delta-
        aminolevulinic acid dehydratase 1-2 and 2-2 isozymes against blood lead with higher hematologic
        parameters. Environ Health Perspect 112: 538-541. http://dx.doi.org/10.1289/ehp.6464

     Kim. JH: Lee. KH: Yoo. DH: Kang. D: Cho. SH:  Hong. YC. (2007). GSTM1 and TNF-alpha gene
        polymorphisms and relations between blood lead and inflammatory markers in a non-occupational
        population. Mutat Res Genet Toxicol Environ Mutagen 629: 32-39.
        http://dx.doi.0rg/10.1016/i.mrgentox.2007.01.004

     Kim. R; Landrigan. C: Mossmann. P; Sparrow. D: Hu. H. (1997). Age and secular trends in bone lead
        levels in middle-aged and elderly men: Three-year longitudinal follow-up in the Normative Aging
        Study. Am J Epidemiol 146: 586-591.

     Kim. R; Rotnitsky. A; Sparrow. D: Weiss. ST; Wager. C: Hu. H. (1996). A longitudinal study of low-level
        lead exposure and impairment of renal function The Normative Aging Study. JAMA 275: 1177-1181.
        http://dx.doi.org/10.1001/iama.1996.03530390043032

     Kim. Y: Kim. BN: Hong. YC: Shin. MS: Yoo. HJ: Kim. JW: Bhang. SY: Cho. SC.  (2009b). Co-exposure
        to environmental  lead and manganese affects the intelligence of school-aged children.
        Neurotoxicology  30: 564-571. http://dx.doi.Org/10.1016/i.neuro.2009.03.012

     Kimber. I; Stonard. MD; Gidlow. DA; Niewola. Z. (1986). Influence of chronic low-level exposure to
        lead on plasma immunoglobulin concentration and cellular immune function in man. Int Arch  Occup
        Environ Health 57: 117-125. http://dx.doi.org/10.1007/BF00381379

     Kiraly. E; Jones. DG. (1982). Dendritic spine changes in rat hippocampal pyramidal cells after postnatal
        lead treatment: Agolgi study. Exp Neurol 77:  236-239.
        http://dx.doi.org/10.1016/0014-4886(82)90158-3

     Kiran Kumar. B; Prabhakara Rao. Y; Noble. T; Weddington. K; McDowell. VP; Rajanna. S: Bettaiya. R.
        (2009). Lead-induced alteration of apoptotic proteins in different regions of adult rat brain. Toxicol
        Lett 184: 56-60. http://dx.doi.Org/10.1016/i.toxlet.2008.10.023

     Kirberger. M; Yang. JJ. (2008). Structural differences between Pb2+- and Ca2+-binding sites in proteins:
        Implications with respect to toxicity. J Inorg Biochem 102:  1901-1909.
        http://dx.doi.0rg/10.1016/i.jinorgbio.2008.06.014

     Kishi. R; Ikeda. T; Miyake. H;  Uchino. E; Tsuzuki. T; Inoue. K. (1983). Effects of low lead exposure on
        neuro-behavioral  function in the rat. Arch Environ Occup Health 38: 25-33.

     Kishikawa. H; Song. R; Lawrence. DA. (1997). Interleukin-12 promotes enhanced resistance to Listeria
        monocytogenes infection of lead-exposed mice. Toxicol Appl Pharmacol 147: 180-189.
        http://dx.doi.org/10.1006/taap.1997.8308

     Kishimoto. T; Oguri. T; Ueda.  D: Tada. M. (1995). Effect of lead on tube formation by cultured human
        vascular endothelial cells. Arch Toxicol 69: 718-721. http://dx.doi.org/10.1007/s002040050238

     Kiziler. AR; Aydemir. B; Onaran. I; Alici. B; Ozkara. H; Gulyasar. T; Akyolcu. MC. (2007). High levels
        of cadmium and lead in seminal fluid and blood of smoking men are associated with high oxidative
        stress and damage in infertile subjects. Biol Trace Elem Res 120: 82-91.
        http://dx.doi.org/10.1007/sl2011-007-8020-8
November 2012                             5-748                  Draft - Do Not Cite or Quote

-------
     Klann. E; Shelton. KR. (1989). The effect of lead on the metabolism of a nuclear matrix protein which
        becomes prominent in lead-induced intranuclear inclusion bodies. J Biol Chem 264: 16969-16972.

     Klein. D; Y-JY. W: Kamyab. S: Okuda. H; Sokol. RZ. (1994). Effects of toxic levels of lead on gene
        regulation in the male axis: increase in messenger ribonucleic acids and intracellular stores of
        gonadotrophs within the central nervous  system. Biol Reprod 50: 802-811. http://dx.doi.org/10.1095/
        biolreprod50.4.802

     Knowles. SO: Donaldson. WE. (1990). Dietary modification of lead toxicity: Effects on fatty acid and
        eicosanoid metabolism in chicks. Comp Biochem Physiol C Toxicol Pharmacol 95: 99-104.
        httrj://dx.doi.org/10.1016/0742-8413(90)90088-O

     Knowles. SO: Donaldson. WE. (1997). Lead disrupts eicosanoid metabolism, macrophage function, and
        disease resistance in birds. Biol Trace Elem Res 60: 13-26. http://dx.doi.org/10.1007/BF02783306

     Kojima. M; Degawa. M. (2006). Gender-related difference in altered gene expression of a sterol
        regulatory element binding protein, SREBP-2, by lead nitrate in rats: Correlation with development of
        hypercholesterolemia. J Appl Toxicol 26: 381-384. http://dx.doi.org/10.1002/jat.1138

     Kojima. M; Sekikawa. K; Nemoto. K; Degawa. M. (2005). Tumor necrosis factor-alpha-independent
        downregulation of hepatic cholesterol 7 alpha-hydroxylase gene in mice treated with lead nitrate.
        Toxicol Sci 87: 537-542. http://dx.doi.org/10.1093/toxsci/kfi267

     Koka. S: Huber. SM; Boini. KM; Lang. C: Foller. M; Lang. F. (2007). Lead decreases parasitemia and
        enhances  survival of Plasmodium berghei-infected mice. Biochem Biophys Res Commun 363:
        484-489. http://dx.doi.0rg/10.1016/i.bbrc.2007.08.173

     Kolesarova. A; Roychoudhury. S:  Slivkova.  J: Sirotkin. A; Capcarova. M; Massanyi. P. (2010). In vitro
        study on the effects of lead and mercury  on porcine ovarian granulosa cells. J Environ  Sci Health A
        Tox Hazard Subst Environ Eng 45: 320-331. http://dx.doi.org/10.1080/10934520903467907

     Koller. LD. (1973). Immunosuppression produced by lead, cadmium, and mercury. Am J Vet Res 34:
        1457-1458.

     Koller. LD; Kerkvliet. NI; Exon. JH. (1985). Neoplasia induced in male rats fed lead acetate, ethyl urea,
        and sodium nitrite. Toxicol Pathol 13: 50-57. http://dx.doi.org/10.1177/019262338501300107

     Koller. LD; Kovacic. S. (1974). Decreased antibody formation in mice exposed to lead. Nature 250:
        148-150. http://dx.doi.org/10.1038/250148aO

     Korashy. HM; Ei-Kadi. AOS. (2008). The role of redox-sensitive transcription factors NF-kappa B and
        AP-1 in the modulation of the Cyplal gene by mercury, lead, and copper. Free Radic Biol Med 44:
        795-806. http://dx.doi.0rg/10.1016/i.freeradbiomed.2007.ll.003

     Korashy. HM; El-Kadi. AOS. (2004). Differential effects of mercury, lead and copper on the constitutive
        and inducible expression of aryl hydrocarbon receptor (AHR)-regulated genes in cultured hepatoma
        Hepa Iclc7 cells. Toxicology 201:  153-172. http://dx.doi.Org/10.1016/i.tox.2004.04.011

     Korashy. HM; El-Kadi. AOS. (2008). NF-kappa B  and AP-1  are key signaling pathways in the
        modulation of NAD(P)H: Quinone oxidoreductase 1 gene by mercury, lead, and copper. J Biochem
        Mol Toxicol 22: 274-283. http://dx.doi.org/10.1002/ibt.20238

     Kordas. K; Canfield. RL; Lopez. P; Rosado. JL; Vargas. GG: Cebrian. ME; Rico. JA; Ronquillo. D;
        Stoltzfus. RJ. (2006). Deficits in cognitive function and achievement in Mexican first-graders with
        low blood lead concentrations. Environ Res 100: 371-386.
        http://dx.doi.0rg/10.1016/i.envres.2005.07.007

     Kordas. K; Casavantes. KA; Mendoza. C: Lopez. P; Ronquillo. D; Rosado. JL; Vargas. GG: Stoltzfus. RJ.
        (2007). The association between lead and micronutrient status, and children's sleep, classroom
        behavior,  and activity. Arch Environ Occup Health 62: 105-112.
        http://dx.doi.Org/10.3200/AEOH.62.2.105-112
November 2012                              5-749                  Draft - Do Not Cite or Quote

-------
     Kordas. K; Ettinger. AS: Bellinger. DC: Schnaas. L; Tellez Rojo. MM; Hernandez-Avila. M; Hu. H;
        Wright. RO. (2011). A dopamine receptor (DRD2) but not dopamine transporter (DAT1) gene
        polymorphism is associated with neurocognitive development of Mexican preschool children with
        lead exposure. J Pediatr 159: 638-643. http://dx.doi.Org/10.1016/j.jpeds.2011.03.043

     Kordas. K; Ettinger. AS: Lamadrid-Figueroa. H; Tellez-Rojo. MM; Hernandez-Avila. M; Hu. H; Wright.
        RO. (2009). Methylenetetrahydrofolate reductase (MTHFR) C677T, A1298C and Gl 793 A genotypes,
        and the relationship between maternal folate intake, tibia lead and infant size at birth. Br J Nutr 102:
        907-914. http://dx.doi.org/10.1017/s0007114509318280

     Kottgen. A; Selvin. E; Stevens. LA; Levey. AS: Van Lente. F; Coresh. J. (2008). Serum cystatin C in the
        United States: The Third National Health and Nutrition Examination Survey (NHANES III). Am J
        Kidney Dis 51: 385-394. http://dx.doi.Org/10.1053/i.ajkd.2007.ll.019

     Kowolenko. M; Tracy. L; Lawrence. D. (1991). Early effects of lead on bone marrow cell responsiveness
        in mice challenged with Listeria monocytogenes. Fundam Appl Toxicol 17:  75-82.
        http://dx.doi.org/10.1093/toxsci/17.L75

     Kowolenko. M; Tracy. L; Mudzinski. S: Lawrence. DA. (1988). Effect of lead on macrophage function. J
        LeukocBiol 43: 357-364.

     Krieg. EF. Jr. (2007). The relationships between blood lead levels and serum follicle stimulating hormone
        and luteinizing  hormone in the third national health and nutrition examination survey. Environ Res
        104: 374-382. http://dx.doi.Org/10.1016/i.envres.2006.09.009

     Krieg. EF. Jr; Butler. MA. (2009). Blood lead, serum homocysteine, and neurobehavioral test
        performance in the third National Health and Nutrition Examination Survey. Neurotoxicology 30:
        281-289. http://dx.doi.0rg/10.1016/i.neuro.2008.12.014

     Krieg. EF. Jr; Butler. MA; Chang. MH; Liu. TB; Yesupriya. A; Lindegren. ML; Dowling. N.  (2009). Lead
        and cognitive function in ALAD genotypes in the Third National Health and Nutrition Examination
        Survey. Neurotoxicol Teratol 31: 364-371. http://dx.doi.Org/10.1016/i.ntt.2009.08.003

     Krieg. EF. Jr; Butler. MA; M-h. C: Liu. T; Yesupriya. A; Dowling. N; Lindegren. ML.  (2010). Lead and
        cognitive function in VDR genotypes in the Third National Health and Nutrition Examination Survey.
        Neurotoxicol Teratol 32: 262-272. http://dx.doi.Org/10.1016/i.ntt.2009.12.004

     Krieg. EF. Jr: Chrislip. DW: Crespo. CJ: Brightwell. WS: Ehrenberg. RL: Otto. DA. (2005). The
        relationship between blood lead levels and neurobehavioral test performance in NHANES III and
        related occupational studies. Public Health Rep 120: 240-251.

     Krishnamoorthv. MS: Parthiban. N; Muthu. P; Paul. V: Balagopal. G: Kumaravel. TS. (1993). Effect of
        acute pretreatment of lead on picrotoxin-induced convulsions in rats. J Appl Toxicol 13: 155-159.

     Kristal-Boneh. E; Coller.  D; Froom. P; Harari. G: Ribak. J. (1999). The association between occupational
        lead exposure and serum cholesterol and lipoprotein levels. Am J Public Health 89:  1083-1087.

     Krocova. Z; Macela. A; Kroca. M; Hernychova. L. (2000).  The immunomodulatory effect(s) of lead and
        cadmium on the cells of immune system in vitro. Toxicol In Vitro 14: 33-40.

     Kuhlmann. AC: McGlothan. JL; Guilarte. TR. (1997). Developmental lead exposure causes spatial
        learning deficits in adult rats. Neurosci Lett 233: 101-104.
        http://dx.doi.org/10.1016/S0304-3940(97)00633-2

     Kuo. CY; Wong. RH; Lin. JY; Lai. JC: Lee. H. (2006). Accumulation of chromium and nickel metals in
        lung tumors from lung cancer patients in Taiwan. J Toxicol Environ Health A 69: 1337-1344.
        http://dx.doi.org/10.1080/15287390500360398

     Kuruvilla. A; Pillav. VV: Adhikari. P; Venkatesh. T; Chakrapani. M; Rao. HTJ; Bastia.  BK; Raieev. A;
        Saralava. KM; Rai. M. (2006). Clinical manifestations of lead workers of Mangalore, India. Toxicol
        Ind Health 22: 405-413. http://dx.doi.org/10.1177/0748233706074174
November 2012                             5-750                  Draft - Do Not Cite or Quote

-------
     Laden. F; Neas. LM; Dockery. DW: Schwartz. J. (2000). Association of fine paniculate matter from
        different sources with daily mortality in six US cities. Environ Health Perspect 108: 941-947.
        http://dx.doi.org/10.1289/ehp.00108941

     Lai. CC: Lin. HH; Chen. CW: Chen. SH; Chiu. TH. (2002). Excitatory action of lead on rat sympathetic
        preganglionic neurons in vitro and in vivo. Life Sci 71: 1035-1045.
        http://dx.doi.org/10.1016/S0024-3205(02)01789-7

     Lai. LH; Chou. SY; Wu. FY; Chen. JJH; Kuo. HW. (2008a). Renal dysfunction and hyperuricemia with
        low blood lead levels and ethnicity in community-based study. Sci Total Environ 401: 39-43.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.04.004

     Lai. B; Murtfay. RC: Anand. M; Chandra. SV: Kumar. R; Tripathi. O: Srimal. RC. (1991). Cardiotoxicity
        and hypertension  in rats after oral lead exposure. Drug Chem Toxicol 14: 305-318.
        http://dx.doi.org/10.3109/01480549109002192

     Lamadrid-Figueroa. H; Tellez-Rojo. MM; Hernandez-Avila. M; Trejo-Valdivia. B; Solano-Gonzalez. M;
        Mercado-Garcia. A; Smith. D; Hu. H; Wright. RO. (2007). Association between the plasma/whole
        blood lead ratio and history of spontaneous abortion: A nested cross-sectional study. BMC Pregnancy
        Childbirth 7: 22. http://dx.doi.org/10.1186/1471-2393-7-22

     Lamb. MR; Janevic. T; Liu. X; Cooper.  T; Kline. J: Factor-Litvak. P. (2008). Environmental lead
        exposure, maternal thyroid function, and childhood growth. Environ Res 106:  195-202.
        http://dx.doi.0rg/10.1016/i.envres.2007.09.012

     Landrigan. PJ; Suk. WA; Amler. RW. (1999). Chemical wastes, children's health, and the Superfund Basic
        Research Program [Review]. Environ Health Perspect 107: 423-427.

     Lang. F; Gulbins. E;  Lerche. H; Huber. SM; Kempe. PS: Foller.  M. (2008). Eryptosis, a window to
        systemic disease [Review]. Cell Physiol Biochem 22: 373-380. http://dx.doi.org/10.1159/000185448

     Lanphear. BP; Dietrich. K; Auinger. P; Cox. C. (2000). Cognitive deficits associated with blood lead
        concentrations <10 microg/dL in US children and adolescents. Public Health Rep 115: 521-529.

     Lanphear. BP; Hornung. R; Khoury. J: Yolton. K; Baghurst. P; Bellinger. DC: Canfield. RL; Dietrich.
        KN; Bornschein. R; Greene. T; Rothenberg. SJ: Needleman. HL; Schnaas. L; Wasserman. G:
        Graziano. J: Roberts. R. (2005). Low-level environmental lead exposure and children's intellectual
        function: An international pooled analysis. Environ Health Perspect 113: 894-899.

     Lara-Tejero. M; Pamer. EG. (2004). T cell responses to Listeria monocytogenes [Review]. Curr Opin
        Microbiol 7: 45-50. http://dx.doi.Org/10.1016/j.mib.2003.12.002

     Larsson. A; Malm. J: Grubb. A; Hansson. LO. (2004). Calculation of glomerular filtration rate expressed
        in mL/min from plasma cystatin C values in mg/L. Scand J Clin Lab Invest 64: 25-30.

     Laschi-Loquerie. A; Decotes. J: Tachon. P; Evreux. JC. (1984). Influence of lead acetate on
        hypersensitivity: Experimental study. J Immunopharm 6: 87-93.
        http://dx.doi.org/10.3109/08923978409026461

     Laslev. SM; Gilbert. ME. (2002). Rat hippocampal glutamate and GABA release exhibit biphasic effects
        as a function of chronic lead exposure level. Toxicol Sci 66:  139-147.
        http://dx.doi.org/10.1093/toxsci/66.1.139

     Laties. AM; Zrenner. E. (2002). Viagra (sildenafil citrate) and ophthalmology [Review]. Prog Retin Eye
        Res 21: 485-506.  http://dx.doi.org/10.1016/sl350-9462(02)00013-7

     Lau. YS: Camoratto. AM; White. LM; Moriartv. CM. (1991). Effect of lead on TRH and GRF binding in
        rat anterior pituitary membranes. Toxicology 68: 169-179.
        http://dx.doi.org/10.1016/0300-483X(91)90019-W

     Laughlin. NK; Luck. ML; Lasky. RE. (2008). Postnatal lead effects on the development of visual spatial
        acuity in rhesus monkeys (Macaca mulatta). Dev Psychobiol  50: 608-614.
        http://dx.doi.org/10.1002/dev.20315
November 2012                             5-751                  Draft - Do Not Cite or Quote

-------
     Laughlin. NK; Luck. ML; Lasky. RE. (2009). Early lead exposure effects on an auditory threshold task in
        the rhesus monkey (Macaca mulatta). Dev Psychobiol 51: 289-300.
        http://dx.doi.org/10.1002/dev.20364

     Lawrence. DA. (1981a). Heavy metal modulation of lymphocyte activities - II. Lead, an in vitro mediator
        of B-cell activation. International Journal of Immunopharmacology 3: 153-161.
        http://dx.doi.org/10.1016/0192-0561(81)90006-0

     Lawrence. DA. (198 Ib). In vivo and in vitro effects of lead on humoral and cell- mediated immunity.
        Infect Immun 31: 136-143.

     Lawton. LJ; Donaldson. WE.  (1991). Lead-induced tissue fatty acid alterations and lipid peroxidation.
        Biol Trace Elem Res 28: 83-97. http://dx.doi.org/10.1007/BF02863075

     Lazarus. JH. (2005). Thyroid disorders associated with pregnancy: Etiology, diagnosis, and management
        [Review]. Treat Endocrinol 4: 31-41.

     Leasure. JL; Giddabasappa. A; Chanev. S: Johnson. JE; Pothakos. K; Lau. YS: Fox. DA. (2008). Low-
        level human equivalent gestational lead exposure produces sex-specific motor and coordination
        abnormalities and late-onset obesity in year-old mice. Environ Health Perspect 116: 355-361.
        http://dx.doi.org/10.1289/ehp.10862

     Ledda-Columbano. GM; Coni. O; Curto. M; Giacomini. O; Faa. G: Sarma. DSR; Columbano. A. (1992).
        Mitogen-induced liver hyperplasia does not substitute for compensatory regeneration during
        promotion of chemical hepatocarcinogenesis. Carcinogenesis 13: 379-383.
        http://dx.doi.0rg/10.1093/carcin/13.3.379

     Lee. BK; Lee. GS: Stewart. WF; Ahn. KD; Simon. D; Kelsev. KT; Todd. AC:  Schwartz. BS.  (200la).
        Associations of blood pressure and hypertension with lead dose measures and polymorphisms in the
        vitamin D receptor and delta-aminolevulinic acid dehydratase genes. Environ Health Perspect 109:
        383-389.

     Lee. DH; Blomhoff. R; Jacobs. DR. Jr. (2004). Is serum gamma glutamyltransferase a marker of oxidative
        stress? [Review]. Free Radic Res 38: 535-539. http://dx.doi.org/10.1080/10715760410001694026

     Lee. DH; Lim. JS: Song. K; Boo. Y; Jacobs. DR. (2006a). Graded associations of blood lead and urinary
        cadmium concentrations with oxidative-stress-related markers in the US population: Results from the
        Third National Health and Nutrition Examination Survey. Environ Health Perspect 114: 350-354.
        http://dx.doi.org/10.1289/ehp.8518

     Lee. JE; Chen. S: Golemboski. KA; Parsons. PJ: Dietert. RR. (200Ib). Developmental windows of
        differential lead-induced immunotoxicity in chickens. Toxicology 156: 161-170.
        http://dx.doi.org/10.1016/S0300-483X(00)00350-4

     Lee. JE; Naqi. SA; Kao. E; Dietert. RR. (2002). Embryonic exposure to lead: Comparison of immune and
        cellular responses in unchallenged and virally stressed chickens. Arch Toxicol 75: 717-724.
        http://dx.doi.org/10.1007/s00204-001-0283-9

     Lee. JJ: Battles. AH. (1994). Lead toxicity via arachidonate signal transduction to growth responses in the
        splenic macrophage. Environ Res 67: 209-219. http://dx.doi.org/10.1006/enrs.1994.1075

     Lee. MK; Cho. SY; Kim. DJ;  Jang. JY; Shin. KH; Park. SA; Park. EM; Lee. JS: Choi. MS: Kim. MJ.
        (2005). Du-zhong (Eucommia ulmoides Oliv.) cortex water extract alters heme biosynthesis and
        erythrocyte antioxidant defense system in lead-administered rats. J Med Food 8: 86-92.
        http://dx.doi.0rg/10.1089/imf.2005.8.86

     Lee. MY; Shin. JH; Han. HS:  Chung. JH. (2006b). In vivo effects of lead on erythrocytes following
        chronic exposure through drinking water. Arch Pharm Res 29:  1158-1163.

     Lee. TH; Tseng. MC: Chen. CJ: Lin. JL. (2009). Association of high body lead store with severe
        intracranial carotid atherosclerosis. Neurotoxicology 30:  876-880.
        http://dx.doi.0rg/10.1016/i.neuro.2009.07.004
November 2012                             5-752                   Draft - Do Not Cite or Quote

-------
     Legate. ME; Barhoumi. R; Burghardt. RC: Tiffany-Castiglioni. E. (1993). Low-level lead exposure in
        cultured astroglia: Identification of cellular targets with vital fluorescent probes. Neurotoxicology 14:
        267-272.

     Lenroot. RK; Giedd. JN. (2006). Brain development in children and adolescents: insights from anatomical
        magnetic resonance imaging [Review]. Neurosci Biobehav Rev 30: 718-729.
        http://dx.doi.0rg/10.1016/i.neubiorev.2006.06.001

     Leret. ML; Garcia-Uceda. F; Antonio. MT. (2002). Effects of maternal lead administration on
        monoaminergic, GABAergic and glutamatergic systems. Brain Res Bull 58: 469-473.
        http://dx.doi.org/10.1016/S0361-9230(02)00819-5

     Lesenev. AM; Deme. D; Legue. O: Ohavon. R; Chanson. P; Sales. JP; Carvalho. DP; Dupuy. C: Virion.
        A. (1999).  Biochemical characterization of a Ca2+/NAD(P)H-dependent H2O2 generator in human
        thyroid tissue. Biochimie 81: 373-380. http://dx.doi.org/10.1016/S0300-9084(99)80084-4

     Levey. AS: Bosc. JP: Lewis. JB; Greene. T; Rogers. N; Roth. D. (1999). A more accurate method to
        estimate glomerular filtration rate from serum creatinine: A new prediction equation. Ann Intern Med
        130: 461-470.

     Levey. AS: Greene. T; Kusek. JW: Beck. G. (2000). A simplified equation to predict glomerular filtration
        rate from serum creatinine [Abstract]. J Am Soc Nephrol 11: A0828.

     Levey. AS: Stevens. LA; Schmid. CH; Zhang. YP; Castro. AF; Feldman. HI; Kusek. JW: Eggers. P; Van
        Lente. F; Greene. T; Coresh. J. (2009). A new equation to estimate glomerular filtration rate. Ann
        Intern Med 150: 604-612.

     Levin. ED; Bowman. RE; Wegert. S: Vuchetich. J. (1987). Psychopharmacological investigations of a
        lead-induced long-term cognitive deficit in monkeys. Psychopharmacology 91: 334-341.

     Leviton. A; Bellinger. D; Allred. EN; Rabinowitz. M; Needleman. H; Schoenbaum. S. (1993). Pre- and
        postnatal low-level lead exposure and children's dysfunction in school. Environ Res 60: 30-43.
        http://dx.doi.org/10.1006/enrs.1993.1003

     Lewis. M; Pitts. D. (2004). Inorganic lead exposure in the rat activates striatal cFOS expression at lower
        blood levels and inhibits amphetamine-induced cFOS expression at higher blood levels. J Pharmacol
        ExpTher310: 815-820. http://dx.doi.org/10.1124/ipet.103.063941

     Li Chen. T; Wise. SS: Kraus. S: Shaffiev. F; Levine. KM; Thompson. WD; Romano. T; O'Hara. T; Wise.
        JP. Sr. (2009). Particulate hexavalent chromium is cytotoxic and genotoxic to the North Atlantic right
        whale (Eubalaena glacialis) lung and skin fibroblasts. Environ Mol Mutagen 50: 387-393.
        http://dx.doi.org/10.1002/em.20471

     Li. JP; Wang. CY; Tang. YA; Lin. YW: Yang. JL. (2008a). Role  of extracellular signal-regulated kinase
        (ERK) signaling in nucleotide excision repair and genotoxicity in response to As(III)  and Pb(II). Pure
        Appl Chem 80: 2735-2750. http://dx.doi.org/10.1351/pac200880122735

     Li. N: Yu. ZL: Wang. L: Zheng. YT: Jia. JX: Wang. O: Zhu. MJ: Liu. XH: Xia. X: Li. WJ. (2009c). Early-
        life lead exposure affects the activity of TNF-alpha and expression of SNARE complex in
        hippocampus of mouse pups. Biol Trace Elem Res 132: 227-238.
        http://dx.doi.org/10.1007/sl2011-009-8551-2

     Li. N; Yu. ZL: Wang. L; Zheng. YT; Jia. JX; Wang. Q; Zhu. MJ: Liu. XL; Xia. X; Li. WJ. (201 Ob).
        Increased tau phosphorylation and beta amyloid in the hipocampus of mouse pups by early life lead
        exposure. Acta Biol Hung 61: 123-134. http://dx.doi.Org/10.1556/ABiol.61.2010.2.l

     Li. S: Chen. W: Srinivasan. SR: Bond. MG: Tang. R; Urbina. EM; Berenson. GS. (2003). Childhood
        cardiovascular risk factors and carotid vascular changes in adulthood: The Bogalusa Heart Study.
        JAMA 290: 2271-2276. http://dx.doi.org/10.1001/iama.290.17.2271

     Li. S: Zhao. Z; Zhou. X; Liu. S. (2008c). The effect of lead on intracellular Ca2+ in mouse
        lymphocytes. Toxicol In Vitro 22: 1815-1819. http://dx.doi.Org/10.1016/i.tiv.2008.08.005
November 2012                             5-753                  Draft - Do Not Cite or Quote

-------
     Li. TY; Zhang. X; Wei. XP; Liu. YF; Qu. P; Liu. YX; Chen. J. (2011). [Impact of antioxidant vitamins
        and heavy metal levels at birth on neurodevelopment of children assessed at two years of age].
        Zhonghua Er Ke Za Zhi 49: 439-444.

     Lidsky. TI; Schneider. JS. (2003).  Lead neurotoxicity in children: basic mechanisms and clinical
        correlates [Review]. Brain 126: 5-19. http://dx.doi.org/10.1093/brain/awg014

     Lidsky. TI; Schneider. JS. (2004).  Lead and public health: Review of recent findings, re-evaluation of
        clinical risks. J Environ Monit 6: 36N-42N. http://dx.doi.org/10.1039/B403259b

     Lilienthal. H; Kohler. K; Turfeld. M; Winneke. G. (1994). Persistent increases in scotopic b-wave
        amplitudes after lead exposure in monkeys. Exp Eye Res 59: 203-209.
        http://dx.doi.org/10.1006/exer. 1994.1098

     Lilienthal. H; Lenaerts. C: Winneke. G: Hennekes. R. (1988). Alteration of the visual evoked potential
        and the electroretinogram in lead-treated monkeys. Neurotoxicol Teratol 10: 417-422.
        http://dx.doi.org/10.1016/0892-0362(88)90002-5

     Lilienthal. H; Winneke. G. (1996). Lead effects on the brain stem auditory evoked potential in monkeys
        during and after the treatment phase. Neurotoxicol Teratol 18: 18:17-32.

     Lilis. R; Fischbein. A; Eisinger. J: Blumberg. WE; Diamond. S: Anderson. HA; Rom. W: Rice. C:
        Sarkozi. L; Kon. S: Selikoff. IJ. (1977). Prevalence of lead disease among secondary lead smelter
        workers and biological indicators of lead exposure. Environ Res 14: 255-285.
        http://dx.doi.org/10.1016/0013-9351(77)90037-8

     Lim. SY: Dohertv. JD: McBride. K:  Miller-Ihli. NJ: Carmona. GN: Stark. KD: Salem. N. (2005). Lead
        exposure and (n-3) fatty acid deficiency during rat neonatal development affect subsequent spatial task
        performance and olfactory discrimination. JNutr 135: 1019-1026.

     Lin-Tan. DT; Lin. JL; Wang. LH; Wang. LM; Huang. LM; Liu. L; Huang. JY; Huang. YL. (2007a).
        Fasting glucose levels in predicting 1-year all-cause mortality in patients who do not have diabetes
        and are on maintenance hemodialysis. J Am Soc Nephrol 18: 2385-2391.
        http://dx.doi.org/10.1681/Asn.2006121409

     Lin-Tan. DT; Lin. JL; Yen. TH; Chen.  KH; Huang. YL. (2007b). Long-term outcome of repeated lead
        chelation therapy in progressive non-diabetic chronic kidney diseases. Nephrol Dial Transplant 22:
        2924-2931.  http://dx.doi.org/10.1093/ndt/gfm342

     Lin. CN; Wang. LH; Shen. KH. (2009). Determining urinary trace elements (Cu, Zn, Pb, As, and Se) in
        patients with bladder cancer. J Clin Lab Anal  23: 192-195. http://dx.doi.org/10.1002/jcla.20318

     Lin. JL: Lin-Tan. DT: Chen. KH; Hsu. CW: Yen. TH; Huang. WH; Huang. YL. (2010). Blood lead levels
        association with 18-month all-cause mortality in patients with chronic peritoneal dialysis. Nephrol
        Dial Transplant 25: 1627-1633. http://dx.doi.org/10.1093/ndt/gfp663

     Lin. JL: Lin-Tan. DT: Hsu. CW: Yen. TH: Chen. KH: Hsu. HH: Ho. TC: Hsu. KH. (2011). Association of
        blood lead levels with mortality in patients on maintenance hemodialysis. Am J Med 124: 350-358.
        http://dx.doi.0rg/10.1016/i.amjmed.2010.10.022

     Lin. JL; Lin-Tan. DT; Hsu. KH; Yu.  CC. (2003).  Environmental lead exposure and progression of chronic
        renal diseases in patients without diabetes. N  Engl J Med 348: 277-286.
        http://dx.doi.org/10.1097/01.ASN.0000118529.01681

     Lin. JL; Lin-Tan. DT; Li. YJ; Chen. KH; Huang. YL. (2006a). Low-level environmental exposure to lead
        and progressive chronic kidney diseases. Am  J Med 119: 1-9.
        http://dx.doi.0rg/10.1016/i.amjmed.2006.01.005

     Lin. JL: Lin-Tan. DT: Yen. TH: Hsu. CW: Jena. CC: Chen. KH: Hsu. KH: Huang. YL. (2008). Blood lead
        levels, malnutrition, inflammation,  and mortality in patients with diabetes treated by long-term
        hemodialysis. Am J Kidney Dis 51: 107-115.  http://dx.doi.Org/10.1053/i.aikd.2007.10.002
November 2012                              5-754                   Draft - Do Not Cite or Quote

-------
     Lin. JL; Lin-Tan. DT; Yu. CC: Li. YJ; Huang. YY; Li. KL. (2006b). Environmental exposure to lead and
        progressive diabetic nephropathy in patients with type II diabetes. Kidney Int 69: 2049-2056.
        http://dx.doi.org/10.1038/sj.ki.5001505

     Lin. RH; Lee. CH; Chen. WK; Lin-Shiau. SY. (1994). Studies on cytotoxic and genotoxic effects of
        cadmium nitrate and lead nitrate in Chinese hamster ovary cells. Environ Mol Mutagen 23: 143-149.
        http://dx.doi.org/10.1002/em.2850230212

     Lin. TA; Tai-yi. J. (2007). Benchmark dose approach for renal dysfunction in workers exposed to lead.
        Environ Toxicol 22: 229-233. http://dx.doi.org/10.1002/tox.20260

     Lind. SE; Park. JS: Drexler. JW. (2009). Pyrithione and 8-hydroxyquinolines transport lead across
        erythrocyte membranes. Transl Res 154: 153-159. http://dx.doi.Org/10.1016/i.trsl.2009.06.002

     Lison. D; Raguzzi. F; Lauwerys. R.  (1990). Comparison of the effects of auranofin, heavy metals and
        retinoids on protein kinase C in vitro and on a protein kinase C mediated response in macrophages.
        Pharmacol Toxicol 67: 239-242.  http://dx.doi.Org/10.llll/i.1600-0773.1990.tb00820.x

     Little. BB; Spalding. S: Walsh. B; Keves. DC: Wainer. J; Pickens. S: Rovster. M; Villanacci. J: Gratton. T.
        (2009). Blood lead levels and growth status among African-American and Hispanic children in Dallas,
        Texas-1980 and 2002: Dallas Lead Project II. Ann Hum Biol 36: 331-341.
        http://dx.doi.org/10.1080/03014460902806615

     Liu. HT; Niu. RY; Wang. JM; He. Y; Wang. JD. (2008). Changes caused by fluoride and lead in energy
        metabolic enzyme activities in the reproductive system of male offspring rats. Fluoride 41: 184-191.

     Liu. J; Han. D; Li. Y; Zheng.  L; Gu. C: Piao. Z; Au. WW: Xu. X; Huo. X. (201 Ob). Lead affects apoptosis
        and related gene XIAP and Smac expression in the hippocampus of developing rats. Neurochem Res
        35: 473-479. http://dx.doi.org/10.1007/sll064-009-0083-9

     Liu. J; Kershaw. WC: Klaassen. CD. (1991). The protective effect of metallothionein on the toxicity of
        various metals in rat primary hepatocyte culture. Toxicol Appl Pharmacol 107: 27-34.
        http://dx.doi.org/10.1016/0041-008X(91)90327-B

     Liu. J: Xu. X: Wu. K: Piao. Z: Huang. J: Quo. Y: Li. W: Zhang. Y: Chen. A: Huo. X. (20lib). Association
        between lead exposure from electronic waste recycling and child temperament alterations.
        Neurotoxicology 32: 458-464. http://dx.doi.Org/10.1016/i.neuro.2011.03.012

     Liu. L; Wong. T; Pozza. M; Lingenhoehl. K; Wang. Y;  Sheng. M; Auberson. Y; Wang. Y.  (2004). Role of
        NMDA receptor subtypes  in governing the direction of hippocampal synaptic plasticity. Science 304:
        1021-1024. http://dx.doi.org/10.1126/science.1096615

     Liu. MY; Hsieh. WC: Yang. BC. (2000). In vitro aberrant gene expression as the indicator of lead-
        induced neurotoxicity in U-373MG cells. Toxicology 147: 59-64.
        http://dx.doi.org/10.1016/S0300-483X(00)00186-4

     Llanos. MN; Ronco. AM. (2009). Fetal growth restriction is related to placenta! levels of cadmium, lead
        and arsenic but not with antioxidant activities. Reprod Toxicol 27: 88-92.
        http://dx.doi.0rg/10.1016/i.reprotox.2008.ll.057

     Lodi. S: Sharma. V: Kansal. L. (2011). The protective effect of Rubia cordifolia against lead nitrate-
        induced immune response impairment and kidney oxidative damage. Indian J Pharmacol 43: 441-444.
        http://dx.doi.org/10.4103/0253-7613.83118

     Lohmann. C: Bonhoeffer. T. (2008). A role for local calcium signaling in rapid synaptic partner selection
        by dendritic filopodia. Neuron 59: 253-260. http://dx.doi.Org/10.1016/j.neuron.2008.05.025

     Lolin. Y;  O'Gorman. P. (1988). An intra-erythrocyctic low molecular weight lead-binding protein in acute
        and chronic lead exposure and its possible protective role in lead toxicity. Ann Clin Biochem 25:
        688-697.

     Long. GJ: Pounds. JG: Rosen. JF (1992). Lead intoxication alters basal and parathyroid hormone-
        regulated cellular calcium homeostasis in rat osteosarcoma (ROS 17/28) cells. Calcif Tissue Int 50:
        451-458. http://dx.doi.org/10.1007/BF00296777
November 2012                              5-755                  Draft - Do Not Cite or Quote

-------
     Long. GJ: Rosen. JF. (1992). Lead perturbs epidermal growth factor (EOF) modulation of intracellular
        calcium metabolism and collagen synthesis in clonal rat osteoblastic (ROS 17/28) cells. Toxicol Appl
        Pharmacol 114: 63-70. http://dx.doi.org/10.1016/0041-008X(92)90097-C

     Long. GJ: Rosen. JF; Pounds. JG. (1990). Cellular lead toxicity and metabolism in primary and clonal
        osteoblastic bone cells. Toxicol Appl Pharmacol 102: 346-361.

     Long. GJ: Rosen. JF: Schanne. FAX. (1994). Lead activation of protein kinase C from rat brain:
        Determination of free calcium, lead, and zinc by 19F NMR. J Biol Chem 269: 834-837.

     Long. Y; Li. Q; Li. J; Cui. Z. (2011). Molecular analysis, developmental function and heavy metal-
        induced expression of ABCC5 in zebrafish. Comp Biochem Physiol B Biochem Mol Biol 158: 46-55.
        http://dx.doi.0rg/10.1016/i.cbpb.2010.09.005

     Lopez-Larrubia. P; Cauli. O. (2011). Alterations of apparent diffusion coefficient (ADC) in the brain of
        rats chronically exposed to lead acetate. Toxicology 281: 1-6.
        http://dx.doi.0rg/10.1016/i.tox.2010.12.009

     Lopez. CM; Pineiro. AE; Nunez. N; Avagnina. AM; Villaamil. EC: Roses. OE. (2000). Thyroid hormone
        changes in males exposed to lead in the Buenos Aires area (Argentina). Pharmacol Res 42: 599-602.
        http://dx.doi.org/10.1006/phrs.2000.0734

     Louis. ED; Applegate. L; Graziano. JH; Parides. M; Slavkovich. V; Bhat. HK. (2005). Interaction
        between blood lead concentration and delta-amino-levulinic acid dehydratase gene polymorphisms
        increases the odds of essential tremor. Mov Disord 20: 1170-1177.
        http://dx.doi.org/10.1002/mds.20565

     Louis. ED; Jurewicz. EC: Applegate. L; Factor-Litvak. P; Parides. M; Andrews. L; Slavkovich. V;
        Graziano. JH; Carroll. S: Todd. A. (2003). Association between essential tremor and blood lead
        concentration. Environ Health Perspect 111: 1707-1711.

     Lu. H; Guizzetti. M; Costa. LG. (2002). Inorganic lead activates the mitogen-activated protein kinase
        kinase-mitogen-activated protein kinase-p90RSK signaling pathway in human astrocytoma cells via a
        protein kinase C-dependent mechanism. J Pharmacol Exp Ther 300: 818-823.

     Lundstrom. NG: Englyst. V: Gerhardsson. L; Jin. T; Nordberg. G. (2006). Lung cancer development in
        primary smelter workers: A nested case-referent study. J Occup Environ Med 48: 376-380.
        http://dx.doi.org/10.1097/01.iom.0000201556.95982.95

     Lustberg. M; Silbergeld. E. (2002). Blood lead levels and mortality. Arch Intern Med 162: 2443-2449.

     Lutz. PM; Wilson. TJ; Ireland. J: Jones. AL; Gorman. JS: Gale. NL; Johnson. JC: Hewett. JE. (1999).
        Elevated immunoglobulin E (IgE) levels in children with exposure to environmental lead. Toxicology
        134: 63-78. http://dx.doi.org/10.1016/S0300-483X(99)00036-0

     Lynes. MA; Fontenot. AP; Lawrence. DA; Rosenspire. AJ; Pollard. KM. (2006). Gene expression
        influences on metal immunomodulation [Review]. Toxicol Appl Pharmacol 210: 9-16.
        http://dx.doi.0rg/10.1016/i.taap.2005.04.021

     MacDonald. JF: Jackson. MF; Beazely. MA. (2006). Hippocampal long-term synaptic plasticity and
        signal amplification of NMDAreceptors [Review]. Crit Rev Neurobiol 18: 71-84.

     Machida. M; Sun. SJ: Oguma. E; Kavama. F (2009).  High bone matrix turnover predicts blood levels of
        lead among perimenopausal women. Environ Res  109: 880-886.
        http://dx.doi.0rg/10.1016/i.envres.2009.06.005

     Maciejczyk. P; Chen. LC. (2005). Effects of subchronic exposures to concentrated ambient particles
        (CAPs) in mice: VIII. Source-related  daily variations in in vitro responses to CAPs. Inhal Toxicol  17:
        243-253. http://dx.doi.org/10.1080/08958370590912914

     Madhavi. D; Devi. KR; Rao. KK; Reddv. PP. (2007). Modulating effect of Phyllanthus fruit extract
        against lead genotoxicity in germ cells of mice. J Environ Biol 28: 115-117.
November 2012                             5-756                   Draft - Do Not Cite or Quote

-------
     Maeda. K; Sugino. H; Hirose. T; Kitagawa. H; Nagai. T; Mizoguchi. H; Takuma. K; Yamada. K. (2007).
        Clozapine prevents a decrease in neurogenesis in mice repeatedly treated with phencyclidine. J
        Pharmacol Sci 103: 299-308. http://dx.doi.org/10.1254/iphs.FP0061424

     Magalhaes. C: Socodato. R; Paes-de-Carvalho. R. (2006). Nitric oxide regulates the proliferation of chick
        embryo retina cells by a cyclic GMP-independent mechanism. Int J Dev Neurosci 24: 53-60.
        http://dx.doi.0rg/10.1016/i.ijdevneu.2005.10.004

     Magyar. JS: Weng. TC: Stern. CM; Dye. DF; Rous. BW: Payne. JC: Bridgewater. BM; Mijovilovich. A;
        Parkin. G: Zaleski. JM; Penner-Hahn. JE; Godwin. HA. (2005). Reexamination of lead(II)
        coordination preferences in sulfur-rich sites: Implications for a critical mechanism of lead poisoning. J
        Am Chem Soc 127: 9495-9505. http://dx.doi.org/10.1021/ia0424530

     Mahaffev. KR; Capar. SG: Gladen. BC: Fowler. BA. (1981). Concurrent exposure to lead, cadmium, and
        arsenic. Effects on toxicity and tissue metal concentrations in the rat. Transl Res 98: 463-481.

     Mahaffev. KR; Fowler. BA. (1977). Effects of concurrent administration of lead, cadmium, and arsenic in
        the rat. Environ Health Perspect 19: 165-171. http://dx.doi.org/10.1289/ehp.7719165

     Mahram. M; Mousavinasab. N; Dinmohammadi. H; Soroush. S: Sarkhosh. F. (2007). Effect of living in
        lead mining area on growth. Indian J Pediatr 74: 555-559. http://dx.doi.org/10.1007/sl2098-007-0107-
        x

     Maitani. T; Watahiki. A; Suzuki. KT. (1986). Induction of metallothionein after lead administration by
        three injection routes in mice. Toxicol Appl Pharmacol 83: 211-217.
        http://dx.doi.org/10.1016/0041-008X(86)90298-X

     Maizlish. NA; Parra. G: Feo. O. (1995). Neurobehavioural evaluation of Venezuelan workers exposed to
        inorganic lead. Occup Environ Med 52: 408-414. http://dx.doi.Org/10.1136/oem.52.6.408

     Malveaux. FJ; Conroy. MC: Adkinson. NF. Jr; Lichtenstein. LM. (1978). IgE receptors on human
        basophils: Relationship to serum IgE concentration. J Clin Invest 62: 176-181.
        http://dx.doi.org/10.1172/JCI109103

     Malvezzi. CK; Moreira. EG: Vassilieff. I; Vassilieff. VS: Cordellini. S. (2001). Effect of L-arginine,
        dimercaptosuccinic acid (DMSA) and the association of L-arginine and DMSA on tissue lead
        mobilization and blood pressure level in plumbism. Braz J Med Biol Res 34: 1341-1346.
        http://dx.doi.org/10.1590/S0100-879X2001001000016

     Marchlewicz. M; Baranowska-Bosiacka. I; Kolasa. A; Kondarewicz. A; Chlubek. D; Wiszniewska. B.
        (2009). Disturbances of energetic metabolism in rat epididymal epithelial cells as  a consequence of
        chronic lead intoxication. Biometals 22: 877-887. http://dx.doi.org/10.1007/sl0534-009-9238-z

     Marcus. DK; Fulton. JJ; Clarke. EJ. (2010). Lead and  conduct problems: Ameta-analysis. J Clin Child
        Adolesc Psychol 39: 234-241. http://dx.doi.org/10.1080/15374411003591455

     Marie. C: Hall. JE. (2011). Obesity, metabolic syndrome and diabetic nephropathy  [Review]. Contrib
        Nephrol 170: 28-35. http://dx.doi.org/10.1159/000324941

     Markovac. J: Goldstein. GW. (1988a).  Lead activates protein kinase C in immature rat brain microvessels.
        Toxicol Appl Pharmacol 96: 14-23. http://dx.doi.org/10.1016/0041-008X(88)90242-6

     Markovac. J: Goldstein. GW. (1988b).  Picomolar concentrations of lead stimulate brain protein kinase
        C. Nature 334: 71-73. http://dx.doi.org/10.1038/334071aO

     Marques. CC: Nunes. AC: Pinheiro. T; Lopes. PA; Santos. MC: Viegas-Crespo. AM; Ramalhinho. MG:
        Mathias. ML. (2006). An assessment of time-dependent effects of lead exposure in algerian mice (Mus
        spretus) using different methodological approaches. Biol Trace Elem Res 109: 75-90.
        http://dx.doi.Org/10.1385/BTER:109:l:075

     Marques. M; Millas. I; Jimenez. A; Garcia-Colis. E; Rodriguez-Feo. JA; Velasco. S: Barrientos. A;
        Casado. S: Lopez-Farre. A. (2001). Alteration of the soluble guanylate cyclase system in the vascular
        wall of lead-induced hypertension in rats. J Am Soc Nephrol 12: 2594-2600.
November 2012                              5-757                  Draft - Do Not Cite or Quote

-------
     Martin. D; Glass. TA; Bandeen-Roche. K; Todd. AC: Shi. WP; Schwartz. BS. (2006). Association of
        blood lead and tibia lead with blood pressure and hypertension in a community sample of older adults.
        Am J Epidemiol 163: 467-478. http://dx.doi.org/10.1093/aje/kwj060

     Martinez-Haro. M; Green. AJ; Mateo. R. (2011). Effects of lead exposure on oxidative stress biomarkers
        and plasma biochemistry in waterbirds in the field. Environ Res 111: 530-538.
        http://dx.doi.0rg/10.1016/i.envres.2011.02.012

     Martinez. T; Lartigue. J: Juarez. F; Avila-perez. P; Marquez. C: Zarazua. G: Tejeda. S. (2007).
        206Pb/207Pb ratios in dry deposit samples from the Metropolitan Zone of Mexico Valley. Journal of
        Radioanal Chem 273: 577-582. http://dx.doi.org/10.1007/sl0967-007-0913-7

     Mas-Oliva. J. (1989). Effect of lead on the erythrocyte (Ca2+,Mg2+)ATPase activity: Calmodulin
        involvement. Mol Cell Biochem 89: 87-93. http://dx.doi.org/10.1007/BF00228283

     Massadeh. AM; Al-Safi. SA; Momani.  IF; Al-Mahmoud. M; Alkofahi. AS. (2007). 115. Biol Trace Elem
        Res 115: 157-167. http://dx.doi.org/10.1007/BF02686027

     Massanyi. P; Lukac. N; Makarevich. AV; Chrenek. P; Forgacs. Z; Zakrzewski. M;  Stawarz. R; Toman. R;
        Lazor. P; Flesarova. S. (2007). Lead-induced alterations in rat kidneys and testes in vivo. J Environ
        Sci Health A Tox Hazard Subst Environ Eng 42: 671-676.
        http://dx.doi.org/10.1080/10934520701244474

     Masso-Gonzalez. EL; Antonio-Garcia.  MT. (2009). Natural antioxidants protect against lead-induced
        damage during pregnancy and lactation in rat's pups. Ecotoxicol Environ Saf 72: 2137-2142.
        http://dx.doi.0rg/10.1016/i.ecoenv.2009.03.013

     Masso. EL; Corredor. L; Antonio. MT.  (2007). Oxidative damage in liver after perinatal intoxication with
        lead and/or cadmium. J Trace Elem Med Biol 21: 210-216.
        http://dx.doi.0rg/10.1016/i.itemb.2007.03.002

     Matovic. V: Buha. A; Bulat. Z; Dukic-Cosic. D. (2011). Cadmium toxicity revisited: Focus on oxidative
        stress induction and interactions with zinc and magnesium [Review]. Arh Hig Rada Toksikol 62:
        65-76. http://dx.doi.org/10.2478/10004-1254-62-2011-2075

     Maue'l. J: Ransijn. A; Buchmuller-Rouiller. Y. (1989). Lead inhibits intracellular killing of Leishmania
        parasites and extracellular cytolysis of target cells by macrophages exposed to macrophage activating
        factor. J Leukoc Biol 45: 401-409.

     Mazumdar. M; Bellinger. DC: Gregas.  M; Abanilla. K; Bacic. J: Needleman. HL. (2011). Low-level
        environmental lead exposure in childhood and adult intellectual function: A follow-up study. Environ
        Health Global Access Sci Source 10: 24. http://dx.doi.org/10.1186/1476-069X-10-24

     McCabe. MJ. Jr; Lawrence. DA.  (1990). The heavy metal lead exhibits B cell-stimulatory factor activity
        by enhancing B cell la expression and differentiation. J Immunol 145: 671-677.

     McCabe. MJ. Jr; Lawrence. DA.  (1991). Lead, a major environmental pollutant, is immunomodulatory by
        its differential effects on CD4+T cell subsets. Toxicol Appl Pharmacol  111: 13-23.
        http://dx.doi.org/10.1016/0041-008X(91)90129-3

     McCabe. MJ. Jr; Singh. KP; Reiners. JJ. Jr. (1999). Lead intoxication impairs the generation of a delayed
        type hypersensitivity response. Toxicology 139: 255-264.
        http://dx.doi.org/10.1016/S0300-483X(99)00147-X

     McElroy. JA; Shafer. MM; Gangnon. RE; Crouch. LA; Newcomb. PA.  (2008). Urinary lead exposure and
        breast cancer risk in a population-based case-control study. Cancer Epidemiol Biomarkers Prev 17:
        2311-2317. http://dx.doi.org/10.1158/1055-9965.epi-08-0263

     McLachlin. JR; Gover. RA; Cherian. MG. (1980). Formation of lead-induced inclusion bodies in primary
        rat kidney epithelial cell cultures: Effect of actinomycin D and cycloheximide. Toxicol Appl
        Pharmacol 56: 418-431. http://dx.doi.org/10.1016/0041-008X(80)90076-9
November 2012                             5-758                   Draft - Do Not Cite or Quote

-------
     McMichael. AJ; Baghurst PA; Vimpani. GV: Robertson. EF; Wigg. NR; Tong. SL. (1992).
        Sociodemographic factors modifying the effect of environmental lead on neuropsychological
        development in early childhood. Neurotoxicol Teratol 14: 321-327.
        http://dx.doi.org/10.1016/0892-0362(92)90038-C

     McMichael. AJ: Baghurst. PA; Wigg. NR; Vimpani. GV: Robertson. EF; Roberts. RJ. (1988). Port Pirie
        cohort study: Environmental exposure to lead and children's abilities at the age of four years. N Engl J
        Med 319: 468-475. http://dx.doi.org/10.1056/NEJM19880825319080

     McNeill. PR; Wong. HK; Naravana. A; Wilson. DM. (2007). Lead promotes abasic site accumulation and
        co-mutagenesis in mammalian cells by inhibiting the major abasic endonuclease Apel. Mol Carcinog
        46: 91-99. http://dx.doi.org/10.1002/mc.20196

     Meek. WH; Williams. CL; Cermak. JM; Blusztajn. JK. (2007). Developmental periods of choline
        sensitivity provide an ontogenetic mechanism for regulating memory capacity and age-related
        dementia. Frontiers in Integrative Neuroscience 1: 7. http://dx.doi.org/10.3389/neuro.07.007.2007

     Meeker. JD; Rossano. MG: Protas. B;  Diamond. MP; Puscheck. E; Daly. D; Paneth. N; Wirth. JJ. (2008).
        Cadmium, lead, and other metals in relation to semen quality: Human evidence for molybdenum as a
        male reproductive toxicant. Environ Health Perspect 116: 1473-1479.
        http://dx.doi.org/10.1289/ehp.11490

     Meeker. JD; Rossano. MG: Protas. B;  Padmanahban. V; Diamond. MP; Puscheck. E; Daly. D; Paneth. N;
        Wirth. JJ. (2010). Environmental exposure to metals and male reproductive hormones: Circulating
        testosterone is inversely associated with blood molybdenum. Fertil Steril 93: 130-140.
        http://dx.doi.0rg/10.1016/i.fertnstert.2008.09.044

     Mendez-Armenta. M; Nava-Ruiz. C: Fernandez-Valverde. F; Sanchez-Garcia. A; Rios. C. (2011).
        Histochemical changes in muscle of rats exposed subchronically to low doses of heavy metals.
        Environ Toxicol Pharmacol 32: 107-112. http://dx.doi.Org/10.1016/i.etap.2011.03.001

     Mendez-Gomez. J; Garcia-Vargas. GG: Lopez-Carrillo. L; Calderon-Aranda. ES: Gomez. A; Vera. E;
        Valverde. M; Cebrian. ME; Rojas.  E. (2008). Genotoxic effects of environmental exposure to arsenic
        and lead on children in region Lagunera, Mexico. Ann N Y Acad  Sci 1140: 358-367.
        http://dx.doi.org/10.1196/annals.1454.027

     Mendiola. J: Moreno. JM; Roca. M; Vergara-Juarez. N; Martinez-Garcia. MJ; Garcia-Sanchez. A; Elvira-
        Rendueles. B;  Moreno-Grau. S: Lopez-Espin. JJ:  Ten. J: Bernabeu. R; Torres-Cantero. AM. (2011).
        Relationships between heavy metal concentrations in three different body fluids and male reproductive
        parameters: A pilot study. Environ  Health Global Access Sci Source  10: 6.
        http://dx.doi.org/10.1186/1476-069X-10-6

     Mendv. A; Gasana. J: Vieira. ER.  (2012). Urinary heavy metals and associated medical conditions in the
        US adult population.  Int J Environ Health Res 22: 105-118.
        http://dx.doi.org/10.1080/09603123.2011.605877

     Menegazzi. M; Carcereri-De Prati. A;  Suzuki. H; Shinozuka. H; Pibiri. M; Piga. R; Columbano. A;
        Ledda-Columbano. GM. (1997). Liver cell proliferation induced by nafenopin and cyproterone acetate
        is not associated with increases in activation  of transcription factors NF-kappaB and AP-1 or with
        expression of tumor necrosis factor alpha. Hepatology 25: 585-592.
        http://dx.doi.org/10.1002/hep.510250316

     Meng. XM; Zhu. DM; Ruan. DY; She. JQ: Luo. L. (2005). Effects of chronic lead exposure on H MRS of
        hippocampus and frontal lobes in children. Neurology 64: 1644-1647.
        http://dx.doi.org/10.1212/01.WNL.0000160391.58004.D4

     Menke. A; Muntner. P; Batuman. V; Silbergeld. EK;  Guallar. E. (2006). Blood lead below 0.48
        micromol/L (10 microg/dL) and mortality among US adults. Circulation 114:  1388-1394.
        http://dx.doi.org/10.1161/circulationaha.106.628321
November 2012                             5-759                  Draft - Do Not Cite or Quote

-------
     Mesdaghinia. A; Yazdanpanah. H; Seddighi. M; Banafshe. HR; Hevdari. A. (2010). Effect of short-term
        lead exposure on PTZ-induced seizure threshold in mice. Toxicol Lett 199: 6-9.
        http://dx.doi.0rg/10.1016/i.toxlet.2010.07.012

     Messerschmidt. A; Huber. R; Wieghardt. K; Poulos. T. (2001). Handbook of metalloproteins. In A
        Messerschmidt; R Huber; K Wieghardt; T Poulos (Eds.). Chichester, U.K.: Wiley.
        http://dx.doi.org/10.1002/0470028637

     Mielzvnska. D; Siwinska. E; Kapka. L; Szyfter. K; Knudsen. LE; Merlo. DF. (2006). The influence of
        environmental exposure to complex mixtures including PAHs and lead on genotoxic effects in
        children living in Upper Silesia, Poland. Mutagenesis 21: 295-304.
        http://dx.doi.org/10.1093/mutage/gel037

     Miller. TE; Golemboski. KA; Ha. RS; Bunn. T; Sanders. FS; Dietert. RR. (1998). Developmental
        exposure to lead causes persistent immunotoxicity in Fischer 344 rats. Toxicol Sci 42: 129-135.
        http://dx.doi.org/10.1006/toxs.1998.2424

     Min. JY; Min. KB; Cho. SI; Kim. R; Sakong. J; Paek. D. (2007). Neurobehavioral function in children
        with low blood lead concentrations. Neurotoxicology 28: 421-425.
        http://dx.doi.0rg/10.1016/i.neuro.2006.3.007

     Min. JY; Min. KB; Kim. R; Cho. SI; Paek. D. (2008a). Blood lead levels and increased bronchial
        responsiveness.  Biol Trace Elem Res  123: 41-46. http://dx.doi.org/10.1007/sl2011-008-8099-6

     Min. KB; Min. JY; Cho. SI; Kim. R; Kim. H; Paek. D. (2008b). Relationship between low blood lead
        levels and growth in children of white-collar civil servants in Korea. Int J Hyg Environ Health 211:
        82-87. http://dx.doi.0rg/10.1016/j.ijheh.2007.03.003

     Min. MYO; Singer. LT; Kirchner. HL; Minnes. S; Short. E; Hussain. Z; Nelson. S. (2009). Cognitive
        development and low-level lead exposure in poly-drug exposed children. Neurotoxicol Teratol 31:
        225-231. http://dx.doi.0rg/10.1016/i.ntt.2009.03.002

     Minozzo. R; Deimling. LI; Gigante. LP; Santos-Mello. R. (2004). Micronuclei in peripheral blood
        lymphocytes of workers exposed to lead. Mutat Res Genet Toxicol Environ Mutagen 565: 53-60.
        http://dx.doi.0rg/10.1016/i.mrgentox.2004.09.003

     Minozzo. R; Deimling. LI; Santos-Mello. R. (2010). Cytokinesis-blocked micronucleus cytome and
        comet assays in peripheral blood lymphocytes  of workers exposed to lead considering folate and
        vitamin B12 status. Mutat Res Genet Toxicol Environ Mutagen 697: 24-32.
        http://dx.doi.0rg/10.1016/i.mrgentox.2010.01.009

     Miranda. ML; Edwards. SE; Swamy. GK; Paul. CJ; Neelon. B.  (2010).  Blood lead levels among pregnant
        women: Historical versus contemporaneous exposures. Int J Environ Res Public Health 7:  1508-1519.
        http://dx.doi.org/10.3390/iierph7041508

     Miranda. ML; Kim. D; Galeano. M; Paul. CJ; Hull. AP; Morgan. SP (2007a). The relationship between
        early childhood blood lead levels and performance  on end of grade tests. Environ Health Perspect 115:
        1242-1247. http://dx.doi.org/10.1289/ehp.9994

     Miranda. ML; Kim. D; Reiter. J; Overstreet Galeano. MA; Maxson. P. (2009). Environmental contributors
        to the achievement gap. Neurotoxicology 30: 1019-1024.
        http://dx.doi.0rg/10.1016/i.neuro.2009.07.012

     Mishra. KP. (2009). Lead exposure and its impact on immune system: A review [Review]. Toxicol In
        Vitro 23: 969-972. http://dx.doi.Org/10.1016/i.tiv.2009.06.014

     Mishra. KP; Chauhan. U.K.; Naik. S. (2006a). Effect of lead exposure on serum immunoglobulins and
        reactive nitrogen and oxygen intermediate. Hum Exp Toxicol 25: 661-665.
        http://dx.doi.org/10.1177/0960327106070453

     Mishra. KP; Rani. R; Yadav. VS; Naik. S. (2010).  Effect of lead exposure on lymphocyte subsets and
        activation markers. Immunopharmacol Immunotoxicol 32: 446-449.
        http://dx.doi.org/10.3109/08923970903503668
November 2012                             5-760                   Draft - Do Not Cite or Quote

-------
     Mishra. KP; Singh. VK; Rani. R; Yadav. VS: Chandran. V; Srivastava. SP; Seth. PK. (2003). Effect of
        lead exposure on the immune response of some occupationally exposed individuals. Toxicology 188:
        251-259. http://dx.doi.org/10.1016/S0300-483X(03)00091-X

     Misonou. H; Morishima-Kawashima. M; Ihara. Y. (2000). Oxidative stress induces intracellular
        accumulation of amyloid beta-protein (Abeta) in human neuroblastoma cells. Biochemistry 39:
        6951-6959. http://dx.doi.org/10.1021M000169p

     Mistry. P; Lucier. GW: Fowler. BA. (1985). High-affinity lead binding proteins in rat kidney cytosol
        mediate cell-free nuclear translocation of lead. J Pharmacol Exp Ther 232: 462-469.

     Mistry. P; Mastri. C: Fowler. BA. (1986). Influence of metal ions on renal cytosolic lead-binding proteins
        and nuclear uptake of lead in the kidney. Biochem Pharmacol 35: 711-713.
        http://dx.doi.org/10.1016/0006-2952(86)90371-0

     Moffitt. TE; Caspi. A; Rutter. M; Silva. PA. (2001). Sex differences in antisocial behavior: Conduct
        disorder, delinquency, and violence in the Dunedin Longitudinal Study. Cambridge, U.K.: University
        of Cambridge Press.

     Mohammad. A; Ali. N; Reza. B; Ali. K. (2010). Effect of ascorbic acid supplementation on nitric oxide
        metabolites and systolic blood pressure  in rats exposed to lead. Indian J Pharmacol 42: 78-81.
        http://dx.doi.org/10.4103/0253-7613.64501

     Mohammad. IK;  Mahdi. AA; Raviraja. A; Naimul. I; Iqbal. A; Thuppil. V. (2008). Oxidative stress in
        painters exposed to low lead levels. Arh Hig Rada Toksikol 59: 161-169.
        http://dx.doi.org/10.2478/10004-1254-59-2008-1883

     Molero. L; Carrasco. C: Marques. M; Vaziri. ND; Mateos-Caceres. PJ: Casado. S: Macava. C: Barrientos.
        A; Lopez-Farre. AJ. (2006). Involvement of endothelium and  endothelin-1 in lead-induced smooth
        muscle cell dysfunction in rats. Kidney  Int 69: 685-690. http://dx.doi.org/10.1038/sj.ki.5000103

     Molina. RM; Phattanarudee. S: Kim. J:  Thompson. K; Wessling-Resnick. M; Maher. TJ; Brain. JD.
        (2011). Ingestion of Mn and Pb by rats during and after pregnancy alters iron metabolism and
        behavior in offspring. Neurotoxicology  32: 413-422. http://dx.doi.Org/10.1016/i.neuro.2011.03.010

     Monteiro. HP; Abdalla. DSP: Augusto.  O: Bechara. EJH. (1989).  Free radical generation during delta-
        aminolevulinic acid autoxidation: Induction by hemoglobin and connections with porphyrinpathies.
        Arch Biochem Biophys 271: 206-216. http://dx.doi.org/10.1016/0003-9861(89)90271-3

     Monteiro. HP; Abdalla. DSP: Faljoni-Alario. A; Bechara. EJH. (1986). Generation of active oxygen
        species during coupled autoxidation of oxyhemoglobin and delta-aminolevulinic acid. Biochim
        Biophys Acta 881: 100-106. http://dx.doi.org/10.1016/0304-4165(86)90102-9

     Monteiro. HP; Bechara. EJH: Abdalla. DSP. (1991). Free radicals involvement in neurological porphyrias
        and lead poisoning. Mol Cell  Biochem 103: 73-83. http://dx.doi.org/10.1007/BF00229595

     Montenegro. MF; Barbosa. F. Jr; Sandrim. VC: Gerlach. RF; Tanus-Santos. JE. (2006). A polymorphism
        in the delta-aminolevulinic acid dehydratase gene modifies plasma/whole blood lead ratio. Arch
        Toxicol 80: 394-398. http://dx.doi.org/10.1007/s00204-005-0056-v

     Moore. CF; Gaiewski. LL; Laughlin. NK; Luck. ML; Larson. JA; Schneider. ML. (2008). Developmental
        lead exposure induces tactile defensiveness in rhesus monkeys (Macaca mulatta). Environ Health
        Perspect 116:  1322-1326. http://dx.doi.org/10.1289/ehp.11203

     Moore. JF; Gover. RA. (1974). Lead-induced inclusion bodies: Composition and probable role in  lead
        metabolism. Environ Health Perspect 7:  121-127. http://dx.doi.org/10.1289/ehp.747121

     Moore. JF; Gover. RA; Wilson. M.  (1973).  Lead-induced inclusion bodies: Solubility, amino acid content,
        and relationship to residual acidic nuclear proteins. Lab Invest 29: 488-494.

     Moorhouse. SR; Garden. S: Drewitt PN; Elev. BP; Hargreaves. PJ: Felling. D. (1988). The effect of
        chronic low level lead exposure on blood-brain barrier function in the developing rat. Biochem
        Pharmacol 37: 4539-4547. http://dx.doi.org/10.1016/0006-2952(88)90670-3
November 2012                              5-761                   Draft - Do Not Cite or Quote

-------
     Moorman. WJ: Skaggs. SR; Clark. JC: Turner. TW: Sharpnack. DP; Murrell. JA; Simon. SD; Chapin.
        RE; Schrader. SM. (1998). Male reproductive effects of lead, including species extrapolation for the
        rabbit model. Reprod Toxicol 12: 333-346. http://dx.doi.org/10.1016/S0890-6238(98)00010-0

     Morita. Y; Sakai. T; Araki. S: Araki. T; Masuvama. Y. (1997). Nicotinamide adenine dinucleotide
        synthetase activity in erythrocytes as a tool for the biological monitoring of lead exposure. Int Arch
        Occup Environ Health 70: 195-198. http://dx.doi.org/10.1007/s004200050206

     Morris. S: van Aardt. WJ: Ahern. MD. (2005). The effect of lead on the metabolic and energetic status of
        the Yabby, Cherax destructor, during environmental hypoxia. Aquat Toxicol 75:  16-31.
        http://dx.doi.0rg/10.1016/i.aquatox.2005.07.001

     Mosad. SM; Ghanem. AA; El-Fallal. HM; El-Kannishy. AM; El Baiomy. AA; Al-Diastv. AM; Arafa. LF.
        (2010). Lens cadmium, lead, and serum vitamins C, E, and beta carotene in cataractous smoking
        patients. Curr Eye Res 35: 23-30. http://dx.doi.org/10.3109/02713680903362880

     Moser. R; Oberlev. TD; Daggett. DA; Friedman. AL; Johnson. JA; Siegel. FL. (1995). Effects of lead
        administration on developing rat kidney: I Glutathione S-transferase isoenzymes. Toxicol Appl
        Pharmacol 131: 85-93. http://dx.doi.org/10.1006/taap.1995.1050

     Moss. ME; Lanphear. BP; Auinger. R (1999). Association of dental caries and blood lead levels. JAMA
        281: 2294-2298. http://dx.doi.org/10.1001/iama.281.24.2294

     Mudipalli. A. (2007). Lead hepatotoxicity & potential health effects [Review]. Indian J Med Res 126:
        518-527.

     Mudzinski.  SP; Rudofsky. UH; Mitchell. DG: Lawrence. DA. (1986). Analysis of lead effects on in vivo
        antibody-mediated immunity in several mouse strains. Toxicol Appl Pharmacol 83: 321-330.
        http://dx.doi.org/10.1016/0041-008X(86)90309-l

     Muldoon. SB; Caulev. JA: Kuller. LH; Morrow. L; Needleman. HL; Scott. J: Hooper. FJ. (1996). Effects
        of blood lead levels on cognitive function of older women. Neuroepidemiology  15: 62-72.
        http://dx.doi.org/10.1159/000109891

     Miiller. S: Gillert. KE; Krause. C: Gross. U: L'Age-Stehr. J: Diamantstein. T. (1977). Suppression of
        delayed type hypersensitivity of mice by lead. Experientia 33: 667-668.
        http://dx.doi.org/10.1007/BF01946564

     Miiller. YM; Kobus. K; Schatz. JC: Ammar. D; Nazari. EM. (2012). Prenatal lead acetate exposure
        induces apoptosis and changes GFAP expression during spinal cord development. Ecotoxicol Environ
        Saf 75: 223229. http://dx.doi.Org/10.1016/i.ecoenv.2011.08.004

     Muntner. P; He. J: Vupputuri. S: Coresh. J: Batuman. V. (2003). Blood lead and chronic kidney disease in
        the general United States population:  Results from NHANES III. Kidney Int 63: 1044-1050.
        http://dx.doi.0rg/10.1046/i.1523-1755.2003.00812.x

     Muntner. P; Menke. A; Batuman. V; Rabito. FA; He. J: Todd. AC. (2007). Association of tibia lead and
        blood lead with end-stage renal disease: A pilot study of African-Americans. Environ Res 104:
        396-401. http://dx.doi.0rg/10.1016/i.envres.2007.04.001

     Muntner. P; Menke. A; DeSalvo. KB; Rabito. FA; Batuman. V. (2005). Continued decline in blood lead
        levels among adults in the United States - The National Health and Nutrition Examination Surveys.
        Arch Intern Med 165: 2155-2161. http://dx.doi.org/10.1001/archinte.165.18.2155

     Murakami. K; Feng. G: Chen. SG. (1993). Inhibition of brain protein kinase C subtypes by lead. J
        Pharmacol Exp Ther 264: 757-761.

     Murata. K; Iwata. T; Dakeishi. M; Karita. K. (2009). Lead toxicity: Does the critical level of lead
        resulting in adverse effects differ between adults and children? [Review]. J Occup Health 51: 1-12.
        http://dx.doi.org/10.1539/ioh.K8003

     Murthy. RC: Gupta. SK; Saxena. DK. (1995). Nuclear alterations during acrosomal cap formation in
        spermatids of lead-treated rats. Reprod Toxicol  9: 483-489.
November 2012                              5-762                   Draft - Do Not Cite or Quote

-------
     Muwakkit. S: Nuwayhid. I; Nabulsi. M; al Haij. R; Khoury. R; Mikati. M; Abboud. MR. (2008). Iron
        deficiency in young Lebanese children: Association with elevated blood lead levels. J Pediatr Hematol
        Oncol 30: 382-386. http://dx.doi.org/10.1097/MPH.Ob013e318165b283

     Naha. N; Chowdhury. AR. (2006). Inorganic lead exposure in battery and paint factory: Effect on human
        sperm structure and functional activity. J UOEH 28: 157-171.

     Naha. N; Manna. B. (2007). Mechanism of lead induced effects on human spermatozoa after occupational
        exposure. Kathmandu Univ Med J 5: 85-94.

     Naicker. N; Norris. SA; Mathee. A; Becker. P; Richter. L. (2010). Lead exposure is associated with a
        delay in the onset of puberty in South African adolescent females: Findings from the birth to twenty
        cohort. Sci Total Environ 408:  4949-4954. http://dx.doi.Org/10.1016/i.scitotenv.2010.07.037

     Nakagawa. K. (1989). Hepatic glutathione metabolism in mice acutely treated with lead acetate. Jpn J
        Pharmacol 51: 173-179. http://dx.doi.org/10.1254/iip.51.173

     Nakagawa. K. (1991). Decreased glutathione S-transferase activity in mice livers by acute treatment with
        lead, independent of alteration in glutathione  content. Toxicol Lett 56: 13-17.
        http://dx.doi.org/10.1016/0378-4274(91)90085-K

     Nakaiima. T; Deguchi. T; Kagawa. K; Hikita. H; Ueda. K; Katagishi. T; Ohkawara. T; Kakusui. M;
        Kimura. H;  Okanoue. T; Kashima. K; Ashihara. T.  (1995). Identification of apoptotic hepatocytes in
        situ in rat liver after lead nitrate administration. J Gastroenterol 30: 725-730.
        http://dx.doi.org/10.1007/BF02349638

     Nakao. K; Kibavashi. K; Taki. T; Koyama. H. (2010).  Changes in the brain after intracerebral
        implantation of a lead pellet in the rat. J Neurotrauma 27: 1925-1934.
        http://dx.doi.org/10.1089/neu.2010.1379

     Nakhoul. F; Kayne. LH; Brautbar. N; Hu. MS: Mcdonough. A; Eggena. P; Golub. MS: Berger. M; Chang.
        CT; Jamgotchian. N; Lee. DBN. (1992). Rapid hypertensinogenic effect of lead: Studies in the
        spontaneously hypertensive rat. Toxicol Ind Health 8: 89-102.

     Nampoothiri. LP; Agarwal. A; Gupta. S. (2007).  Effect of co-exposure to lead and cadmium on
        antioxidant  status in rat ovarian granulose cells. Arch Toxicol 81: 145-150.
        http://dx.doi.org/10.1007/s00204-006-0133-x

     Nampoothiri. LP; Gupta. S. (2006). Simultaneous effect of lead and cadmium on granulosa cells: A
        cellular model for ovarian toxicity. Reprod Toxicol 21: 179-185.
        http://dx.doi.0rg/10.1016/i.reprotox.2005.07.010

     Nampoothiri. LP; Gupta. S. (2008). Biochemical effects of gestational coexposure to lead and cadmium
        on reproductive performance, placenta, and ovary. J Biochem Mol Toxicol 22: 337-344.
        http://dx.doi.org/10.1002/ibt.20246

     Nandi.  S: Gupta. PS: Selvaraju. S: Roy. SC: Ravindra. JP (2010). Effects of exposure to heavy metals on
        viability, maturation, fertilization, and embryonic development of buffalo (Bubalus bubalis) oocytes in
        vitro. Arch Environ Contam Toxicol 58: 194-204. http://dx.doi.org/10.1007/s00244-009-9342-7

     Naravana. K; Al-Bader. M. (2011). Ultrastructural and DNA damaging effects of lead nitrate in the liver
        [Review]. Exp Toxicol Pathol 63: 43-51. http://dx.doi.Org/10.1016/i.etp.2009.09.007

     Nation. JR; Gleaves. DH. (2001). Low-level lead exposure and intelligence in children [Comment]. Arch
        ClinNeuropsychol 16: 375-388.

     Nava-Hernandez. MP; Hauad-Marroquin. LA; Bassol-Mavagoitia. S: Garcia-Arenas. G: Mercado-
        Hernandez.  R; Echavarri-Guzman. MA; Cerda-Flores. RM. (2009). Lead-, cadmium-, and arsenic-
        induced DNA damage in rat germinal cells. DNA Cell Biol 28: 241-248.
        http://dx.doi.org/10.1089/dna.2009.0860
November 2012                             5-763                   Draft - Do Not Cite or Quote

-------
     Navarro-Moreno. LG: Quintanar-Escorza. MA; Gonzalez. S: Mondragon. R; Cerbon-Solorzano. J;
        Valdes. J: Calderon-Salinas. JV. (2009). Effects of lead intoxication on intercellular junctions and
        biochemical alterations of the renal proximal tubule cells. Toxicol In Vitro 23: 1298-1304.
        http://dx.doi.0rg/10.1016/i.tiv.2009.07.020

     Navas-Acien. A; Schwartz. BS: Rothenberg. SJ: Hu. H; Silbergeld. EK; Guallar. E. (2008). Bone lead
        levels and blood pressure endpoints: Ameta-analysis. Epidemiology 19: 496-504.
        http://dx.doi.org/10.1097/EDE.Ob013e31816a2400

     Navas-Acien. A; Selvin. E; Sharrett. AR; Calderon-Aranda. E; Silbergeld. E; Guallar. E. (2004). Lead,
        cadmium, smoking, and increased risk of peripheral arterial disease. Circulation 109: 3196-3201.
        http://dx.doi.org/10.1161/01.CIR.0000130848.18636.B2

     Navas-Acien. A; Silbergeld. EK; Sharrett. AR; Calderon-Aranda. E; Selvin. E; Guallar. E. (2005). Metals
        in urine and peripheral arterial disease. Environ Health Perspect 113: 164-169.
        http://dx.doi.org/10.1289/ehp.7329

     Navas-Acien. A; Tellez-Plaza. M; Guallar. E; Muntner. P; Silbergeld. E; Jaar. B; Weaver. V. (2009). Blood
        cadmium and lead and chronic kidney disease in US adults: Ajoint analysis. Am J Epidemiol  170:
        1156-1164. http://dx.doi.org/10.1093/aie/kwp248

     Naves. M; Diaz-Lopez. JB; Gomez. C: Rodriguez-Rebollar. A; Cannata-Andia. JB. (2005). Determinants
        of incidence of osteoporotic fractures in the female Spanish population older than 50. Osteoporos Int
        16: 2013-2017. http://dx.doi.org/10.1007/s00198-005-1983-4

     Nawrot TS: Thijs. L; Den Hond. EM; Roels. HA; Staessen. JA. (2002). An epidemiological re-appraisal
        of the association between blood pressure and blood lead: Ameta-analysis. J Hum Hypertens  16:
        123-131. http://dx.doi.org/10.1038/si/ihh/1001300

     Navak. BN; Ray. M; Persaud. TVN. (1989a).  Maternal and fetal chromosomal aberrations in mice
        following prenatal exposure to subembryotoxic doses of lead nitrate. Acta Anat 135:  185-188.
        http://dx.doi.org/10.1159/000146751

     Navak. BN; Ray. M; Persaud. TVN; Nigli. M. (1989b). Relationship of embryotoxicity to genotoxicity of
        lead nitrate in mice. Exp Pathol  36: 65-73.

     Neal. A; Stansfield. K; Worlev. P; Thompson. R; Guilarte. T.  (2010a).  Lead exposure during
        synaptogenesis alters vesicular proteins and impairs vesicular release: Potential role of NMD A
        receptor-dependent BDNF signaling. Toxicol Sci 116: 249-263.
        http://dx.doi.org/10.1093/toxsci/kfqlll

     Neal. RE; Lin. C: Isom. R; Vaishnav. K; Zigler. JS. Jr.  (2010b). Opacification of lenses cultured in the
        presence of Pb. Mol Vis 16: 2137-2145.

     Neaton. JD; Kuller.  L; Stamler. J: Wentworth. DN. (1995). Impact of systolic and diastolic blood pressure
        on cardiovascular mortality. In JH Laragh; BM Brenner (Eds.), Hypertension: Pathophysiology,
        diagnosis, and management (2nd ed., pp. 127-144). New York, NY: Raven Press Ltd.

     Needleman. HL; Bellinger. D. (2001). Studies of lead exposure and the developing central nervous
        system: A reply to Kaufman [Comment]. Arch Clin Neuropsychol  16: 359-374.
        http://dx.doi.org/10.1016/S0887-6177(00)00087-l

     Needleman. HL; Gatsonis. CA. (1990). Low-level lead exposure and the IQ of children: Ameta-analysis
        of modern studies. JAMA 263: 673-678. http://dx.doi.org/10.1001/iama.1990.03440050067035

     Needleman. HL; Gunnoe. C: Leviton. A; Reed. R: Peresie. H; Maher.  C: Barrett. P. (1979). Deficits in
        psychologic and classroom performance of children with elevated dentine lead levels. N Engl J Med
        300: 689-695. http://dx.doi.org/10.1056/NEJM197903293001301

     Needleman. HL; Leviton. A; Bellinger. D. (1982). Lead-associated intellectual deficit [letter] [Letter]. N
        Engl J Med 306: 367. http://dx.doi.org/10.1056/NEJM198202113060619
November 2012                             5-764                   Draft - Do Not Cite or Quote

-------
    Needleman. HL; McFarland. C: Ness. RB; Fienberg. SE; Tobin. MJ. (2002). Bone lead levels in
        adjudicated delinquents: A case control study. Neurotoxicol Teratol 24: 711-717.
        http://dx.doi.org/10.1016/S0892-0362(02)00269-6

    Needleman. HL; Riess. JA; Tobin. MJ: Biesecker. GE; Greenhouse. JB. (1996). Bone lead levels and
        delinquent behavior. JAMA 275: 363-369. http://dx.doi.org/10.1001/iama.1996.03530290033034

    Needleman. HL: Schell. A; Bellinger. D; Leviton. A; Allred. EN. (1990). The long-term effects of
        exposure to low doses of lead in childhood: An 11-year follow-up report. N Engl J Med 322: 83-88.
        http://dx.doi.org/10.1056/NEJM199001113220203

    Nehez. M; Lorencz. R; Desi. I. (2000). Simultaneous action of cypermethrin and two environmental
        pollutant metals, cadmium and lead, on bone marrow cell chromosomes of rats in subchronic
        administration. Ecotoxicol Environ Saf 45: 55-60. http://dx.doi.org/10.1006/eesa.1999.1831

    Nelson. AE; Chaudhary. S: Kraus. VB; Fang. F; Chen. JC: Schwartz. TA; Shi. XA; Renner. JB; Stabler.
        TV: Helmick. CG: Caldwell. K: Poole. AR: Jordan. JM. (2011). Whole blood lead levels are
        associated with biomarkers of joint tissue metabolism in African American and white men and
        women: The Johnston County Osteoarthritis Project. Environ Res 111: 1208-1214.
        http://dx.doi.0rg/10.1016/i.envres.2011.08.002

    Nelson. MM; Espy. KA. (2009). Low-Level Lead Exposure and Contingency-Based Responding in
        Preschoolers: An Exploratory Study. Dev Neuropsychol 34: 494-506.
        http://dx.doi.org/10.1080/87565640902964565

    Nemoto. K; Ito. S: Yoshida. C: Miyata. M; Kojima. M; Degawa. M. (2011). Hepatic expression of
        spermatogenic genes and their transiently remarkable downregulations in Wistar-Kyoto rats in
        response to lead-nitrate administration: strain-difference in the gene expression patterns. J Toxicol Sci
        36: 357-364.

    Nemsadze. K; Sanikidze. T; Ratiani. L; Gabunia. L; Sharashenidze. T.  (2009). Mechanisms of lead-
        induced poisoning [Review]. Georgian Med News92-96.

    Nenov. VD; Taal. MW: Sakharova. OV: Brenner. BM.  (2000). Multi-hit nature of chronic renal disease
        [Review]. Curr Opin Nephrol Hypertens 9: 85-97.

    Nestmann. ER; Zhang. BW. (2007). Chromosome aberration test of Pigment Yellow 34 (lead chromate) in
        Chinese hamster ovary cells. Mutat Res  Genet Toxicol Environ Mutagen 633: 126-132.
        http://dx.doi.0rg/10.1016/i.mrgentox.2007.05.012

    Neuberger. JS: Hu. SC: Drake. KD; Jim. R. (2009). Potential health impacts of heavy-metal exposure at
        the Tar Creek Superfund site, Ottawa County, Oklahoma. Environ Geochem Health 31: 47-59.
        http://dx.doi.org/10.1007/sl0653-008-9154-0

    Newland. MC: Yezhou. S: Logdberg. B; Berlin. M. (1994). Prolonged behavioral effects of in utero
        exposure to lead or methyl mercury: Reduced sensitivity to changes in reinforcement contingencies
        during behavioral transitions and in steady state. Toxicol Appl Pharmacol  126: 6-15.
        http://dx.doi.org/10.1006/taap.1994.1084

    Ni. Z; Hou. S: Barton. CH; Vaziri. ND. (2004). Lead exposure raises superoxide and hydrogen peroxide
        in human endothelial and vascular smooth muscle cells. Kidney Int 66: 2329-2336.
        http://dx.doi.0rg/10.llll/i.1523-1755.2004.66032.x

    Nicolescu. R: Petcu. C: Cordeanu. A; Fabritius. K; Schlumpf. M; Krebs. R: Kramer. U; Winneke. G.
        (2010). Environmental exposure to lead, but not other neurotoxic metals, relates to  core elements of
        ADHD in Romanian children: Performance and questionnaire data. Environ Res 110: 476-483.
        http://dx.doi.0rg/10.1016/i.envres.2010.04.002

    Nigg. J; Knottnerus. G: Martel. M; Nikolas. M; Cavanagh. K; Karmaus. W: Rapplev. M. (2008). Low
        blood lead levels associated with clinically diagnosed attention-deficit/hyperactivity disorder and
        mediated by weak cognitive control. Biol Psychiatry 63: 325-331.
        http://dx.doi.0rg/10.1016/i.biopsvch.2007.07.013
November 2012                             5-765                  Draft - Do Not Cite or Quote

-------
     Nigg. JT; Nikolas. M; Mark Knottnerus. G: Cavanagh. K; Friderici. K. (2010). Confirmation and
        extension of association of blood lead with attention-deficit/hyperactivity disorder (ADHD) and
        ADHD symptom domains at population-typical exposure levels. J Child Psychol Psychiatry 51: 58-65.
        http://dx.doi.0rg/10.llll/i.1469-7610.2009.02135.x

     Nikolova. P; Kavaldzhieva. B. (1991). The effect of certain heavy metals (Mn and Pb) on parameters of
        erythrocyte energy metabolism. J Hyg Epidemiol Microbiol Immunol 35: 361-365.

     Niu. RY; Sun. ZL; Cheng. ZT; Li. ZG: Wang. JD. (2009). Decreased learning ability and low
        hippocampus glutamate in offspring rats exposed to fluoride and lead. Environ Toxicol Pharmacol 28:
        254-258. http://dx.doi.0rg/10.1016/i.etap.2009.04.012

     NKF (National Kidney Foundation). (2002). K/DOQI clinical practice guidelines for chronic kidney
        disease: Evaluation, classification, and stratification, Part 4: Definition and stages of chronic kidney
        disease. Am J Kidney Dis 39: S46-S75. http://dx.doi.org/10.1053/ajkd.2002.30943

     Nordberg. M; Winblad. B; Fratiglioni. L; Basun. H. (2000). Lead concentrations in elderly urban people
        related to blood pressure and mental performance: Results from a population-based study. Am J Ind
        Med 38: 290-294. httrj://dx.doi.org/10.1002/1097-0274(200009)38:3<290::AID-AJIM7>3.0.CO:2-T

     Novak. J; Banks. RO. (1995). Lead and nickel alter the cardiorenal actions of endothelin in the rat. Proc
        Soc Exp Biol Med 208: 191-198.

     Nowak. P; Szczerbak. G: Nitka. D; Kostrzewa. RM; Sitkiewicz. T; Brus. R. (2008). Effect of prenatal
        lead exposure on nigrostriatal neurotransmission and hydroxyl radical formation in rat neostriatum:
        Dopaminergic-nitrergic interaction. Toxicology 246: 83-89.
        http://dx.doi.0rg/10.1016/i.tox.2007.12.026

     NRC (National Research Council). (2000). Toxicological effects of methylmercury. Washington, DC:
        National Academy Press, http://books.nap.edu/catalog.php7record id=9899

     Nriagu. J: Afeiche. M; Linder. A; Arowolo. T; Ana. G: Sridhar.  MKC: Oloruntoba. EO; Obi. E; Ebenebe.
        JC:  Orisakwe. OE; Adesina. A. (2008). Lead poisoning associated with malaria in children of urban
        areas of Nigeria. Int J Hyg Environ Health 211: 591-605. http://dx.doi.org/10.1016/Ujheh.2008.05.001

     NTP (National Toxicology Program). (2011). Report on carcinogens: Twelfth edition. Washington, DC:
        U.S. Department of Health and Human Services, http://ntp.niehs.nih.gov/ntp/roc/twelfth/rocl2.pdf

     Nuvts.  GD: Van Vlem. E; Thys. J: De Leersnijder. D; D'Haese.  PC: Elseviers. MM; De Broe. ME. (1995).
        New occupational risk factors for chronic renal failure. Lancet 346: 7-11.
        http://dx.doi.org/10.1016/S0140-6736(95)92648-8

     O'Byrne. PM; Dolovich. M; Dirks. R; Roberts. RS: Newhouse. MT. (1984). Lung epithelial permeability:
        Relation to nonspecific airway responsiveness. Journal of Applied Physiology: Respiratory,
        environmental  and exercise physiology 57: 77-84.

     O'Byrne. PM; Dolovich. M; Duvall. A; Newhouse. MT. (1982). Lung epithelial permeability after
        histamine challenge [Abstract]. Am Rev Respir Dis 125: 280.

     Oberlev. TD; Friedman. AL; Moser. R; Siegel. FL. (1995). Effects of lead administration on developing
        rat kidney: II Functional, morphologic, and immunohistochemical studies. Toxicol Appl Pharmacol
        131: 94-107. http://dx.doi.org/10.1006/taap.1995.1051

     Oberto. A; Marks.  N; Evans. HL; Guidotti. A. (1996). Lead (Pb+2) promotes apoptosis in newborn rat
        cerebellar neurons: Pathological implications. J Pharmacol Exp Ther 279: 435-442.

     Obhodas. J: Tucak-Zoric. S: Kutle. A; Valkovic. V. (2007). Indications for synergetic and antagonistic
        effects between trace elements in the environment to human health. Coll Antropol 31: 209-219.

     Oishi. H; Nakashima. M; Totoki. T; Tomokuni. K. (1996).  Chronic lead exposure may inhibit
        endothelium-dependent hyperpolarizing factor in rats. J Cardiovasc Pharmacol 28: 558-563.
        http://dx.doi.org/10.1097/00005344-199610000-00013
November 2012                              5-766                   Draft - Do Not Cite or Quote

-------
     Olafiranye. O: Zizi. F; Brimah. P; Jean-Louis. G: Makarvus. AN; Mcfarlane. S: Ogedegbe. G. (2011).
        Management of hypertension among patients with coronary heart disease. Int J Hypertens 2011:
        653903. http://dx.doi.org/10.4061/2011/653903

     Olaleve. SB; Adaramoye. OA; Erigbali. PP; Adeniyi. OS. (2007). Lead exposure increases oxidative
        stress in the gastric mucosa of HCl/ethanol-exposed rats. World J Gastroenterol 13: 5121-5126.

     Olewinska. E; Kasperczvk. A; Kapka. L; Kozlowska. A;  Pawlas. N; Dobrakowski. M; Birkner. E;
        Kasperczyk. S. (2010). Level of DNA damage in lead-exposed workers. Ann Agric Environ Med 17:
        231-236.

     Oliveira. H; Spano. M; Santos. C: Pereira. MDL. (2009). Lead chloride affects sperm motility and
        acrosome reaction in mice: Lead affects mice sperm motility and acrosome reaction. Cell Biol Toxicol
        25: 341-353. http://dx.doi.org/10.1007/sl0565-008-9088-4

     Olivero-Verbel. J: Duarte. D; Echenique. M; Guette. J: Johnson-Restrepo. B; Parsons. PJ. (2007). Blood
        lead levels in children aged 5-9 years living in Cartagena, Colombia. Sci Total Environ 372: 707-716.
        http://dx.doi.0rg/10.1016/i.scitotenv.2006.10.025

     Olszewski. J: Latusinski. J: Kita. A; Pietkiewicz. P; Starostecka. B; Majak. J. (2006). Comparative
        assessment of aluminum and lead concentrations in serum and tissue bioptates in patients with
        laryngeal papilloma or cancer. B-ENT 2: 47-49.

     Ong. CN; Lee. WR. (1980a). Distribution of lead-203 in  human peripheral blood in vitro. Br J Ind Med
        37: 78-84. http://dx.doi.Org/10.1136/oem.37.l.78

     Ong. CN; Lee. WR. (1980b). High affinity of lead for fetal haemoglobin. Br J Ind Med 37: 292-298.
        http://dx.doi.0rg/10.1136/oem.37.3.292

     Onuegbu. AJ; Olisekodiaka. MJ; Nwaba. El; Adeveve. AD; Akinola. FFS. (2011). Assessment of some
        renal indices in people occupationally exposed to lead. Toxicol Ind Health 27: 475-479.
        http://dx.doi.org/10.1177/0748233710390020

     Opler. MGA; Brown. AS: Graziano. J: Desai. M; Zheng. W: Schaefer. C: Factor-Litvak. P; Susser. ES.
        (2004). Prenatal lead exposure, delta-aminolevulinic acid, and schizophrenia. Environ Health Perspect
        112: 548-552. http://dx.doi.org/10.1289/ehp.10464

     Opler. MGA: Buka. SL; Groeger. J: McKeague. I; Wei. C: Factor-Litvak. P; Bresnahan. M; Graziano. J:
        Goldstein. JM; Seidman. LJ; Brown. AS: Susser. ES.  (2008). Prenatal exposure to lead, delta-
        aminolevulinic acid, and schizophrenia: Further evidence. Environ Health Perspect 116: 1586-1590.
        http://dx.doi.org/10.1289/ehp.10464

     Orisakwe. OE; Nwachukwu. E; Osadolor. HB; Afonne. OJ: Okocha. CE. (2007). Liver and kidney
        function tests amongst paint factory workers in Nkpor, Nigeria. Toxicol Ind Health 23: 161-165.
        http://dx.doi.org/10.1177/0748233707081908

     Oriin. E; Yalcm. SS: Aykut. O: Orhan. G: Morgil. GK; Yurdakok. K; Uzun. R. (2011).  Breast milk lead
        and cadmium levels from suburban areas of Ankara. Sci Total Environ 409: 2467-2472.
        http://dx.doi.0rg/10.1016/i.scitotenv.2011.02.035

     Oskarsson. A; Fowler. BA. (1985). Effects of lead inclusion bodies on subcellular distribution of lead in
        rat kidney: The relationship to mitochondria! function. Exp Mol Pathol 43: 397-408.
        http://dx.doi.org/10.1016/0014-4800(85)90076-0

     Oskarsson. A; Squibb. KS: Fowler. BA. (1982).  Intracellular binding of lead in the kidney: The partial
        isolation and characterization of postmitochondrial lead binding components. Biochem Biophys Res
        Commun 104: 290-298. http://dx.doi.org/10.1016/0006-291X(82)91973-8

     Osman. K; Pawlas. K; Schutz. A; Gazdzik. M; Sokal. JA; Vahter. M. (1999).  Lead exposure and hearing
        effects in children in Katowice, Poland. Environ Res  80: 1-8. http://dx.doi.org/10.1006/enrs.1998.3886

     Osterode. W: Barnas. D; Geissler. K. (1999). Dose dependent reduction of erythroid progenitor cells and
        inappropriate erythropoietin response in exposure to lead: new aspects of anaemia  induced by lead.
        Occup Environ Med 56: 106-109.
November 2012                             5-767                   Draft - Do Not Cite or Quote

-------
     Osterode. W: Ulberth. F. (2000). Increased concentration of arachidonic acid in erythrocyte membranes in
        chronically lead-exposed men. J Toxicol Environ Health A 59: 87-95.
        http://dx.doi.org/10.1080/009841000156998

     Ostro. B; Feng. WY; Broadwin. R; Green. S: Lipsett. M. (2007). The effects of components of fine
        paniculate air pollution on mortality in California: Results from CALFINE. Environ Health Perspect
        115: 13-19. http://dx.doi.org/10.1289/ehp.9281

     Oteiza. PI; Kleinman. CG: Demasi. M; Bechara. EJH. (1995). 5-Aminolevulinic acid induces iron release
        from ferritin. Arch Biochem Biophys 316: 607-611. http://dx.doi.org/10.1006/abbi.1995.1080

     Otto. DA; Fox. DA. (1993). Auditory and visual dysfunction following lead exposure [Review].
        Neurotoxicology 14: 191-207.

     Overmann. SR. (1977). Behavioral effects of asymptomatic lead exposure during neonatal development
        in rats. Toxicol Appl Pharmacol 41: 459-471. http://dx.doi.org/10.1016/S0041-008X(77)80002-l

     Ozsov. SY; Ozsov. B; Ozyildiz. Z; Aytekin. I. (2010). Protective effect of L-c ami tine on experimental
        lead toxicity in rats: A clinical, histopathological and immunohistochemical study. Biotech Histochem
        86: 436-443. http://dx.doi.org/10.3109/10520295.2010.529825

     Pace. BM; Lawrence. DA; Behr. MJ; Parsons. PJ: Dias. JA. (2005). Neonatal lead exposure changes
        quality of sperm and number of macrophages in testes of BALB/c mice. Toxicology 210: 247-256.
        http://dx.doi.0rg/10.1016/i.tox.2005.02.004

     Paglia. DE; Valentine. WN. (1975).  Characteristics of a pyrimidine-specific 5'-nucleotidase in human
        erythrocytes. J Biol Chern 250: 7973-7979.

     Palaniappan. K; Roy. A; Balakrishnan. K; Gopalakrishnan. L; Mukherjee. B; Hu. H; Bellinger. DC.
        (2011). Lead exposure and visual-motor abilities in children from Chennai, India. Neurotoxicology 32:
        465-470. http://dx.doi.0rg/10.1016/i.neuro.2011.03.011

     Palus. J: Rydzynski. K; Dziubaltowska.  E; Wvszynska. K; Nataraian. AT; Nilsson. R. (2003). Genotoxic
        effects of occupational exposure to lead and cadmium. Mutat Res Genet Toxicol Environ Mutagen
        540: 19-28. http://dx.doi.org/10.1016/S1383-5718(03)00167-0

     Pan. SY; Morrison. H; Gibbons. L; Zhou. J: Wen. SW: DesMeules. M; Mao. Y. (2011). Breast cancer risk
        associated  with residential proximity to industrial plants in Canada. J Occup Environ Med 53:
        522-529. http://dx.doi.org/10.1097/JOM.Ob013e318216dOb3

     Pandva. C: Pillai. P: Nampoothiri. LP; Bhatt. N; Gupta. S: Gupta. S. (2012). Effect of lead and cadmium
        co-exposure on testicular steroid metabolism and antioxidant system of adult male rats. Andrologia
        44: 813-822. http://dx.doi.Org/10.llll/i.1439-0272.2010.01137.x

     Pandva. CD; Pillai. PP; Gupta. SS. (2010). Lead and cadmium co-exposure mediated toxic insults on
        hepatic steroid metabolism and antioxidant system of adult male rats. Biol Trace Elem Res 134:
        307-317. http://dx.doi.org/10.1007/sl2011-009-8479-6

     Park. SK; Elmarsafawv. S: Mukherjee. B; Spiro. A. Ill; Vokonas. PS: Nie. H; Weisskopf. MG: Schwartz.
        J; Hu. H. (2010).  Cumulative lead exposure and age-related hearing loss: The VA Normative Aging
        Study. Hear Res 269: 48-55. http://dx.doi.Org/10.1016/i.heares.2010.07.004

     Park. SK; Hu. H; Wright. RO: Schwartz. J: Cheng. Y; Sparrow. D; Vokonas. PS: Weisskopf. MG.
        (2009b). Iron metabolism genes, low-level lead exposure, and QT interval. Environ Health Perspect
        117: 80-85. http://dx.doi.org/10.1289/ehp.11559

     Park. SK; Mukherjee. B; Xia. X;  Sparrow. D; Weisskopf. MG: Nie. H; Hu. H. (2009c). Bone lead level
        prediction models and their application to examine the relationship of lead exposure and hypertension
        in the third National Health and Nutrition Examination Survey. J Occup Environ Med 51: 1422-1436.
        http://dx.doi.org/10.1097/JOM.Ob013e3181bf6c8d

     Park. SK; Schwartz. J: Weisskopf. M; Sparrow. D: Vokonas. PS: Wright. RO: Coull. B; Nie. H; Hu. H.
        (2006). Low-level lead exposure, metabolic syndrome, and heart rate variability: The VA Normative
        Aging Study. Environ Health Perspect 114: 1718-1724.
November 2012                             5-768                  Draft - Do Not Cite or Quote

-------
     Parkinson. DK; Ryan. C: Bromet EJ; Connell. MM. (1986). A psychiatric epidemiologic study of
        occupational lead exposure. Am J Epidemiol 123: 261-269.

     Parr. PR; Harris. EJ. (1976). The effect of lead on the calcium-handling capacity of rat heart
        mitochondria. Biochemistry 158: 289-294.

     Pasha. Q: Malik. SA; Iqbal. J: Shah. MH. (2007). Characterization and distribution of the selected metals
        in the scalp hair of cancer patients in comparison with normal donors. Biol Trace Elem Res 118:
        207-216. http://dx.doi.org/10.1007/sl2011-007-0035-7

     Pasha. Q: Malik. SA; Iqbal. J: Shaheen. N; Shah. MH. (2008a). Comparative distribution of the scalp hair
        trace metal contents in the benign tumour patients and normal donors. Environ Monit Assess  147:
        377-388. http://dx.doi.org/10.1007/sl0661-007-0127-z

     Pasha. Q: Malik. SA; Iqbal. J: Shaheen. N; Shah. MH. (2008b). Comparative evaluation of trace metal
        distribution and correlation in human malignant and benign breast tissues.  Biol Trace Elem Res 125:
        30-40. http://dx.doi.org/10.1007/sl2011-008-8158-z

     Pasha. Q: Malik. SA; Shah. MH. (2008c). Statistical analysis of trace metals in the plasma of cancer
        patients versus controls. J Hazard Mater 153: 1215-1221.
        http://dx.doi.0rg/10.1016/i.jhazmat.2007.09.115

     Pasha. Q: Malik. SA; Shaheen. N; Shah.  MH. (2010). Investigation of trace metals in the blood plasma
        and scalp hair of gastrointestinal cancer patients in comparison with controls. Clin Chim Acta 411:
        531-539. http://dx.doi.0rg/10.1016/i.cca.2010.01.010

     Pasha Shaik. A;  Sankar. S: Reddv. SC: Das. PG: Jamil. K. (2006). Lead-induced genotoxicity in
        lymphocytes from peripheral blood samples of humans: In vitro studies. Drug Chem Toxicol  29:
        111-124. http://dx.doi.org/10.1080/01480540500408739

     Pastor-Barriuso. R; Banegas. JR; Damian. J: Appel. LJ; Guallar. E. (2003). Systolic blood pressure,
        diastolic blood pressure, and pulse pressure: An evaluation of their joint effect on mortality. Ann
        Intern Med 139: 731-739.

     Patel. AB; Prabhu. AS. (2009). Determinants of lead level in umbilical cord blood. Indian Pediatr 46:
        791-793.

     Patil. AJ; Bhagwat VR; Patil. JA; Dongre. NN; Ambekar. JG: Das. KK. (2006a). Biochemical aspects of
        lead exposure in silver jewelry workers in western Maharashtra (India). J Basic Clin Physiol
        Pharmacol 17: 213-229.

     Patil. AJ; Bhagwat. VR: Patil. JA; Dongre. NN; Ambekar. JG: Das. KK. (2007). Occupational lead
        exposure in battery manufacturing workers,  silver jewelry workers, and spray painters in western
        Maharashtra (India): Effect on liver and kidney function. J Basic  Clin Physiol Pharmacol 18:  87-100.
        http://dx.doi.Org/10.1515/JBCPP.2007.18.2.87

     Patil. AJ: Bhagwat. VR: Patil. JA; Dongre. NN; Ambekar. JG: Jailkhani. R; Das. KK. (2006b). Effect of
        lead (Pb) exposure on the activity of superoxide dismutase and catalase in battery manufacturing
        workers (BMW) of Western Maharashtra  (India) with reference to heme biosynthesis. Int J Environ
        Res Public Health 3: 329-337.

     Pavanello. S: Bollati. V: Pesatori. AC: Kapka. L; Bolognesi. C: Bertazzi. PA; Baccarelli. A. (2009).
        Global and gene-specific promoter methylation changes are related to anti-B[a]PDE-DNA adduct
        levels and influence micronuclei levels in polycyclic aromatic hydrocarbon-exposed individuals. Int J
        Cancer 125:  1692-1697. http://dx.doi.org/10.1002/iic.24492

     Pavanello. S: Pesatori. AC: Dioni. L; Hoxha. M; Bollati. V: Siwinska. E; Mielzynska. D; Bolognesi. C:
        Bertazzi. PA; Baccarelli. A. (2010). Shorter telomere length in peripheral blood lymphocytes  of
        workers exposed to polycyclic aromatic hydrocarbons. Carcinogenesis 31: 216-221.
        http://dx.doi.org/10.1093/carcin/bgp278

     Pavton. M; Hu. H; Sparrow.  D; Weiss. ST. (1994). Low-level lead exposure and renal function in the
        Normative Aging Study. Am J Epidemiol  140: 821-829.
November 2012                              5-769                   Draft - Do Not Cite or Quote

-------
     Pavton. M; Riggs. KM; Spiro. A. Ill: Weiss. ST; Hu. H. (1998). Relations of bone and blood lead to
        cognitive function: The VA Normative Aging Study. Neurotoxicol Teratol 20: 19-27.

     Pekcici. R; Kavlakoglu. B; Yilmaz. S: Sahin. M; Delibasi. T. (2010). Effects of lead on thyroid functions
        in lead-exposed workers. Central European Journal of Medicine 5: 215-218.
        http://dx.doi.org/10.2478/sll536-009-0092-8

     Peng. S: Hajela. RK: Atchison. WD. (2002). Characteristics of block by Pb2+ of function of human
        neuronal L-, N-, and R-type Ca2+ channels transiently expressed in human embryonic kidney 293
        cells. Mol Pharmacol 62: 1418-1430. http://dx.doi.org/10.1124/mol.62.6.1418

     Pentvala. S: Ruggeri. J; Veerraiu. A; Yu. Z; Bhatia. A;  Desaiah. D; Vig. P. (2010). Microsomal Ca2+ flux
        modulation as an indicator of heavy metal toxicity. Indian J Exp Biol 48: 737-743.

     Perez-Bravo. F; Ruz. M; Moran-Jimenez. MJ; Olivares. M; Rebolledo. A; Codoceo. J: Sepulveda. V:
        Jenkin. A; Santos. JL; Fontanellas. A. (2004). Association between aminolevulinate dehydrase
        genotypes and blood lead levels in children from a lead-contaminated area in Antofagasta, Chile. Arch
        Environ Contam Toxicol 47: 276-280. http://dx.doi.org/10.1007/s00244-004-2215-l

     Perlstein. T; Weuve. J: Schwartz. J: Sparrow. D; Wright. R; Litonjua. A; Nie. H; Hu. H. (2007).
        Cumulative community-level lead exposure and pulse pressure: The Normative Aging Study. Environ
        Health Perspect 115: 1696-1700. http://dx.doi.org/10.1289/ehp.10350

     Peters. JL; Kubzansky. L; McNeely. E;  Schwartz. J: Spiro. A. Ill; Sparrow. D: Wright. RO: Nie. H; Hu. H.
        (2007). Stress as a potential modifier of the impact of lead levels on blood pressure: The Normative
        Aging Study. Environ Health Perspect 115: 1154-1159. http://dx.doi.org/10.1289/ehp.10002

     Petit. TL; Leboutillier. JC. (1979). Effects of lead exposure during development on neocortical dendritic
        and synaptic structure.  Exp Neural 64: 482-492. http://dx.doi.org/10.1016/0014-4886(79)90226-7

     Piao. F; Cheng. F; Chen. H;  Li. G: Sun. X; Liu. S: Yamauchi. T; Yokoyama. K. (2007). Effects of zinc
        coadministration on lead toxicities in rats. Ind Health 45: 546-551.

     Piccinini. F; Favalli. L; Chiari. MC.  (1977). Experimental investigations on the conctraction induced by
        lead in arterial smooth muscle. Toxicology 8: 43-51. http://dx.doi.org/10.1016/0300-483X(77)90022-l

     Pillai. P: Pandva. C: Bhatt N; Gupta. SS.  (2012). Biochemical and reproductive effects of
        gestational/lactational exposure to lead and cadmium with respect to testicular steroidogenesis,
        antioxidant system, endogenous sex steroid and cauda-epididymal functions. Andrologia 44: 92-101.
        http://dx.doi.0rg/10.llll/i.1439-0272.2010.01109.x

     Pillai. P; Pandva. C: Gupta. S: Gupta. S. (2010). Biochemical and molecular effects of gestational and
        lactational coexposure to lead and cadmium on ovarian steroidogenesis are associated with oxidative
        stress in Fl  generation rats. J Biochem Mol Toxicol 24: 384-394. http://dx.doi.org/10.1002/ibt.20351

     Pillai. P: Patel. R: Pandva. C: Gupta. S. (2009). Sex-specific effects of gestational and lactational
        coexposure to lead and cadmium on hepatic phase I and phase II xenobiotic/steroid-metabolizing
        enzymes and antioxidant status. J Biochem Mol Toxicol 23: 419-431.
        http://dx.doi.org/10.1002/ibt.20305

     Pilsner. JR; Hu. H; Ettinger.  A; Sanchez. BN; Wright. RO: Cantonwine. D: Lazarus. A; Lamadrid-
        Figueroa. H; Mercado-Garcia. A; Tellez-Rojo. MM; Hernandez-Avila. M. (2009). Influence of
        prenatal lead exposure  on genomic methylation of cord blood DNA. Environ Health Perspect 117:
        1466-1471.  http://dx.doi.org/10.1289/ehp.0800497

     Pilsner. JR; Hu. H; Wright. RO: Kordas. K; Ettinger. AS: Sanchez. BN; Cantonwine. D: Lazarus. AL;
        Cantoral. A; Schnaas. L;  Tellez-Rojo. MM; Hernandez-Avila. M. (2010). Maternal MTHFR genotype
        and haplotype predict deficits in early  cognitive development in a lead-exposed birth cohort in Mexico
        City. Am J Clin Nutr 92:  226-234. http://dx.doi.org/10.3945/aicn.2009.28839

     Pine. MD; Hinev. JK; Dearth. RK; Bratton. GR; Dees. WL. (2006). IGF-1 administration to prepubertal
        female rats can overcome delayed puberty caused by maternal Pb exposure. Reprod Toxicol 21:
        104-109. http://dx.doi.0rg/10.1016/i.reprotox.2005.07.003
November 2012                              5-770                  Draft - Do Not Cite or Quote

-------
     Pineda-Zavaleta. AP; Garcia-Vargas. G: Borja-Aburto. VH; Acosta-Saavedra. LC: Aguilar. EV; Gomez-
        Mufioz. A; Cebrian. ME; Calderon-Aranda. ES. (2004). Nitric oxide and superoxide anion production
        in monocytes from children exposed to arsenic and lead in region Lagunera, Mexico. Toxicol Appl
        Pharmacol 198: 283-290. http://dx.doi.Org/10.1016/i.taap.2003.10.034

     Pinkerton. LE; Biagini. RE; Ward. EM; Hull. RD; Deddens. JA; Boeniger. MF; Schnorr. TM; MacKenzie.
        BA; Luster. MI. (1998). Immunologic findings among lead-exposed workers. Am J Ind Med 33:
        400-408. httrj://dx.doi.org/10.1002/(SICI)1097-0274(199804)33:4<400::AID-AJIMll>3.0.CO:2-2

     Pinto. D; Ceballos. JM; Garcia. G: Guzman. P; Del Razo. LM; Vera. E; Gomez. H; Garcia. A; Gonsebatt.
        ME. (2000). Increased cytogenetic damage in outdoor painters. Mutat Res Genet Toxicol Environ
        Mutagen467: 105-111. http://dx.doi.org/10.1016/S1383-5718(00)00024-3

     Pirkle. JL; Brodv. DJ: Gunter. EW: Kramer. RA; Paschal. DC: Flegal. KM; Matte. TD. (1994). The
        decline in blood lead levels in the United States: The National Health and Nutrition Examination
        Surveys (NHANES). JAMA 272: 284-291. http://dx.doi.org/10.1001/jama.1994.03520040046039

     Pizent. A; Colak. B; Kljakovic Gaspic. Z; Telisman. S. (2009). Prostate-specific antigen (PSA) in serum
        in relation to blood lead concentration and alcohol consumption in men. Arh Hig Rada Toksikol 60:
        69-78. http://dx.doi.org/10.2478/10004-1254-60-2009-1901

     Pizent. A; Macan. J; Jurasovic. J: Varnai. VM; Milkovic-Kraus. S: Kanceljak-Macan. B. (2008).
        Association of toxic and essential metals with atopy markers and ventilatory lung function in women
        and men. Sci Total Environ 390: 369-376. http://dx.doi.Org/10.1016/j.scitotenv.2007.10.049

     Plusquellec. P; Muckle. G: Dewailly. E; Ayotte. P; Begin. G: Desrosiers. C: Despres. C: Saint-Amour. D;
        Poitras. K. (2010). The relation of environmental contaminants exposure to behavioral indicators in
        Inuit preschoolers in Arctic Quebec. Neurotoxicology 31: 17-25.
        http://dx.doi.0rg/10.1016/i.neuro.2009.10.008

     Plusquellec. P: Muckle. G: Dewailly. E; Ayotte. P: Jacobson. SW: Jacobson. JL. (2007). The relation of
        low-level prenatal lead exposure to behavioral indicators of attention in Inuit infants in Arctic Quebec.
        Neurotoxicol Teratol 29: 527-537. http://dx.doi.Org/10.1016/i.ntt.2007.07.002

     Pocock.  SJ: Smith. M; Baghurst P (1994). Environmental lead and children's intelligence: A systematic
        review of the epidemiological evidence [Review]. Br Med J 309: 1189-1197.
        http://dx.doi.org/10.1136/bmj.309.6963.1189

     Pollack. AZ; Schisterman. EF; Goldman. LR; Mumford. SL; Albert. PS: Jones. RL; Wactawski-Wende. J.
        (2011). Cadmium, lead, and mercury in relation to reproductive hormones and anovulation in
        premenopausal women. Environ Health Perspect 119:  1156-1161.
        http://dx.doi.org/10.1289/ehp.1003284

     Pollitt. E. (2005). Are the psychological tests valid? [Letter]. Am J Clin Nutr 82: 201; author reply
        201-201; author reply 202.

     Porgba. R; Gac. P; Poreba. M; Antonowicz-Juchniewicz. J; Andrzejak. R. (2011 a). Relationship between
        occupational exposure to lead and local arterial stiffness and left ventricular diastolic function in
        individuals with arterial hypertension. Toxicol Appl Pharmacol 254: 342-348.
        http://dx.doi.0rg/10.1016/i.taarj.2011.05.009

     Poreba. R; Gac. P; Poreba. M; Derkacz. A; Pilecki. W; Antonowicz-Juchniewicz. J; Andrzejak. R. (2010).
        [Relationship between chronic exposure to lead, cadmium and manganese, blood pressure values and
        incidence of arterial hypertension]. Med Pr 61: 5-14.

     Poreba. R; Poreba. M; Gac. P; Andrzejak. R. (2011). Ambulatory blood pressure monitoring and
        structural changes in carotid arteries in normotensive workers occupationally exposed to lead. Hum
        Exp Toxicol 30: 1174-1180. http://dx.doi.org/10.1177/0960327110391383

     Poreba. R; Poreba. M; Gac. P; Pilecki. W; Andrzejak. R. (2010). Impaired  endothelial function assessed
        by flow-mediated vasodilatation in workers occupationally exposed to lead. Environ Toxicol
        Pharmacol 29: 266-270. http://dx.doi.Org/10.1016/i.etap.2010.03.001
November 2012                              5-771                   Draft - Do Not Cite or Quote

-------
     Porgba. R; Porgba. M; Gac. P; Steinmetz-Beck. A; Beck. B; Pilecki. W: Andrzejak. R; Sobieszczanska.
        M. (20lib). Electrocardiographic changes in workers occupationally exposed to lead. Ann
        Noninvasive Electrocardiol 16: 33-40. http://dx.doi.Org/10.llll/i.1542-474X.2010.00406.x

     Poulos. L; Qammaz. S: Athanaselis. S: Maravelias. C: Koutselinis. A. (1986). Statistically significant
        hematopoietic effects of low blood lead levels. Arch Environ Occup Health 41: 384-386.

     Prentice. RC: Kopp. SJ. (1985). Cardiotoxicity of lead at various perfusate calcium concentrations:
        Functional and metabolic responses of the perfused rat heart. Toxicol Appl Pharmacol 81: 491-501.
        http://dx.doi.org/10.1016/0041-008X(85)90420-X

     Prins. JM; Park. S: Lurie. PL (2010). Decreased expression of the voltage-dependent anion channel in
        differentiated PC-12 and SH-SY5Ycells following low-level Pb exposure. Toxicol Sci 113: 169-176.
        http://dx.doi.org/10.1093/toxsci/kfp249

     Prospective Studies Collaboration. (2002). Age-specific relevance of usual blood pressure to vascular
        mortality: A meta-analysis of individual data for one million adults in 61 prospective studies. Lancet
        360: 1903-1913. http://dx.doi.org/10.1016/80140-6736(02)11911-8

     Prozialeck. WC: Grunwald. GB; Dev. PM; Reuhl. KR; Parrish. AR. (2002). Cadherins and NCAM as
        potential targets in metal toxicity [Review]. Toxicol Appl Pharmacol 182: 255-265.
        http://dx.doi.org/10.1006/taap.2002.9422

     Pugh Smith.  P; Nriagu. JO. (2011). Lead poisoning and asthma among low-income and African American
        children in Saginaw, Michigan. Environ Res 111:  81-86.
        http://dx.doi.0rg/10.1016/i.envres.2010.ll.007

     Purdv. RE; Smith. JR; Ding. Y;  Oveisi. F; Vaziri. ND; Gonick. HC. (1997). Lead-induced hypertension is
        not associated with altered vascular reactivity in vitro. Am J Hypertens 10: 997-1003.
        http://dx.doi.org/10.1016/S0895-7061(97)00108-8

     Putnam. CD; Arvai. AS: Bourne. Y; Tainer. JA. (2000). Active and inhibited human catalase structures:
        Ligand and NADPH binding and catalytic mechanism. J Mol Biol 296: 295-309.
        http://dx.doi.org/10.1006/imbi.1999.3458

     Pvatt. DW: Zheng. JH; Stillman. WS: Irons. RD. (1996). Inorganic lead activates NF-kappa B in primary
        human CD4+ T lymphocytes. Biochem Biophys Res Commun 227: 380-385.
        http://dx.doi.org/10.1006/bbrc. 1996.1516

     Qian. Y; Harris. ED; Zheng. Y; Tiffany-Castiglioni. E. (2000). Lead targets GRP78, a molecular
        chaperone, in C6 rat glioma cells. Toxicol Appl Pharmacol 163: 260-266.
        http://dx.doi.org/10.1006/taap.1999.8878

     Qian. Y; Zheng. Y; Ramos. KS: Tiffany-Castiglioni. E. (2005). GRP78 compartmentalized redistribution
        in Pb-treated glia: Role of GRP78 in lead-induced oxidative stress. Neurotoxicology 26: 267-275.
        http://dx.doi.0rg/10.1016/i.neuro.2004.09.002

     Qian. ZM; Morgan. EH. (1990). Effect of lead on the transport of transferrin-free and transferrin-bound
        iron into rabbit reticulocytes. Biochem Pharmacol 40:  1049-1054.
        http://dx.doi.org/10.1016/0006-2952(90)90492-4

     Qiao. YF; Jiang. YS: Pang. DZ. (2006). [Expression of renal nuclear factor-kappaB, transforming growth
        factor-beta and fibronectin of rats exposed to lead]. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za
        Zhi 24: 139-142.

     Qu. W: Diwan. BA; Liu. J: Gover. RA; Dawson. T; Horton. JL; Cherian. MG: Waalkes. MP (2002). The
        metallothionein-null phenotype is associated with heightened sensitivity to lead toxicity and an
        inability to form inclusion bodies. Am J Pathol 160: 1047-1056.
        http://dx.doi.org/10.1016/S0002-9440(10)64925-5

     Queirolo. El; Ettinger. AS: Stoltzfus. RJ: Kordas. K. (2010). Association of anemia, child and family
        characteristics with elevated blood lead concentrations in preschool children from Montevideo,
        Uruguay. Arch Environ Occup Health 65: 94-100. http://dx.doi.org/10.1080/19338240903390313
November 2012                              5-772                   Draft - Do Not Cite or Quote

-------
     Queiroz. MLS: Almeida. M; Gallao. MI; Hoehr. NF. (1993). Defective neutrophil function in workers
        occupationally exposed to lead. Pharmacol Toxicol 72: 73-77.
        http://dx.doi.0rg/10.llll/i.1600-0773.1993.tb00293.x

     Queiroz. MLS: Costa. FF; Bincoletto. C: Perlingeiro. RCR: Dantas. PCM; Cardoso. MP; Almeida. M.
        (1994a). Engulfment and killing capabilities of neutrophils and phagocytic splenic function in persons
        occupationally exposed to lead. International Journal of Immunopharmacology 16: 239-244.
        http://dx.doi.org/10.1016/0192-0561(94)90018-3

     Queiroz. MLS: da Rocha. MC: Torello. CO: de Souza Queiroz. J; Bincoletto. C: Morgano. MA; Romano.
        MR; Paredes-Gamero. EJ; Barbosa. CMV; Calgarotto. AK. (2011). Chlorella vulgaris restores bone
        marrow cellularity and cytokine production in lead-exposed mice. Food Chem Toxicol 49: 2934-2941.
        http://dx.doi.0rg/10.1016/i.fct.2011.06.056

     Queiroz. MLS: Perlingeiro. RCR: Bincoletto. C: Almeida. M; Cardoso. MP; Dantas. PCM. (1994b).
        Immunoglobulin levels and cellular immune function in lead exposed workers. Immunopharmacol
        Immunotoxicol 16: 115-128. http://dx.doi.org/10.3109/08923979409029904

     Quinlan. GJ: Halliwell. B; Moorhouse. CP; Gutteridge. JMC. (1988). Action of lead(II) and
        aluminium(III) ions on iron-stimulated lipid peroxidation in liposomes, erythrocytes and rat liver
        microsomal fractions. Biochim Biophys Acta 962: 196-200.
        http://dx.doi.org/10.1016/0005-2760(88)90159-2

     Quintanar-Escorza.  MA; Gonzalez-Martinez. MT; del Pilar. IQM; Calderon-Salinas. JV. (2010).
        Oxidative damage increases intracellular free calcium [Ca2+]i concentration in human erythrocytes
        incubated with lead. Toxicol In Vitro 24: 1338-1346. http://dx.doi.Org/10.1016/i.tiv.2010.05.002

     Quintanar-Escorza.  MA; Gonzalez-Martinez. MT; Navarro. L; Maldonado. M; Arevalo. B; Calderon-
        Salinas. JV. (2007). Intracellular free calcium concentration and calcium transport in human
        erythrocytes of lead-exposed workers. Toxicol Appl Pharmacol 220: 1-8.
        http://dx.doi.0rg/10.1016/i.taap.2006.10.016

     Quintanilla-Vega. B; Smith. PR; Kahng. MW: Hernandez. JM; Albores. A; Fowler. BA. (1995). Lead-
        binding proteins in brain tissue of environmentally lead-exposed humans. Chem Biol Interact 98:
        193-209.  http://dx.doi.org/10.1016/0009-2797(95)03646-6

     Rabbani-Chadegani. A; Abdosamadi. S: Fani. N; Mohammadian. S. (2009). A comparison of the effect of
        lead nitrate on rat liver chromatin, DNA and histone proteins in solution. Arch Toxicol 83: 565-570.
        http://dx.doi.org/10.1007/s00204-008-0362-2

     Rabbani-Chadegani. A. Ill: Fani. N; Abdossamadi. S: Shahmir. N. (2011). Toxic effects of lead and nickel
        nitrate on rat liver chromatin components. J Biochem Mol Toxicol 25: 127-134.
        http://dx.doi.org/10.1002/ibt.20368

     Rabinowitz. MB.  (1988). Lead and pregnancy. Birth 15: 236-241.
        http://dx.doi.Org/10.llll/i.1523-536X.1988.tb01117.x

     Rabinowitz. MB; Alfred. EN; Bellinger. DC: Leviton. A; Needleman. HL. (1990). Lead and childhood
        propensity to infectious and allergic disorders:  Is there an association? Bull Environ Contam Toxicol
        44: 657-660. http://dx.doi.org/10.1007/BF01701784

     Rabinowitz. MB; Wang. JD; Soong. WT. (1992). Children's classroom behavior and lead in Taiwan. Bull
        Environ Contam Toxicol 48: 282-288. http://dx.doi.org/10.1007/BF00194385

     Rafalowska.  U: Struzynska. L; Dabrowska-Bouta. B; Lenkiewicz. A. (1996). Is lead toxicosis a reflection
        of altered energy metabolism in brain synaptosomes? [Review]. Acta Neurobiol Exp (Wars) 56:
        611-617.

     Rafique. M; Khan. N; Perveen. K; Naqvi. A. (2009). The effects of lead and zinc on the quality of semen
        of albino  rats.  J Coll Physicians Surg Pak 19: 510-513. http://dx.doi.org/08.2009/JCPSP.510513
November 2012                             5-773                   Draft - Do Not Cite or Quote

-------
     Raghavan. SRV: Culver. BD; Gonick. HC. (1980). Erythrocyte lead-binding protein after occupational
        exposure: I Relationship to lead toxicity. Environ Res 22: 264-270.
        http://dx.doi.org/10.1016/0013-9351(80)90138-3

     Raghavan. SRV: Culver. BD; Gonick. HC. (1981). Erythrocyte lead-binding protein after occupational
        exposure: II. Influence on lead inhibition of membrane Na+, K+ - adenosinetriphosphatase. J Toxicol
        Environ Health 7: 561-568. http://dx.doi.org/10.1080/15287398109530001

     Raghavan. SRV: Gonick. HC. (1977). Isolation of low-molecular-weight lead-binding protein from
        human erythrocytes. Proc Soc Exp Biol Med 155: 164-167.

     Raine. A; Lencz. T; Bihrle. S: LaCasse. L; Colletti. P. (2000). Reduced prefrontal gray matter volume and
        reduced autonomic activity in antisocial personality disorder. Arch Gen Psychiatry 57: 119-127.

     Raitakari. OT; Juonala. M; Kahonen. M; Taittonen. L; Laitinen. T; Maki-Torkko. N; Jarvisalo. MJ; Uhari.
        M; Jokinen. E; Ronnemaa. T; Akerblom. HK; Viikari. JS. (2003).  Cardiovascular risk factors in
        childhood and carotid artery intima-media thickness in adulthood: The Cardiovascular Risk in Young
        Finns Study. JAMA290:  2277-2283. http://dx.doi.org/10.1001/iama.290.17.2277

     Rajah. T; Ahuja. YR. (1995). In vivo genotoxic effects of smoking and occupational lead exposure in
        printing press workers. Toxicol Lett 76: 71-75. http://dx.doi.org/10.1016/0378-4274(94)03200-9

     Rajah. TT; Ahuja. YR. (1996). In vivo genotoxicity of alcohol consumption and lead exposure in printing
        press workers. Alcohol 13: 65-68. http://dx.doi.org/10.1016/0741-8329(95)02014-4

     Rajan. P; Kelsev. KT; Schwartz. JD; Bellinger. DC: Weuve. J: Sparrow. D; Spiro. A. Ill; Smith. TJ; Nie.
        H; Hu. H; Wright. RO. (2007). Lead burden and psychiatric symptoms and the modifying influence of
        the delta-aminolevulinic acid dehydratase (ALAD) polymorphism: The VA Normative Aging Study.
        Am J Epidemiol 166:  1400-1408. http://dx.doi.org/10.1093/aie/kwm220

     Rajan. P; Kelsev. KT; Schwartz. JD; Bellinger. DC: Weuve. J: Spiro. A. Ill; Sparrow. D; Smith. TJ: Nie.
        H; Weisskopf. MG: Hu. H; Wright. RO. (2008). Interaction of the delta-aminolevulinic acid
        dehydratase polymorphism and lead burden on cognitive function: The VA Normative Aging Study. J
        Occup Environ Med 50: 1053-1061. http://dx.doi.org/10.1097/JOM.Ob013e3181792463

     Rajaraman. P; Stewart. PA; Samet. JM; Schwartz. BS: Linet. MS: Zahm. SH; Rothman. N; Yeager. M;
        Fine. HA; Black. PM; Loeffler. J: Shapiro. WR; Selker. RG: Inskip. PP. (2006). Lead, genetic
        susceptibility, and risk of adult brain tumors. Cancer Epidemiol Biomarkers Prev 15: 2514-2520.
        http://dx.doi.org/10.1158/1055-9965.EPI-06-0482

     Ramesh. GT; Manna. SK; Aggarwal. BB; Jadhav. AL. (1999).  Lead activates nuclear transcription factor -
        kB, activator protein-1, and amino-terminal c-Jun kinase in pheochromocytoma cells. Toxicol Appl
        Pharmacol 155: 280-286. http://dx.doi.org/10.1006/taap.1999.8624

     Ramesh. GT; Manna. SK; Aggarwal. BB; Jadhav. AL. (2001).  Lead exposure activates nuclear factor
        kappa B, activator protein-1, c-Jun N-terminal kinase and caspases in the rat brain. Toxicol Lett 123:
        195-207. http://dx.doi.org/10.1016/S0378-4274(01)00395-2

     Rana. SVS. (2008). Metals and apoptosis: Recent developments. J Trace Elem Med Biol 22: 262-284.
        http://dx.doi.0rg/10.1016/i.itemb.2008.08.002

     Razmiafshari. M; Kao. J: dAvignon. A; Zawia. NH. (2001). NMR identification of heavy metal-binding
        sites in a synthetic zinc finger peptide: Toxicological implications for the interactions of xenobiotic
        metals with zinc finger proteins. Toxicol Appl Pharmacol 172: 1-10.
        http://dx.doi.org/10.1006/taap.2001.9132

     Razmiafshari. M; Zawia. NH. (2000). Utilization of a synthetic peptide as a tool to study the interaction
        of heavy metals with the  zinc finger domain of proteins critical for gene expression in the developing
        brain. Toxicol Appl Pharmacol 166: 1-12. http://dx.doi.org/10.1006/taap.2000.8950

     Reddv. GR; Zawia. NH. (2000). Lead exposure alters Egr-1 DNA-binding in the neonatal rat brain. Int J
        DevNeurosci 18: 791-795. http://dx.doi.org/10.1016/S0736-5748(00)00048-4
November 2012                             5-774                   Draft - Do Not Cite or Quote

-------
     Rees. PJ: Shelton. D; Chan. TB; Eiser. N; Clark. TJH; Maisev. MN. (1985). Effects of histamine on lung
        permeability in normal and asthmatic subjects. Thorax 40: 603-606.

     Regunathan. S: Sundaresan. R. (1984). Pyruvate metabolism in the brain of young rats intoxicated with
        organic and inorganic lead. JNeurochem 43: 1346-1351.
        http://dx.doi.org/10.1111/i.l471-4159.1984.tb05393.x

     Rehman. S: Rehman. S: Chandra. O: Abdulla. M. (1995). Evaluation of malondialdehyde as an index of
        lead damage in rat brain homogenates. Biometals  8: 275-279. http://dx.doi.org/10.1007/BF00141599

     Reif. A; Fritzen. S: Finger. M; Strobel. A; Lauer. M; Schmitt. A; Lesch. K. (2006). Neural stem cell
        proliferation is decreased in schizophrenia, but not in depression. Mol Psychiatry 11: 514-522.
        http://dx.doi.org/10.1038/sj.mp.4001791

     Rendon-Ramirez. A; Cerbon-Solorzano. J; Maldonado-Vega. M; Quintanar-Escorza. MA; Calderon-
        Salinas. JV. (2007).  Vitamin-E reduces the oxidative damage on 5-aminolevulinic dehydratase induced
        by lead intoxication in rat erythrocytes. Toxicol In Vitro 21: 1121-1126.
        http://dx.doi.0rg/10.1016/i.tiv.2007.04.019

     Reuhl. KR; Puce.  DC: Gilbert. SG: Mallett J. (1989). Effects of chronic developmental lead exposure on
        monkey neuroanatomy: Visual system. Toxicol Appl Pharmacol 99: 501-509.
        http://dx.doi.org/10.1016/0041-008X(89)90157-9

     Revis. NW: Zinsmeister. AR; Bull. R. (1981). Atherosclerosis and hypertension induction by lead and
        cadmium ions: An effect prevented by calcium ion. PNAS 78: 6494-6498.

     Reza. B; Ali. N; Azhdar. H; Alireza. A; Ali. K. (2008). Effects of low-level lead exposure on blood
        pressure and function of the rat isolated heart. Indian  J Pharmacol 40: 69-72.
        http://dx.doi.org/10.4103/0253-7613.41041

     Rhodes.  D; Spiro. A. Ill; Aro. A; Hu. H. (2003). Relationship of bone and blood lead levels to psychiatric
        symptoms: The Normative Aging Study. J Occup Environ Med  45: 1144-1151.
        http://dx.doi.org/10.1097/01.jom.0000094995.23808.7b

     Rice. C:  Ghorai. JK; Zalewski. K; Weber. DN. (2011). Developmental lead exposure causes startle
        response deficits in  zebrafish. Aquat Toxicol 105:  600-608.
        http://dx.doi.0rg/10.1016/i.aquatox.2011.08.014

     Rice. D: Barone. S. (2000). Critical periods of vulnerability for the developing nervous system: Evidence
        from humans and animal models [Review]. Environ Health Perspect 108: 511-533.
        http://dx.doi.org/10.1289/ehp.00108s3511

     Rice. DC. (1990). Lead-induced behavioral impairment on a spatial discrimination reversal task in
        monkeys exposed during different periods of development. Toxicol Appl Pharmacol 106:  327-333.
        http://dx.doi.org/10.1016/0041-008X(90)90251-O

     Rice. DC. (1992a). Effect of lead during different developmental periods in the monkey on concurrent
        discrimination performance. Neurotoxicology 13:  583-592.

     Rice. DC. (1992b). Lead exposure during different developmental periods produces different effects on FI
        performance in monkeys tested as juveniles  and adults. Neurotoxicology 13: 757-770.

     Rice. DC. (1997). Effects of lifetime lead exposure in monkeys on  detection of pure tones. Fundam Appl
        Toxicol 36: 112-118. http://dx.doi.org/10.1006/faat.1996.2268

     Rice. DC. (1998). Effects of lifetime lead exposure on spatial and temporal visual function in monkeys.
        Neurotoxicology 19: 893-902.

     Rice. DC: Gilbert. SG.  (1985). Low lead exposure from birth produces behavioral toxicity (DRL) in
        monkeys. Toxicol Appl Pharmacol 80: 421-426. http://dx.doi.org/10.1016/0041-008X(85)90386-2

     Rice. DC: Gilbert. SG.  (1990a). Lack of sensitive period for lead-induced behavioral impairment on a
        spatial delayed alternation task in monkeys.  Toxicol Appl Pharmacol 103: 364-373.
        http://dx.doi.org/10.1016/0041-008X(90)90236-N
November 2012                             5-775                   Draft - Do Not Cite or Quote

-------
     Rice. DC: Gilbert. SG. (1990b). Sensitive periods for lead-induced behavioral impairment (nonspatial
        discrimination reversal) in monkeys. Toxicol Appl Pharmacol 102: 101-109.
        http://dx.doi.org/10.1016/0041-008X(90)90087-B

     Rice. DC: Karpinski. KF. (1988). Lifetime low-level lead exposure produces deficits in delayed
        alternation in adult monkeys. Neurotoxicol Teratol 10: 207-214.
        http:/7dx.doi.org/10.1016/0892-0362(88)90019-0

     Richardt. G: Federolf. G: Habermann. E. (1986). Affinity of heavy metal ions to intracellular Ca2+-
        binding proteins. Biochem Pharmacol 35: 1331-1335.
        httrj://dx.doi.org/10.1016/0006-2952(86)90278-9

     Richetti. S: Rosemberg. D; Ventura-Lima. J; Monserrat. J; Bogo. M; Bonan. C. (2010).
        Acetylcholinesterase activity and antioxidant capacity of zebrafish brain is altered by heavy metal
        exposure. Neurotoxicology 32: 116-122. http://dx.doi.Org/10.1016/i.neuro.2010.ll.001

     Riddell. TJ; Solon. O; Quimbo. SA; Tan. CM; Butrick.  E; Peabodv. JW. (2007). Elevated blood-lead
        levels among children living in the rural Philippines. Bull World Health Organ 85: 674-680.

     Pus. MD; Dietrich. KN; Succop. PA; Berger. OG: Bornschein. RL. (2004). Early exposure to lead and
        neuropsychological outcome in adolescence. J Int Neuropsychol Soc 10: 261-270.
        http://dx.doi.org/10.1017/S1355617704102154

     Rizzi. E; Castro. MM; Fernandes. K; Barbosa. F; Arisi. GM; Garcia-Cairasco. N; Bendhack. LM; Tanus-
        Santos. JE; Gerlach. RF. (2009). Evidence of early involvement of matrix metalloproteinase-2 in lead-
        induced hypertension. Arch Toxicol 83: 439-449. http://dx.doi.org/10.1007/s00204-008-0363-l

     Robins. JM; Cullen. MR; Connors. BB; Kayne. RD. (1983). Depressed thyroid indexes associated with
        occupational exposure to inorganic lead. Arch Intern Med 143: 220-224.
        http://dx.doi.0rg/10.1001/archinte.143.2.220

     Robinson. G: Baumann. S: Kleinbaum. D; Barton. C: Schroeder. S: Mushak. P; Otto. D.  (1985). Effects
        of low to moderate lead exposure on brainstem auditory evoked potentials in children. In
        Neurobehavioral methods in occupational and environmental health: extended abstracts from the
        second international symposium; August; Copenhagen, Denmark. Copenhagen, Denmark: World
        Health Organization.

     Rodriguez-Iturbe. B; Sindhu. RK; Quiroz. Y; Vaziri. ND. (2005). Chronic exposure to low doses of lead
        results in renal infiltration of immune cells, NF-kappaB  activation, and overexpression of
        tubulointerstitial angiotensin II. Antioxid Redox Signal 7: 1269-1274.
        http://dx.doi.0rg/10.1089/ars.2005.7.1269

     Rodriguez-Iturbe. B; Vaziri. ND; Herrera-Acosta. J; Johnson. RJ. (2004). Oxidative stress, renal
        infiltration of immune cells, and salt-sensitive hypertension: All for one and one for all [Review]. Am
        J Physiol 286: F606-F616. http://dx.doi.org/10.1152/aiprenal.00269.2003

     Roels. H; Lauwerys. R; Konings. J; Buchet. JP; Bernard. A; Green. S; Bradley. D; Morgan. W; Chettle.
        EX (1994). Renal function and hyperfiltration capacity in lead smelter workers with high bone lead.
        Occup Environ Med 51: 505-512.  http://dx.doi.Org/10.1136/oem.51.8.505

     Romeo. R; Aprea. C; Boccalon. P; OrsiD; Porcelli. B; Sartorelli. P. (1996).  Serum erthropoietin and blood
        lead concentrations. Int Arch Occup Environ Health 69:  73-75.

     Roncal. C; Mu. W;  Reungjui. S; Kim. KM; Henderson. GN; Ouyang. X; Nakagawa. T; Johnson. RJ.
        (2007). Lead, at low levels, accelerates arteriolopathy and tubulointerstitial injury in chronic kidney
        disease. Am J Physiol Renal Physiol 293: F1391-F1396.
        http://dx.doi.org/10.1152/ajprenal.00216.2007

     Rondo. PH; Carvalho. MDF; Souza. MC; Moraes. F. (2006). Lead, hemoglobin, zinc protoporphyrin and
        ferritin concentrations in children. Rev Saude Publica 40: 71-76.
        http://dx.doi.org/10.1590/S0034-89102006000100012
November 2012                              5-776                  Draft - Do Not Cite or Quote

-------
     Ronis. MJ; Badger. TM; Shema. SJ: Roberson. PK; Shaikh. F. (1998a). Effects on pubertal growth and
        reproduction in rats exposed to lead perinatally or continuously throughout development. J Toxicol
        Environ Health A 53: 327-341. http://dx.doi.org/10.1080/009841098159312

     Ronis. MJJ; Aronson. J; Gao. GG: Hogue. W: Skinner. RA; Badger. TM; Lumpkin. CK. Jr. (2001).
        Skeletal effects of developmental lead exposure in rats. Toxicol Sci 62: 321-329.
        http://dx.doi.0rg/10.1093/toxsci/62.2.321

     Ronis. MJJ: Badger. TM: Shema. SJ: Roberson. PK: Shaikh. F. (1996). Reproductive toxicity and growth
        effects in rats exposed to lead at different periods during development. Toxicol Appl Pharmacol 136:
        361-371. http://dx.doi.org/10.1006/taap.1996.0044

     Ronis. MJJ: Gandv. J: Badger. T.  (1998b). Endocrine mechanisms underlying reproductive toxicity in the
        developing rat chronically exposed to dietary lead. J Toxicol Environ Health A 54: 77-99.
        http://dx.doi.org/10.1080/009841098158935

     Roshan. VD; Assali. M; Moghaddam. AH; Hosseinzadeh. M; Myers. J. (2011). Exercise training and
        antioxidants: effects on rat heart tissue exposed to lead acetate. Int J Toxicol 30: 190-196.
        http://dx.doi.org/10.1177/1091581810392809

     Rossi-George. A; Virgolini. M; Weston. D; Thiruchelvam. M; Corv-Slechta. D. (2011). Interactions of
        lifetime lead exposure and stress: Behavioral, neurochemical and HPA axis effects. Neurotoxicology
        32: 83-99. http://dx.doi.Org/10.1016/i.neuro.2010.09.004

     Rossi-George. A; Virgolini. MB; Weston. D; Corv-Slechta. DA. (2009). Alterations in glucocorticoid
        negative feedback following maternal Pb, prenatal stress and the combination: A potential biological
        unifying mechanism for their corresponding disease profiles. Toxicol Appl Pharmacol 234: 117-127.
        http://dx.doi.0rg/10.1016/i.taap.2008.10.003

     Rothenberg. SJ: Poblano. A; Garza-Morales. S. (1994b). Prenatal and perinatal low level lead exposure
        alters brainstem auditory evoked responses in infants. Neurotoxicology 15: 695-699.

     Rothenberg. SJ: Poblano. A; Schnaas. L. (2000). Brainstem auditory evoked response at five years and
        prenatal and postnatal blood lead. Neurotoxicol Teratol 22: 503-510.
        http://dx.doi.org/10.1016/S0892-0362(00)00079-9

     Rothenberg. SJ: Rothenberg. JC. (2005). Testing the dose-response specification in  epidemiology: Public
        health and policy consequences for lead. Environ Health Perspect 113:  1190-1195.
        http://dx.doi.org/10.1289/ehp.7691

     Rothenberg. SJ: Schnaas. L; Salgado-Valladares. M; Casanueva. E; Geller. AM; Hudnell. HK; Fox. DA.
        (2002b). Increased ERG a- and b-wave amplitudes in 7- to 10-year-old children resulting from
        prenatal lead exposure. Invest Ophthalmol Vis Sci 43: 2036-2044.

     Rotruck. JT; Pope. AL; Ganther. HE; Swanson. AB; Hafeman. DG: Hoekstra. WG.  (1973). Selenium:
        Biochemical role as a component of glutathione peroxidase. Science 179: 588-590.
        http://dx.doi.org/10.1126/science.179.4073.588

     Rousseau. MC: Parent. ME; Nadon. L; Latrellie. B; Siemiatvcki. J. (2007). Occupational exposure to lead
        compounds and risk of cancer among men: A population-based case-control study. Am J Epidemiol
        166: 1005-1014. http://dx.doi.org/10.1093/aie/kwml83

     Rousseau. MC: Straif. K; Siemiatvcki. J. (2005). IARC carcinogen update [Letter].  Environ Health
        Perspect 113: A580-A581. http://dx.doi.org/10.1289/ehp.113-a580

     Roy. A; Bellinger. D; Hu. H; Schwartz. J: Ettinger. AS: Wright. RO: Bouchard. M; Palaniappan. K;
        Balakrishnan. K. (2009a). Lead exposure and behavior among young children in Chennai, India.
        Environ Health Perspect 117:  1607-1611. http://dx.doi.org/10.1289/ehp.0900625

     Roy. A; Hu. H; Bellinger. DC: Mukherjee. B; Modali. R; Nasaruddin. K; Schwartz. J: Wright. RO:
        Ettinger. AS: Palaniapan. K; Balakrishnan. K. (2011). Hemoglobin, lead exposure, and intelligence
        quotient: Effect modification by the DRD2 Taq IA polymorphism. Environ Health Perspect 119:
        144-149. http://dx.doi.org/10.1289/ehp.0901878
November 2012                              5-777                   Draft - Do Not Cite or Quote

-------
     Roy. AA; Baxla. SP; Gupta. T; Bandvopadhyava. R; Tripathi. SN. (2009b). Particles emitted from indoor
        combustion sources: Size distribution measurement and chemical analysis. Inhal Toxicol 21: 837-848.
        http://dx.doi.org/10.1080/08958370802538050

     Roy. AK; Dhir. H; Sharma. A. (1992). Modification of metal-induced micronuclei formation in mouse
        bone marrow erythrocytes by Phyllanthus fruit extract and ascorbic acid. Toxicol Lett 62: 9-17.
        http://dx.doi.org/10.1016/0378-4274(92)90072-R

     Roy. NK; Rossman. TG. (1992). Mutagenesis and comutagenesis by lead compounds. Mutat Res 298:
        97-103. http://dx.doi.org/10.1016/0165-1218(92)90034-W

     Rubio. J: Riqueros. MI; Gasco. M; Yucra. S: Miranda. S: Gonzales. GF. (2006). Lepidium meyenii
        (Maca) reversed the lead acetate induced - Damage on reproductive function in male rats. Food Chem
        Toxicol 44: 1114-1122. http://dx.doi.Org/10.1016/i.fct.2006.01.007

     Rybicki. BA; Johnson. CC: Uman. J; Gorell. JM. (1993). Parkinson's disease mortality and the industrial
        use of heavy metals in Michigan. Mov Disord 8: 87-92. http://dx.doi.org/10.1002/mds.870080116

     Sa. I; da Costa. MJP; Cunha. EM. (2012). Lead hepatotoxicology: A study in an animal model. Toxicol
        Ind Health 28:  108-113. http://dx.doi.org/10.1177/0748233711407240

     Sabbioni. E; Marafante. E. (1976). Identification of lead-binding components in rat liver: In vivo study.
        Chem Biol Interact 15: 1-20. http://dx.doi.org/10.1016/0009-2797(76)90124-l

     Saimi. Y; Kung. C. (2002). Calmodulin as an ion channel subunit [Review]. Annu Rev Physiol 64:
        289-311. http://dx.doi.org/10.1146/annurev.phvsiol.64.100301.111649

     Sainath. SB; Meena. R; Supriya.  C. h; Reddv. KP; Reddv. PS. (2011). Protective role of Centella asiatica
        on lead-induced oxidative stress and suppressed reproductive health in male rats. Environ Toxicol
        Pharmacol32:  146-154. http://dx.doi.Org/10.1016/i.etap.2011.04.005

     Sakai. T; Morita. Y. (1996). delta-Aminolevulinic acid in plasma or whole blood as a sensitive indicator
        of lead effects, and its relation to the other heme-related parameters. Int Arch Occup Environ Health
        68: 126-132. http://dx.doi.org/10.1007/BF00381245

     Sakata. S: Shimizu. S: Ogoshi. K; Hirai. K; Ohno. Y; Kishi. T; Sherchand. JB;  Utsumi. M; Shibata. M;
        Takaki. M; Ueda. M; Mori. I.  (2007). Inverse relationship between serum erythropoietin and blood
        lead concentrations in Kathmandu tricycle taxi drivers. Int Arch Occup Environ Health 80: 342-345.
        http://dx.doi.org/10.1007/s00420-006-0125-4

     Sakuma. H; Kusama. M; Yamaguchi. K; Matsuki. T; Sugawara. S. (1984). The distribution of cigarette
        smoke components between mainstream and sidestream smoke: II. Bases. Beitraege zur
        Tabakforschung International 12: 199-209.

     Salawu. EO; Adeevo. OA; Falokun. OP; Yusuf. UA; Oyerinde. A; Adeleke. AA. (2009). Tomato
        (Lycopersicon esculentum) prevents lead-induced testicular toxicity. Journal of Human Reproductive
        Sciences 2: 30-34. http://dx.doi.org/10.4103/0974-1208.51346

     Saldiva. PHN; Clarke. RW: Coull. BA; Stearns. RC: Lawrence. J: Murthy. GGK; Diaz. E; Koutrakis. P;
        Suh. H; Tsuda. A; Godleski. JJ. (2002). Lung inflammation induced by concentrated ambient air
        particles is related to particle composition. Am J Respir Crit Care Med 165: 1610-1617.
        http://dx.doi.org/10.1164/rccm.2106102

     Saleh. M; Mathison. JC: Wolinski. MK; Bensinger. SJ: Fitzgerald. P; Droin. N; Ulevitch. RJ; Green. PR;
        Nicholson. DW. (2006). Enhanced bacterial clearance and sepsis resistance  in caspase-12-deficient
        mice. Nature 440: 1064-1068. http://dx.doi.org/10.1038/nature04656

     Sanchez-Fructuoso. AI; Blanco. J: Cano. M; Ortega. L; Arroyo. M; Fernandez. C: Prats. D; Barrientos. A.
        (2002a). Experimental lead nephropathy: Treatment with calcium disodium
        ethylenediaminetetraacetate. Am J Kidney Dis 40: 59-67. http://dx.doi.org/10.1053/ajkd.2002.33936

     Sanchez-Fructuoso. AI; Cano. M; Arroyo. M; Fernandez. C: Prats. D; Barrientos. A. (2002b). Lead
        mobilization during calcium disodium ethylenediaminetetraacetate chelation therapy in treatment of
        chronic lead poisoning. Am J  Kidney Dis 40: 51-58. http://dx.doi.org/10.1053/aikd.2002.33913
November 2012                              5-778                   Draft - Do Not Cite or Quote

-------
     Sandercock. GRH; Brodie. DA. (2006). The role of heart rate variability in prognosis for different modes
        of death in chronic heart failure [Review]. Pacing Clin Electrophysiol 29: 892-904.
        http://dx.doi.0rg/10.llll/i.1540-8159.2006.00457.x

     Sandhir. R; Gill. KD. (1995). Effect of lead on lipid peroxidation in liver of rats. Biol Trace Elem Res 48:
        91-97. http://dx.doi.org/10.1007/BF02789081

     Sandhir. R; Julka. D; Gill. KD. (1994). Lipoperoxidative damage on lead exposure in rat brain and its
        implications on membrane bound enzymes. Pharmacol Toxicol 74: 66-71.
        http://dx.doi.0rg/10.llll/i.1600-0773.1994.tb01077.x

     Sanna. E; Vallascas. E. (2011). Hair lead levels to evaluate the subclinical impact of lead on growth in
        Sardinian children (Italy). Am J Hum Biol 23: 740-746. http://dx.doi.org/10.1002/ajhb.21203

     Santibanez. M; Vioque. J: Alguacil. J: Barber. X; de la Hera. G: Kauppinen. T. (2008). Occupational
        exposures and risk of oesophageal cancer by histological type: a case-control study in eastern Spain.
        Occup Environ Med 65: 774-781. http://dx.doi.org/10.1136/oem.2007.037929

     Santos. MRV; Marchioro. M; Antoniolli. AR. (2006). Lead effects on non-adrenergic non-cholinergic
        relaxations in the rat gastric fundus. Toxicol In Vitro  20: 38-42.
        http://dx.doi.0rg/10.1016/i.tiv.2005.05.005

     Saraiva. MCP; Taichman. RS: Braun. T; Nriagu. J: Eklund.  SA; Burt. BA. (2007). Lead exposure and
        periodontitis in US  adults. J Periodontal Res 42: 45-52.
        http://dx.doi.0rg/10.llll/i.1600-0765.2006.00913.x

     Sarasua. SM; Mueller.  P; Kathman. S: Campagna. D; Uddin. MS: White. MC. (2003). Confirming the
        utility of four kidney biomarker tests in a longitudinal follow-up study. Ren Fail 25: 797-817.

     Sarasua. SM; Vogt RF; Henderson. LO; Jones. PA; Lybarger. JA. (2000). Serum immunoglobulins and
        lymphocyte subset distributions in children and adults living in communities assessed for lead and
        cadmium exposure. J  Toxicol Environ Health A 60: 1-15. http://dx.doi.org/10.1080/009841000156556

     Sarnat. JA; Marmur. A; Klein. M; Kim. E; Russell. AG:  Sarnat. SB; Mulholland. JA; Hopke. PK; Tolbert.
        PE. (2008). Fine particle sources and cardiorespiratory morbidity: An application of chemical mass
        balance and factor analytical source-apportionment methods. Environ Health Perspect 116: 459-466.
        http://dx.doi.org/10.1289/ehp.10873

     Saverv. LC: Grlickova-Duzevik. E; Wise. SS: Thompson. WD; Hinz. JA: Thompson. LH; Wise. JP, Sr.
        (2007). Role of the  Fancg gene in protecting cells from paniculate chromate-induced chromosome
        instability. Mutat Res  Genet Toxicol Environ Mutagen 626: 120-127.
        http://dx.doi.0rg/10.1016/i.mrgentox.2006.09.005

     Schanne. FAX; Dowd.  TL; Gupta. RK; Rosen. JF. (1989). Lead increases free Ca2+ concentration  in
        cultured osteoblastic bone cells: simultaneous  detection of intracellular free Pb2+ by 19F NMR.
        PNAS 86: 5133-5135.

     Schanne. FAX; Long. GJ: Rosen. JF. (1997). Lead induced rise in intracellular free calcium is mediated
        through activation of protein kinase C in osteoblastic bone cells. Biochim Biophys Acta 136: 247-254.
        http://dx.doi.org/10.1016/S0925-4439(97)00006-9

     Schell. LM; Denham. M; Stark. AD; Gomez. M; Ravenscroft. J: Parsons. PJ: Aydermir. A; Samelson. R.
        (2003). Maternal blood lead concentration, diet during pregnancy, and anthropometry predict neonatal
        blood lead in a socioeconomically disadvantaged population. Environ Health Perspect 111: 195-200.
        http://dx.doi.org/10.1289/ehp.5592

     Schell. LM; Denham. M; Stark. AD; Parsons. PJ:  Schulte. EE. (2009). Growth of infants' length, weight,
        head and arm circumferences in relation to low levels of blood lead measured serially. Am J Hum Biol
        21: 180-187. http://dx.doi.org/10.1002/aihb.20842

     Schirrmacher. K; Wiemann. M; Bingmann. D; Busselberg. D. (1998). Effects of lead, mercury, and
        methyl mercury on  gap junctions and [CA2+]i in bone cells. Calcif Tissue Int 63:  134-139.
        http://dx.doi.org/10.1007/s002239900503
November 2012                              5-779                  Draft - Do Not Cite or Quote

-------
     Schmitt. CJ: Whyte. JJ: Brumbaugh. WG: Tillitt DE. (2005). Biochemical effects of lead, zinc, and
        cadmium from mining on fish in the Tri-States District of northeastern Oklahoma, USA. Environ
        Toxicol Chem 24:  1483-1495. http://dx.doi.Org/10.1897/04-332R.l

     Schmitt. CJ: Whyte. JJ: Roberts. AP; Annis. ML; May. TW: Tilitt. DE. (2007b). Biomarkers of metals
        exposure in fish from lead-zinc mining areas of southeastern Missouri, USA. Ecotoxicol Environ Saf
        67: 31-47. http://dx.doi.Org/10.1016/i.ecoenv.2006.12.011

     Schnaas. L; Rothenberg.  SJ: Flores. MF; Martinez. S: Hernandez. C: Osorio. E; Velasco. SR; Perroni. E.
        (2006). Reduced intellectual development in children with prenatal lead exposure. Environ Health
        Perspect 114: 791-797. http://dx.doi.org/10.1289/ehp.8552

     Schnaas. L; Rothenberg.  SJ: Perroni. E; Martinez. S: Hernandez. C: Hernandez. RM. (2000). Temporal
        pattern in the effect of postnatal blood lead level on intellectual development of young children.
        Neurotoxicol Teratol 22: 805-810. http://dx.doi.org/10.1016/S0892-0362(00)00101-X

     Schneider. JS: Anderson. DW: Wade. TV; Smith. MG: Leibrandt. P; Zuck. L; Lidsky. TI. (2005).
        Inhibition of progenitor cell proliferation in the dentate gyrus of rats following post-weaning lead
        exposure. Neurotoxicology 26: 141-145. http://dx.doi.Org/10.1016/j.neuro.2004.06.006

     Schober. SE; Mirel. LB; Graubard. BI; Brodv. DJ: Flegal. KM.  (2006). Blood lead levels and death from
        all causes, cardiovascular disease, and cancer: Results from  the NHANES III Mortality Study. Environ
        Health Perspect 114: 1538-1541. http://dx.doi.org/10.1289/ehp.9123

     Schrauzer. GN. (1987). Effects of selenium antagonists on cancer susceptibility: New aspects of chronic
        heavy metal toxicity. J UOEH 9: 208-215.

     Schrauzer. GN. (2008). Effects of selenium and low levels of lead on mammary tumor development and
        growth in MMTV-infected female mice. Biol Trace Elem Res 125: 268-275.
        http://dx.doi.org/10.1007/sl2011-008-8172-l

     Schwark. WS: Haluska. M; Blackshear. P; Magana. T. (1985). Lifetime lead intoxication: influence on the
        amygdaloid kindling model of epileptogenesis. Toxicology 36: 49-60.
        http://dx.doi.org/10.1016/0300-483X(85)90006-X

     Schwartz. BS: Chen. S: Caffo. B; Stewart. WF; Bolla. KI; Yousem. D; Davatzikos. C. (2007). Relations
        of brain volumes with cognitive function in males 45 years and older with past lead exposure.
        Neuroimage 37: 633-641. http://dx.doi.Org/10.1016/j.neuroimage.2007.05.035

     Schwartz. BS: Lee. BK; Bandeen-Roche. K; Stewart. W: Bolla. K; Links. J: Weaver. V; Todd. A. (2005).
        Occupational lead exposure and longitudinal decline in neurobehavioral test scores. Epidemiology 16:
        106-113. http://dx.doi.org/10.1097/01.ede.0000147109.62324.51

     Schwartz. BS: Lee. BK; Lee. GS: Stewart. WF; Lee. SS: Hwang. KY; Ahn. KD; Kim. YB; Bolla. KI;
        Simon. D; Parsons. PJ; Todd. AC. (2001). Associations of blood lead, dimercaptosuccinic acid-
        chelatable lead, and tibia lead with neurobehavioral test scores in South Korean lead workers. Am J
        Epidemiol  153: 453-464. http://dx.doi.Org/10.1093/aie/153.5.453

     Schwartz. BS; Lee. BK; Lee. GS; Stewart. WF; Simon. D; Kelsev. K; Todd. AC. (2000a). Associations of
        blood lead, dimercaptosuccinic acid-chelatable lead, and tibia lead with polymorphisms in the vitamin
        D receptor and delta-aminolevulinic acid dehydratase genes. Environ Health Perspect 108: 949-954.
        http://dx.doi.org/10.2307/3435053

     Schwartz. BS; Lee. BK; Stewart. W; Ahn. KD; Kelsev. K; Bresssler. J. (1997a). Associations of subtypes
        of hemoglobin with delta-aminolevulinic acid dehydratase genotype and dimercaptosuccinic acid-
        chelatable lead levels. Arch Environ Occup Health 52: 97-103.
        http://dx.doi.org/10.1080/00039899709602871

     Schwartz. BS; Lee. BK; Stewart. W; Sithisarankul. P; Strickland. PT; Ahn. KD; Kelsev. K. (1997b).
        delta-Aminolevulinic acid dehydratase genotype modifies four hour urinary lead excretion after oral
        administration of dimercaptosuccinic acid. Occup Environ Med 54: 241-246.
November 2012                              5-780                   Draft - Do Not Cite or Quote

-------
     Schwartz. J. (1991). Lead, blood pressure, and cardiovascular disease in men and women. Environ Health
        Perspect 91: 71-75. http://dx.doi.org/10.1289/ehp.919171

     Schwartz. J. (1994). Low-level lead exposure and children's IQ: A meta-analysis and search for a
        threshold. Environ Res 65: 42-55. http://dx.doi.org/10.1006/enrs.1994.1020

     Schwartz. J: Landrigan. PJ; Baker. EL. Jr; Orenstein. WA; von Lindern. IH. (1990). Lead-induced
        anemia: dose-response relationships and evidence for a threshold. Am J Public Health 80: 165-168.

     Schwartz. J: Otto. D. (1987). Blood lead, hearing thresholds, and neurobehavioral development in
        children and youth. Arch Environ Occup Health 42: 153-160.
        http://dx.doi.org/10.1080/00039896.1987.9935814

     Schwartz. J: Otto. D. (1991). Lead and minor hearing impairment. Arch Environ Occup Health 46:
        300-305. http://dx.doi.org/10.1080/00039896.1991.9934391

     Sciarillo. WG: Alexander. G: Farrell. KP (1992). Lead exposure and child behavior. Am J Public Health
        82: 1356-1360.

     Scinicariello. F; Murray. HE; Moffett. DB; Abadin. HG: Sexton. MJ; Fowler. BA. (2007). Lead and delta-
        aminolevulinic acid dehydratase polymorphism: Where does it lead? A meta-analysis. Environ Health
        Perspect 115:  35-41. http://dx.doi.org/10.1289/ehp.9448

     Scinicariello. F; Yesupriya. A; Chang. MH; Fowler. BA. (2010). Modification by ALAD of the
        association between blood lead and blood pressure in the U.S. population: Results from the Third
        National Health and Nutrition Examination Survey. Environ Health Perspect 118: 259-264.
        http://dx.doi.org/10.1289/ehp.0900866

     Seki. T; Arai. Y. (1993). Highly polysialylated NCAM expression in the developing and adult rat spinal
        cord [Letter].  Dev Brain Res 73: 141-145. httrj://dx.doi.org/10.1016/0165-3806(93)90056-G

     Selevan. SG: Rice. DC: Hogan. KA; Euling. SY; Pfahles-Hutchens. A; Bethel. J.  (2003). Blood lead
        concentration  and delayed puberty in girls. N Engl J Med 348: 1527-1536.
        http://dx.doi.org/10.1056/NEJMoa020880

     Selvin-Testa. A; Lopez-Costa. JJ: Nessi-de Avinon. AC: Saavedra. JP (1991). Astroglial alterations in rat
        hippocampus  during chronic lead exposure. Glia 4: 384-392. http://dx.doi.org/10.1002/glia.440040406

     Selye. H; Tuchweber. B; Bertok. L. (1966). Effect of lead acetate on the susceptibility of rats to bacterial
        endotoxins. J Bacteriol 91: 884-890.

     Senger. MR; Rico. EP; Arizi. MDB; Frazzon. APG: Dias. RD; Bogo. MR; Bonan. CD. (2006). Exposure
        to Hg2+ and Pb2+ changes NTPDase and ecto-5 '-nucleotidase activities in central nervous system of
        zebrafish (Danio rerio). Toxicology 226: 229-237. http://dx.doi.Org/10.1016/j.tox.2006.07.012

     Shabani. A; Rabbani. A. (2000). Lead nitrate induced apoptosis in alveolar macrophages from rat lung.
        Toxicology 149: 109-114. http://dx.doi.org/10.1016/S0300-483X(00)00232-8

     Shadick. NA; Kim. R; Weiss.  S: Liang. MH; Sparrow. D: Hu. H. (2000). Effect of low level lead exposure
        on hyperuricemia and gout among middle aged and elderly men: the normative aging study. J
        Rheumatol27: 1708-1712.

     Shah. F; Kazi. TG: Afridi. HI; Baig. JA; Khan. S:  Kolachi. NF; Wadhwa. SK; Shah. AQ. (2010).
        Environmental exposure of lead and iron deficit anemia in children age ranged 1-5 years: A cross
        sectional study. Sci Total Environ 408: 5325-5330. http://dx.doi.Org/10.1016/i.scitotenv.2010.07.091

     Shaik. AP; Jamil. K.  (2009). Individual susceptibility and genotoxicity in workers exposed to hazardous
        materials like  lead.  J Hazard Mater 168: 918-924. http://dx.doi.Org/10.1016/i.jhazmat.2009.02.129

     Shakoor. A; Gupta. PK; Singh. YP; Kataria. M. (2000). Beneficial effects of aluminum on the progression
        of lead-induced nephropathy in rats. Basic Clin Pharmacol Toxicol 87: 258-260.
        http://dx.doi.0rg/10.1034/i.1600-0773.2000.pto870603.x
November 2012                              5-781                   Draft - Do Not Cite or Quote

-------
     Shan. G: Tang. T; Zhang. X. (2009). The protective effect of ascorbic acid and thiamine supplementation
        against damage caused by lead in the testes of mice. J Huazhong Univ Sci Technolog Med Sci 29:
        68-72. http://dx.doi.org/10.1007/sll596-009-0114-4

     Sharifi. AM; Darabi. R; Akbarloo. N; Larijani. B; Khoshbaten. A. (2004). Investigation of circulatory and
        tissue ACE activity during development of lead-induced hypertension. Toxicol Lett 153: 233-238.
        http://dx.doi.0rg/10.1016/i.toxlet.2004.04.013

     Sharifi. AM; Mousavi. SH. (2008). Studying the effects of lead on DNA fragmentation and proapoptotic
        bax and antiapoptotic Bcl-2 protein expression in PC12 cells. Toxicol Mech Meth 18:  75-79.
        http://dx.doi.org/10.1080/15376510701665814

     Sharifi. AM; Mousavi. SH; Jorjani. M. (2010). Effect of chronic lead exposure on pro-apoptotic Bax and
        anti-apoptotic Bcl-2 protein expression in rat hippocampus in vivo. Cell Mol Neurobiol 30: 769-774.
        http://dx.doi.org/10.1007/sl0571-010-9504-l

     Sharma. A; Sharma. V: Kansal. L. (2010a). Amelioration of lead-induced hepatotoxicity by Allium
        sativum extracts in Swiss albino mice. Libyan Journal  of Medicine 5: 4621.
        http://dx.doi.org/10.4176/091107

     Sharma. V: Sharma. A; Kansal. L. (201 Ob). The effect of oral administration of Allium sativum extracts
        on lead nitrate induced toxicity in male mice. Food Chem Toxicol 48: 928-936.
        http://dx.doi.0rg/10.1016/i.fct.2010.01.002

     Shelkovnikov. SA; Gonick. HC. (2001). Influence of lead on rat thoracic aorta contraction and relaxation.
        Am J Hypertens 14: 873-878. http://dx.doi.org/10.1016/S0895-7061(01)02149-5

     Shelton. KR: Cunningham. JG: Klann. E; Merchant. RE; Egle. PM; Bigbee. JW. (1990). Low-abundance
        32-kilodalton nuclear protein specifically enriched in the central nervous system. J Neurosci Res 25:
        287-294. http://dx.doi.org/10.1002/inr.490250304

     Shelton. KR: Egle. PM. (1982). The proteins of lead-induced intranuclear inclusion bodies. J Biol Chem
        257: 11802-11807.

     Shih. RA; Glass. TA; Bandeen-Roche. K; Carlson. MC: Bolla. KI; Todd. AC: Schwartz. BS. (2006).
        Environmental lead exposure and cognitive function in community-dwelling older adults. Neurology
        67:  1556-1562. http://dx.doi.org/10.1212/01.wnl.0000239836.26142.c5

     Shin. JH; Lim. KM; Noh. JY; Bae. ON: Chung. SM; Lee. MY; Chung. JH. (2007). Lead-induced
        procoagulant activation of erythrocytes through phosphatidylserine exposure may lead to thrombotic
        diseases. Chem Res Toxicol 20: 38-43. http://dx.doi.org/10.1021/tx060114+

     Shinkai. Y; Yamamoto. C: Kaii. T. (2010). Lead induces the expression of ER chaperones GRP78 and
        GRP94 in vascular endothelial cells via the JNK-AP-1 pathway.  Toxicol Sci 114: 378-386.
        http://dx.doi.org/10.1093/toxsci/kfq008

     Shinozuka. H; Ohmura. T; Katval. SL; Zedda. AI; Ledda-Columbano. GM; Columbano. A. (1996).
        Possible roles of nonparenchymal cells in hepatocyte proliferation induced by lead nitrate and by
        tumor necrosis factor  alpha. Hepatology 23: 1572-1577. http://dx.doi.org/10.1002/hep.510230638

     Siddiqui. MKJ; Jyoti; Singh. S: Mehrotra. PK; Singh. K; Sarangi. R. (2006). Comparison of some trace
        elements concentration in blood, tumor free breast and tumor tissues of women with benign and
        malignant breast lesions: An Indian study. Environ Int  32: 630-637.
        http://dx.doi.0rg/10.1016/i.envint.2006.02.002

     Silberstein. T; Saphier. O; Paz-Tal. O: Trimarchi. JR; Gonzalez. L; Keefe. PL. (2006). Lead concentrates
        in ovarian follicle compromises pregnancy. J Trace Elem Med Biol 20: 205-207.
        http://dx.doi.0rg/10.1016/i.jtemb.2006.05.001

     Silkin. YA; Silkina. EN; Sherstobitov. AO; Gusev. GP (2001). Activation of potassium channels in
        erythrocytes  of marine teleost Scorpaena porcus. Membr Cell Biol 14: 773-782.
November 2012                             5-782                   Draft - Do Not Cite or Quote

-------
     Silva. PA; Hughes. P; Williams. S: Faed. JM. (1988). Blood lead, intelligence, reading attainment, and
        behaviour in eleven year old children in Dunedin, New Zealand. J Child Psychol Psychiatry 29: 43-52.
        http://dx.doi.0rg/10.llll/i.1469-7610.1988.tb00687.x

     Silveira. EA; Lizardo. JH; Souza. LP; Stefanon. I; Vassallo. DV. (2010). Acute lead-induced
        vasoconstriction in the vascular beds of isolated perfused rat tails is endothelium-dependent. Braz J
        Med Biol Res 43: 492-499. http://dx.doi.org/10.1590/S0100-879X2010007500027

     Simoes. MR; Ribeiro Junior. RF; Vescovi. MV; de Jesus. HC: Padilha. AS: Stefanon. I; Vassallo. DV:
        Salaices. M; Fioresi. M. (2011). Acute lead exposure increases arterial pressure: role of the Renin-
        Angiotensin system. PLoS ONE 6: el8730. http://dx.doi.org/10.1371/iournal.pone.0018730

     Simons. TJB. (1986). Passive transport and binding of lead by human red blood cells. J Physiol 378:
        267-286.

     Simons. TJB. (1993a). Lead-calcium interactions in cellular lead toxicity [Review]. Neurotoxicology 14:
        77-85.

     Simons. TJB. (1993b). Lead transport and binding by human erythrocytes in vitro. Pflugers Arch 423:
        307-313. http://dx.doi.org/10.1007/BF00374410

     Simons. TJB. (1995). The affinity of human erythrocyte porphobilinogen synthase for Zn2+ and Pb2+.
        FEES J 234: 178-183. http://dx.doi.Org/10.llll/i.1432-1033.1995.178 c.x

     Simsek. N; Karadeniz. A; Kalkan. Y; Keles. ON; Unal. B. (2009). Spirulina platensis feeding inhibited
        the anemia- and leucopenia-induced lead and cadmium in rats. J Hazard Mater 164: 1304-1309.
        http://dx.doi.0rg/10.1016/i.jhazmat.2008.09.041

     Singh. A; Cullen. C: Dykeman. A; Rice. D; Foster. W. (1993a).  Chronic lead exposure induces
        ultrastructural alterations in the monkey testis. J Submicrosc Cytol Pathol 25: 479-486.

     Singh. B; Chandran. V; Bandhu. HK; Mittal. BR; Bhattacharya. A; Jindal. SK; Varma. S. (2000). Impact
        of lead exposure on pituitary-thyroid axis in humans. Biometals  13: 187-192.
        http://dx.doi.Org/10.1023/A:1009201426184

     Singh. J; Pritchard. DE; Carlisle. PL; Mclean. JA; Montaser. A; Orenstein. JM; Patierno. SR. (1999).
        Internalization of carcinogenic lead chromate particles by cultured normal human lung epithelial cells:
        Formation of intracellular lead-inclusion bodies and induction of apoptosis. Toxicol Appl Pharmacol
        161: 240-248. http://dx.doi.org/10.1006/taap.1999.8816

     Singh. US: Saxena. DK; Singh. C: Murthy. RC: Chandra. SV. (1991). Lead-induced fetal nephrotoxicity
        in iron-deficient rats. Reprod Toxicol 5:211-217. http://dx.doi.org/10.1016/0890-6238(91)90053-I

     Sivaprasad. R; Nagarai. M; Varalakshmi. P. (2003). Combined efficacies of lipoic acid and meso-2,3-
        dimercaptosuccinic acid on lead-induced erythrocyte membrane lipid peroxidation and antioxidant
        status in rats. Hum Exp Toxicol 22: 183-192. http://dx.doi.org/10.1191/0960327103ht335oa

     Skoczynska. A; Stojek. E. (2005). The impact of subchronic lead poisoning on the vascular effect of nitric
        oxide in rats. Environ Toxicol Pharmacol 19: 99-106. http://dx.doi.Org/10.1016/j.etap.2004.05.004

     Slivkova. J: Popelkova. M; Massanyi. P; Toporcerova. S: Stawarz. R; Formicki. G: Lukac. N; Putaa. A;
        Guzik. M. (2009). Concentration of trace elements in human semen and relation to spermatozoa
        quality. J Environ Sci Health A Tox Hazard Subst Environ Eng 44: 370-375.
        http://dx.doi.org/10.1080/10934520802659729

     Smejkalova. J.  (1990). The chromosomal aberrations investigation in children permanently living in the
        lead polluted area. Sb Ved Pr Fak Karlovy Univ Hradci Kralovea 33: 539-564.

     Smith. CM; Hu. H; Wang. X; Kelsev. KT. (1995a). ALA-D genotype is not associated with HT or HB
        levels among workers exposed to low levels of lead. Med Lav 86: 229-235.

     Smith. CM; Wang. X; Hu. H; Kelsev. KT. (1995b). A polymorphism in the delta-aminolevulinic acid
        dehydratase gene may modify the pharmacokinetics and toxicity of lead. Environ Health Perspect 103:
        248-253. http://dx.doi.org/10.1289/ehp.95103248
November 2012                              5-783                   Draft - Do Not Cite or Quote

-------
     Smith. PR; Kahng. MW: Quintanilla-Vega. B; Fowler. BA. (1998). High-affinity renal lead-binding
        proteins in environmentally-exposed humans. Chem Biol Interact 115: 39-52.
        http://dx.doi.org/10.1016/S0009-2797(98)00060-X

     Snyder. JE; Filipov. NM; Parsons. PJ: Lawrence. DA. (2000). The efficiency of maternal transfer of lead
        and its influence on plasma IgE and splenic cellularity of mice. Toxicol Sci 57: 87-94.
        http://dx.doi.0rg/10.1093/toxsci/57.l.87

     Snyder. RD; Lachmann. PJ. (1989). Thiol involvement in the inhibition of DNA repair by metals in
        mammalian cells. Mol Toxicol 2:  117-128.

     Sobekova. A; Holovska. K; Lenartova. V: Legatfa. J: Javorsky. P. (2009). The alteration of glutathione
        peroxidase activity in rat organs after lead exposure. Acta Physiol Hung 96: 37-44.
        http://dx.doi.Org/10.1556/APhysiol.96.2009.l.4

     Sokol. RZ. (1987). Hormonal effects of lead acetate in the male rat: mechanism of action. Biol Reprod
        37: 1135-1138. http://dx.doi.0rg/10.1095/biolreprod37.5.1135

     Sokol. RZ. (1989). Reversibility of the toxic effect of lead on the male reproductive axis. Reprod Toxicol
        3:  175-180.

     Sokol. RZ; Berman. N. (1991). The effect of age of exposure on lead-induced testicular toxicity.
        Toxicology 69: 269-278. http://dx.doi.org/10.1016/0300-483X(91)90186-5

     Sokol. RZ; Madding. CE; Swerdloff. RS. (1985). Lead toxicity and the hypothalamic-pituitary-testicular
        axis.  Biol Reprod 33: 722-728. http://dx.doi.org/10.1095/biolreprod33.3.722

     Sokol. RZ; Okuda. H; Nagler. HM; Berman. N. (1994). Lead exposure in vivo alters the fertility potential
        of sperm in vitro. Toxicol Appl Pharmacol 124: 310-316.

     Solliwav. BM; Schaffer. A; Pratt. H; Yannai. S. (1996). Effects of exposure to lead on selected
        biochemical and haematological variables. Basic  Clin Pharmacol Toxicol 78: 18-22.

     Solon. O; Riddell. TJ; Quimbo. SA; Butrick. E; Aylward. GP; Lou Bacate. M; Peabodv. JW. (2008).
        Associations between cognitive function, blood lead concentration, and nutrition among children in
        the central Philippines. J Pediatr 152: 237-243. http://dx.doi.Org/10.1016/i.jpeds.2007.09.008

     Songdej. N; Winters. PC: McCabe. MJ. Jr; van Wijngaarden. E. (2010). A population-based assessment of
        blood lead levels in relation to inflammation. Environ Res 110: 272-277.
        http://dx.doi.0rg/10.1016/i.envres.2009.12.008

     Spector. JT; Navas-Acien. A; Fadrowski. J: Guallar. E; Jaar. B; Weaver. VM. (2011). Associations of
        blood lead with estimated glomerular filtration rate using MDRD, CKD-EPI  and serum cystatin C-
        based equations. Nephrol Dial Transplant 26: 2786-2792. http://dx.doi.org/10.1093/ndt/gfq773

     SRC (Syracuse Research Corporation).  (2002). Baseline human health risk assessment: Eureka Mills -
        Eureka, Utah. Denver, CO: U.S. Environmental Protection Agency.
        http://www.epa.gov/region8/r8risk/pdf/hhra_eureka.pdf

     Srivastava. D; Hurwitz. RL; Fox. DA. (1995). Lead- and  calcium-mediated inhibition of bovine rod
        cGMPphosphodiesterase: Interactions with magnesium. Toxicol Appl Pharmacol 134: 43-52.
        http://dx.doi.org/10.1006/taap.1995.1167

     Srivastava. V: Dearth. RK;  Hinev. JK; Ramirez. LM; Bratton. GR; Dees. W. (2004). The effects of
        low-level Pb on steroidogenic acute regulatory protein (StAR) in the prepubertal rat ovary. Toxicol Sci
        77: 35-40. http://dx.doi.org/10.1093/toxsci/kfg249

     Stacchiotti. A; Morandini. F; Bettoni. F; Schena. I; Lavazza. A; Grigolato. PG: Apostoli. P; Rezzani. R;
        Aleo. MF. (2009).  Stress proteins and oxidative, damage in a renal derived cell line exposed to
        inorganic mercury and lead. Toxicology 264: 215-224. http://dx.doi.Org/10.1016/j.tox.2009.08.014
November 2012                              5-784                  Draft - Do Not Cite or Quote

-------
     Stackpole. MM; Wise. SS: Goodale. BC: Duzevik. EG: Munroe. RC: Thompson. WD; Thacker. J;
        Thompson. LH; Hinz. JM; Wise. JP. (2007). Homologous recombination repair protects against
        paniculate chromate-induced chromosome instability in Chinese hamster cells. Mutat Res-Fundam
        Mol Mech Mutagen 625: 145-154. http://dx.doi.Org/10.1016/i.mrfmmm.2007.06.003

     Staessen. JA; Lauwerys. RR; Buchet. JP: Bulpitt. CJ: Rondia. D; Van Renterghem. Y; Amery. A. (1992).
        Impairment of renal function with increasing blood lead concentrations in the general population. N
        EnglJ Med 327: 151-156. http://dx.doi.org/10.1056/NEJM199207163270303

     Staessen. JA; Nawrot. T; Den Hond. E; Thijs. L; Fagard. R; Hoppenbrouwers. K; Koppen. G: Nelen. V;
        Schoeters. G: Vanderschueren. D; Van Hecke. E; Verschaeve. L; Vlietinck. R; Roels. HA. (2001).
        Renal function, cytogenetic measurements, and sexual developments in adolescents in relation to
        environmental pollutants: Afeasibility study of biomarkers. Lancet 357: 1660-1669.
        http://dx.doi.org/10.1016/S0140-6736(00)04822-4

     Stangle. DE; Smith. PR; Beaudin. SA; Strawderman. MS: Levitsky. DA; Strupp. BJ. (2007). Succimer
        chelation improves learning, attention, and arousal regulation in lead-exposed rats but produces lasting
        cognitive impairment in the absence of lead exposure. Environ Health Perspect 115: 201-209.
        http://dx.doi.org/10.1289/ehp.9263

     Stayner. L; Steenland. K; Dosemeci. M; Hertz-Picciotto. I. (2003). Attenuation of exposure-response
        curves in occupational cohort studies at high exposure levels. Scand J Work Environ Health 29:
        317-324.

     Steenland. NK; Thun. MJ; Ferguson. CW: Port. FK. (1990). Occupational and other exposures associated
        with male end-stage renal disease: A case/control study. Am J Public Health 80: 153-157.

     Steinthorsdottir. SD; Eliasdottir. SB; Indridason. OS: Agustsdottir. IM; Palsson. R; Edvardsson. VO.
        (2011). Prevalence of hypertension  in 9- to 10-year-old Icelandic school children. J Clin Hypertens
        (Greenwich) 13: 774-779. http://dx.doi.Org/10.llll/j.1751-7176.2011.00496.x

     Sterling. G: O'Neill. K; McCaffertv. M; O'Neill. J. (1982). Effect of chronic lead ingestion by rats on
        glucose metabolism and acetylcholine synthesis in cerebral cortex slices. JNeurochem  39: 592-596.
        http://dx.doi.org/10.1111/i.l471-4159.1982.tb03989.x

     Stevens. LA; Coresh. J: Schmid. CH; Feldman. HI; Froissart. M; Kusek. J: Rossert. J: Van Lente. F;
        Bruce. RD; Zhang. YP; Greene. T; Levey. AS.  (2008). Estimating GFR using serum cystatin C alone
        and in combination with serum creatinine: A pooled analysis of 3,418 individuals with CKD. Am J
        Kidney Dis 51: 395-406. http://dx.doi.Org/10.1053/i.ajkd.2007.ll.018

     Stewart. PW: Elaine. C: Cohen. M; Burright. RG: Donovick. PJ. (1996). Acute and longer term effects of
        meso-2,3 dimercaptosuccinic acid (DMSA) on the behavior of lead-exposed and control mice. Physiol
        Behav 59: 849-855. http://dx.doi.org/10.1016/0031-9384(95)02185-X

     Stewart. WF; Schwartz. BS: Davatzikos. C: Shen. D; Liu. D; Wu. X; Todd. AC: Shi. W: Bassett. S:
        Youssem. D. (2006). Past adult lead exposure is linked to neurodegeneration measured  by brain MRI.
        Neurology 66: 1476-1484. http://dx.doi.org/10.1212/01.wnl.0000216138.69777.15

     Stewart. WF; Schwartz. BS: Simon. D; Kelsev. K; Todd. AC. (2002). ApoE genotype, past adult lead
        exposure, and neurobehavioral function. Environ Health Perspect 110: 501-505.
        http://dx.doi.org/10.1289/ehp.02110501

     Stiles. KM; Bellinger. DC. (1993). Neuropsychological correlates of low-level lead exposure in school-
        age children: a prospective study. Neurotoxicol Teratol 15: 27-35.

     Stoleski. S: Karadzinska-Bislimovska.  J: Stikova. E; Risteska-Kuc. S: Mijakoski. D;  Minov. J. (2008).
        Adverse effects in workers exposed to inorganic  lead. Arh Hig Rada Toksikol 59: 19-29.
        http://dx.doi.org/10.2478/10004-1254-59-2008-1866

     Stromberg. U; Schiitz. A; Skerfving. S. (1995). Substantial decrease of blood lead in  Swedish children,
        1978-94, associated with petrol lead. Occup Environ Med 52: 764-769.
        http://dx.doi.org/10.1136/oem.52.ll.764
November 2012                             5-785                   Draft - Do Not Cite or Quote

-------
     Struzvnska. L; Dabrowska-Bouta. B; Koza. K; Sulkowski. G. (2007). Inflammation-like glial response in
        lead-exposed immature rat brain. Toxicol Sci 95: 156-162. http://dx.doi.org/10.1093/toxsci/kfl 134

     Struzynska. L; Dabrowska-Bouta. B; Rafalowska. U. (1997a). Acute lead toxicity and energy metabolism
        in rat brain synaptosomes. Acta Neurobiol Exp (Wars) 57: 275-281.

     Struzynska. L; Walski. M; Gadamski. R; Dabrowska-Bouta. B; Rafalowska. U. (1997b). Lead-induced
        abnormalities in blood-brain barrier permeability in experimental chronic toxicity. Mol Chem
        Neuropathol 31: 207-224. http://dx.doi.org/10.1007/BF02815125

     Su. Y; Ding. Y; Jiang. M; Hu. X; Zhang. Z. (2007). Protein 4.2 Komatsu (D175Y) associated with the
        lack of interaction with ankyrin in human red blood cells. Blood Cells Mol Dis 38: 221-228.
        http://dx.doi.0rg/10.1016/i.bcmd.2006.ll.004

     Sun. H; Wang. HL; Wang. S. (2007). D-serine relieves chronic lead exposure-impaired long-term
        potentiation in the CA1 region of the rat hippocampus in vitro. Neurosci Lett 417: 118-122.
        http://dx.doi.0rg/10.1016/i.neulet.2007.01.085

     Sun. L; Hu. J; Zhao. Z; Li. L; Cheng.  H. (2003). Influence of exposure to environmental lead on serum
        immunoglobulin in preschool children. Environ Res 92: 124-128.
        http://dx.doi.org/10.1016/80013-9351(02)00090-7

     Sun. LR; Suszkiw. JB. (1995). Extracellular inhibition and intracellular enhancement of Ca2+ currents by
        Pb2+ in bovine adrenal chromaffin cells. J Neurophysiol 74: 574-581.

     Sun. Y; Sun. D; Zhou. Z; Zhu. G: Zhang. H; Chang. X; Lei. L; Jin. T. (2008a). Osteoporosis in a Chinese
        population due to occupational exposure to lead. Am J Ind Med 51: 436-442.
        http://dx.doi.org/10.1002/aiim.20567

     Sun. Y; Sun. DH; Zhou. ZJ; Zhu. GY; Lei. LJ; Zhang. HY; Chang. XL; Jin. TY. (2008b). Estimation of
        benchmark dose for bone damage  and renal dysfunction in a Chinese male population occupationally
        exposed to lead. Ann Occup  Hyg 52: 527-533.  http://dx.doi.org/10.1093/annhyg/men031

     Sundstrom.  R; Muntzing. K; Kalimo.  H; Sourander. P. (1985). Changes in the integrity of the blood-brain
        barrier in suckling rats with low dose lead encephalopathy. Acta Neuropathol 68:  1-9.
        http://dx.doi.org/10.1007/BF00688948

     Surkan. PJ:  Schnaas. L; Wright. RJ: Tellez-Rojo. MM; Lamadrid-Figueroa. H; Hu. H; Hernandez-Avila.
        M; Bellinger. DC: Schwartz. J: Perroni. E; Wright. RO. (2008). Maternal self-esteem, exposure to
        lead, and child neurodevelopment. Neurotoxicology 29: 278-285.
        http://dx.doi.0rg/10.1016/i.neuro.2007.ll.006

     Surkan. PJ: Zhang. A; Trachtenberg. F; Daniel. DB; McKinlav. S: Bellinger. DC. (2007).
        Neuropsychological function in children with blood lead levels <10 ^g/dL. Neurotoxicology 28:
        1170-1177. http://dx.doi.0rg/10.1016/i.neuro.2007.07.007

     Suszkiw. J: Toth. G: Murawsky. M; Cooper. GP (1984). Effects of Pb2+ and Cd2+ on acetylcholine
        release and Ca2+ movements in synaptosomes and subcellular fractions from rat brain and Torpedo
        electric organ. Brain Res  323: 31-46. http://dx.doi.org/10.1016/0006-8993(84)90262-2

     Suszkiw. JB. (2004). Presynaptic disruption of transmitter release by lead. Neurotoxicology 25: 599-604.
        http://dx.doi.0rg/10.1016/i.neuro.2003.09.009

     Suzuki. T; Morimura. S: Diccianni. MB; Yamada. R: Hochi. SI; Hirabavashi. M; Yuki. A; NomuraK;
        Kitagawa. T; Imagawa. M; Muramatsu. M.  (1996). Activation of glutathione transferase P gene by
        lead requires glutathione transferase P enhancer I. J Biol Chem 271: 1626-1632.
        http://dx.doi.0rg/10.1074/ibc.271.3.1626

     Swanson. J: Baler. RD; Volkow. ND. (2011). Understanding the effects of stimulant medications on
        cognition in individuals with attention-deficit hyperactivity disorder:  a decade of progress [Review].
        Neuropsychopharmacology 36: 207-226. http://dx.doi.org/10.1038/npp.2010.160
November 2012                             5-786                   Draft - Do Not Cite or Quote

-------
     Swarup. D: Naresh. R; Varshnev. VP; Balagangatharathilagar. M; Kumar. P; Nandi. D; Patra. RC. (2007).
        Changes in plasma hormones profile and liver function in cows naturally exposed to lead and
        cadmium around different industrial areas. Res Vet Sci 82: 16-21.
        http://dx.doi.0rg/10.1016/i.rvsc.2006.05.002

     Taketani. S: Tanaka. A; Tokunaga. R. (1985). Reconstitution of heme-synthesizing activity from ferric ion
        and porphyrins, and the effect of lead on the activity. Arch Biochem Biophys 242: 291-296.
        http://dx.doi.org/10.1016/0003-9861(85)90504-l

     Tanner. CM; Chen. B; Wang. W: Peng. M; Liu. Z; Liang. X; Kao. LC: Gillev. DW: Goetz. CG:
        Schoenberg. BS. (1989). Environmental factors and Parkinson's disease: A case-control study in
        China. Neurology 39: 660-664.

     Tapisso. JT; Marques. CC: Mathias. ML; Ramalhinho. MG. (2009).  Induction of micronuclei and sister
        chromatid exchange in bone-marrow cells and abnormalities in sperm of Algerian mice (Mus spretus)
        exposed to cadmium, lead and zinc. Mutat Res Genet Toxicol Environ Mutagen 678: 59-64.
        http://dx.doi.0rg/10.1016/i.mrgentox.2009.07.001

     Tarantini. L;  Bonzini.  M; Apostoli. P; Pegoraro. V; Bollati. V: Marinelli. B; Cantone. L; Rizzo. G: Hou.
        L; Schwartz. J; Bertazzi. PA; Baccarelli. A. (2009). Effects of paniculate matter on genomic DNA
        methylation content and iNOS promoter methylation. Environ Health Perspect 117: 217-222.
        http://dx.doi.org/10.1289/ehp.11898

     Task Force of the European Society of Cardiology and the North American Society of Pacing and
        Electrophysiologv. (1996). Heart rate variability: Standards of measurement, physiological
        interpretation and clinical use. Circulation 93: 1043-1065.

     Tavakoli-Nezhad. M; Barron. AJ; Pitts. DK.  (2001). Postnatal inorganic lead exposure decreases the
        number of spontaneously active midbrain dopamine neurons in the rat. Neurotoxicology 22: 259-269.
        http://dx.doi.org/10.1016/S0161-813X(01)00010-9

     te Velde. SJ:  Ferreira.  I; Twisk. JW: Stehouwer. CD; van Mechelen. W: Kemper. HC: Study. AGaHL.
        (2004). Birthweight and arterial stiffness and blood pressure in adulthood—results from the
        Amsterdam Growth and Health Longitudinal Study. Int J Epidemiol 33: 154-161.
        http://dx.doi.org/10.1093/iie/dvh011

     Technische Universitat Braunschweig. (2011). BRENDA:  Comprehensive enzyme information system.
        Available online at http://www.brenda-enzymes.org (accessed January 10, 2011).

     Teijon. C: Blanco. MD; Romero. CS: Beneit. JV: Villarino. AL; Guerrero. S: Olmo. R. (2010). Study of
        response of thymic and submaxillary lymph node lymphocytes to administration of lead by different
        routes. Biol Trace Elem Res 135: 74-85. http://dx.doi.org/10.1007/sl2011-009-8495-6

     Teijon. C: Olmo. R; Blanco. D: Romero. A; Teijon. JM.  (2006). Low doses of lead: Effects on
        reproduction and development in rats. Biol Trace Elem Res 111:  151-165.
        http://dx.doi.Org/10.1385/BTER:lll:l:151

     Telisman. S:  Colak. B; Pizent. A; Jurasovic. J: Cvitkovic. P. (2007). Reproductive toxicity of low-level
        lead exposure in men. Environ Res  105: 256-266. http://dx.doi.Org/10.1016/i.envres.2007.05.011

     Tellez-Rojo. M. (2008).  Email  from Martha Tellez Rojo, Direction of Evaluation of Programs and
        Quantitative Methods, Center of Investigation in Evaluation and  Polls, National institute of Public
        Health to Jee-Young  Kim, USEPA, Subject - Questions on the Tellez-Rojo et al. (2006) paper.  (EPA-
        HQ-OAR-2006-0735-5123). U.S. Environmental Protection Agency.
        http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OAR-2006-0735-5123

     Tellez-Rojo. MM; Bellinger. DC: Arroyo-Quiroz. C: Lamadrid-Figueroa. H; Mercado-Garcia. A;
        Schnaas-Arrieta. L; Wright. RO: Hernandez-Avila. M; Hu. H. (2006). Longitudinal associations
        between blood lead concentrations lower than 10 microg/dL and  neurobehavioral development in
        environmentally exposed children in Mexico City. Pediatrics 118: e323-e330.
        http://dx.doi.org/10.1542/peds.2005-3123
November 2012                             5-787                  Draft - Do Not Cite or Quote

-------
     Tepper. LB; Levin. LS. (1975). A survey of air and population lead levels in selected American
        communities. In TB Griffin; JH Knelson (Eds.), Lead (pp. 152-196). Stuttgart, Germany: Georg
        Thieme Publishers.

     Terziivanov. D; Filipova. M; Janku. I; Balik. J; Filip. V; Stika. L. (1982). Changes in electroretinogram
        and serum potassium during L-DOPA treatment in parkinsonism. Arch Psychiatr Z Neurol 232:
        507-513. http://dx.doi.org/10.1007/BF00344065

     Teschke. K; Olshan. AF; Daniels. JL; De Roos. AJ; Parks. CG: Schulz. M; Vaughan. TL. (2002).
        Occupational exposure assessment in case-control studies: Opportunities for improvement [Review].
        Occup Environ Med 59: 575-594. http://dx.doi.Org/10.1136/oem.59.9.575

     Tessitore. L; Sesca. E; Pani. P; Dianzani. MU. (1995). Sexual dimorphism in the regulation of cell
        turnover during liver hyperplasia. Chem Biol Interact 97: 1-10.
        http://dx.doi.org/10.1016/0009-2797(94)03602-l

     Thind. IS: Khan. MY. (1978). Potentiation of the neurovirulence of langat virus infection by lead
        intoxication in mice. Exp Mol Pathol 29: 342-347. http://dx.doi.org/10.1016/0014-4800(78)90076-X

     Tian. L; Lawrence. DA. (1995). Lead inhibits nitric oxide production in vitro by murine splenic
        macrophages. Toxicol Appl Pharmacol 132:  156-163. http://dx.doi.org/10.1006/taap.1995.1096

     Tian. L; Lawrence. DA. (1996). Metal-induced modulation of nitric oxide production in vitro by murine
        macrophages: Lead, nickel, and cobalt utilize different mechanisms. Toxicol Appl Pharmacol 141:
        540-547. http://dx.doi.org/10.1006/taap.1996.0320

     Tian. Y; Green. PG: Stamova. B; Hertz-Picciotto. I; Pessah. IN; Hansen. R; Yang. X; Gregg. JP;
        Ashwood. P; Jickling. G: Van de Water. J; Sharp.  FR. (2011). Correlations of gene expression with
        blood lead levels in children with autism compared to typically developing controls. Neurotox Res 19:
        1-13. http://dx.doi.org/10.1007/sl2640-009-9126-x

     Tiffany-Castiglioni. E; Sierra. EM; Wu. JN; Rowles.  TK. (1989). Lead toxicity in neuroglia [Review].
        Neurotoxicology 10: 417-443.

     Tokar. EJ; Diwan. BA; Waalkes. MP (2010). Early life inorganic lead exposure induces testicular
        teratoma and renal and urinary bladder preneoplasia in adult metallothionein-knockout mice but not in
        wild type mice. Toxicology 276: 5-10. http://dx.doi.Org/10.1016/i.tox.2010.06.006

     Tomokuni. K; Ichiba.  M. (1990). Effect of lead on the activity of erythrocyte porphobilinogen deaminase
        in vivo and in vitro. Toxicol Lett 50: 137-142. http://dx.doi.org/10.1016/0378-4274(90)90003-5

     Tomoum. HY; Mostafa. GA; Ismail. NA; Ahmed. SM.  (2010). Lead exposure and its association with
        pubertal development in school-age Egyptian children: Pilot study. Pediatrics International 52: 89-93.
        http://dx.doi.Org/10.llll/i.1442-200X.2009.02893.x

     Tomsig. JL; Suszkiw.  JB. (1993). Intracellular mechanism of Pb(2+)-induced norepinephrine release from
        bovine chromaffin cells. Am J Physiol Cell Physiol 265: C1630-C1636.

     Tomsig. JL; Suszkiw.  JB. (1995). Multisite interactions between Pb2+ and protein kinase C and its role in
        norepinephrine release from bovine adrenal chromaffin cells. JNeurochem 64: 2667-2673.
        http://dx.doi.0rg/10.1046/i.1471-4159.1995.64062667.x

     Tonelli. M; Wiebe. N; Hemmelgarn. B; Klarenbach. S:  Field. C: Manns. B; Thadhani. R; Gill. J. (2009).
        Trace elements in  hemodialysis patients: A systematic review and meta-analysis  [Review]. BMC Med
        7: 25. http://dx.doi.org/10.1186/1741-7015-7-25

     Tong. IS: Lu. Y. (2001). Identification of confounders in the assessment of the relationship between lead
        exposure and child development. Ann Epidemiol  11: 38-45.
        http://dx.doi.org/10.1016/S1047-2797(00)00176-9

     Tong. S: Baghurst. P:  McMichael. A; Sawyer. M; Mudge. J. (1996). Lifetime exposure to environmental
        lead and children's intelligence at 11-13 years: The Port Pirie cohort study. Br Med J 312: 1569-1575.
        http://dx.doi.org/10.1136/bmi.312.7046.1569
November 2012                              5-788                  Draft - Do Not Cite or Quote

-------
     Tong. S: Baghurst. PA; Sawyer. MG: Burns. J: McMichael. AJ. (1998). Declining blood lead levels and
        changes in cognitive function during childhood: the Port Pirie cohort study. JAMA 280: 1915-1919.
        http://dx.doi.org/10.1001/iama.280.22.1915

     Tong. S: McMichael. AJ: Baghurst. PA. (2000). Interactions between environmental lead exposure and
        sociodemographic factors on cognitive development. Arch Environ Health 55: 330-335.
        http://dx.doi.org/10.1080/00039890009604025

     Trope. I; Lopez-Villegas. D; Cecil. KM; Lenkinski. RE. (2001). Exposure to lead appears to selectively
        alter metabolism of cortical gray matter. Pediatrics 107: 1437-1443.
        http://dx.doi.org/10.1542/peds. 107.6.1437

     Tsaih. SW: Korrick. S: Schwartz. J; Amarasiriwardena. C: Aro. A; Sparrow. D; Hu. H. (2004). Lead,
        diabetes, hypertension, and renal function: The Normative Aging Study. Environ Health Perspect 112:
        1178-1182. http://dx.doi.org/10.1289/ehp.7024

     Tsao. DA; Yu. HS: Cheng. JT; Ho. CK; Chang. HR. (2000). The change of beta-adrenergic system in
        lead-induced hypertension. ToxicolAppl Pharmacol 164: 127-133.
        http://dx.doi.org/10.1006/taap.1999.8871

     Tsuang. M. (2000). Schizophrenia: Genes and environment [Review]. Biol Psychiatry 47: 210-220.
        http://dx.doi.org/10.1016/S0006-3223(99)00289-9

     Tuppurainen. M; Wagar. G: Kurppa. K; Sakari. W: Wambugu. A; Froseth. B; Alho. J: Nykyri. E. (1988).
        Thyroid function as assessed by routine laboratory tests of workers with long-term lead exposure.
        Scand J Work Environ Health 14: 175-180.

     Turgut. S: Polat. A; Inan. M; Turgut. G: Emmungil. G: Bican. M; Karakus. TY; Gene. O. (2007).
        Interaction between anemia and blood levels of iron, zinc, copper,  cadmium and lead in children.
        Indian J Pediatr 74: 827-830. http://dx.doi.org/10.1007/sl2098-007-0147-2

     Turi. GJ: Ellis. R; Wattie. JN; Labiris. NR; Inman. MD. (2011). The effects of inhaled house dust mite on
        airway barrier function and sensitivity to inhaled methacholine in mice. Am J Physiol Lung Cell Mol
        Physiol 300:  L185-L190. http://dx.doi.org/10.1152/ajplung.00271.2010

     Turkez. H; Geyikoglu. F; Tatar. A; Keles. MS: Kaplan. I. (2011). The effects of some  boron compounds
        against heavy metal toxicity in human blood. Exp Toxicol Pathol 64: 93-101.
        http://dx.doi.0rg/10.1016/i.etp.2010.06.011

     U.S. EPA (U.S. Environmental Protection Agency). (1978). National ambient air quality standard for
        lead: final rules and proposed rulemaking. Fed Reg 43: 46246-46263.

     U.S. EPA (U.S. Environmental Protection Agency). (1986a). Air quality criteria for lead [EPA Report].
        (EPA/600/8-83/028aF-dF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1986b). Air quality criteria for lead: Volume I of IV
        [EPA Report]. (EPA-600/8-83/028aF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1986c). Lead effects on cardiovascular function,
        early development, and stature: An addendum to U.S. EPAAir Quality Criteria for Lead (1986) [EPA
        Report]. (EPA-600/8-83/028aF). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (1990a). Air quality criteria for lead: Supplement to
        the 1986 addendum [EPA Report]. (EPA/600/8-89/049F). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2006b). Air quality criteria for lead: Volume I of II
        [EPA Report]. (EPA/600/R-05/144aF). Research Triangle Park, NC.
        http ://cfpub. epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823

     U.S. EPA (U.S. Environmental Protection Agency). (2006c). Air quality criteria for lead: Volume II of II
        [EPA Report]. (EPA/600/R-05/144bF). Research Triangle Park, NC.
        http ://cfpub .epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823
November 2012                             5-789                  Draft - Do Not Cite or Quote

-------
     U.S. EPA (U.S. Environmental Protection Agency). (2006g). Table 5-7, studies reporting lead-induced
        shifts in Thl versus Th2 cytokines [EPA Report]. In Air quality criteria for lead: Volume I of II (pp.
        5-225  - 225-226). (EPA/600/R-5/144aF). Research Triangle Park, NC.

     U.S. EPA (U.S. Environmental Protection Agency). (2006h). Table AX5.9.6., pattern of lead-induced
        macrophage immunotoxicity [EPA Report]. In Air quality criteria for lead: Volume II of II (pp.
        AX5-155). (EPA/600/R-05/144bF). Research Triangle Park, NC.

     U.S. Renal Data System. (2009). USRDS 2009 annual data report: Atlas of chronic kidney disease and
        end-stage renal disease in the United States. Bethesda, MD: National Institutes of Health, National
        Institute of Diabetes and Digestive and Kidney Diseases,  http://www.usrds.org/adr 2009.htm

     Ueda. D; Kishimoto. T; Dekio. S: Tada. M. (1997). Inhibitory effect of lead on tube formation by cultured
        human vascular endothelial cells. Hum Cell 10: 283-291.

     Ujihara. H; Sasa. M; Ban. T. (1995). Selective blockade of P-type calcium channels by lead in cultured
        hippocampal neurons. Jpn J Pharmacol 67: 267-269. http://dx.doi.org/10.1254/ijp.67.267

     Ukaejiofo. EO; Thomas. N; Ike. SO. (2009). Haematological assessment of occupational exposure to lead
        handlers in Enugu urban, Enugu State, Nigeria. Niger J Clin Pract 12: 58-64.

     Undeger. U; Basaran. N; Canpinar. H; Kansu. E. (1996). Immune alterations in lead-exposed workers.
        Toxicology 109: 167-172. http://dx.doi.org/10.1016/0300-483X(96)03333-l

     Upadhyav. AK; Mathur. R; Bhadauria. M; Nirala. SK. (2009). Therapeutic influence of zinc and ascorbic
        acid against lead induced biochemical alterations. Therapie 64: 383-388.
        http://dx.doi.org/10.2515/therapie/2009055

     Urch. B; Brook. JR; Wasserstein. D; Brook. RD; Raiagopalan. S: Corey. P; Silverman. F. (2004).  Relative
        contributions of PM2.5 chemical constituents to acute arterial vasoconstriction in humans. Inhal
        Toxicol 16: 345-352. http://dx.doi.org/10.1080/08958370490439489

     Ustundag. A; Duydu. Y. (2007). The influence of melatonin and N-acetylcysteine in delta-aminolevulinic
        acid and lead induced genotoxicity in lymphocytes in vitro. Biol Trace Elem Res 117: 53-64.
        http://dx.doi.org/10.1007/BF02698083

     Uzbekov. MG: Bubnova. NI; Kulikova. GV (2007). Effect of prenatal lead exposure on superoxide
        dismutase activity in the brain and liver of rat fetuses. Bull Exp Biol Med 144: 783-785.
        http://dx.doi.org/10.1007/sl0517-007-0431-l

     Uzu. G: Sauvain. JJ; Baeza-Squiban. A; Riediker. M; Hohl. MSS: Val. S: Tack. K; Denys. S: Pradere. P;
        Dumat. C. (2011). In vitro assessment of the pulmonary toxicity and gastric availability of lead-rich
        particles from a lead recycling plant.  Environ Sci Technol 45: 7888-7895.
        http://dx.doi.org/10.1021/es200374c

     Vaglenov. A; Carbonell. E; Marcos. R. (1998). Biomonitoring of workers exposed to lead genotoxic
        effects, its modulation by polyvitamin treatment and evaluation of the induced radioresistance. Mutat
        Res Genet Toxicol Environ Mutagen 418: 79-92. http://dx.doi.org/10.1016/S1383-5718(98)00111-9

     Vaglenov. A; Creus. A; Laltchev. S: Petkova. V: Pavlova. S: Marcos. R. (2001). Occupational exposure to
        lead and induction of genetic damage. Environ Health Perspect 109: 295-298.

     Valencia.  I; Castillo. EE; Chamorro. G: Bobadilla. RA; Castillo. C. (2001). Lead induces endothelium-
        and Ca2+-independent contraction in rat aortic rings. Pharmacol Toxicol  89: 177-182.
        http://dx.doi.0rg/10.llll/i.0901-9928.2001.890406.x

     Valentino. M; Governa. M; Marchiseppe. I; Visona. I. (1991). Effects of lead on polymorphonuclear
        leukocyte (PMN) functions in occupationally exposed workers. Arch Toxicol 65: 685-688.
        http://dx.doi.org/10.1007/BF02098038

     Valentino. M; Rapisarda. V: Santarelli. L; Bracci. M; Scorcelletti. M; Di Lorenzo. L; Cassano. F; Soleo.
        L_. (2007). Effect of lead on the levels of some immunoregulatory cytokines in occupationally exposed
        workers. Hum Exp Toxicol 26: 551-556. http://dx.doi.org/10.1177/0960327107073817
November 2012                              5-790                  Draft - Do Not Cite or Quote

-------
     Valverde. M; Fortoul. TI; Diaz-Barriga. F; Maiia. J: Del Castillo. ER. (2002). Genotoxicity induced in
        CD-I mice by inhaled lead: Differential organ response. Mutagenesis 17: 55-61.
        http://dx.doi.0rg/10.1093/mutage/17.l.55

     Van de Vvver. FL; D'Haese. PC: Visser. WJ: Elseviers. MM; Knippenberg. LJ; Lamberts. LV; Wedeen.
        RP; de Broe. ME. (1988). Bone lead in dialysis patients. Kidney Int 33: 601-607.

     Van Esch. GJ: Kroes. R. (1969). The induction of renal tumours by feeding basic lead acetate to mice and
        hamsters. Br J Cancer 23: 765-771. http://dx.doi.org/10.1038/bjc. 1969.95

     Van Larebeke. N; Koppen. G: Nelen. V; Schoeters. G: Van Loon. H; Albering. H; Riga. L; Vlietinck. R;
        Kleinjans. J. (2004). Differences in HPRT mutant frequency among middle-aged Flemish women in
        association with area of residence and blood lead levels. Biomarkers 9: 71-84.
        http://dx.doi.org/10.1080/13547500310001652160

     van Wijngaarden. E; Campbell. JR; Cory-Slechta. DA. (2009). Bone lead levels are associated with
        measures of memory impairment in older adults. Neurotoxicology 30: 572-580.
        http://dx.doi.0rg/10.1016/i.neuro.2009.05.007

     van Wijngaarden. E; Dosemeci. M. (2006). Brain cancer mortality and potential occupational exposure to
        lead: Findings from the National Longitudinal Mortality Study, 1979-1989. Int J Cancer 119:
        1136-1144. http://dx.doi.org/10.1002/ijc.21947

     Vander. AJ. (1988). Chronic effects of lead on the renin-angiotensin system [Review]. Environ Health
        Perspect 78: 77-83. http://dx.doi.org/10.1289/ehp.887877

     Vargas-Robles. H; Romo. E; Sanchez-Mendoza. A; Rios. AP;  Soto. V: Avila-Casado. C: Medina. A;
        Escalante. B. (2007). Lead exposure effect on angiotensin  II renal vasoconstriction. Hum Exp Toxicol
        26: 499-507. http://dx.doi.org/10.1177/0960327106077597

     Vassallo. DV: Lebarch. EC: Moreira. CM; Wiggers. GA; Stefanon. I. (2008). Lead reduces tension
        development and the myosin ATPase activity of the rat right ventricular myocardium. Braz J Med Biol
        Res 41: 789-795. http://dx.doi.org/10.1590/S0100-879X2008000900008

     Vaziri. ND. (2008a). Causal link between oxidative stress, inflammation, and hypertension. Iran J Kidney
        Dis2: 1-10.

     Vaziri. ND. (2008b). Mechanisms of lead-induced hypertension and cardiovascular disease [Review]. Am
        J Physiol Heart Circ Physiol 295: H454-H465. http://dx.doi.org/10.1152/ajpheart.00158.2008

     Vaziri. ND; Ding. Y. (2001). Effect of lead on nitric oxide synthase expression in coronary endothelial
        cells: Role of superoxide. Hypertension 37: 223-226. http://dx.doi.Org/10.1161/01.HYP.37.2.223

     Vaziri. ND; Ding. Y; Ni. Z. (1999a). Nitric oxide synthase expression in the course of lead-induced
        hypertension. Hypertension 34: 558-562.

     Vaziri. ND; Ding. Y; Ni. Z; Barton. CH. (2005). Bradykinin down-regulates, whereas arginine analogs
        up-regulates, endothelial nitric-oxide synthase expression in coronary endothelial cells.  J Pharmacol
        ExpTher313:  121-126. http://dx.doi.org/10.1124/ipet.104.076497

     Vaziri. ND; Ding. Y; Ni. Z; Gonick. HC. (1997). Altered nitric oxide metabolism and increased oxygen
        free radical activity in lead-induced hypertension: effect of lazaroid therapy. Kidney Int 52:
        1042-1046. http://dx.doi.org/10.1038/ki.1997.426

     Vaziri. ND; Khan. M. (2007). Interplay of reactive oxygen species and nitric oxide in the pathogenesis of
        experimental lead-induced hypertension [Review]. Clin Exp Pharmacol Physiol 34: 920-925.
        http://dx.doi.0rg/10.llll/i.1440-1681.2007.04644.x

     Vaziri. ND; Liang. K; Ding. Y. (1999b). Increased nitric oxide inactivation by reactive oxygen species in
        lead-induced hypertension. Kidney Int 56: 1492-1498.
        http://dx.doi.0rg/10.1046/i.1523-1755.1999.00670.x
November 2012                              5-791                   Draft - Do Not Cite or Quote

-------
     Vaziri. ND; Lin. CY; Farmand. F; Sindhu. RK. (2003). Superoxide dismutase, catalase, glutathione
        peroxidase and NADPH oxidase in lead-induced hypertension. Kidney Int 63: 186-194.
        http://dx.doi.0rg/10.1046/i.1523-1755.2003.00711.x

     Vaziri. ND; Wang. XQ. (1999). cGMP-mediated negative-feedback regulation of endothelial nitric oxide
        synthase expression by nitric oxide. Hypertension 34: 1237-1241.

     Vaziri. ND; Wang. XQ: Oveisi. F; Rad. B. (2000). Induction of oxidative stress by glutathione depletion
        causes severe hypertension in normal rats. Hypertension 36: 142-146. http://dx.doi.org/10.1161/
        Ol.HYP.36.1.142

     Vega-Dienstmaier. JM; Salinas-Pielago. JE; Gutierrez-Campos. MDR; Mandamiento-Ayquipa. RD; Yara-
        Hokama. MDC: Ponce-Canchihuaman. J; Castro-Morales. J. (2006). Lead levels and cognitive
        abilities in Peruvian children. Rev Bras Psiquiatr 28: 33-39.
        http://dx.doi.org/10.1590/S1516-44462006000100008

     Veranth. JM: Moss. TA: Chow. JC: Labban. R: Nichols. WK: Walton. JC: Walton. JG: Yost. GS. (2006).
        Correlation of in vitro cytokine responses with the chemical composition of soil-derived paniculate
        matter. Environ Health Perspect 114: 341-349. http://dx.doi.org/10.1289/ehp.8360

     Verina. T; Rohde. CA; Guilarte. TR. (2007). Environmental lead exposure during early life alters granule
        cell neurogenesis and morphology in the hippocampus of young adult rats. Neuroscience 145:
        1037-1047. http://dx.doi.0rg/10.1016/i.neuroscience.2006.12.040

     Verma. SK; Dua. R; Gill. KD. (2005). Impaired energy metabolism after co-exposure to lead and ethanol.
        Basic Clin Pharmacol Toxicol 96: 475-479. http://dx.doi.Org/10.llll/i.1742-7843.2005.pto 96611.x

     Vetter. SW: Leclerc. E. (2003). Novel aspects of calmodulin target recognition and activation. Eur J
        Biochem 270: 404-414. http://dx.doi.Org/10.1046/i.1432-1033.2003.03414.x

     Vigeh. M; Yokoyama. K; Kitamura. F; Afshinrokh. M; Bevgi. A; Niroomanesh. S. (2010). Early
        pregnancy blood lead and spontaneous abortion. Women Health 50: 756-766.
        http://dx.doi.org/10.1080/03630242.2010.532760

     Vigeh. M; Yokoyama. K; Sevedaghamiri. Z; Shinohara. A; Matsukawa. T; Chiba. M; Yunesian. M.
        (2011). Blood lead at currently acceptable levels may cause preterm labour. Occup Environ Med 68:
        231-234. http://dx.doi.org/10.1136/oem.2009.050419

     Vimpani. GV: Wigg. NR; Robertson. EF; McMichael. AJ; Baghurst. PA; Roberts. RJ. (1985). The Port
        Pirie cohort study: blood lead concentration and childhood developmental assessment. In LJ
        Goldwater; LM Wysocki; RA Volpe (Eds.), Edited proceedings: Lead environmental health - the
        current issues (pp. 139-146). Durham, NC: Duke University.

     Vinceti. M; Guidetti. D; Bergomi. M; Caselgrandi. E; Vivoli. R; Olmi. M; Rinaldi. L; Rovesti. S: Solime.
        K (1997). Lead, cadmium, and selenium in the blood of patients with sporadic amyotrophic lateral
        sclerosis. Ital JNeurol Sci 18: 87-92. http://dx.doi.org/10.1007/BF01999568

     Virgolini. MB; Chen. K; Weston. DP; Bauter. MR; Cory-Slechta. DA. (2005). Interactions of chronic
        lead exposure and intermittent stress:  Consequences for brain catecholamine systems and associated
        behaviors and HPA axis function. Toxicol Sci 87: 469-482. http://dx.doi.org/10.1093/toxsci/kfi269

     Virgolini. MB; Rossi-George. A; Lisek. R; Weston. DP; Thiruchelvam. M; Cory-Slechta. DA. (2008a).
        CNS effects of developmental Pb exposure are enhanced by combined maternal and offspring stress.
        Neurotoxicology 29: 812-827. http://dx.doi.Org/10.1016/i.neuro.2008.03.003

     Virgolini. MB; Rossi-George. A; Weston. D; Cory-Slechta. DA. (2008b). Influence  of low level maternal
        Pb exposure  and prenatal stress on offspring stress challenge responsivity. Neurotoxicology 29:
        928-939. http://dx.doi.0rg/10.1016/i.neuro.2008.09.010

     Vogetseder. A; Picard. N; Gaspert. A; Walch. M; Kaissling. B; Le Hir. M. (2008). Proliferation capacity of
        the renal proximal tubule involves the bulk of differentiated epithelial cells. Am J Physiol Cell Physiol
        294: C22-C28. http://dx.doi.org/10.1152/aipcell.00227.2007
November 2012                              5-792                   Draft - Do Not Cite or Quote

-------
     Vvskocil. A; Semecky. V; Fiala. Z; Cizkova. M; Viau. C. (1995). Renal alterations in female rats
        following subchronic lead exposure. J Appl Toxicol 15: 257-262.
        http://dx.doi.org/10.1002/iat.2550150405

     Waalkes. MP: Diwan. BA; Ward. JM; Devor. DE; Gover. RA. (1995). Renal tubular tumors and atypical
        hyperplasias in B6C3F1 mice exposed to lead acetate during gestation and lactation occur with
        minimal chronic nephropathy. Cancer Res 55: 5265-5271.
     Waalkes. MP: Liu. J; Gover. RA; Diwan. BA. (2004). Metallothionein-I/II double knockout mice are
        hypersensitive to lead-induced kidney carcinogenesis: Role of inclusion body formation. Cancer Res
        64: 7766-7772. http://dx.doi.org/10.1158/0008-5472.CAN-04-2220

     Wananukul. W: Sura. T; Salaitanawatwong. P. (2006). Polymorphism of delta-aminolevulinic acid
        dehydratase and its effect on blood lead levels in Thai workers. Arch Environ Occup Health 61: 67-72.
        http://dx.doi.Org/10.3200/AEOH.61.2.67-72

     Wang. A; Wang. Q; Song. Q; Xu. J. (2009a). [Study of ALAD and VDR gene polymorphisms associated
        with lead nephrotoxicity susceptibility]. Wei Sheng Yan Jiu 38: 326-329.
     Wang. CH; Zhang. Y; Liang. JC: Shan. G: Wang. Y; Shi. Q. (2006a). Impacts of ascorbic acid and
        thiamine supplementation at different concentrations on lead toxicity in testis. Clin Chim Acta 370:
        82-88. http://dx.doi.0rg/10.1016/j.cca.2006.01.023

     Wang. CY; Lin. YW: Yang. JL. (2008c). Activation of protein kinase C alpha signaling prevents
        cytotoxicity and mutagenicity following lead acetate in CL3 human lung cancer cells. Toxicology 250:
        55-61. http://dx.doi.0rg/10.1016/j.tox.2008.06.001

     Wang. FT; Hu. H; Schwartz. J: Weuve. J: Spiro. AS. Ill; Sparrow. D; Nie. HL; Silverman. EK; Weiss. ST;
        Wright. RO. (2007a). Modifying effects of the HFE polymorphisms on the association between lead
        burden and cognitive decline. Environ Health Perspect 115:  1210-1215.
        http://dx.doi.org/10.1289/ehp.9855

     Wang. G: Chen. XQ; Lipsky. MM; Whittaker. MH; Fowler. BA. (2005a). Effect of combined lead,
        cadmium and arsenic exposure on the expression of cellular protective factors in rat kidneys at LOEL
        dose levels [Abstract]. Toxicologist 85: 237.

     Wang. GS: Fowler. BA.  (2008). Roles of biomarkers in evaluating interactions among mixtures of lead,
        cadmium and arsenic [Review]. Toxicol Appl Pharmacol 233: 92-99.
        http://dx.doi.0rg/10.1016/i.taap.2008.01.017

     Wang. HL; Chen. XT; Yang. B; Ma. FL; Wang. S: Tang. ML; Hao. MG: Ruan. DY. (2008d). Case-control
        study of blood lead levels and attention deficit hyperactivity disorder in Chinese children. Environ
        Health Perspect 116: 1401-1406. http://dx.doi.org/10.1289/ehp.11400
     Wang. L; Chen. DW: Wang. H; Liu. ZP (2009b). Effects of lead and/or cadmium on the expression of
        metallothionein in the kidney of rats. Biol Trace Elem Res 129: 190-199.
        http://dx.doi.org/10.1007/sl2011-008-8288-3

     Wang. L; Wang. H; Hu.  MZ; Cao. J: Chen. DW:  Liu. Z. (2009c). Oxidative stress and apoptotic changes
        in primary cultures of rat proximal tubular cells exposed to lead. Arch Toxicol 83: 417-427.
        http://dx.doi.org/10.1007/s00204-009-0425-z

     Wang. L; Wang. H; Li. J: Chen. D; Liu. Z. (20 lib). Simultaneous effects of lead and cadmium on primary
        cultures of rat proximal tubular cells: interaction of apoptosis and oxidative stress. Arch Environ
        Contam Toxicol 61: 500-511. http://dx.doi.org/10.1007/s00244-011-9644-4

     Wang. L; Wang. Z; Liu.  J. (2010d). Protective effect of N-acetylcysteine on experimental chronic lead
        nephrotoxicity in immature female rats. Hum Exp Toxicol 29: 581-591.
        http://dx.doi.org/10.1177/0960327109357270

     Wang. Q; Luo. WJ: Zhang. WB; Dai.  ZM; Chen. YM; Chen. JY (2007b). Iron supplementation protects
        against lead-induced apoptosis through MAPK pathway in weanling rat cortex. Neurotoxicology 28:
        850-859. http://dx.doi.0rg/10.1016/i.neuro.2007.04.004
November 2012                             5-793                   Draft - Do Not Cite or Quote

-------
     Wang. Q; Zhao. HH; Chen. JW: Hao. QL; Gu. KD; Zhu. YX; Zhou. YK; Ye. LX. (201 Of), delta-
        Aminolevulinic acid dehydratase activity, urinary delta-aminolevulinic acid concentration and zinc
        protoporphyrin level among people with low level of lead exposure. Int J Hyg Environ Health 213:
        52-58. http://dx.doi.0rg/10.1016/j.ijheh.2009.08.003

     Wang. TL; Kao. TH; Inbarai. BS: Su. YT; Chen. BH. (2010g). Inhibition effect of poly(y-glutamic acid)
        on lead-induced toxicity in mice. J Agric Food Chem 58: 12562-12567.
        http://dx.doi.org/10.1021/ifl034509

     Wang. YY; Sui. KX; Hong. LI; Ma. HY. (2009e). The effects of lead exposure on placenta! NF-kappaB
        expression and the consequences for gestation. Reprod Toxicol 27: 190-195.
        http://dx.doi.0rg/10.1016/i.reprotox.2008.12.006

     Wasserman. G: Graziano. JH; Factor-Litvak. R; Popovac. D; Morina. N; Musabegovic. A; Vrenezi. N;
        Capuni-Paracka. S: Lekic. V: Preteni-Redjepi. E; Hadzialjevic. S: Slavkovich. V; Kline. J: Shrout. P;
        Stein. Z. (1992). Independent effects of lead exposure and iron deficiency anemia on developmental
        outcome at age 2 years.  J Pediatr 121: 695-703. http://dx.doi.org/10.1016/80022-3476(05)81895-5

     Wasserman. GA; Factor-Litvak. P. (2001). Methodology, inference and causation: Environmental lead
        exposure and childhood intelligence [Comment]. Arch Clin Neuropsychol 16: 343-352.
        http://dx.doi.org/10.1016/S0887-6177(00)00085-8

     Wasserman. GA; Factor-Litvak. P; Liu. X; Todd. AC: Kline. JK; Slavkovich. V; Popovac. D; Graziano.
        JH. (2003). The relationship between blood lead, bone lead and child intelligence. Child Neuropsychol
        9: 22-34. http://dx.doi.Org/10.1076/chin.9.l.22.14497

     Wasserman. GA: Graziano. JH: Factor-Litvak. P; Popovac. D: Morina. N;  Musabegovic. A; Vrenezi. N;
        Capuni-Paracka. S: Lekic. V: Preteni-Redjepi. E; Hadzialjevic. S: Slavkovich. V; Kline. J: Shrout. P:
        Stein. Z. (1994). Consequences of lead exposure and iron supplementation on childhood development
        at age 4 years. Neurotoxicol Teratol 16: 233-240. http://dx.doi.org/10.1016/0892-0362(94)90044-2

     Wasserman. GA: Liu. X; Lolacono. NJ; Factor-Litvak. P; Kline. JK;  Popovac. D: Morina. N;
        Musabegovic. A; Vrenezi. N; Capuni-Paracka. S: Lekic. V: Preteni-Redjepi. E; Hadzialjevic. S:
        Slavkovich. V; Graziano. JH. (1997). Lead exposure and intelligence in 7-year-old children: The
        Yugoslavia prospective study. Environ Health Perspect 105: 956-962.

     Wasserman. GA: Liu. X; Pine. DS: Graziano. JH.  (2001). Contribution of maternal smoking during
        pregnancy and lead exposure to early child behavior problems. Neurotoxicol Teratol 23: 13-21.
        http://dx.doi.org/10.1016/S0892-0362(00)00116-l

     Wasserman. GA: Musabegovic. A; Liu. X; Kline. J: Factor-Litvak. P; Graziano. JH. (2000). Lead
        exposure and motor functioning in 4 1/2-year-old children: The Yugoslavia prospective study. J
        Pediatr 137: 555-561. http://dx.doi.org/10.1067/mpd.2000.109111

     Wasserman. GA: Staghezza-Jaramillo. B;  Shrout. P: Popovac. D: Graziano. J. (1998). The effect of lead
        exposure on behavior problems in preschool children. Am J Public Health 88: 481-486.
        http://dx.doi.Org/10.2105/AJPH.88.3.481

     Waterman. SJ: El-Fawal. HAN; Snyder. CA. (1994). Lead alters the immunogenicity of two neural
        proteins: A potential mechanism for the progression of lead-induced neurotoxicity. Environ Health
        Perspect 102: 1052-1056.

     Watts. SW: Chai. S: Webb.  RC. (1995). Lead acetate-induced contraction in rabbit mesenteric artery:
        Interaction with calcium and protein kinase C. Toxicology 99: 55-65.
        http://dx.doi.org/10.1016/0300-483X(94)03003-K

     Weaver. V; Jaar. B. (2010).  UpToDate: Lead nephropathy and lead-related nephrotoxicity. Available
        online at http://www.uptodate.com/contents/lead-nephropathv-and-lead-related-nephrotoxicitv
        (accessed October 25, 2010).

     Weaver. VM. (2010). Further analysis of Korean lead worker data set with a focus on determining the
        functional form of the exposure-response relationship. Available online
November 2012                              5-794                  Draft - Do Not Cite or Quote

-------
     Weaver. VM; Ellis. LR; Lee. BK; Todd. AC: Shi. W: Ahn. KD; Schwartz. BS. (2008). Associations
        between patella lead and blood pressure in lead workers. Am J Ind Med 51: 336-343.
        http://dx.doi.org/10.1002/aiim.20573

     Weaver. VM: Griswold. M: Todd. AC: Jaar. BG: Ahn. KD: Thompson. CB: Lee. BK. (2009).
        Longitudinal associations between lead dose and renal function in lead workers. Environ Res 109:
        101-107. http://dx.doi.org/10.1016/i.envres.20Q8.09.005

     Weaver. VM: Jarr. BG: Schwartz. BS: Todd. AC: Ahn. KD: Lee. SS: Wen. J; Parsons. PJ: Lee. BK.
        (2005a). Associations among lead dose biomarkers, uric acid, and renal function in Korean lead
        workers. Environ Health Perspect 113: 36-42. http://dx.doi.org/10.1289/ehp.7317

     Weaver. VM: Kim. NS: Jaar. BG: Schwartz. BS: Parsons. PJ: Steuerwald. AJ; Todd. AC: Simon. D; Lee.
        BK. (2011). Associations of low-level urine cadmium with kidney function in lead workers. Occup
        Environ Med 68:  250-256. http://dx.doi.org/10.1136/oem.2010.056077

     Weaver. VM: Lee. BK: Ahn. KD: Lee. GS: Todd. AC: Stewart. WF: Wen. J: Simon. DJ: Parsons. PJ:
        Schwartz. BS. (2003a). Associations of lead biomarkers with renal function in Korean lead workers.
        Occup Environ Med 60: 551-562. http://dx.doi.Org/10.1136/oem.60.8.551

     Weaver. VM; Lee. BK: Todd. AC: Ahn. KD: Shi. W: Jaar. BG: Kelsev. KT; Lustberg. ME; Silbergeld.
        EK; Parsons. PJ: Wen. J: Schwartz. BS. (2006). Effect modification by delta-aminolevulinic acid
        dehydratase, vitamin D receptor, and nitric oxide synthase gene polymorphisms on associations
        between patella lead and renal function in lead workers. Environ Res 102: 61-69.
        http://dx.doi.0rg/10.1016/i.envres.2006.01.001

     Weaver. VM; Schwartz. BS: Ahn. KD: Stewart. WF; Kelsev. KT; Todd. AC: Wen. J: Simon. DJ: Lustberg.
        ME; Parsons. PJ:  Silbergeld. EK; Lee. BK. (2003b). Associations of renal function with
        polymorphisms in the "delta"-aminolevulinic acid dehydratase, vitamin D receptor, and nitric oxide
        synthase genes in Korean lead workers. Environ Health Perspect  111: 1613-1619.

     Weaver. VM: Schwartz. BS: Jaar. BG: Ahn. KD: Todd. AC: Lee.  SS: Kelsev. KT: Silbergeld. EK:
        Lustberg. ME: Parsons. PJ: Wen. J: Lee. BK. (2005b). Associations of uric acid with polymorphisms
        in the "delta"-aminolevulinic acid dehydratase, vitamin D receptor, and nitric oxide synthase genes in
        Korean lead workers. Environ Health Perspect 113: 1509-1515. http://dx.doi.org/10.1289/ehp.7927

     Wedrychowski. A: Schmidt. WN: Hnilica. LS. (1986). The in vivo cross-linking of proteins and DNAby
        heavy metals. J Biol Chem 261: 3370-3376.

     Wei. H: Wei. D: Yi. S: Zhang. F: Ding. W. (2011). Oxidative stress induced by urban fine particles in
        cultured EA.hy926 cells. Hum Exp Toxicol 30: 579-590. httrj://dx.doi.org/10.1177/0960327110374207

     Weiss. B. (1988). Neurobehavioral toxicity as a basis for risk assessment [Review]. Trends Pharmacol Sci
        9: 59-62.

     Weiss. B. (1990). Risk assessment: The insidious nature of neurotoxicity and the aging brain [Review].
        Neurotoxicology  11: 305-314.

     Weisskopf. MG: Jain. N; Nie. HL: Sparrow. D: Vokonas. P: Schwartz. J: Hu. H. (2009). A prospective
        study of bone lead concentration and death from all causes, cardiovascular diseases, and cancer in the
        department  of veterans  affairs normative aging study. Circulation 120: 1056-1064.
        http://dx.doi.org/10.1161/circulationaha.108.827121

     Weisskopf. MG: Proctor. SP: Wright. RO: Schwartz. J: Spiro. A. Ill: Sparrow. D: Nie. HL: Hu. H.
        (2007b). Cumulative lead exposure and cognitive performance among elderly men. Epidemiology 18:
        59-66. htrp://dx.doi.org/10.1097/01.ede.0000248237.35363.29

     Weisskopf. MG: Weuve. J: Nie. H: Saint-Hilaire. MH: Sudarsky.  L: Simon. DK: Hersh. B: Schwartz. J:
        Wright. RO: Hu. H. (2010). Association of cumulative lead exposure with Parkinson's Disease.
        Environ Health Perspect 118: 1609-1613. http://dx.doi.org/10.1289/ehp.1002339
November 2012                             5-795                  Draft - Do Not Cite or Quote

-------
     Weisskopf. MG: Wright. RO: Schwartz. J; Spiro. A. Ill: Sparrow. D; Aro. A; Hu. H. (2004). Cumulative
        lead exposure and prospective change in cognition among elderly men: The VA Normative Aging
        Study. Am J Epidemiol 160: 1184-1193. http://dx.doi.org/10.1093/aje/kwh333

     Wellenius. GA; Coull. BA; Godleski. JJ; Koutrakis. P; Okabe. K; Savage. ST. (2003). Inhalation of
        concentrated ambient air particles exacerbates myocardial ischemia in conscious dogs. Environ Health
        Perspect 111: 402-408. http://dx.doi.org/10.1289/ehp.5775

     Wells. E; Jarrett. J: Hong Lin. Y; Caldwell. K; Hibbeln. J: Apelberg. B; Herbstman. J: Halden. R; Witter.
        F; Goldman. L. (2011 a). Body burdens of mercury, lead, selenium and copper among Baltimore
        newborns. Environ Res 111: 411-417. http://dx.doi.Org/10.1016/j.envres.2010.12.009

     Wells. EM; Navas-Acien. A; Herbstman. JB; Apelberg. BJ; Silbergeld. EK; Caldwell. KL; Jones. RL;
        Halden. RU; Witter. FR; Goldman. LR. (20lib). Low level lead exposure and elevations in blood
        pressure during pregnancy. Environ Health Perspect 119: 664-669.
        http://dx.doi.org/10.1289/ehp.1002666

     Westerink. RH; Vijverberg. HP.  (2002). Ca2+-independent vesicular catecholamine release in PC 12 cells
        by nanomolar concentrations of Pb2+. JNeurochem 80: 861-867.
        http://dx.doi.0rg/10.1046/i.0022-3042.2001.00751.x

     Wetmur. JG. (1994). Influence of the common human delta-aminolevulinate dehydratase polymorphism
        on lead body burden [Review]. Environ Health Perspect 102: 215-219.

     Wetmur. JG: Lehnert. G: Desnick. RJ. (1991b). The delta-aminolevulinate dehydratase polymorphism:
        Higher blood lead levels in lead workers and environmentally exposed children with the 1-2 and 2-2
        isozymes. Environ Res 56: 109-119. http://dx.doi.org/10.1016/80013-9351(05)80001-5

     Weuve. J: Kelsev. KT; Schwartz. J: Bellinger. D; Wright. RO: Raian. P; Spiro. A. Ill; Sparrow. D; Aro. A;
        Hu. H. (2006). Delta-aminolevulinic acid dehydratase polymorphism and the relation between low
        level lead exposure and the Mini-Mental Status  Examination in older men: The Normative Aging
        Study. Occup Environ Med 63: 746-753. http://dx.doi.org/10.1136/oem.2006.027417

     Weuve. J: Korrick. SA; Weisskopf. MA; Ryan. LM; Schwartz. J: Nie. HL; Grodstein. F; Hu. H. (2009).
        Cumulative exposure to lead in relation to cognitive function in older women. Environ Health
        Perspect 117: 574-580. http://dx.doi.org/10.1289/ehp.11846

     Whitworth. JA; Mangos. GJ: Kelly. JJ. (2000). Gushing, cortisol, and cardiovascular disease [Review].
        Hypertension 36: 912-916. http://dx.doi.Org/10.1161/01.HYP.36.5.912

     WHO (World Health Organization).  (1948). Preamble to the Constitution of the World Health
        Organization as adopted by the International Health Conference, New York, 19-22 June, 1946. In
        Constitution of the World Health Organization (pp. 2). Geneva, Switzerland.
        http://whqlibdoc.who.int/hist/official records/constitution.pdf

     WHO (World Health Organization).  (1992). International statistical classification of diseases and related
        health problem, tenth revision.  Geneva, Switzerland, http ://www.who, int/classifications/icd/en/

     Wiebe. JP; Barr. KJ. (1988). Effect of prenatal and neonatal exposure to lead on the affinity and number
        of estradiol receptors in the uterus. J Toxicol Environ Health 24: 451-460.
        http://dx.doi.org/10.1080/15287398809531176

     Wiebe. JP: Barr. KJ: Buckingham. KD. (1988). Effect of prenatal and neonatal exposure to lead on
        gonadotropin receptors and steroidogenesis in rat ovaries. J Toxicol Environ Health 24: 461-476.

     Wiemann. M; Schirrmacher. K;  Busselberg. D. (1999). Interference of lead with the calcium release
        activated calcium flux of osteoblast-like cells. Calcif Tissue Int 65: 479-485.
        http://dx.doi.org/10.1007/s002239900736

     Wilker. E; Korrick. S: Nie. LH;  Sparrow. D: Vokonas. P: Coull. B; Wright. RO: Schwartz. J: Hu. H.
        (2011). Longitudinal changes in bone lead levels: The VA Normative Aging Study. J Occup Environ
        Med 53: 850-855. http://dx.doi.org/10.1097/JOM.Ob013e31822589a9
November 2012                             5-796                   Draft - Do Not Cite or Quote

-------
     Williams. BL; Pennock-Roman. M; Suen. HK; Magsumbol. MS: Ozdenerol. E. (2007). Assessing the
        impact of the local environment on birth outcomes: A case for HLM. J Expo Sci Environ Epidemiol
        17: 445-457. http://dx.doi.org/10.1038/sj.jes.7500537

     Williams. PL; Sergevev. O: Lee. MM; Korrick. SA; Burns. JS: Humblet. O: DelPrato. J: Revich. B;
        Hauser. R. (2010). Blood lead levels and delayed onset of puberty in a longitudinal study of Russian
        boys. Pediatrics 125: 1088-1096. http://dx.doi.org/10.1542/peds.2009-2575

     Wise. JR Sr; Stearns. DM; Wetterhahn. KE; Patierno. SR. (1994). Cell-enhanced dissolution of
        carcinogenic lead chromate particles: the role of individual dissolution products in clastogenesis.
        Carcinogenesis 15: 2249-2254. http://dx.doi.org/10.1093/carcin/15.10.2249

     Wise. JP. Sr: Wise. SS: Lacerte. C: Wise. JP; Aboueissa. AM. (2011). The genotoxicity of paniculate and
        soluble chromate in sperm whale (physeter macrocephalus) skin fibroblasts. Environ Mol Mutagen 52:
        43-49. http://dx.doi.org/10.1002/em.20579

     Wise. SS: Holmes. AL; Ketterer. ME; Hartsock. WJ: Fomchenko. E; Katsifis. S: Thompson. WD; Wise.
        JP. Sr. (2004). Chromium is the proximate clastogenic species for lead chromate-induced
        clastogenicity in human bronchial cells. Mutat Res Genet Toxicol Environ Mutagen 560: 79-89.
        http://dx.doi.0rg/10.1016/i.mrgentox.2004.02.009

     Wise. SS: Holmes. AL; Moreland. JA; Xie. H; Sandwick. SJ: Stackpole. MM; Fomchenko. E; Teufack. S:
        May. AJ. Jr; Katsfis.  SP; Wise. JP. Sr. (2005). Human lung cell growth is not stimulated by lead ions
        after lead chromate-induced genotoxicity. Mol Cell Biochem 279: 75-84.
        http://dx.doi.org/10.1007/sll010-005-8217-0

     Wise. SS: Holmes. AL: Qin. Q; Xie. H; Katsifis. SP: Thompson. WD: Wise. JR Sr. (2010).  Comparative
        genotoxicity and cytotoxicity of four hexavalent chromium compounds in human bronchial cells.
        Chem Res Toxicol 23: 365-372. http://dx.doi.org/10.1021/tx900363j

     Wise. SS: Holmes. AL; Wise. JP. Sr. (2006a).  Paniculate and soluble hexavalent chromium ate cytotoxic
        and genotoxic to human lung epithelial cells. Mutat Res Genet Toxicol Environ Mutagen 610: 2-7.
        http://dx.doi.0rg/10.1016/i.mrgentox.2006.06.005

     Wise. SS: Holmes. AL: Xie. H; Thompson. WD: Wise. JP. Sr. (2006b).  Chronic exposure to paniculate
        chromate induces spindle assembly checkpoint bypass in human lung cells. Chem Res Toxicol 19:
        1492-1498. http://dx.doi.org/10.1021/tx0601410

     Wise. SS: Shafflev. F; Lacerte. C: Goertz. CEC: Dunn. JL; Gulland. FMD; Aboueissa. AE; Zheng. TZ;
        Wise. JP. (2009). Paniculate and soluble hexavalent chromium are cytotoxic and genotoxic to Steller
        sea lion lung cells. Aquat Toxicol 91: 329-335. http://dx.doi.Org/10.1016/i.aquatox.2008.12.004

     Wiwanitkit. V: Suwansaksri. J: Soogarun. S. (2008). White blood cell sister chromatid exchange among a
        sample of Thai subjects exposed to lead: Lead-induced genotoxicity. Toxicol Environ Chem 90:
        765-768. http://dx.doi.org/10.1080/02772240701712758

     Wolf. J: Daley. AJ. (2007). Microbiological aspects of bacterial lower respiratory tract illness in children:
        Atypical pathogens. Paediatr Respir Rev 8: 212-220. http://dx.doi.Org/10.1016/i.prrv.2007.07.004

     Wolff. MS: Britton. JA; Boguski. L; Hochman. S: Malonev. N; Serra. N; Liu. ZS: Berkowitz. G: Larson.
        S: Forman. J. (2008). Environmental exposures and puberty in inner-city girls. Environ Res 107:
        393-400. http://dx.doi.0rg/10.1016/i.envres.2008.03.006

     Wozniak. K; Blasiak. J. (2003). In vitro genotoxicity of lead acetate: Induction of single and double DNA
        strand breaks and DNA-protein cross-links. Mutat Res Genet Toxicol Environ Mutagen 535: 127-139.
        http://dx.doi.org/10.1016/S1383-5718(02)00295-4

     Wright. JP: Dietrich. KN; Ris. MD; Hornung. RW: Wessel. SD; Lanphear. BP; Ho. M; Rae. MN. (2008).
        Association of prenatal and childhood blood lead concentrations with criminal arrests in early
        adulthood. PLoS Med 5: 732-740. http://dx.doi.org/10.1371/iournal.pmed.0050101
November 2012                             5-797                   Draft - Do Not Cite or Quote

-------
     Wright. RO: Schwartz. J; Wright. RJ: Bollati. V: Tarantini. L; Park. SK; Hu. H; Sparrow. D; Vokonas. P;
        Baccarelli. A. (2010). Biomarkers of lead exposure and DNA methylation within retrotransposons.
        Environ Health Perspect 118: 790-795. http://dx.doi.org/10.1289/ehp.0901429

     Wright. RO: Tsaih. SW: Schwartz. J; Spiro. A; McDonald. K; Weiss. ST; Hu. H. (2003). Lead exposure
        biomarkers and mini-mental status exam scores in older men. Epidemiology 14: 713-718.
        htrp://dx.doi.org/10.1097/01.EDE.0000081988.85964.db

     Wu. J: Basha. MR; Brock. B; Cox. DP;  Cardozo-Pelaez. F; Mcpherson. CA; Harry. J: Rice. DC: Malonev.
        B; Chen. D; Lahiri. DK; Zawia. NH. (2008a). Alzheimer's disease (AD)-like pathology in aged
        monkeys after infantile exposure to environmental metal lead (pb): Evidence for a developmental
        origin and environmental link for AD. JNeurosci 28: 3-9.
        http://dx.doi.org/10.1523/ineurosci.4405-07.2008

     Wu. J: Basha. R; Zawia. NH.  (2008b). The environment, epigenetics and amyloidogenesis [Review]. J
        Mol Neurosci 34: 1-7. http://dx.doi.org/10.1007/sl2031-007-0009-4

     Wu. MT; Kelsev. K; Schwartz. J: Sparrow. D: Hu. H. (2003a). A delta-aminolevulinic acid dehydratase
        (ALAD) polymorphism may modify the relationship of low-level lead exposure to uricemia and renal
        function: The normative aging study. Environ Health Perspect 111: 335-341.
        http://dx.doi.org/10.1289/ehp.5504

     Wu. T; Buck.  GM; Mendola. P. (2003b). Blood lead levels and sexual maturation in U.S. girls: The Third
        National Health and Nutrition Examination Survey,  1988-1994. Environ Health Perspect 111:
        737-741. http://dx.doi.org/10.1289/ehp.6008

     Wu. TN; Shen. CY; Lai. JS: Goo. CF; Ko. KN; Chi. HY; Chang. PY; Liou. SH. (2000). Effects of lead
        and noise exposures on hearing ability. Arch Environ Health 55: 109-114.
        http://dx.doi.org/10.1080/00039890009603396

     Xie. H; Holmes. AL; Wise. SS: Huang.  S: Peng. C: Wise. JP. Sr. (2007). Neoplastic transformation of
        human bronchial cells by lead chromate particles. Am J Respir Cell Mol Biol 37: 544-552.
        htrp://dx.doi.org/10.1165/rcmb.2007-0058OC

     Xie. H; Wise. SS: Wise. JP. Sr. (2008). Deficient repair  of particulate hexavalent chromium-induced DNA
        double strand breaks leads to neoplastic transformation. Mutat Res Genet Toxicol Environ Mutagen
        649: 230-238. http://dx.doi.Org/10.1016/i.mrgentox.2007.09.008

     Xie. Y; Chiba. M; Shinohara. A; Watanabe. H; Inaba. Y. (1998). Studies on lead-binding protein and
        interaction between lead and selenium in the human erythrocytes. Ind Health 36: 234-239.
        http://dx.doi.org/10.2486/indhealth.36.234

     Xu. J; Ji. LD;  Xu. LH. (2006). Lead-induced apoptosis in PC 12 cells: Involvement of p53, Bcl-2 family
        and caspase-3. Toxicol Lett 166:  160-167. http://dx.doi.Org/10.1016/i.toxlet.2006.06.643

     Xu. J; Lian. LJ; Wu. C: Wang. XF; Fu. WY; Xu. LH. (2008). Lead induces oxidative stress, DNA damage
        and alteration of p53, Bax and Bcl-2 expressions in mice. Food Chem Toxicol 46:  1488-1494.
        http://dx.doi.0rg/10.1016/i.fct.2007.12.016

     Xu. J: Yan. CH: Yang. B: Xie. HF: Zou. XY: Zhong. L: Gao. Y: Tian. Y: Shen. XM. (2009c). The role of
        metabotropic glutamate receptor  5 in developmental lead neurotoxicity. Toxicol Lett 191: 223-230.
        http://dx.doi.0rg/10.1016/i.toxlet.2009.09.001

     Xu. SZ; Raj anna. B. (2006). Glutamic acid reverses Pb2+-induced reductions of NMDA receptor subunits
        in vitro.  Neurotoxicology 27: 169-175. http://dx.doi.Org/10.1016/i.neuro.2005.08.005

     Yamanaka. K; Chun. SJ: Boillee. S: Fujimori-Tonou. N; Yamashita. H; Gutmann. DH; Takahashi. R;
        Misawa. H; Cleveland. DW. (2008).  Astrocytes as determinants of disease progression in inherited
        amyotrophic lateral sclerosis. Nat Neurosci 11: 251-253. http://dx.doi.org/10.1038/Nn2047

     Yanez. L; Garcia-Nieto. E; Rojas. E; Carrizales. L; Mejia. J: Calderon. J: Razo. I; Diaz-Barriga. F. (2003).
        DNA damage in blood cells from children exposed to arsenic and lead in a mining area. Environ Res
        93: 231-240. http://dx.doi.Org/10.1016/i.envres.2003.07.005
November 2012                              5-798                   Draft - Do Not Cite or Quote

-------
     Yang. Y; Ma. Y; Ni. L; Zhao. S: Li. L; Zhang. J; Fan. M; Liang. C: Cao. J: Xu. L. (2003). Lead exposure
        through gestation-only caused long-term learning/memory deficits in young adult offspring. Exp
        Neural 184: 489-495. http://dx.doi.org/10.1016/S0014-4886(03)00272-3

     Yang. Y; Raine. A; Lencz. T; Bihrle. S: LaCasse. L; Colletti. P. (2005). Volume reduction in prefrontal
        gray matter in unsuccessful criminal psychopaths. Biol Psychiatry 57: 1103-1108.
        http://dx.doi.0rg/10.1016/i.biopsvch.2005.01.021

     Yazbeck. C: Thiebaugeorges. O: Moreau. T; Goua. V: Debotte. G: Sahuquillo. J: Forhan. A; Foliguet B;
        Magnin. G: Slama. R; Charles. MA; Huel. G. (2009). Maternal blood lead levels and the risk of
        pregnancy-induced hypertension: The EDEN cohort study. Environ Health Perspect 117: 1526-1530.
        htrp://dx.doi.org/10.1289/ehp.080Q488

     Ye. SH. (1993). Hypoxanthine phosphoribosyl transferase assay of lead mutagenicity on keratinocytes.
        Acta Pharmacol Sin 14:  145-147.

     Yeates. DB; Hameister. WM. (1992). Alveolar epithelial permeability in baboons: Histamine and
        capsaicin. J Physiol 450: 363-374.

     Yedjou. CG: Milner. JN; Howard. CB; Tchounwou. PB. (2010). Basic apoptotic mechanisms of lead
        toxicity in human leukemia (HL-60) cells. Int J Environ Res Public Health 7: 2008-2017.
        http://dx.doi.org/10.3390/ijerph7052008

     Yen. TH: Lin. JL: Lin-Tan. DT: Hsu. CW: Chen. KH: Hsu. HH. (2011). Blood cadmium level's
        association with 18-month mortality in diabetic patients with maintenance haemodialysis. Nephrol
        Dial Transplant 26: 998-1005. http://dx.doi.org/l0.1093/ndt/gfq448

     Yiin. SJ: Lin. TH. (1995). Lead-catalyzed peroxidation of essential unsaturated fatty acid. Biol Trace
        Elem Res 50: 167-172. http://dx.doi.org/10.1007/BF02789419

     Yin. Y; Zhang. T; Dai. Y; Bao. Y; Chen. X; Lu. X. (2008). The effect of plasma lead on anembryonic
        pregnancy. AnnN YAcad Sci 1140:  184-189. http://dx.doi.org/10.1196/annals.1454.042

     Youssef. SAH; El-Sanousi. AA; Afifi. NA; El-Brawv. AMA. (1996).  Effect of subclinical lead toxicity on
        the immune response of chickens to Newcastle disease virus vaccine. Res Vet Sci 60: 13-16.
        http://dx.doi.org/10.1016/S0034-5288(96)90123-X

     Yu. CC: Lin. JL; Lin-Tan. DT. (2004). Environmental exposure to lead and progression of chronic renal
        diseases: A four-year prospective longitudinal study. JAm Soc Nephrol 15: 1016-1022.
        http://dx.doi.org/10.1097/01.ASN.0000118529.01681.4F

     Yu. DY; Li. WF; Deng. B; Mao. XF. (2008). Effects of lead on hepatic antioxidant status and transcription
        of superoxide dismutase gene in pigs. Biol Trace Elem Res 126: 121-128.
        http://dx.doi.org/10.1007/sl2011-008-8198-4

     Yu. J; Fujishiro. H; Miyataka. H; Oyama. TM; Hasegawa. T; Seko. Y; Miura. N; Himeno. S. (2009).
        Dichotomous effects of lead acetate on the expression of metallothionein in the liver and kidney of
        mice. Biol Pharm Bull 32: 1037-1042. http://dx.doi.org/10.1248/bpb.32.1037

     Yuan. W: Holland. SK; Cecil. KM; Dietrich. KN; Wessel. SD; Altave. M; Hornung. RW: Ris. MD;
        Egelhoff. JC: Lanphear.  BP (2006). The impact of early childhood lead exposure on brain
        organization: A functional magnetic resonance imaging study of language function. Pediatrics 118:
        971-977. http://dx.doi.org/10.1542/peds.2006-0467

     Yiicesov. B; Turhan. A; lire. M; Jmir. T; Karakava. A. (1997a). Effects of occupational lead and cadmium
        exposure on some immunoregulatory cytokine levels in man. Toxicology  123: 143-147.
        http://dx.doi.org/10.1016/S0300-483X(97)00107-8

     Yiicesov. B; Turhan. A; lire. M; Jmir. T; Karakava. A. (1997b). Simultaneous effects of lead and
        cadmium on NK cell activity and some phenotypic parameters. Immunopharmacol Immunotoxicol 19:
        339-348. http://dx.doi.org/10.3109/08923979709046980
November 2012                             5-799                   Draft - Do Not Cite or Quote

-------
     Yun. SW: Hover. S. (2000). Effects of low-level lead on glycolytic enzymes and pyruvate dehydrogenase
        of rat brain in vitro: Relevance to sporadic Alzheimer's disease? Journal of Neural Transmission 107:
        355-368. http://dx.doi.org/10.1007/s007020050030

     Zailina. H; Junidah. R; Josephine. Y; Jamal. HH. (2008). The influence of low blood lead concentrations
        on the cognitive and physical development of primary school children in Malaysia. Asia Pac J Public
        Health 20: 317-326. http://dx.doi.org/10.1177/1010539508322697

     Zawia. NH; Basha. MR. (2005). Environmental risk factors and the developmental basis for Alzheimer's
        disease [Review]. RevNeurosci 16: 325-337.

     Zawia. NH; Harry. GJ. (1995). Exposure to lead-acetate modulates the developmental expression of
        myelin genes in the rat frontal lobe. Int J Dev Neurosci 13: 639-644.
        http://dx.doi.org/10.1016/0736-5748(95)00032-C

     Zawia. NH: Sharan. R; Brydie. M; Oyama. T; Crumpton. T. (1998). Spl as a target site for metal-induced
        perturbations of transcriptional regulation of developmental brain gene expression. Dev Brain Res
        107: 291-298. http://dx.doi.org/10.1016/S0165-3806(98)00023-6

     Zeisel. S: Blusztajn. J. (1994). Choline and human nutrition [Review]. Annu Rev Nutr 14: 269-296.
        http://dx.doi.org/10.1146/annurev.nu.14.070194.001413

     Zelikoff. JT; Li. JH; Hartwig. A; Wang. XW: Costa. M; Rossman. TG. (1988). Genetic toxicology of lead
        compounds. Carcinogenesis 9: 1727-1732. http://dx.doi.Org/10.1093/carcin/9.10.1727

     Zelikoff. JT; Parsons. E; Schlesinger. RB.  (1993). Inhalation of paniculate lead oxide disrupts pulmonary
        macrophage-mediated functions important for host defense and tumor surveillance in  the lung.
        Environ Res 62: 207-222. http://dx.doi.org/10.1006/enrs.1993.1106

     Zeller. I; Knoflach. M; Seubert. A; Kreutmaver. S: Stelzmiiller. M; Wallnoefer. E; Blunder. S: Frotschnig.
        S: Messner. B; Willeit. J: Debbage. P; Wick. G: Kiechl. S: Laufer. G: Bernhard. D. (2010). Lead
        contributes to arterial intimal hyperplasia through nuclear factor erythroid 2-related factor-mediated
        endothelial interleukin 8 synthesis and subsequent invasion of smooth muscle cells. Arterioscler
        Thromb Vase Biol 30: 1733-1740. http://dx.doi.org/10.1161/ATVBAHA.110.211011

     Zeng. H; Chattarji. S: Barbarosie. M; Rondi-Reig. L; Philpot. B; Miyakawa. T; Bear. M; Tonegawa.  S.
        (2001). Forebrain-specific calcineurin knockout selectively impairs bidirectional synaptic plasticity
        and working/episodic-like memory. Cell 107: 617-629.
        http://dx.doi.org/10.1016/S0092-8674(01)00585-2

     Zentner. LE; Rondo. PH; Mastroeni. SS. (2006). Lead contamination and anthropometry  of the newborn
        baby. J Trap Pediatr 52: 369-371. http://dx.doi.org/10.1093/tropej/fml009

     Zentner. LEA; Rondo. PHC: Duran. MC: Oliveira. JM.  (2008). Relationships of blood lead to calcium,
        iron, and vitamin C intakes in Brazilian pregnant women. Clin Nutr 27: 100-104.
        http://dx.doi.0rg/10.1016/i.clnu.2007.10.005

     Zeyrek. D; Soran. M; Cakmak. A; Kocyigit. A; Iscan. A. (2009). Serum copper and ainc levels in mothers
        and cord blood of their newborn infants with neural  tube defects: A case-control study. Indian Pediatr
        46: 675-680.

     Zhang. A; Hu. H; Sanchez. BN; Ettinger. AS: Park. SK; Cantonwine. D: Schnaas. L; Wright. RO:
        Lamadrid-Figueroa. H; Tellez-Rojo. MM. (2011 a). Association between prenatal lead exposure and
        blood  pressure in female offspring. Environ Health Perspect 120: 445-450.
        http://dx.doi.org/10.1289/ehp.1103736

     Zhang. A; Park. SK; Wright. RO: Weisskopf. MG: Mukherjee. B; Nie. H; Sparrow. D: Hu. H. (2010a).
        HFE H63D polymorphism as a modifier of the effect of cumulative lead exposure on pulse pressure:
        The Normative Aging Study. Environ Health Perspect 118: 1261-1266.
        http://dx.doi.org/10.1289/ehp. 1002251
November 2012                              5-800                   Draft - Do Not Cite or Quote

-------
     Zhang. LF; Peng. SQ: Sheng. W: Li. BL; Gang. H; Dong. YS. (2007a). Direct effects of lead (Pb2+) on
        the relaxation of in vitro cultured rat aorta to acetylcholine. Toxicol Lett 170: 104-110.
        http://dx.doi.0rg/10.1016/i.toxlet.2007.02.004

     Zhang. LF; Peng. SQ: Wang. S. (2005). Influence of lead (Pb2+) on the reactions of in vitro cultured rat
        aorta to 5-hydroxytryptamine. Toxicol Lett 159: 71-82. http://dx.doi.Org/10.1016/i.toxlet.2005.04.009

     Zhang. LF; Peng. SQ: Wang. S. (2009a). Decreased aortic contractile reaction to 5-hydroxytryptamine in
        rats with long-term hypertension induced by lead (Pb2+) exposure. Toxicol Lett 186: 78-83.
        http://dx.doi.0rg/10.1016/i.toxlet.2009.01.004

     Zhao. Y; Wang. L; Shen. HB; Wang. ZX; Wei. QY; Chen. F. (2007). Association between delta-
        aminolevulinic acid dehydratase (ALAD) polymorphism and blood lead levels: A meta-regression
        analysis. J Toxicol Environ Health 70: 1986-1994. http://dx.doi.org/10.1080/15287390701550946

     Zhao. ZY: Li. R; Sun. L; Li. ZY: Yang. RL. (2004).  Effect of lead exposure on the immune function of
        lymphocytes and erythrocytes in preschool children. J Zhejiang Univ Sci 5: 1001-1004.
        http://dx.doi.org/10.1007/BF02947614

     Zheng. C: Skold. M; Li. J: Nennesmo. I; Fadeel. B; Henter. J.  (2007). VEGF reduces astrogliosis and
        preserves neuromuscular junctions in ALS transgenic mice. Biochem Biophys Res Commun 363:
        989-993. http://dx.doi.0rg/10.1016/i.bbrc.2007.09.088

     Zhou. J; Xu. YH; Chang. HF.  (1985). The effects of lead ion on immune function of rabbit alveolar
        macrophages: Quantitation of immune phagocytosis and rosette formation by 51Cr in vitro. Toxicol
        Appl Pharmacol 78: 484-487.

     Zhu. M; Fitzgerald. EF; Gelberg. KH; Lin.  S: Druschel. CM. (2010). Maternal low-level lead exposure
        and fetal growth. Environ Health Perspect 118: 1471-1475. http://dx.doi.org/10.1289/ehp.0901561

     Zimmermann. MB; Muthawa. S: Moretti. D; Kurpad. A; Hurrell. RF. (2006). Iron fortification reduces
        blood lead levels in children in Bangalore, India. Pediatrics 117: 2014-2021.
        http://dx.doi.org/10.1542/peds.2005-2440

     Zoeger. N; Roschger. P: Hofstaetter. JG: Jokubonis. C: Pepponi. G: Falkenberg. G: Fratzl. P:
        Berzlanovich. A; Osterode. W: Streli. C: Wobrauschek. P. (2006). Lead accumulation in tidemark of
        articular cartilage. Osteoarthritis Cartilage 14: 906-913. http://dx.doi.Org/10.1016/j.joca.2006.03.001

     Zuo. PJ; Qu. W: Cooper. RN; Gover. RA; Diwan. BA; Waalkes. MP (2009). Potential role of alpha-
        synuclein and metallothionein in lead-induced inclusion body formation. Toxicol Sci 111: 100-108.
        http://dx.doi.org/10.1093/toxsci/kfpl32

     Zuscik. MJ; Ma. L; Buckley. T; Puzas. JE; Drissi. H; Schwarz. EM; O'Keefe. RJ. (2007). Lead induces
        chondrogenesis and alters transforming  growth factor-beta and bone morphogenetic protein signaling
        in mesenchymal cell populations. Environ Health Perspect 115: 1276-1282.
        http://dx.doi.org/10.1289/ehp.10028
November 2012                              5-801                   Draft - Do Not Cite or Quote

-------
       6 POTENTIALLY  AT-RISK  POPULATIONS

 1                  The NAAQS are intended to protect public health with an adequate margin of safety. In
 2                  so doing, protection is provided for both the population as a whole and those groups
 3                  potentially at increased risk for health effects from exposure to the air pollutant for which
 4                  each NAAQS is set (Preface to this ISA). To facilitate the identification of populations at
 5                  increased risk for Pb-related health effects, studies have evaluated various factors that
 6                  may contribute to susceptibility and/or vulnerability to Pb. The definitions of
 7                  susceptibility and vulnerability vary across studies, but in most instances "susceptibility"
 8                  refers to biological or intrinsic factors (e.g., age, sex) while "vulnerability" refers to
 9                  nonbiological or extrinsic factors (e.g., socioeconomic status [SES]) (U.S. EPA. 2010.
10                  2009a). Additionally, in some cases, the terms "at-risk" and "sensitive" populations have
11                  been used to encompass these concepts more generally. In this ISA, "at-risk" groups are
12                  defined as those with characteristics that increase the risk of Pb-related health effects in a
13                  population. These characteristics include various factors, such as genetic background,
14                  race and ethnicity, sex, age, diet, pre-existing disease, SES, and characteristics that may
15                  modify exposure or the response to Pb.

16                  Individuals, and ultimately populations, could experience increased risk for air pollutant
17                  induced health effects via multiple avenues. A group with intrinsically increased risk
18                  would have one or more factors that increase risk for an effect through a biological
19                  mechanism. In general, people in this category would have a steeper concentration-
20                  risk relationship, compared to those not in the  category. Potential factors that are often
21                  considered intrinsic include genetic background  and sex. A group of people could also
22                  have extrinsically increased risk, which would be through an external, non-biological
23                  factor. Examples of extrinsic factors include SES and diet. Some groups are at risk of
24                  increased internal dose at a given exposure concentration. In addition, some groups
25                  could have greater  exposure (concentration x time), regardless of the delivered dose.
26                  Finally, there are those who might be placed at increased risk for experiencing a
27                  greater exposure by being exposed at a higher  concentration. For example, groups of
28                  people living near Pb smelters.

29                  Some factors described above are multifaceted and may influence the risk of an air
30                  pollutant related health effect through a combination of avenues. For example, SES may
31                  affect access to medical care, which itself may contribute to the presence of preexisting
32                  diseases and conditions considered as intrinsic factors. Additionally, children's outdoor
33                  activities can lead to more hand-to-mouth contact with contaminated soil than adults,
34                  which leads to increased intake dose and exposure. At the same time, children have
      November 2012                              6-1                   Draft - Do Not Cite or Quote

-------
 1                   biological (i.e., intrinsic) differences from adults that may influence their uptake,
 2                   metabolism, storage, and excretion.

 3                   The emphasis of this chapter is to identify and understand the factors that potentially
 4                   increase the risk of Pb-related health effects, regardless of whether the increased risk is
 5                   due to intrinsic factors, extrinsic factors, increased dose/exposure, or a combination due
 6                   to the often interconnectedness of factors. The following sections examine factors that
 7                   potentially lead to increased risk of Pb-related health effects and characterize the overall
 8                   weight of evidence for each factor.

 9                   Approach to Classifying Potential At-Risk Factors

10                   To identify factors that potentially lead to some populations being at greater risk of
11                   Pb-related health effects, the evidence across relevant scientific disciplines (i.e., exposure
12                   sciences, dosimetry, toxicology, and epidemiology) was evaluated. In this systematic
13                   approach, the collective evidence  is used to examine coherence of effects across
14                   disciplines and determine biological plausibility. The collective results across the
15                   scientific disciplines comprise the overall weight of evidence that is used to determine
16                   whether a specific factor results in a population being at increased risk of an air pollutant
17                   related health effect. The first section of this chapter (Section 6.1) summarizes
18                   physiological factors that influence Pb levels in the body. The second section of this
19                   chapter (Section 6.2) summarizes  information on factors potentially related to differential
20                   Pb exposure. The studies presented in this section supplement the material provided in
21                   Chapter 3_ and Chapter_4_by examining how factors such as age, sex, race and ethnicity,
22                   SES, proximity to Pb sources, and residential factors may affect Pb exposure or blood Pb
23                   levels. The third section of this chapter (Section 6.3) discusses the epidemiologic and
24                   toxicological studies evaluated in  Chapter_5 that provide information on factors
25                   potentially related to increased risk of Pb-induced health effects. To examine whether Pb
26                   differentially affects certain populations, epidemiologic studies conduct stratified
27                   analyses to identify the presence or absence of effect measure modification. A thorough
28                   evaluation  of potential effect measure modifiers may help identify populations  that are at
29                   increased risk for Pb-related health effects. Highlighted studies include only those where
30                   the population was stratified into subgroups (e.g.,  males versus females or smokers
31                   versus nonsmokers) for comparative analysis. In the case of many biomarker studies and
32                   the epidemiologic studies considered, this approach allowed for a comparison between
33                   populations exposed to similar Pb concentrations and within the same study design.
34                   Toxicological studies also provide evidence of Pb effects and biological plausibility for
35                   factors that may lead to increased  risk for Pb-related health effects. Included
36                   toxicological studies may have categorized the study populations by different factors,
37                   such as age, sex, diet/nutrition status, and genetics, or are those that examined animal
      November 2012                               6-2                   Draft - Do Not Cite or Quote

-------
 1                   models of disease. These epidemiologic and toxicological studies provide the scientific
 2                   basis for an overall weight of the evidence evaluation for the identification of specific
 3                   populations potentially at risk of Pb-related health effects. Details on the magnitude of
 4                   effects for studies in this third section (Section 6.3) are included in summaries of the
 5                   studies presented in Chapter_5.

 6                   Numerous studies that focused on only one potentially at-risk population were described
 7                   in previous chapters (Chapter_5) but are not discussed in detail in this chapter because
 8                   they lacked stratified analysis with adequate comparison groups. For example, pregnancy
 9                   is a lifestage with potentially increased risk for mothers and fetuses, but because there are
10                   no comparison groups for stratified analyses, these studies were presented in Chapter_5
11                   but are not included here. Additionally, it is understood that some of the stratified
12                   variables/factors discussed in this third section (Section 6.3) may not be effect measure
13                   modifiers but instead may be mediators of Pb-related health effects. Mediators are factors
14                   that fall on the causal pathway between Pb exposure and health outcomes, whereas effect
15                   measure modifiers are factors that result in changes in the measured associations between
16                   Pb exposure and health effects. Because mediators are caused by Pb exposure and are
17                   also intermediates in the disease pathway that is studied, mediators are not correctly
18                   termed "at-risk" factors. Some of the factors discussed in this third section could be
19                   mediators and/or modifiers. These are noted in Table 6-5.

20                   Building on the causal framework discussed in detail in the Preamble and used
21                   throughout the ISA, conclusions are made regarding the strength of evidence for each
22                   factor that may contribute to increased risk of a Pb-related health effect based on the
23                   evaluation and synthesis of evidence across scientific disciplines. The conclusions drawn
24                   considered the "Aspects to Aid in Judging Causality" discussed in Table I of the
25                   Preamble. The categories considered for evaluating the potential increased risk of an air
26                   pollutant-related health effect are "adequate evidence," "suggestive evidence,"
27                   "inadequate evidence," and "evidence  of no effect." They are described in more detail in
28                   Table 6-1.
      November 2012                               6-3                    Draft - Do Not Cite or Quote

-------
     Table 6-1     Classification of evidence for potential at-risk factors.
Health Effects
Adequate
evidence
Suggestive
evidence
Inadequate
evidence
Evidence of no
effect
There is substantial, consistent evidence within a discipline to conclude that a factor results
in a population or lifestage being at increased risk of air pollutant-related health effect(s)
relative to some reference population or lifestage. Where applicable this includes
coherence across disciplines. Evidence includes multiple high-quality studies.
The collective evidence suggests that a factor results in a population or lifestage being at
increased risk of an air pollutant-related health effect relative to some reference population
or lifestage, but the evidence is limited due to some inconsistency within a discipline or,
where applicable, a lack of coherence across disciplines.
The collective evidence is inadequate to determine if a factor results in a population or
lifestage being at increased risk of an air pollutant-related health effect relative to some
reference population or lifestage. The available studies are of insufficient quantity, quality,
consistency, and/or statistical power to permit a conclusion to be drawn.
There is substantial, consistent evidence within a discipline to conclude that a factor does
not result in a population or lifestage being at increased risk of air pollutant-related health
effect(s) relative to some reference population or lifestage. Where applicable this includes
coherence across disciplines. Evidence includes multiple high-quality studies.
     6.1       Physiological Factors that Influence the Internal
                Distribution of Pb

 1                  Blood and bone Pb measures are influenced to varying degrees by biokinetic processes
 2                  (e.g., absorption, distribution, metabolism, excretion), which are discussed in detail in
 3                  Chapter_4. These processes can be affected by multiple factors, such as age, genetics,
 4                  diet, and co-exposure with other metals and non-metals.

 5                  Age influences the biokinetic response to Pb within the body. Infants may be considered
 6                  an at-risk population because Pb easily crosses the placental barrier and accumulates in
 7                  fetal tissue during gestation (Pillai et al., 2009; Wang et al., 2009e; Uzbekov et al., 2007).
 8                  This transfer of Pb from mother to fetus is partly due to the remobilization of the
 9                  mother's bone stores (O'Flaherty. 1998; Franklin et al., 1997). This also results in
10                  increased maternal blood Pb levels (Lamadrid-Figueroa et al.. 2006; Gulson et al.. 2004a;
11                  Hertz-Picciotto  et al.. 2000; Gulson etal.. 1997; Lagerkvist et al., 1996; Schuhmacher et
12                  al.. 1996; Rothenberg et al.. 1994a). Bone growth rate is high during childhood. The
13                  majority of a child's Pb body burden is not permanently incorporated in the bone, but
14                  some Pb does remain in the bone until older age (McNeill et al.. 2000; O'Flaherty. 1995;
15                  Leggett. 1993).  Older adults are more likely to have age-related degeneration of bones
16                  and organ systems and a possible redistribution of Pb stored in the bones into the blood
17                  stream (Popovic et al., 2005; Garrido Latorre et al., 2003; Gulson et al., 2002).
     November 2012
6-4
Draft - Do Not Cite or Quote

-------
 1                  Various genes can also affect Pb biomarker concentrations. Genetic variants of the
 2                  vitamin D receptor (VDR) in humans have been associated with varied bone and plasma
 3                  Pb levels (Rezende et al. 2008; Theppeang et al.. 2004; Schwartz et al. 2000a). Multiple
 4                  studies have also examined the association between the aminolevulinate dehydratase
 5                  (ALAD) polymorphism and blood Pb levels and found that the ALAD-2 polymorphism
 6                  may be biologically related to varying Pb levels, although some studies report no
 7                  difference for ALAD alleles (see also Section 5.2.3) (Mivaki et al., 2009; Shaik and
 8                  Jamil. 2009: Sobin et al.. 2009:  Chen et al.. 2008c: Rabstein et al.. 2008: Scinicariello et
 9                  al.. 2007: Zhao et al.. 2007: Montenegro et al.. 2006: Wananukul et al.. 2006).

10                  It is well established that diets sufficient in minerals such as calcium, iron, and zinc offer
11                  some protection from Pb exposure by preventing or competing with Pb for absorption in
12                  the GI tract. A study in China reported that children who regularly consumed breakfast
13                  had lower blood Pb levels than those children that did not eat breakfast (Liu et al..
14                  201 la). Diets designed to limit or reduce caloric intake and induce weight loss have been
15                  associated with increased blood Pb levels in adult animals (Han et al.. 1999).  A
16                  toxicological study reported negative effects of Pb on osmotic fragility, TEARS
17                  production, catalase activity, and other oxidative parameters, but most of these effects
18                  were reduced to the levels observed in the control group when the rats were given
19                  supplementation of zinc and vitamins (Masso-Gonzalez and Antonio-Garcia.  2009).
20                  Toxicological studies by Jamieson et al.  (2008: 2006) also reported that a zinc-deficient
21                  diet increases bone and renal Pb content and impairs skeletal growth and mineralization.
22                  A zinc-supplemented diet attenuated bone and renal Pb content. Toxicological studies
23                  have shown that dietary deficiency of calcium induces increased Pb absorption and
24                  retention (Fullmer. 1992: Mykkanen and Wasserman. 1981: SixandGoyer. 1970).
25                  Increased calcium intake reduces accumulation of Pb in bone and mobilization of Pb
26                  during pregnancy and lactation (Bogden et al.. 1995). Additionally, studies have reported
27                  that iron deficiencies may result in higher Pb absorption or altered biokinetics (Schell et
28                  al.. 2004: Marcus and Schwartz. 1987: Mahaffev and Annest. 1986).

29                  In summary, age, genetics, and diet affect the biokinetics of Pb, which in turn affects the
30                  internal distribution of Pb. These factors were discussed in greater detail in Chapter_4
31                  where more information on overall biokinetic and physiological factors affecting Pb
32                  distribution is provided.
      November 2012                              6-5                   Draft - Do Not Cite or Quote

-------
      6.2        Population Characteristics Potentially Related to Differential
                 Pb Exposure

 1                  Elevated or differential Pb exposure and related biomarker levels (such as blood Pb),
 2                  have been shown to be statistically related to several population characteristics, including
 3                  age, sex, race and ethnicity, SES, proximity to Pb sources, and residential factors (U.S.
 4                  EPA, 2006b). In most cases, exposure, absorption, and biokinetics of Pb are all
 5                  influenced to varying degrees by such characteristics. Additionally, the relative
 6                  importance of such population characteristics in affecting exposure, absorption, and
 7                  biokinetics varies among individuals in a population and is difficult to quantify. This
 8                  section presents recent studies demonstrating a relationship between each population
 9                  characteristic and exposure status. The studies presented in this section build upon the
10                  current body of literature suggesting that population characteristics differentially
11                  influence Pb exposure; the new literature does not alter previous understanding of the
12                  differential influence of population characteristics on Pb exposure. Differential response
13                  to given Pb exposures is discussed in Section 6.3.
      6.2.1       Age
                    6.2.1.1      Early Childhood

14                  Typically, children have increased exposure to Pb compared with adults because
15                  children's behaviors and activities include increased hand-to-mouth contact, crawling,
16                  and poor hand-washing that typically result in increased ingestion compared with adults
17                  (U.S. EPA. 2006b). Children can also have increased Pb exposure because outdoor
18                  activities can lead to hand-to-mouth contact with contaminated soil. For example, Zahran
19                  et al. (2010) observed that a 1% reduction in soil Pb concentration led to a 1.55 (ig/dL
20                  reduction in median blood Pb levels (p <0.05) among New Orleans children.

21                  Age of the children may influence blood Pb levels through a combination of behavioral
22                  and biokinetic factors. The 2009-2010 NHANES data are presented in Table 6-2 by age
23                  and sex. Among children, highest blood Pb levels occurred in the 1-5 year age group
24                  (children under age 1 were not included), and within this subgroup (not shown on the
25                  table), 1-year old children had the highest blood Pb  levels (99th percentile: 9.47 ug/dL)
26                  (NCHS. 2010).  It is possible that high blood Pb levels among these young children may
27                  also be related to in utero exposures resulting from maternal Pb remobilization from bone
28                  stores from historic exposures (Miranda et al.. 2010) or from contemporaneous Pb
29                  exposures if the mothers had appreciable current Pb exposure. Jones et al. (2009a)
      November 2012                             6-6                   Draft - Do Not Cite or Quote

-------
 1                  analyzed the NHANES datasets for the years 1999-2004 to study trends in blood Pb
 2                  among two different age groups of children over time (see Table 6-3). They observed
 3                  greater percentages of children aged 1-2 years having blood Pb levels of 2.5 to <5 (ig/dL,
 4                  5 to <7.5 (ig/dL, and > 10 (ig/dL, compared with 3-5 year-old children, but no age
 5                  difference was noted for the 7.5 to <10 (ig/dL bracket. At the same time, 1-2 year-old
 6                  children had lower percentages of blood Pb levels <1 (ig/dL and 1 to <2.5 (ig/dL
 7                  compared with children ages 3-5 years old. This implies that there is a shift in the
 8                  distribution of blood Pb levels as children age, even during early childhood. These
 9                  distribution differences may be attributable to differences in exposure (including
10                  behavioral influences, such as hand-to-mouth contact and crawling in proximity to
11                  contaminated  surfaces), residual contributions from the mother's Pb burden, age-
12                  dependent variability in biokinetics or diet (e.g., milk versus solid diets). Yapici et al.
13                  (2006) studied the relationship between blood Pb level and age among a cohort of
14                  children between 6 and 73 months of age with elevated blood Pb levels (87.6% of study
15                  group with blood Pb greater than 20 (ig/dL) living near a Turkish coal mine. They
16                  observed a low but statistically significant negative correlation between blood Pb and age
17                  (r = -0.38,p<0.001).
      November 2012                              6-7                   Draft - Do Not Cite or Quote

-------
Table 6-2 Blood Pb levels by age and sex, 2009-2010 NHANES.
Age
1-5 yr
6-11 yr
12-19 yr
20-59 yr
60+ yr
Overall
Sex
Total
Male
Female
Total
Male
Female
Total
Male
Female
Total
Male
Female
Total
Male
Female
Total
Male
Female
N
836
429
407
1009
521
488
1183
632
551
3856
1843
2013
1909
941
968
8793
4366
4427
Avg.
1.61
1.59
1.64
1.05
1.10
0.99
0.84
0.98
0.69
1.50
1.88
1.15
2.09
2.46
1.73
1.50
1.75
1.25
Std. Dev.
1.49
1.32
1.65
0.74
0.73
0.75
0.68
0.69
0.62
1.83
2.33
1.10
1.51
1.78
1.07
1.57
1.88
1.13
5%
0.53
0.51
0.54
0.42
0.45
0.38
0.33
0.40
0.30
0.44
0.56
0.40
0.72
0.87
0.65
0.43
0.50
0.39
25%
0.85
0.83
0.90
0.61
0.66
0.58
0.50
0.58
0.44
0.72
0.92
0.61
1.16
1.39
1.01
0.72
0.84
0.63
50%
1.21
1.22
1.20
0.83
0.88
0.79
0.69
0.80
0.57
1.08
1.37
0.89
1.69
1.99
1.43
1.10
1.29
0.96
75%
1.81
1.84
1.77
1.22
1.30
1.12
0.96
1.11
0.79
1.70
2.12
1.35
2.53
2.90
2.14
1.76
2.05
1.48
95%
4.00
4.09
3.69
2.36
2.37
2.35
1.82
2.09
1.31
3.53
4.49
2.63
4.79
5.56
3.75
3.66
4.31
2.97
99%
8.03
7.49
9.59
4.29
4.18
3.98
3.10
3.91
2.25
7.27
9.68
4.41
8.28
9.89
5.42
7.21
8.62
5.17
Source: (NCHS. 2010)
November 2012
6-8
Draft - Do Not Cite or Quote

-------

Table 6-3
Pb Units:
M9/dL
(95% Cl)
Overall 2
Sex
Female 1
Male 1
Age
1-2 yr 1
3-5 yr 1
Race/Ethnicity
Non-
Hispanic
Black
Mexican
American
Non-
Hispanic
White
Percentage of children within six categories/brackets of blood Pb
levels, 1999-2004 NHANES.
7.5 to
Geometric 1to<2.5 2.5to<5 5to<7.5 <10 >10
N mean <1 |jg/dL, % |jg/dL,% |Jg/dU% MO/dU % MO/dU % |jg/dL,%
,532 1.9(1.8-2.0) 14.0 55.0 23.6 4.5 1.5 1.4
(11.6-16.6) (52.1-57.9) (21.1-26.1) (3.3-5.9) (1.0-2.1) (1.0-2.0)

,211 1.9(1.7-2.0) 14.1 54.5 23.9 4.5 1.4 1.7
(10.8-17.7) (51.1-57.8) (20.3-27.8) (3.3-5.8) (0.8-2.3) (0.9-2.6)
,321 1.9(1.7-2.0) 14.0 55.5 23.2 4.6 1.5 1.3
(11.4-16.7) (51.4-59.5) (20.3-26.3) (3.0-6.5) (0.9-2.3) (0.7-2.6)

,231 2.1 (2.0-2.2) 10.6 51.0 27.9 6.7 1.4 2.4
(7.7-13.9) (46.7-55.3) (24.9-31.0) (5.0-8.6) (0.8-2.2) (1.4-3.5)
,301 1.7(1.6-1.9) 16.2 57.6 20.7 3.1 1.5 0.9
(12.9-19.9) (53.8-61.4) (17.9-23.7) (1.9-4.6) (0.8-2.3) (0.4-1.5)

755 2.8 (2.5-3.0) 4.0 (2.5-5.7) 42.5 36.2 9.4 4.6 3.4
(37.8-47.2) (33.1-39.3) (6.9-12.2) (3.0-6.5) (1.8-5.5)
812 1.9(1.7-2.0) 10.9 61.0 22.1 3.4 1.3 1.2
(8.6-13.4) (56.9-65.1) (18.0-26.5) (2.2-5.0) (0.6-2.2) (0.4-2.6)
731 1.7(1.6-1.8) 17.6 57.1 19.7 3.6 0.8 1.2
(14.0-21.5) (52.4-61.7) (16.1-23.5) (1.9-5.8) (0.3-1.6) (0.6-2.0)
Poverty-Income Ratio (PIR)
1
2
3
4
5
6
£1.3 1
>1.3 1
Source: Reprinted
,302 2.4(2.2-2.5) 6.7(4.6-9.2) 49.3 32.5 6.9 2.8 1.8
(44.9-53.7) (28.6-36.4) (2.2-8.8) (1.7-4.1) (1.1-2.7)
,070 1.5(1.4-1.6) 19.9 60.4 16.0 2.3 0.6 0.8
(16.3-23.8) (56.9-63.8) (12.9-19.3) (1.2-3.7) (0.1-1.4) (0.3-1.6)
with permission of the American Academy of Pediatrics; Jones et al. (2009a)
Fetal and child Pb biomarkers have been demonstrated to relate to maternal Pb
biomarkers as reported in the 2006 Pb AOCD (U.S. EPA. 2006b). Kordas et al. (2010)
observed that maternal hair Pb concentration was a statistically significant predictor of
child hair Pb concentration ((3 = 0.37 ± 0.07, p <0.01). Elevated blood Pb levels among
mothers present a potential exposure route to their children in utero or through breast
milk; see Miranda et al. (2010).
November 2012
6-9
Draft - Do Not Cite or Quote

-------
                    6.2.1.2     Older Adulthood

 1                  Blood Pb levels tend to be higher in older adults compared with the general adult
 2                  population, as described in the 2006 Pb AQCD (U.S. EPA. 2006b). Table 6-2 presents
 3                  2009-2010 NHANES data broken down by age group and shows that blood Pb levels
 4                  were highest in the among participants 65 years old or older, in comparison with adults
 5                  aged 20-64 years and with adolescents. In a study of blood Pb and saliva Pb in a mostly
 6                  female population in Detroit, Nriagu et al. (2006)  found that age was a statistically
 7                  significant positive predictor of blood Pb (p <0.001). Average blood Pb levels among 14-
 8                  to 24-year-old subjects was 2.60 ± 0.16 ug/dL compared with 4.29 ± 0.56 ug/dL among
 9                  subjects aged 55 years or older. Higher average and median levels among older adults
10                  could potentially be due to a shared experience of higher historical Pb exposures stored in
11                  bone in conjunction with remobilization of stored Pb during bone loss (Section 4.2).

12                  Theppeang et al. (2008b) studied Pb concentrations in the blood, tibia, and patella of
13                  subjects age 50-70 as part of the Baltimore Memory Study. They found a statistically
14                  significant relationship between age and tibia Pb ((3 = 0.37, p <0.01 in a model including
15                  age, race/ethnicity, Yale energy index, and 2 diet variables; (3 = 0.57, p <0.01  in a model
16                  including age, sex, and an interaction term for sex and age, which was also statistically
17                  significant at p  = 0.03). Theppeang et al.  (2008b) also noted that patella Pb
18                  concentrations were also positively associated with age, although the authors did not
19                  present the data or significance levels.  A statistically significant relationship was not
20                  observed between the log-transform of blood Pb and age ((3 = 0.007, p = 0.11), although
21                  the age range of subjects may not have been sufficient to discern a difference  in blood Pb
22                  level.

23                  Miranda et al. (2010) observed that older pregnant women (ages 30-34 years and 35-39
24                  years) had statistically significant higher odds of having greater blood Pb levels than
25                  younger pregnant women (25- to 29-year-olds) in the reference age category.  These
26                  results could be related to a historical component to Pb exposure among mothers. These
27                  findings were also consistent with observations that Pb storage in bones  increased with
28                  age before subsequent release with bone loss occurring during pregnancy, as described in
29                  Section 4.2 and summarized in Section 6.1.
      November 2012                             6-10                  Draft - Do Not Cite or Quote

-------
      6.2.2       Sex

 1                  The AQCD (U.S. EPA. 2006b) described several studies showing higher Pb biomarker
 2                  levels in male adults compared with female adults. The 2009-2010 NHANES showed
 3                  that overall, males have significantly higher blood Pb levels (average: 1.75 ug/dL) than
 4                  females (average: 1.25 ug/dL) (p <0.0005). Among adults aged 20-59 years,  average
 5                  blood Pb levels were 64% higher for males compared with females (p <0.0005). Among
 6                  adults 60 years or older, average blood Pb levels were 30% higher for males  compared
 7                  with females (p <0.0005) (NCHS. 2010). In their study of Pb burden among Baltimore
 8                  adults aged 50-70 years, Theppeang et al. (2008b) observed that average blood Pb levels
 9                  were statistically significantly higher (p <0.01) among men (4.4 (ig/dL) than women
10                  (3.1 (ig/dL). For average tibia Pb levels, Theppeang et al. (2008b) noted no difference
11                  (p = 0.12) between men (18.0 (ig/g) and women (19.4 (ig/g).

12                  Among U.S. children, the 2009-2010 NHANES data showed that blood Pb levels were
13                  higher among girls than boys for the 1- to 5-years age group (Table 6-2). Blood Pb levels
14                  became slightly higher among boys for the 6- to 11-years age group, and levels were
15                  substantially higher among adolescent males than females 12- to 19-years old. The
16                  2009-2010 NHANES data suggest that sex-based differences in blood Pb levels are not
17                  substantial until adolescence.
      6.2.3       Race and Ethnicity

18                  Higher blood Pb and bone Pb levels among African Americans have been well
19                  documented (U.S. EPA. 2006b). Model results presented in the 2006 Pb AQCD have
20                  demonstrated not just elevations in blood Pb among African Americans but also
21                  significant associations between blood Pb and race (U.S. EPA. 2006b). Recent studies are
22                  consistent with those previous findings. For instance, Levin et al. (2008) and Jones et al.
23                  (2009a) both analyzed NHANES survey data to examine trends in childhood blood Pb
24                  levels. Data from the Jones et al. (2QQ9a) study, using NHANES data (NCHS. 2009.
25                  2008) from 1988-1991 and 1999-2004 are shown in Figure 6-1. The authors found that
26                  differences among children from different racial/ethnic groups with regard to the
27                  percentage with blood Pb levels > 2.5 ug/dL over the period 1999-2004 have decreased
28                  since the period of 1988-1991. The non-Hispanic black group still had higher percentages
29                  with blood Pb levels > 2.5 ug/dL compared with non-Hispanic whites and Mexican
30                  Americans, with large observable differences for blood Pb levels between 2.5 and
31                  <10 ug/dL. It is notable that the distributions of blood Pb levels among Mexican
32                  American and non-Hispanic white children were nearly identical in the 1999-2004
33                  dataset. Theppeang et al. (2008b) also explored the effect of race and ethnicity  on several
      November 2012                             6-11                  Draft - Do Not Cite or Quote

-------
1                  Pb biomarkers in a study of older adults living in Baltimore, MD. They observed a
2                  statistically significant difference between African American (AA) and Caucasian (C)
3                  subjects with respect to tibia Pb (AA: 21.8 (ig/g, C: 16.7 (ig/g, p <0.01) but not patella Pb
4                  (AA: 7.1 ng/g, C: 7.1 ng/g, p = 0.46) or blood Pb  levels (AA: 3.6 ng/dL, C: 3.6 ng/dL,
5                  p = 0.69). Greater tibia (but lower patella) Pb levels may indicate greater historical
6                  exposure among African Americans compared to  Caucasians in the Baltimore population
7                  studied by Theppeang et al. (2008b).
     November 2012                             6-12                  Draft - Do Not Cite or Quote

-------
                        70
                        60 -
                     C  50
                     OJ
                     5  40
                     M—
                     O

                     £  30
                     OJ
                     u
                     i_

                     o!  20
                        10 -
                                                             1988-1991
                              <1     l-<2.5   2.5-< 5   5-<7.5  7.5-<10    > 10
                                                           1999 - 2004
                     (U
                     1_

                     •a
                     u
                     M-
                     o
                     (U
                     u
                        70 n
                        60 -
                        50 -
                        40 -
                        30 -
                        20
                        10 -
                         o 4-
                               <1     l-<2.5   2.5-<5   5-<7.5   7.5 - <10    > 10


                                           Blood Pb Level (ug/dL)


                     •••+•• Non-Hispanic black^^^Mexican American —^— Non-Hispanic white




Note: from the NHANES survey, 1988-1991 (top) and 1999-2004 (bottom).

Data used with permission of the American Academy of Pediatrics, Jones et al. (2009a)



Figure 6-1       Percent distribution of blood Pb levels by race/ethnicity among

                   U.S. children (1-5 years).
November 2012
6-13
Draft - Do Not Cite or Quote

-------
 1                  Differences in potential exposure among ethnic and racial groups have also been noted in
 2                  a study in the greater metropolitan New Orleans area. Campanella and Mielke (2008)
 3                  found that, in Census blocks where surface soil Pb levels were less than 20 mg/kg, the
 4                  population was 36% black, 55% white, 3.0% Asian, and 6.0% Hispanic, based on the
 5                  2000 Census, with the percentage based on the total number living in Census blocks with
 6                  the same soil Pb levels. In contrast, they found that for Census blocks in which soil Pb
 7                  levels were between 1,000 and 5,000 mg/kg, the population was 62% black, 34% white,
 8                  1% Asian,  and 4% Hispanic (Figure 6-2), although the total population size generally
 9                  declined with soil Pb concentration, with the Census blocks with soil Pb of 1,000-5,000
10                  mg/kg having less than half the population of that in the <20 mg/kg blocks. As described
11                  in Section 6.2.4. the differences observed by Campanella and Mielke (2008) may also be
12                  attributable to SES factors, or SES may be a confounding factor in the relationship
13                  between Pb soil levels and race/ethnicity of nearby residents.
     November 2012                             6-14                   Draft - Do Not Cite or Quote

-------
50,000 -i
45,000
40,000
35,000
30,000
o
| 25,000
Q.
O
°~ 20,000
15,000
10,000
5,000 -
0 -



i— I






Jl
—






Jl
i— i




Jl




-n


_n

• Black
• White
D Asian
D Hispanic
^ ^ In
V «,?
Soil Pb Concentration (mg/kg)
Note: By Census 2000 race/ethnicity demographic groups.
Source: Data used with permission of Springer Science; Campanella and Mielke (2008).

Figure 6-2      Soil Pb concentration exposure among the population of three
                 parishes within greater metropolitan New Orleans.
November 2012
6-15
Draft - Do Not Cite or Quote

-------
      6.2.4        Socioeconomic Status (SES)

 1                  Socioeconomic factors have sometimes been associated with Pb exposure biomarkers.
 2                  Previous results reported in the 2006 Pb AQCD found negative associations between
 3                  income or other SES metrics and blood Pb, although these relationships were not always
 4                  statistically significant (U.S. EPA. 2006b). Nriagu et al. (2006) performed a multiple
 5                  regression analysis of blood Pb and saliva Pb levels on various Socioeconomic,
 6                  demographic, and exposure variables among an adult population in Detroit, Michigan.
 7                  Blood and saliva Pb were both used as indicators of Pb in unbound plasma that is
 8                  available to organs. Nriagu et al. (2006) found that education (p <0.001), income
 9                  (p <0.001), and employment status (p = 0.04) were all statistically significant predictors
10                  of blood Pb levels, with blood Pb decreasing with some scatter as education and income
11                  level increased. Statistically significant relationships were also reported by Nriagu et al.
12                  (2006) for saliva Pb level with respect to education (p <0.001), income (p <0.001), and
13                  employment (p = 0.06). However, the highest educational attainment and income
14                  categories had higher saliva Pb levels compared with other groups; Nriagu et al. (2006)
15                  attributed these inconsistencies to small sample sizes among the high educational
16                  attainment and income categories.

17                  On a national level, the difference in blood Pb levels that have historically been seen to
18                  exist between different income levels has been decreasing. For example, Levin et al.
19                  (2008) cited 1991 -1994 NHANES data [analyzed in Pirkle et al. (1994)1 that the
20                  percentage of children aged 1-5 years with blood Pb levels > 10 ug/dL was 4.5% for the
21                  lowest income group compared with 0.7% for the highest income group. Levin et al.
22                  (2008) also analyzed data from the 1999-2002 NHANES and found no statistically
23                  significant difference between the percent of children with blood Pb levels above
24                  10 ug/dL for Medicaid-enrolled children (1.7%) compared with non-enrolled children
25                  (1.3%). However, Medicaid-enrolled children did have higher median blood Pb levels
26                  (2.6 ug/dL) compared to children not enrolled in Medicaid (1.7 ug/dL).  Adding data for
27                  2003-2004 to the analysis (i.e., for 1999-2004), widened the difference between Medicaid
28                  enrolled and non-enrolled children with regard to percentage having blood Pb levels
29                  > 10 ug/dL (1.9% versus 1.1%), but the difference was still not statistically significant (p
30                  >0.05) and median blood Pb levels for the two groups did not change (Levin et al., 2008).
31                  Likewise, Jones et al. (2009a)  analyzed blood Pb levels with respect to poverty-income
32                  ratio (PIR), which is the ratio of family income to the poverty threshold appropriate for a
33                  given family size. They found statistically significant differences in median blood Pb for
34                  PIR < 1.3 compared with PIR  >1.3. The percentage of 1- to 5-year-old children having
35                  blood Pb > 10  ug/dL was higher for PIR < 1.3 (1.8 versus 0.8); however, this difference
36                  was not statistically significant. Additionally, in residential areas of metropolitan New
37                  Orleans with soil concentrations below 20,000  mg/kg, Campanella and Mielke (2008)

      November 2012                             6-16                   Draft - Do Not Cite or Quote

-------
 1                  observed a linear increase in surface soil Pb concentration with decreasing median
 2                  household income, suggesting a relationship of potential exposure with household
 3                  income. The census block-averaged median household income in areas with soil Pb
 4                  between 2.5 and 20 mg/kg was $40,000 per year, while the corresponding median income
 5                  in areas with soil Pb between 5,000 and 20,000 mg/kg was $24,000 per year. The highest
 6                  soil concentrations (20,000 mg/kg and above) was associated with a median income of
 7                  $27,000.
      6.2.5        Proximity to Pb Sources

 8                  Air and soil Pb concentrations are higher in some industrialized and urbanized areas, as
 9                  described in Sections 3.2. 3.3. 3.5 and 4.1. as a result of historical and contemporary Pb
10                  sources. The highest air Pb concentrations measured using the Pb-TSP monitoring
11                  network have been measured at monitors located near sources emitting Pb. Elevated soil
12                  Pb concentrations have also been measured in urbanized areas compared with less
13                  urbanized or rural locations (Filippelli et al.. 2005). Air Pb concentrations exhibit high
14                  spatial variability even at low concentrations (-0.01 ug/m3) (Martuzevicius et al.. 2004).
15                  Proximity to an industrial source likely contributes to higher Pb exposures, as described
16                  in the 2006 Pb AQCD (U.S.  EPA. 2006^) for several studies of Superfund and other
17                  industrial sites. This is consistent with the observation of higher air concentrations at
18                  source oriented Pb monitoring sites compared with non-source oriented sites in the
19                  2008-2010 data presented in Section 3.5.

20                  Jones et al. (2010) found that neonates born near a Pb-contaminated hazardous waste site
21                  had significantly higher umbilical cord blood Pb levels (median: 2.2 ug/dL [95% CI: 1.5,
22                  3.3 ug/dL]) compared with a reference group of neonates not living near a potentially
23                  contaminated site (median: 1.1 ug/dL [95% CI:  0.8, 1.3 ug/dL]), suggesting that
24                  Pb-contaminated hazardous waste sites contribute to neonatal Pb levels. The population
25                  studied in Jones et al. (2010) was 88% African American; 75% had a high school degree
26                  or equivalent, while 20% had a college degree and 5% attended but did not graduate from
27                  high school. However, Jones et al. (2010) did not analyze covariation between exposure
28                  and maternal characteristics, so it cannot be determined if differences in characteristics
29                  among the maternal groups (which did and did not report nearby hazardous waste sites)
30                  confounded these results.

31                  Studies have suggested that concentration of Pb in soil, a potential exposure media, is
32                  related to land use type and historical sources, as described in  Section 3.6.1. For instance,
33                  Wu et al., (2010) observed that bioavailable  Pb concentrations in Los Angeles surface
34                  soil samples were significantly associated with traffic-related variables and parcel age
      November 2012                             6-17                  Draft - Do Not Cite or Quote

-------
 1                  (i.e., length of time since the parcel was first developed), with parcel age being a highly
 2                  significant predictor of bioavailable soil Pb in most models (p <0.0001). Zahran et al.
 3                  (2010) observed that surface soil Pb levels in 46 Census tracts of metropolitan New
 4                  Orleans dropped following Hurricanes Katrina and Rita, from 330 mg/kg to 200 mg/kg
 5                  (averages of median measurements across all Census tracts for 2000 and 2006) and
 6                  attributed this observation to coverage by relatively cleaner river sediments. Blood Pb
 7                  levels obtained from  children (ages 0-6 years) also declined subsequent to the hurricanes;
 8                  statistical modeling of the changes in soil and blood Pb estimated the decline to be
 9                  1.55 (ig/dL for each 1% reduction in soil Pb (p < 0.05).
      6.2.6        Residential Factors

10                  Findings from a recent study of the association between blood Pb and housing factors by
11                  Dixon et al. (2009). which analyzed data from the NHANES national survey for
12                  1999-2004, are consistent with those from previous studies presented in the
13                  2006 Pb AQCD that observed positive associations between increased blood Pb and
14                  increased house dust Pb levels (U.S. EPA. 2006b; Lanphear et al.. 1998; Laxen et al..
15                  1987). Dixon et al. (2009) used NHANES data from 1999-2004 to perform a linear
16                  regression of blood Pb among children 12-60 months old on several factors including
17                  year of home construction, floor surface condition, floor dust Pb level, windowsill dust
18                  Pb level, and renovation in homes built before 1978. They found that blood Pb (log
19                  transformed) was significantly associated with homes being built after 1950 (p = 0.014),
20                  windowsill Pb level (p = 0.002), dust Pb level (p <0.001), and occurrence of renovation
21                  in pre-1978 homes (p = 0.045). Detailed results of this regression are shown in Table 6-4.
22                  As part of the same study, Gaitens et al. (2009) performed a regression analysis of floor
23                  dust Pb (PbD) and windowsill dust Pb on several factors. Floor dust Pb (log transformed)
24                  was significantly associated with the following housing-related factors: floor surface
25                  condition (p <0.001), windowsill dust Pb (log transformed)  (p O.001), year of
26                  construction (p <0.001), and renovation in a pre-1950 home (p <0.001). Windowsill dust
27                  Pb (log transformed) was significantly associated with the following housing-related
28                  factors: year of construction (p O.001), window surface condition (0.001), and
29                  deteriorated indoor paint (p = 0.028).
      November 2012                             6-18                  Draft - Do Not Cite or Quote

-------
Table 6-4    Regression of log-transformed blood Pb level of children 12-60
              months old on various factors related to housing condition, from
              1999-2004 NHANES dataset.
Variables
Overall p-value   Levels3
                                                                 Estimate (SE)
                     p-Value
Intercept
Age (in years)
Year of construction
PIR
Race/ethnicity
Country of birth
Floor surface/condition x log floor PbD
Floor surface/condition x (log floor PbD)2
Floor surface/condition x (log floor PbD)3
Log windowsill PbD
Home-apartment type
Anyone smoke inside the home
Log cotinine concentration (ng/dL) in blood
Window cabinet or wall renovation in a
pre-1 978 home
0.172
<0.001 Age
Age2
Age3
Age4
0.014 Intercept for missing
1990-present
1978-1989
1960-1977
1950-1959
1940-1949
Before 1940
<0.001 Intercept for missing
Slope
<0.001 Non-Hispanic white
Non-Hispanic black
Hispanic
Other
0.002 Missing
U.S."
Mexico
Elsewhere
<0.001 Intercept for missing
Not smooth and cleanable
Smooth and cleanable or carpeted
Not smooth and cleanable
Smooth and cleanable or carpeted
Uncarpeted not smooth and cleanable
Smooth and cleanable or carpeted
0.002 Intercept for missing
Slope
<0.001 Intercept for missing
Mobile home or trailer
One family house detached
One family house attached
Apartment (1-9 units)
Apartment (a 10 units)
0.015 Missing
Yes
No
0.004 Intercept for missing
Slope
0.045 Missing
Yes
No
-0.517 (0.373)
2.620 (0.628)
-1.353(0.354)
0.273 (0.083)
-0.019 (0.007)
-0.121 (0.052)
-0.198(0.058)
-0.196(0.060)
-0.174(0.056)
-0.207 (0.065)
-0.012 (0.072)
0.000
0.053 (0.065)
-0.053 (0.012)
0.000
0.247 (0.035)
-0.035 (0.030)
0.128(0.070)
-0.077 (0.219)
0.000
0.353 (0.097)
0.154(0.121)
0.178(0.094)
0.386 (0.089)
0.205 (0.032)
0.023(0.015)
0.027 (0.008)
-0.020 (0.014)
-0.009 (0.004)
0.053 (0.040)
0.041 (0.011)
-0.064 (0.097)
0.127(0.067)
-0.025 (0.046)
0.000
0.069 (0.060)
-0.133(0.056)
0.138(0.140)
0.100(0.040)
0.000
-0.150(0.063)
0.039(0.012)
-0.008(0.061)
0.097 (0.047)
0.000
0.172
<0.001
<0.001
0.002
0.008
0.024
0.001
0.002
0.003
0.003
0.870
—
0.420
<0.001
—
<0.001
0.251
0.073
0.728
-
<0.001
0.209
0.065
<0.001
<0.001
0.124
0.001
0.159
0.012
0.186
<0.001
0.511
0.066
0.596
—
0.256
0.022
0.331
0.015
—
0.023
0.002
0.896
0.045
-
"Children: n = 2,155 (age 10-60 months); R2 = 40%
""Includes the 50 states and the District of Columbia
Source: Dixon et al. (2009).
November 2012
           6-19
Draft - Do Not Cite or Quote

-------
 1                  Renovation activities on older homes have been shown to produce excess Pb dust
 2                  concentrations. Gaitens et al. (2009) performed a regression analysis on dust Pb
 3                  concentrations from 1994-2004 NHANES on demographic and housing variables and
 4                  found that renovation of windows, cabinets, or walls in a pre-1950 home was
 5                  significantly associated with floor dust Pb concentration (p <0.001). Paint scraping within
 6                  the last twelve months  was nearly significantly associated with windowsill dust Pb
 7                  concentration (p = 0.053). Dixon et al. (2009) performed a regression analysis on log-
 8                  transformed blood Pb levels from NHANES (1999-2004) on several demographic and
 9                  housing variables and found that renovation of windows, cabinets, or walls in pre-1978
10                  homes was significantly associated with blood Pb concentration (p = 0.045). A case study
11                  by Mielke et al. (2001) reports on elevated indoor and outdoor dust Pb levels at two
12                  houses where exterior paint has been either power sanded (without confinement of
13                  released material) or hand scraped (with collection of released material) to prepare for
14                  repainting. The latter approach appeared to yield lower dust Pb levels, although given the
15                  extremely limited dataset, conclusions are uncertain. In an occupational study of men
16                  performing home renovations in the  U.K., window renovation and wood-stripping
17                  workers specializing in renovation of old houses had significantly higher median blood
18                  Pb levels compared with all workers in similar occupations (wood strippers: 37 (ig/dL;
19                  window renovators: 32 (ig/dL; all workers:  13.7 (ig/dL; p <0.001) (Mason et al..  2005).
      6.3        Factors Potentially Related to Increased  Risk of Pb-lnduced
                 Health Effects

20                  This section evaluates factors examined in recent studies as effect measure modifiers that
21                  potentially increase the risk of various Pb-related health effects. There was limited
22                  evidence from the 2006 Pb AQCD (U.S. EPA. 2006b) for many of potential at-risk
23                  factors described below. Where available, information on conclusions regarding at-risk
24                  populations from the 2006 Pb AQCD is included in the subsections.
      6.3.1       Age

25                  Below is information from epidemiologic and toxicological studies regarding studies of
26                  increased risk for Pb-related health effects among children and older adults. Other age
27                  groups, such as adolescents, have not been evaluated here, if they were not part of
28                  stratified studies of lifestage.
      November 2012                             6-20                 Draft - Do Not Cite or Quote

-------
                     6.3.1.1     Childhood

 1                   According to the 2000 Census, 28.6% of individuals living in the U.S. were under the age
 2                   of 20, with 6.8% aged 0-4 years, 7.3% aged 5-9 years, 7.3% aged 10-14 years, and 7.2%
 3                   aged 15-19 years (SSDAN CensusScope. 2010a). It is recognized that Pb can cross the
 4                   placenta and affect the developing nervous system of the fetus (Sections 4.2.2.4 and
 5                   5.3.2.1) and there is strong evidence of increased risk to the neurocognitive effects of Pb
 6                   exposure during several lifestages throughout gestation, childhood, and into adolescence
 7                   (for more detail, Section 5.3.2.1). However, most recent studies among children do not
 8                   have adequate comparison groups between children of various age groups or between
 9                   children and adults, and were therefore only presented in Chapter_5.

10                   A study including multiple U.S. locations examined associations of blood Pb levels with
11                   various immune parameters among individuals living near Pb industrial sites and
12                   matched controls (Sarasua et al.. 2000).  For several of these endpoints, the association in
13                   the youngest group (ages 6-35 months) and the oldest group (ages 16-75 years) were in
14                   opposite directions. For example, among children ages 6-35 months, the associations
15                   between blood Pb levels and Immunoglobulin A (IgA), Immunoglobulin M (IgM), and
16                   B-cell abundance were positive, whereas the associations among 16-75 year olds were
17                   negative. The opposite  associations were also present for T cell abundance. Ig antibodies,
18                   which are produced by activated B cells, are important mediators of the humoral immune
19                   response to antigens. T cells are important mediators of cell-mediated immune responses
20                   that involve activation of other immune cells and cytokines. These findings by Sarasua et
21                   al. (2000) indicate that very young children may be at increased risk  for Pb-associated
22                   activation of humoral immune responses and perturbations in cell-to-cell interactions that
23                   underlie allergic, asthma, and inflammatory responses (for more information, see
24                   Sections 5.6.2.1 and 5.6.3).

25                   A study among Lebanese children examined the association between blood Pb levels and
26                   transferrin saturation (TS) less than 12% and iron-deficiency anemia (IDA) (Muwakkit et
27                   al.. 2008). A positive association was detected for blood Pb levels >  10 (ig/dL and both
28                   TS less than 12% and IDA among children aged 11-23 months old; however, null
29                   associations were observed among children 24-35 months old. Calculations were not
30                   performed for children aged 36-75 months  because there were no children in the highest
31                   Pb group (> 10 (ig/dL) with either TS <12% or IDA. The authors noted that it is difficult
32                   to know whether the Pb levels were "a cause or a result of IDA levels since previous
33                   studies linked iron deficiency with Pb toxicity.

34                   Overall evidence indicates early childhood as a lifestage of increased risk for Pb-related
35                   health effects. Both epidemiologic studies summarized above reported associations
36                   among the youngest age groups, although different age cut-points were used with one

      November 2012                             6-21                   Draft - Do Not Cite or Quote

-------
 1                  study including only infants 35 months of age and younger. Toxicological studies provide
 2                  support for increased health effects of Pb among younger age groups. Toxicological
 3                  studies have reported that younger animals, whose nervous systems are developing
 4                  (i.e., laying down and pruning neuronal circuits) and whose junctional barrier systems in
 5                  the brain (i.e., the blood brain barrier) and GI system (i.e., gut closure) are immature, are
 6                  more at risk from the effects Pb exposure (Fullmer et al..  1985). In sum, there are
 7                  consistent findings, coherent across disciplines that adequate evidence exists to conclude
 8                  that children are an at-risk population.
                    6.3.1.2     Older Adulthood

 9                  The number of Americans over the age of 65 will be increasing in upcoming years
10                  (estimated to increase from 12.4% of the U.S. population to 19.7% between 2000 to
11                  2030, which is approximately 35 million and 71.5 million individuals, respectively)
12                  (SSDAN CensusScope. 2010a: U.S. Census Bureau. 2010). As of the 2000 Census, 7.2%
13                  of the U.S. population were ages 60-69, 5.8% were 70-79, and 3.3% were age 80 and
14                  older (SSDAN CensusScope. 2010a).

15                  A study using the NHANES III cohort examined blood Pb levels and mortality among
16                  individuals less than 60 years old and individuals 60 years and older (Menke et al.,  2006).
17                  Positive hazard ratios were observed in both age groups but the hazard ratios were greater
18                  in those less than 60 years old. The interactions terms were not statistically significant. A
19                  similar study using the NHANES III cohort examined the relationship between blood Pb
20                  levels and mortality from all-cause, cardiovascular disease, and cancer broken down into
21                  more specific age groups (Schober et al.. 2006). Point estimates were elevated for the
22                  association comparing blood Pb levels > 10 (ig/dL to blood Pb levels <5 (ig/dL and all-
23                  cause mortality for all age groups (40-74, 75-84, and 85+ year olds), although the
24                  association for 75-84 year olds did not reach statistical significance. The association was
25                  also present when comparing blood Pb levels of 5-9 (ig/dL to blood Pb levels <5 (ig/dL
26                  among 40-74 year olds and 75-84 year olds, but not among those 85 years and older.
27                  None of the associations between blood Pb and cardiovascular disease-related mortality
28                  reached statistical significance but the point estimates for cardiovascular disease-related
29                  mortality comparing blood Pb levels > 10 (ig/dL to blood Pb levels <5 (ig/dL were
30                  elevated among all age groups. Finally, the association between blood Pb levels
31                  > 10 (ig/dL and cancer mortality was positive among those 40-74 years old and 85 years
32                  and older but the association was null for those 75-84 years old. Among 75-84 year olds
33                  the association was positive comparing blood Pb levels of 5-9 (ig/dL to <5 (ig/dL. The
34                  other age groups had similar point estimates but the associations were not statistically
35                  significant.


      November 2012                             6-22                  Draft - Do Not Cite or Quote

-------
 1                  A study using the Normative Aging Study cohort reported an interaction between Pb and
 2                  age (Wright et al.. 2003). The inverse association between age and cognitive function was
 3                  greater among those with high blood or patella Pb levels. Effect estimates were in the
 4                  same direction for tibia Pb but the interaction was not statistically significant.

 5                  Finally, a study of current and former Pb workers reported that an interaction term of Pb
 6                  and age (dichotomous cutpoint at 67th percentile but exact age not given) examined in
 7                  models of Pb (measured from blood and patella) and blood pressure was not statistically
 8                  significant (Weaver et al., 2008). Thus, no modification by age was observed in this study
 9                  of Pb and blood pressure.

10                  Toxicological studies have demonstrated Pb-related health effects among older
11                  populations. The kidneys of older animals appear to be more at-risk for Pb-related health
12                  effects from the same dose of Pb (i.e., continuous 50 mg/L Pb acetate drinking water)
13                  than younger animals (Berrahal et al.. 2011). Increased risk related to older age is also
14                  observed for effects on the brain. Recent studies have demonstrated the importance of Pb
15                  exposure during early development in promoting the emergence of Alzheimer's like
16                  pathologies in aged animals. Development of pathologies of old age in brains of aged
17                  animals that were exposed to Pb earlier in life has been documented in
18                  psychopathological effects in  adults (mice and monkeys), (for more details  see
19                  Section 5.3.10.1). These pathologies include the development of neurofibrillary tangles
20                  and increased amyloid precursor protein and its product beta-amyloid (Bashaet al., 2005;
21                  Zawia and Basha. 2005). Some of these findings were seen in animals that no longer had
22                  elevated blood Pb levels.

23                  In summary, results for age-related modification of the association between Pb and
24                  mortality had mixed results. Limited evidence was available for the associations between
25                  Pb and cognitive function or other health effects among older adults. Toxicological
26                  studies have shown increases in Pb-related health effects by age that may be relevant in
27                  humans. Future studies will be instrumental in understanding older age as a factor that
28                  potentially affects the risk of Pb-related outcomes.
      6.3.2        Sex

29                  The distribution of males and females in the U.S. is similar. In 2000, 49.1% of the U.S.
30                  population was male and 50.9% was female. The distribution of sex varied by age with a
31                  greater prevalence of females > 65 years old compared to males (SSDAN CensusScope.
32                  2010a). The 2006 Pb AQCD reported that boys are often found to have higher blood Pb
33                  levels than girls, but findings were "less clear" regarding differences in Pb-related health
34                  effects between males and females (U.S. EPA, 2006b).

      November 2012                             6-23                   Draft - Do Not Cite or Quote

-------
 1                  Multiple epidemiologic studies have examined Pb-related effects on cognition stratified
 2                  by sex. In previous studies using the Cincinnati Lead Study cohort, Dietrich et al. Q987b)
 3                  and Ris et al. (2004) observed interactions between blood Pb (prenatal and postnatal) and
 4                  sex; associations of prenatal and postnatal blood Pb and subsequent decrements in
 5                  memory, attention, and visuoconstruction were observed only among male adolescents.
 6                  More recently, Wright et al. (2008) examined early life blood Pb levels and criminal
 7                  arrests in adulthood. The risks attributable to Pb exposure were greater among males than
 8                  females. Additionally, the association between childhood blood Pb levels and adult gray
 9                  matter volume loss was greater among males than females (Cecil et al.. 2008). In an
10                  expanded analysis of the developmental trajectory of childhood blood Pb levels on adult
11                  gray matter, researchers found that associations between yearly mean blood Pb levels and
12                  volume of gray matter loss were more pronounced in the frontal lobes of males than
13                  females (Brubakeretal.. 2010). Multiple studies were also conducted in Port Pirie,
14                  Australia that examined blood Pb levels at various ages throughout childhood and
15                  adolescence (Tong et al.. 2000; Baghurst et al..  1992; McMichael et al..  1992). These
16                  studies observed Pb effects on cognition deficits were stronger in girls throughout
17                  childhood and into early adolescence. A study in Poland also investigated the association
18                  between umbilical cord blood Pb levels and cognitive deficits and  reported  a positive
19                  association for boys at 36 months but not for girls (Jedrychowski et al.. 2009a). No
20                  association was detected for boys or girls at 24 months.

21                  An epidemiologic study examined the association between concurrent blood Pb levels
22                  and kidney function among 12-20 year olds using the NHANES III study cohort
23                  (Fadrowski et al.. 2010). The results were stratified by sex and no effect measure
24                  modification was apparent.

25                  Similarly, a study of current and former Pb workers examined an interaction term
26                  between sex and Pb for the study of blood Pb and blood pressure (Weaver et al.. 2008).
27                  No modification by sex was present.

28                  Epidemiologic studies have also been performed to assess differences between males and
29                  females for Pb-related effects on various biomarkers. A study comprised mostly of
30                  females reported positive associations between blood Pb and total  immunoglobulin E
31                  (IgE) for women not taking hormone replacement therapy or oral contraceptives (Pizent
32                  et al.. 2008). No association was reported in males, but other associations, such as
33                  bronchial reactivity and reactive skin prick tests were observed in the opposite of the
34                  expected direction, which questions the validity of the results among the male study
35                  participants. Analysis of an NHANES dataset detected no association between blood Pb
36                  levels and inflammatory markers (Songdei et al.. 2010). Although  there was no clear
37                  pattern, a few of the associations were positive between blood Pb and C-reactive protein
      November 2012                             6-24                  Draft - Do Not Cite or Quote

-------
 1                  for males but not females. A study of children living at varying distances from a Pb
 2                  smelter in Mexico reported that blood Pb was associated with increased release of
 3                  superoxide anion from macrophages, which was greater among males than females
 4                  (Pineda-Zavaleta et al.. 2004).

 5                  Epidemiologic investigations of cancer have also examined the associations by sex. A
 6                  study of the association between occupational exposure to Pb and brain tumors reported
 7                  no  sex-specific associations for gliomas, but a positive association for cumulative Pb
 8                  exposure and meningiomas for males but not females (Rajaraman et al.. 2006). An
 9                  ecologic analysis of Pb pollution levels and cancer incidence among children reported
10                  weak correlations overall and the weak correlations were more apparent among males,
11                  whereas no correlation was observed among females (Absalon and Slesak. 2010).

12                  A study of all-cause and cardiovascular mortality using the NHANES III cohort reported
13                  no  modification of the association between blood Pb and all-cause or cardiovascular
14                  mortality by sex (Menke et al., 2006). This did not differ among women when classified
15                  as pre- or post-menopausal.

16                  Toxicological studies have also reported sex differences in Pb-related effects to various
17                  organ systems. Donald et al. (1986) reported a different time course of enhanced social
18                  investigatory behavior between male and female mice exposed to Pb. In a subsequent
19                  publication, Donald et al. (1987) showed that non-social  behavior in mice decreased in
20                  females and increased in males exposed to Pb. Males also had a shorter latency to
21                  aggression with Pb treatment versus controls. Pb affected mood disorders differently for
22                  males and females. Behavioral testing in rats showed males experienced emotional
23                  changes and females depression-like changes with Pb exposure (de Souza Lisboa et al..
24                  2005). In another study, gestational exposure to Pb impaired memory retrieval in male
25                  rats at all 3 doses of Pb exposure; memory retrieval was only impaired in low-dose
26                  female rats (Yang et al.. 2003). Sex-specific differences in mice were also observed for
27                  gross motor skills; at the lowest Pb dose, balance and coordination were most affected
28                  among males (Leasure et al.. 2008).

29                  Pb  and  stress are co-occurring factors that act in a sex-divergent manner to affect
30                  behavior, neurochemistry, and corticosterone levels. Pb and stress act synergistically to
31                  affect fixed interval operant behavior and corticosterone  in female rat offspring. Virgolini
32                  et al.  (2008a) found that effects on the offspring's central nervous system by
33                  developmental Pb exposure (maternal exposure and transferred to the offspring through
34                  lactation) were enhanced by combined maternal and offspring stress and females were
35                  most  at risk. Behavioral related outcomes after gestational and lactational Pb exposure
36                  (with and without stress) exhibited sex-differences in exposed offspring (Virgolini et al..
37                  2008b). Pb-induced changes in brain neurochemistry, with or without concomitant stress


      November 2012                             6-25                   Draft - Do Not Cite or Quote

-------
 1                  exposure, are complex with differences varying by brain region, neurotransmitter type,
 2                  and sex of the animal.

 3                  The brain is known to have a sexually dimorphic area in the hypothalamus, termed the
 4                  sexually dimorphic nucleus (SDN). Lesions in this area affect sex-specific phenotypes
 5                  including behavior. Across species the SDN has a greater cell number and larger size in
 6                  males versus females. This sexually dichotomous area is especially vulnerable to
 7                  perturbation during fetal life and the early postnatal period. This may be one area of the
 8                  brain that could explain some of the sexually dichotomous effects that are seen with Pb
 9                  exposure. One study supporting this line of thought showed that high-dose in utero Pb
10                  exposure (pup blood Pb level 64 (ig/dL at birth) induced reductions in SDN volume in
11                  35% of Pb-exposed male rats (McGivern et al., 1991). Interestingly, another chemical
12                  that is known to cause a hypothalamic lesion in this area, monosodium glutamate, is
13                  associated with adult onset obesity (Olney. 1969); adult onset obesity is seen in the Pb
14                  literature.

15                  Obesity in adult offspring exposed to low-dose Pb in utero was reported for male but not
16                  female mice (Leasure et al., 2008). Obesity was also found in male rat offspring exposed
17                  in utero to high doses of Pb that persisted to 5 weeks of age/end of the study, but among
18                  female rats, body weight remained elevated over controls only to 3 weeks of age (Yang et
19                  al.. 2003). Additionally, low-dose Pb exposure induced retinal decrements in exposed
20                  male mice offspring  (Leasure et al., 2008).

21                  A toxicological study of Pb and antioxidant enzymes in heart and kidney  tissue reported
22                  that male and female rats had differing enzymatic responses, although the amount of Pb
23                  in the heart tissue or the disposition of Pb  also varied between males and  females
24                  (Sobekova et al.. 2009; Alghazal et al.. 2008a). The authors reported these results could
25                  be due to greater deposition of Pb in female rats or greater clearance  of Pb by males
26                  (Sobekova et al.. 2009).

27                  Multiple associations between Pb  and various health endpoints have been examined for
28                  effect measure modification by sex and results have been inconsistent. Although not
29                  observed in all endpoints, some studies reported differences between the  associations for
30                  males and females, especially in neurological studies. However, studies on cognition
31                  from the Cincinnati Lead Study cohort and a study in Poland reported males to be an at-
32                  risk population, whereas studies from Australia pointed to females as an at-risk
33                  population. A difference in sex is supported by toxicological studies. Further research is
34                  needed to confirm the presence or absence of sex-specific associations between Pb and
35                  various health outcomes and to determine in which sex the associations are greater.
      November 2012                             6-26                  Draft - Do Not Cite or Quote

-------
      6.3.3       Genetics

 1                  The 2006 Pb AQCD stated that, "genetic polymorphisms in certain genes have been
 2                  implicated as influencing the absorption, retention, and toxicokinetics of Pb in humans"
 3                  (U.S. EPA. 2006b). The majority of discussion there focused on the aminolevulinate
 4                  dehydratase (ALAD) and vitamin D receptor (VDR) polymorphisms. These two genes, as
 5                  well as additional genes examined in recent studies, are discussed below.
                    6.3.3.1     Aminolevulinate Dehydratase

 6                  The aminolevulinate dehydratase (ALAD) gene encodes for an enzyme that catalyzes the
 7                  second step in the production of heme and is also the principal Pb-binding protein (U.S.
 8                  EPA. 2006b). Studies have examined whether ALAD variants altered associations
 9                  between Pb and various health effects.

10                  Associations between Pb and brain tumors observed in an epidemiologic study varied by
11                  ALAD genotype status (Raiaraman et al.. 2006). Positive associations between Pb
12                  exposure (determined via interview about occupational exposures) and meningioma were
13                  reported among ALAD2 individuals, but this association was not found among
14                  individuals who had the ALAD1 allele. No associations were observed between Pb and
15                  glioma regardless of ALAD genotype.

16                  Studies investigating the association between Pb levels and cognitive function have also
17                  examined modification by ALAD polymorphisms. The evidence is provided by an
18                  NHANES  analysis (Krieg et al.. 2009) as well as multiple analyses from the NAS cohort
19                  examining different tests of cognitive function (Rajan et al.. 2008; Weuve et al.. 2006). In
20                  the study using a cohort from NHANES III, for several indices of cognitive function,
21                  associations with concurrent blood Pb levels were more pronounced in groups with CC
22                  and CG ALAD genotypes (i.e., ALAD2 carriers) (Krieg et al.. 2009). In the NAS cohort
23                  of men, Weuve et al. (2006) found that concurrent blood Pb level but not bone Pb level
24                  was associated with a larger decrease in a test of general cognitive function among
25                  ALAD2 carriers. Another NAS study examined functioning of specific cognitive
26                  domains (e.g., vocabulary, memory, visuospatial skills) and found variable evidence for
27                  effect modification by ALAD genotype across tests (Rajan et al.. 2008). For example,
28                  among ALAD2 carriers, concurrent blood Pb level was associated with a more
29                  pronounced decrease in vocabulary score but less pronounced decrease in a memory
30                  index and no difference in the associations with other cognitive tests. For tibia and patella
31                  Pb levels, ALAD genotype was found to modify associations with different tests, for
32                  example, executive function and perceptual speed. It is not clear why the direction of
      November 2012                             6-27                  Draft - Do Not Cite or Quote

-------
 1                  effect modification would vary among different cognitive domains. The limited number
 2                  of populations examined, and the different cognitive tests performed in each study, make
 3                  it difficult to conclusively summarize findings for effect modification by ALAD variants.
 4                  However, in the limited available body of evidence, blood and bone Pb levels were
 5                  generally associated with lower cognitive function in ALAD2 carriers.

 6                  A study of current and former workers exposed to Pb examined the association between
 7                  blood Pb and blood pressure and reported no modification by ALAD genotype  (Weaver
 8                  et al., 2008). However, another study of blood Pb and blood pressure reported
 9                  interactions between blood Pb and ALAD, but this varied by race/ethnicity (non-Hispanic
10                  white, non-Hispanic black, and Mexican American) (Scinicariello et al.. 2010).

11                  Individuals with ALAD2 variants had greater associations between Pb and kidney
12                  effects; among those with the variant, higher Pb was associated with higher glomerular
13                  filtration measures  (Weaver et al.. 2006; Weaver et al.. 2005b: Weaver et al.. 2003b). A
14                  study of workers at a battery plant storage facility in China reported workers with the
15                  ALAD2 allele demonstrated greater associations between blood Pb levels and renal
16                  injury (Gao et al.. 2010a). Another study of renal function among Pb workers in Asia also
17                  reported greater associations between blood Pb concentrations and renal function by
18                  ALAD,  especially at high blood Pb levels (Chia et al.. 2006).
                    6.3.3.2      Vitamin D Receptor

19                  The vitamin D receptor (VDR) is a regulator of calcium absorption and metabolism. A
20                  recent study of the NHANES III population examined the association between blood Pb
21                  levels and various neurocognitive tests with assessment of effect measure modification
22                  by SNPs and haplotypes of VDR (Krieg et al.. 2010). The results were varied, even
23                  among specific SNPs and haplotypes, with some variants being associated with greater
24                  modification of the relationship between Pb and one type of neurocognitive test
25                  compared to the modification of the relationship between Pb and other neurocognitive
26                  tests. In an epidemiologic study of blood Pb levels and blood pressure among a group of
27                  current and former Pb-exposed workers, no modification was reported by VDR (Weaver
28                  et al.. 2008).
      November 2012                            6-28                  Draft - Do Not Cite or Quote

-------
                    6.3.3.3      Methylenetetrahydrofolate reductase

 1                  Methylenetetrahydrofolate reductase (MTHFR) catalyzes the conversion of
 2                  5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate, which in turn, is involved in
 3                  homocysteine remethylation to the amino acid methionine. A study in Mexico of the
 4                  association between Pb and Bayley's Mental Development Index (MDI) score at 24
 5                  months reported no effect measure modification by MTHFR 677T allele (Pilsner et al..
 6                  2010). Another study in Mexico examined the association between maternal Pb and birth
 7                  weight (Kordas et al.. 2009). No modification of the Pb-birth weight association by
 8                  MTHFR was observed.
                    6.3.3.4      Hemochromatosis

 9                  The hemochromatosis (HFE) gene encodes a protein believed to be involved in iron
10                  absorption. A difference was observed between the association of tibia Pb levels and
11                  cognitive function for men with and without HFE allele variants (Wang et al.. 2007a). No
12                  association between tibia Pb and cognitive function was present for men with HFE
13                  wildtype, but a decline in function was associated with tibia Pb levels among men with
14                  any HFE allele variant. A study of bone Pb levels and HFE reported no difference in
15                  effect estimates for bone Pb and pulse pressure between different HFE variants and HFE
16                  wild-type (Zhang et al.. 2010a). An interaction was observed between an HFE variant in
17                  mothers and maternal tibia Pb in a study of maternal Pb and birth weight (Cantonwine et
18                  al.. 2010b). The inverse association between maternal tibia Pb levels and birth weight
19                  was stronger for those infants whose mothers had the HFE variant. The interaction was
20                  not present between the HFE variants and maternal blood Pb or cord blood Pb
21                  concentrations.
                    6.3.3.5     Other Genetic Polymorphisms

22                  Some other genetic polymorphisms were also examined as to whether they modify
23                  Pb-related health effects, but only limited data were available for these polymorphisms.
24                  These include dopamine receptor D4 (DRD4), dopamine receptor D2 (DRD2), dopamine
25                  transporter (DAT1), glutathione S-transferase Mu 1 (GSTM1), tumor necrosis factor-
26                  alpha (TNF-a), endothelial nitric oxide synthase (eNOS), and various SNPS.
27                  A prospective birth cohort reported that increasing blood Pb levels were associated with
28                  poorer rule learning and reversal, spatial span, and planning in their study population
29                  (Froehlich et al.. 2007). These inverse associations were exacerbated among those
      November 2012                             6-29                  Draft - Do Not Cite or Quote

-------
 1                  lacking DRD4-7. A study of prenatal and postnatal Pb levels in Mexico City reported no
 2                  modification of the associations between Pb levels and neurocognitive development by
 3                  DRD2 or DAT 1 (Kordas et al.. 2011).

 4                  A study of university students in South Korea reported blood Pb levels to be associated
 5                  with biomarkers of inflammation among individuals with GSTM1 null genotype and not
 6                  among individuals with GSTM1 present (Kim et al.. 2007). This study of blood Pb levels
 7                  and inflammation also examined individuals with TNF-a GG, GA, or AA alleles. An
 8                  association was present for those with TNF-a GG but not for those with TNF-a GA or
 9                  AA.

10                  A study of blood Pb and plasma NOX reported no overall association but did report an
11                  inverse correlation among subjects with the eNOS TC+CC genotype (Barbosa et al..
12                  2006c). No correlation was observed for subjects with the eNOS TT genotype; however
13                  the number of subjects in this group was small, especially for those with high blood Pb
14                  levels.

15                  One study examined how the association between occupational Pb exposure and brain
16                  tumors varied among multiple single nucleotide polymorphisms (SNPs) (Bhatti et al..
17                  2009). No effect measure modification of the association between Pb and glioma was
18                  observed for any of the SNPs. GPX1 (the gene encoding for glutathione peroxidase 1)
19                  modified the association for glioblastoma multiforme and meningioma. The association
20                  between Pb and glioblastoma multiforme was also modified by a RAC2 (the gene
21                  encoding for Rac2) variant, and the association between Pb and meningioma was also
22                  modified by XDH (the gene encoding for xanthine dehydrogenase) variant.

23                  Overall, studies of ALAD observed increased Pb-related health effects associated with
24                  certain gene variants. Other genes, such as VDR, HFE, DRD4, GSTM1, TNF-a, and
25                  eNOS, may also affect the risk of Pb-related health effects but conclusions are limited
26                  due to the small number of studies.
      6.3.4       Pre-existing Diseases/Conditions

27                  Studies have also been performed to examine whether certain morbidities increase an
28                  individual's risk of Pb-related effects on health. Recent studies have explored
29                  relationships for autism, diabetes, and hypertension.
      November 2012                             6-30                  Draft - Do Not Cite or Quote

-------
                     6.3.4.1     Autism

 1                   Rates of individuals with autism have increased in recent years. A study reported a
 2                   prevalence rate in 2006 of 9.0 per 1,000 individuals (95% CI: 8.6, 9.3) determined from a
 3                   monitoring network (Autism and Developmental Disabilities Monitoring Network) with
 4                   11 sites across the U.S. (CDC. 2009).
 5                   A cross-sectional study of children with and without autism examined the association
 6                   between blood Pb levels and various immune function and inflammation genes (Tian et
 7                   al., 2011). Blood Pb levels of children with and without autism were associated with
 8                   expression of the genes under study; however, the associations observed were in opposite
 9                   directions (for children with autism, increased blood Pb levels were associated with
10                   increased expression, whereas for children without autism, increased blood Pb levels
11                   were associated with decreased expression).
                     6.3.4.2      Diabetes

12                   Approximately 8% of U.S. adults have diabetes (Pleis et al., 2009). A few studies have
13                   been conducted to investigate the possibility of diabetes as a modifying factor for Pb and
14                   various health outcomes.

15                   Differences in the association between bone and blood Pb levels and renal function for
16                   individuals with and without diabetes at baseline were examined using the Normative
17                   Aging Study cohort (Tsaih et al., 2004). Tibia and blood Pb levels were positively
18                   associated with measures of poor renal function among individuals with diabetes but not
19                   among individuals without diabetes. However, this association was no longer statistically
20                   significant after the exclusion of individuals who were hypertensive or who used diuretic
21                   medications. Another study with this cohort reported no associations between bone Pb
22                   and heart rate variability, which did not differ among those with and without diabetes
23                   (Park et al.  2006V

24                   The NHANES III data were used to evaluate whether the association between blood Pb
25                   and both all-cause and cardiovascular mortality varied among individuals with and
26                   without diabetes (Menke et al.. 2006). The 95% CIs among those with diabetes were
27                   large and no difference was apparent among those with and without diabetes.

28                   Overall, recent epidemiologic studies found that associations between Pb concentrations
29                   and health outcomes did not differ for individuals with and without diabetes. However,
30                   results from the 2006 Pb AQCD found that individuals with diabetes are at "increased
31                   risk of Pb-associated declines in renal function" (U.S. EPA. 2006b). Future research
      November 2012                             6-31                  Draft - Do Not Cite or Quote

-------
 1                   examining associations between Pb and renal function, as well as other health outcomes,
 2                   among individuals with and without diabetes will inform further on the potential for
 3                   increased risk among individuals with diabetes.
                     6.3.4.3      Hypertension

 4                   Hypertension affects approximately 24% of adults in the U.S. and the prevalence of
 5                   hypertension increases with age (61% of individuals > 75 years old have hypertension)
 6                   (Pleis et al. 2009).

 7                   The Normative Aging Study mentioned above evaluating modification of the association
 8                   between Pb levels and renal function by  diabetes also examined modification by
 9                   hypertensive status (Tsaih et al.. 2004). The association between tibia Pb and renal
10                   function, measured by change in serum creatinine, was present among individuals with
11                   hypertension but not among individuals that were normotensive. Models of the follow-up
12                   serum creatinine levels demonstrated an  association with blood Pb for individuals with
13                   hypertension but not individuals without hypertension (this association was not present
14                   when using tibia or patella Pb). Another  study using this population examined
15                   modification of the association between bone Pb and heart rate variability,  measured by
16                   low frequency power, high frequency power, and their ratio (Park et al., 2006). Although
17                   a statistically significant association between bone Pb and heart rate variability was not
18                   observed among individuals with or without hypertension, the estimates were different,
19                   with greater odds for individuals with hypertension (bone Pb levels were positively
20                   related to low frequency power and the ratio of low frequency to high frequency power
21                   and were inversely related to high frequency power).

22                   A study using the NHANES III cohort reported a positive association between blood Pb
23                   levels and both all-cause and cardiovascular mortality for individuals with  and without
24                   hypertension but the associations did not differ based on hypertensive status (Menke et
25                   al.. 2006).

26                   The 2006 Pb AQCD reported that individuals with hypertension had increased risk of
27                   Pb-related effects on renal function (U.S. EPA. 2006b).  This is supported by recent
28                   epidemiologic studies. As described above, studies of Pb-related effects on renal function
29                   and heart rate variability have observed some differences among individuals with
30                   hypertension, but the difference between adults with and without hypertension was not
31                   observed for Pb-related mortality.
      November 2012                             6-32                  Draft - Do Not Cite or Quote

-------
 1                  Overall, studies of Pb-related health effects related to pre-existing conditions have some
 2                  evidence of a potential increased risk of Pb-related health effects. The evidence is
 3                  consistent for Pb-related renal effects and hypertension but is limited for other pre-
 4                  existing conditions.
      6.3.5        Smoking Status

 5                  The rate of smoking among adults 18 years and older in the U.S. is approximately 20%
 6                  and about 21% of individuals identify as former smokers (Pleis et al., 2009). Studies of
 7                  Pb and various health effects have examined smoking as an effect measure modifier.

 8                  A study of blood Pb levels and all-cause and cardiovascular mortality reported no
 9                  modification of this association by smoking status, measured as current, former, or never
10                  smokers (Menke et al.. 2006).  The Normative Aging Study also examined the association
11                  between blood and bone Pb levels and renal function and also reported no interaction
12                  with smoking status (Tsaih et al.. 2004).

13                  A study of Pb-exposed workers and controls reported similar levels of absolute neutrophil
14                  counts (ANC) across Pb exposure categories among non-smokers (Di Lorenzo et al..
15                  2006). However, among current smokers, higher Pb exposure was associated with higher
16                  ANC. Additionally, a positive relationship was observed between higher blood Pb levels
17                  and TNF-a and granulocyte colony-stimulating factor (G-CSF) among both smokers and
18                  nonsmokers, but this association was greater among smokers (Di Lorenzo et al.. 2007). A
19                  recent study of fertile and infertile men examined blood and seminal plasma Pb levels for
20                  smokers and non-smokers (Kiziler et al.. 2007). The blood and seminal plasma Pb levels
21                  were higher for smokers of both fertile and infertile groups. Additionally, the Pb levels
22                  were lowest among non-smoking fertile men and highest among smoking infertile men.

23                  Prenatal smoking exposure was examined in a study of children's concurrent blood Pb
24                  levels and prevalence of attention-deficit/hyperactivity disorder (ADHD) among children
25                  aged 8-15 years.  An interaction was observed between children's current blood Pb  levels
26                  and prenatal tobacco smoke exposure; those children with high Pb levels and prenatal
27                  tobacco smoke exposure had the highest odds of ADHD (Troehlich et al.. 2009).

28                  Overall, the studies have inconsistent findings on whether smoking modifies the
29                  relationship between Pb levels and health effects. Future studies of Pb-related health
30                  effects and current, former, and prenatal smoking exposures among various health
31                  endpoints will aid in determining changes in risk by this factor.
      November 2012                             6-33                  Draft - Do Not Cite or Quote

-------
      6.3.6       Socioeconomic Status

 1                  Based on the 2000 Census data, 12.4% of Americans live in poverty (poverty threshold
 2                  for family of 4 was $17,463) (SSDAN CensusScope. 2010c). Few studies have compared
 3                  blood Pb level effect estimates among groups in different sociodemographic strata.
 4                  Larger blood Pb-associated decreases in cognitive function were found with lower SES in
 5                  some studies (Ris et al.. 2004; Tong et al.. 2000; Bellinger et al.. 1990). In contrast, a
 6                  meta-analysis of eight studies found a smaller decrement in Full Scale Intelligence
 7                  Quotient (FSIQ) for studies in disadvantaged populations than for studies in advantaged
 8                  populations (Schwartz. 1994). While results indicate that blood Pb level is associated
 9                  with FSIQ deficits in both higher and lower sociodemographic groups, they do not clearly
10                  indicate whether groups with different socioeconomic status differ in Pb-related changes
11                  for cognitive function.
      6.3.7        Race/Ethnicity

12                  Based on the 2000 Census, 69.1% of the U.S. population is comprised of non-Hispanic
13                  whites. Approximately 12.1% of people reported their race/ethnicity as non-Hispanic
14                  black and 12.6% reported being Hispanic (SSDAN CensusScope. 2010b).  Studies of
15                  multiple Pb-related health outcomes examined effect measure modification by
16                  race/ethnicity.

17                  A study of adults from the NHANES III cohort examined the association between blood
18                  Pb levels and all-cause and cardiovascular mortality (Menke et al.. 2006).  Stratified
19                  analyses were conducted for non-Hispanic whites, non-Hispanic blacks, and Mexican
20                  Americans and no interaction for race/ethnicity was reported. Other studies have also
21                  used NHANES cohorts to study blood Pb levels and hypertension (Scinicariello et al..
22                  2010; Muntner et al.. 2005). While no association was observed between blood Pb and
23                  hypertension for non-Hispanic whites or Hispanics, a positive association was reported
24                  for non-Hispanic blacks in a study using the NHANES III cohort (Scinicariello et  al..
25                  2010). In another study, although none of the associations between blood Pb levels and
26                  hypertension were statistically significant, increased odds were observed among
27                  non-Hispanic blacks and Mexican Americans but not for non-Hispanic whites (Muntner
28                  et al.. 2005).
      November 2012                            6-34                  Draft - Do Not Cite or Quote

-------
 1                   A study of girls aged 8-18 years from the NHANES III cohort reported an inverse
 2                   association between blood Pb levels and pubertal development among blacks and
 3                   Mexican Americans (Selevan et al., 2003). For non-Hispanic whites, the associations
 4                   were in the same direction but did not reach statistical significance. Of note, less than 3%
 5                   of non-Hispanic whites had blood Pb levels over 5 (ig/dL, whereas 11.6% and 12.8% of
 6                   blacks and Mexican Americans, respectively, had blood Pb levels greater than 5 (ig/dL.

 7                   A study linking educational testing data for 4th grade students in North Carolina reported
 8                   declines in reading and mathematics scores with increasing levels of blood Pb (Miranda
 9                   et al.. 2007a). Although not quantitatively reported, a figure in the study depicted the
10                   association stratified by race, and the slopes appeared to be similar for white and black
11                   children.

12                   Blood Pb and asthma incidence was examined for white and black children living in
13                   Michigan (Joseph  et al.. 2005). When utilizing  separate referent groups for the two races,
14                   the only association is  an increase among whites (although not statistically significant),
15                   but when restricting to the highest blood Pb levels, the association was no longer
16                   apparent.  Whites with low blood Pb levels were used as the referent group for both races
17                   in additional analysis. Although the estimates were elevated for black children compared
18                   to white children (including at the lowest blood Pb levels), the confidence intervals for
19                   the associations overlapped indicating a lack of a difference by race.

20                   The results of these recent epidemiologic studies provide some evidence that there may
21                   be race/ethnicity-related increased risk with higher Pb levels for certain outcomes,
22                   although the overall understanding of potential effect measure modification by
23                   race/ethnicity is limited by the small number of studies. Additionally, these results may
24                   be confounded by  other factors, such as socioeconomic status or nutritional factors.
      6.3.8        Body Mass Index

25                   In the U.S. self-reported rates of obesity were 26.7% in 2009, up from 19.8% in 2000
26                   (Sherry etal.. 2010). The NHANES III cohort was utilized in a study of blood Pb levels
27                   and all-cause and cardiovascular mortality, which included assessment of the associations
28                   by obesity (Menke et al.. 2006). Positive associations were observed among individuals
29                   within both categories of body mass index (BMI; normal [<25 kg/m2] and
30                   overweight/obese [> 25 kg/m2], determined using measured values of height and weight)
31                   but there was no difference in the association between the two categories. Using the
32                   Normative Aging Study data, an investigation of bone Pb levels and heart rate variability
33                   was performed and reported slight changes in the association based on the presence of
34                   metabolic syndrome; however, none of the changes resulted in associations that were

      November 2012                             6-35                   Draft - Do Not Cite or Quote

-------
 1                  statistically significant (Park et al.. 2006). Overall, no modification by BMI or obesity
 2                  was observed among recent epidemiologic studies, but the available epidemiologic and
 3                  supporting toxicological studies are limited.
      6.3.9       Alcohol Consumption

 4                  There are a limited number of studies examining alcohol as a factor affecting Pb-related
 5                  risk. A study using the Normative Aging Study cohort investigated whether the
 6                  association between blood and bone Pb levels and renal function would be modified by
 7                  an individual's alcohol consumption (Tsaih et al.. 2004). No interaction with alcohol
 8                  consumption was observed. However, a toxicological study reported that ethanol
 9                  potentiated the effect of Pb exposure by decreasing renal total protein sulfhydryls
10                  (endogenous antioxidants) in rats. Pb and ethanol also decreased other endogenous renal
11                  antioxidants (glutathione and non-protein sulfhydryls) (Jurczuk et al.. 2006).  Overall,
12                  evidence to determine if alcohol consumption is a potential at-risk factor is of limited
13                  quantity and consistency.
      6.3.10       Nutritional Factors

14                  Different components of diet may affect the association between Pb concentrations and
15                  health outcomes. Recent epidemiologic and toxicological studies of specific mineral
16                  intakes/dietary components are detailed below.
                    6.3.10.1    Calcium

17                  Using the Normative Aging Study (NAS) cohort, researchers examined the association
18                  between Pb levels and hypertension, modified by calcium intake (Elmarsafawy et al..
19                  2006). The associations between Pb levels (measured and modeled separately for blood,
20                  patella, and tibia) and hypertension did not differ based on dichotomized calcium intake
21                  (800 mg/day).
      November 2012                             6-36                  Draft - Do Not Cite or Quote

-------
                    6.3.10.2    Iron

 1                  The 2006 Pb AQCD included studies that indicated individuals with iron-deficiency and
 2                  malnourishment had greater inverse associations between Pb and cognition (U.S. EPA,
 3                  2006b). A recent epidemiologic study of pubertal development among girls observed
 4                  inverse associations between blood Pb and inhibin  B. This association was modified by
 5                  iron deficiency; girls with iron deficiency had a stronger inverse association between Pb
 6                  and inhibin B than those who were iron sufficient (Gollenberg et al., 2010). Toxicological
 7                  studies also reported that iron-deficient diets exacerbate or potentiate the effect of Pb. A
 8                  study of pregnant rats given an iron-deficient diet and exposed to Pb through drinking
 9                  water over GD6-GD14, had decreased litter size, more pups with reduced fetal weight
10                  and reduced crown-rump length, increased litter resorption, and a higher dam blood Pb
11                  level in the highest exposure groups (Singh et al.. 1993b: Saxenaetal.. 1991). Thus, in
12                  this model, iron deficiency makes rat dams more at risk for Pb-dependent embryo and
13                  fetotoxicity (Singh et al.. 1993b).
                    6.3.10.3     Folate

14                  A study by Kordas et al. (2009) examined Pb levels and birth size among term births in
15                  Mexico City. The authors reported no interaction between maternal tibia Pb and folate
16                  levels.
                    6.3.10.4     Protein

17                  No recent epidemiologic studies have evaluated protein intake as a factor affecting
18                  Pb-related health effects. However, a toxicological study demonstrated that differences in
19                  maternal protein intake levels could affect the extent of Pb-induced immunotoxicity
20                  among offspring (Chen et al., 2004).

21                  In sum, the evidence is limited for most dietary factors but evidence for iron deficiency as
22                  a factor that potentially increases risk of Pb-induced effects is present and coherent in
23                  epidemiologic and toxicological studies.
      November 2012                             6-37                  Draft - Do Not Cite or Quote

-------
      6.3.11       Stress

 1                   A study of bone (tibia and patella) Pb levels and hypertension reported modification of
 2                   the association by perceived stress levels (Peters et al.. 2007). Among individuals with
 3                   greater perceived stress levels, stronger associations between blood Pb levels and
 4                   hypertension were present. Among the same study population, higher perceived stress
 5                   was also reported to affect the association between blood Pb levels and cognitive
 6                   function; the higher stress group showed a greater inverse association between Pb and
 7                   cognitive function than those in the low  stress group (Peters et al.. 2008). In another
 8                   study, the inverse association between tibia Pb levels and some measures of cognitive
 9                   function were similarly strengthened by  neighborhood psychosocial hazards (Glass et al..
10                   2009).

11                   Toxicological studies have demonstrated that early life exposure to Pb and maternal
12                   stress can result in toxicity related to multiple  systems (Rossi-George et al.. 2009: Cory-
13                   Slechtaetal.. 2008: Virgolini et al.. 2008a: Virgolini et al.. 2008b). including
14                   dysfunctional corticosterone responses (Rossi-George et al.. 2009: Virgolini et al..
15                   2008b). Additionally, toxicological studies have demonstrated that stressors to the
16                   immune system can also affect associations with Pb exposure. Chickens with low Pb
17                   exposure in ovo, with additional viral stressors, had increased immune cell mobilization
18                   and trafficking dysfunction (Lee et al.. 2002).  Similarly, mice with neonatal Pb exposure,
19                   and an additional immune challenge, had a sickness behavior phenotype, likely driven by
20                   IL-6 production (Dyatlov and Lawrence. 2002).

21                   Although examined in  a limited number of studies, recent epidemiologic studies observed
22                   modification of the association between  Pb and various nervous system health effects by
23                   stress-level. Increased risk of Pb-related health effects by stress is further supported by
24                   toxicological studies.
      6.3.12       Maternal Self-Esteem

25                   Maternal self-esteem has been shown to modify associations between blood Pb levels and
26                   health effects in children. Surkan et al. (2008) studied the association between children's
27                   blood Pb levels and Bayley's MDI and Psychomotor Development Index (PDI) among
28                   mother-child pairs. High maternal self-esteem was independently associated with higher
29                   MDI score and also appeared to attenuate the negative effects of the child's increased
30                   blood Pb levels on MDI and PDI scores. Greater decreases in MDI and PDI were
31                   associated with increased blood Pb levels among children whose mothers were in the
32                   lower quartiles of self-esteem.
      November 2012                             6-38                  Draft - Do Not Cite or Quote

-------
      6.3.13       Cognitive Reserve

 1                  Cognitive reserve has been defined as "the maintenance of cognitive performance in spite
 2                  of ongoing underlying brain pathology" (Bleecker et al.. 2007a). A study of Pb smelter
 3                  workers reported that an inverse association between lifetime weighted blood Pb levels
 4                  and cognitive function was present among workers with low cognitive reserve (measured
 5                  using a reading achievement test) but no association was present in workers with high
 6                  cognitive reserve (Bleecker et al.. 2007a). Inverse associations between lifetime-weighted
 7                  blood Pb levels and motor functions existed among all workers regardless of cognitive
 8                  reserve. No other recent epidemiologic studies were performed examining cognitive
 9                  reserve as a factor affecting risk of Pb-related health outcomes, thus providing limited
10                  evidence to conclude that cognitive reserve is a potential at-risk factor.
      6.3.14       Other Metal Exposure

11                  The 2006 Pb AQCD reported that the majority of studies that examined other toxicants
12                  did so as confounders and not as effect measure modifiers (U.S. EPA. 2006b). Recent
13                  epidemiologic studies have begun to explore the possible interaction between Pb
14                  exposure and co-exposures with other metals. These studies, as well as toxicological
15                  studies of these metals, are described below.
                    6.3.14.1    Cadmium

16                  In a study of girls in the NHANES III cohort, inverse associations were observed
17                  between blood Pb and inhibin B concentrations (Gollenberg et al.. 2010). These inverse
18                  associations were stronger among girls with high cadmium (Cd) and high Pb compared to
19                  those with high Pb and low Cd. Additionally, higher blood Pb and Cd levels together
20                  were positively associated with albuminuria and reduced estimated glomerular filtration
21                  rate, compared to those with the lowest levels of Pb and Cd (Navas-Acien et al.. 2009).

22                  Toxicological studies reported that in rats, the addition of Cd to Pb exposure reduced the
23                  histological signs of renal toxicity from each element alone; however, urinary excretion
24                  of porphyrins were increased, indicating that although measured tissue burdens of Pb
25                  were reduced, the biologically available fraction of Pb was actually increased (Wang and
26                  Fowler. 2008). In other studies, Cd synergistically exacerbated Pb-dependent renal
27                  mitochondrial dysfunction (Wang et al.. 2009c).
      November 2012                             6-39                  Draft - Do Not Cite or Quote

-------
 1                  Overall, epidemiologic and toxicological studies have reported increased risk of
 2                  Pb-related health effects among those with high Cd levels as well; however, the number
 3                  of studies examining both metals is small.
                    6.3.14.2    Manganese

 4                  Among children in South Korea taking part in a study of IQ, an interaction was reported
 5                  between Pb and manganese (Mn) blood levels (Kim et al.. 2009b). Children with high
 6                  blood Mn levels were observed to have reductions in full  scale IQ and verbal IQ
 7                  associated with increased blood Pb levels, whereas no association between blood Pb
 8                  levels and full scale IQ and verbal IQ were noted among those children with low blood
 9                  Mn levels. No effect measure modification by Mn was observed for the association
10                  between blood Pb levels and performance IQ. A study performed among children in
11                  Mexico City observed greater decreases in neurodevelopment with increases in blood
12                  levels of Pb and Mn at 12 months, compared to decreases in neurodevelopment observed
13                  for increased Pb levels with low levels of Mn (Claus Henn et al.. 2012). No interaction
14                  was observed between the two metals and neurodevelopment at 24 months.

15                  Overall, studies have reported increased risk of various health effects with exposure to
16                  other metals in addition to Pb; however, this is limited by the small number of studies.
17                  Toxicological studies, when available, have provided support for these findings.
      6.4        Summary

18                  Table 6-5 provides an overview of the factors examined as potentially increasing the risk
19                  of Pb-related health effects based on the recent evidence integrated across disciplines.
20                  They are classified according to the criteria discussed in the introduction to this chapter.
      November 2012                            6-40                  Draft - Do Not Cite or Quote

-------
      Table 6-5    Summary of evidence for factors that potentially increase the risk of
                    Pb-related health effects.
 i
 2
 o
 3
 4
 5
 6
 1
 8
 9
10
11

12
13
14
Factor Evaluated
Childhood (Sections 6.2.1 , 6.3.1)

Older Adulthood (Sections 6.2.1, 6.3.1)

Sex (Sections 6.2.2, 6.3.2)

Genetics (Section 6.3.3)
Pre-existing Disease3 (Section 6.3.4)
Smoking Status (Section 6.3.5)
Socioeconomic Status (SES) (Sections 6.2.4, 6.3.6)

Race/Ethnicitv (Sections 6.2.3, 6.3.7)

Proximity to Pb Sources (Section 6.2.5)
Residential Factors (Section 6.2.6)
Body Mass Index (BMI) (Section 6.3.8)
Alcohol Consumption (Section 6.3.9)
Nutrition (Section 6.3.10)
Stress (Section 6.3.11)
Maternal Self-Esteem (Section 6.3.12)
Cognitive Reserve3 (Section 6.3.13)
Other Metals (Section 6.3.14)
Classification
Adequate
Suggestive
Suggestive
Suggestive
Suggestive
Inadequate
Suggestive
Adequate
Adequate
Adequate
Inadequate
Inadequate
Adequate
Suggestive
Inadequate
Inadequate
Suggestive
      Possible mediator
There are consistent findings, coherent across disciplines that adequate evidence exists to
conclude that childhood is an at-risk lifestage. Among children, the youngest age groups
were observed to be most at risk of elevated blood Pb levels, with levels decreasing with
increasing age of the children. Children may have increased exposure to Pb compared
with adults because children's behaviors and activities (including increased hand-to-
mouth contact, crawling, and poor hand-washing), differences in diets, and biokinetic
factors. Recent epidemiologic  studies of infants/children detected increased risk of
Pb-related health effects, and this was supported by toxicological studies, providing
adequate evidence to conclude that children are an at-risk population. However, this is
based on a limited number of epidemiologic studies, and more studies are needed for
comparing various age groups and examining adolescents.

For adults, elevated Pb biomarkers were associated with increasing age. It is generally
thought that these elevated levels are related to remobilization of stored Pb during bone
loss and/or higher historical Pb exposures. Studies of older adults had inconsistent
     November 2012
                            6-41
Draft - Do Not Cite or Quote

-------
 1                   findings for effect measure modification of Pb-related mortality but no difference was
 2                   observed for other health effects. However, toxicological studies support the possibility
 3                   of age-related differences in Pb-related health effects. The overall evidence is suggestive
 4                   that older adults are a potential at-risk population based on limited epidemiologic
 5                   evidence but support from toxicological  studies and differential exposure studies.

 6                   Some studies suggest that males at some ages have higher blood Pb levels than
 7                   comparably aged females; this was supported by stratifying the total sample of NHANES
 8                   subjects. Sex-based differences  appeared to be prominent among the adolescent and adult
 9                   age groups but were not observed among the youngest age groups (1-5 years and 6-11
10                   years). Studies of effect measure modification of Pb and various health endpoints by sex
11                   were inconsistent; although it appears that there are some differences in associations for
12                   males and females. This is also  observed in toxicological studies. Overall, there is
13                   suggestive evidence to conclude that sex is a potential at-risk factor, limited due to
14                   inconsistencies between whether males or females are at greater risk of certain outcomes.

15                   Regarding race and ethnicity, recent data suggest that the difference in blood Pb levels
16                   between black and white subjects is decreasing over time, but black subjects still tend to
17                   have higher Pb body burden and Pb exposures than white subjects. Compared to whites,
18                   non-white populations were observed to  be more at risk of Pb-related health effects;
19                   however, this could be related to confounding by factors  such as SES or differential
20                   exposure levels, which was noted in some of the epidemiologic studies. Studies of
21                   race/ethnicity provide adequate  evidence that race/ethnicity is an at-risk factor based on
22                   the higher exposure observed among non-white populations and some modification
23                   observed in studies of associations between Pb levels and health effects.

24                   Similar to race and ethnicity, the gap between SES groups with respect to Pb body burden
25                   appears to be diminishing. Studies of SES and its relationship with Pb-related health
26                   effects are limited and different studies demonstrate increased risk among higher or lower
27                   SES groups, providing limited evidence to determine if SES is an at-risk factor for
28                   Pb-related health effects. However, biomarkers of Pb exposure have been shown to be
29                   higher among lower SES groups even in recent studies in which differences among SES
30                   groups have lessened. Therefore, the  evidence is suggestive to conclude that low SES is a
31                   potential at-risk factor for Pb-related  health effects.

32                   There is evidence associating proximity to areas with Pb  sources, including areas with
33                   large industrial sources, with increased Pb body burden and risk of Pb exposure. High
34                   concentrations of ambient air Pb have been measured near sources, compared with large
35                   urban areas without sources. Additionally, high Pb exposures have been documented near
36                   Superfund sites.
      November 2012                             6-42                  Draft - Do Not Cite or Quote

-------
 1                   Studies utilizing the NHANES dataset have reported increased Pb biomarker measures
 2                   related to increase house dust Pb levels, homes built after 1950, and renovation of
 3                   pre-1978 homes. These findings were consistent with those of several high quality
 4                   studies. Thus, there is adequate evidence that residing in a residence with Pb exposures
 5                   will increase the risk of Pb-related health effects.

 6                   There is suggestive evidence to conclude that various genes are potentially modifying the
 7                   associations between Pb and health effects. Epidemiologic and toxicological studies
 8                   reported that ALAD variants may increase the risk of Pb-related health effects. Other
 9                   genes examined that may also affect risk of Pb-related health effects were VDR, DRD4,
10                   GSTM1, TNF-a, eNOS, and HFE, although the number of studies examining effect
11                   measure modification by these genes was small.

12                   Among nutritional factors, diets sufficient in minerals such as Ca2+, Fe, and Zn offer
13                   some protection from Pb exposure by preventing or competing with Pb for absorption in
14                   the GI tract. Additionally, those with iron deficiencies were observed to be an at-risk
15                   population for Pb-related health effects in both epidemiologic and toxicological studies.
16                   Thus, there is adequate evidence across  disciplines that some nutritional factors
17                   contribute to a population being at increased risk. Other nutritional factors, such as Ca2+,
18                   Zn, and protein intake, demonstrated the potential to modify associations between Pb and
19                   health effects in toxicological studies. Recent epidemiologic studies of these factors were
20                   either not performed or observed no effect modification. Folate was also examined in an
21                   epidemiologic study of birth size but no interaction was reported between Pb and folate.

22                   There was suggestive evidence for several other factors as potentially increasing the risk
23                   of Pb-related health effects: pre-existing diseases/conditions, stress, and co-exposure with
24                   other metals. Pre-existing diseases/conditions have the potential to affect the risk of
25                   Pb-related health effects. Recent epidemiologic studies did not support modification of
26                   associations between Pb and health endpoints by the prevalence of diabetes; however,
27                   past studies have found individuals with diabetes to be an at-risk population with regard
28                   to renal function. Hypertension was observed to be a factor affecting risk in both past and
29                   recent epidemiologic studies. Studies of Pb levels and both renal effects and heart rate
30                   variability demonstrated greater odds of the associations among hypertensive individuals
31                   compared to those that are normotensive. Epidemiologic studies also examined autism as
32                   potential factors affecting Pb-related health effects; differences were observed but few
33                   studies were available to examine this factor. Stress was evaluated as a factor that
34                   potentially increases the risk of Pb-related health outcomes and although limited by the
35                   small number of epidemiologic studies,  increased stress was observed to negatively
36                   impact the association between Pb and health endpoints. Toxicological studies supported
37                   this finding. Finally, interactions between Pb and co-exposure with other metals were
      November 2012                              6-43                   Draft - Do Not Cite or Quote

-------
 1                  evaluated in recent epidemiologic and toxicological studies of health effects. High levels
 2                  of other metals, such as Cd and Mn, were observed to result in greater effects for the
 3                  associations between Pb and various health endpoints but evidence was limited due to the
 4                  small number of studie s.

 5                  Finally, there was inadequate evidence to conclude that smoking, BMI, alcohol
 6                  consumption, maternal self-esteem, and cognitive reserve are potential at-risk factors due
 7                  to limited quantities of studies regarding their effect on Pb-related health outcomes.
 8                  Epidemiologic studies examining smoking as a factor potentially affecting risk reported
 9                  mixed findings. It is possible that smoking modifies the effects of only some Pb-related
10                  health outcomes. In the limited number of studies, modification of associations between
11                  Pb and various health effects (mortality and heart rate variability) was not observed for
12                  BMI/obesity. Also, no modification was observed in an epidemiologic study of renal
13                  function examining alcohol consumption as a modifier, but a toxicological study
14                  supported the potential of alcohol to affect risk. Maternal self-esteem was examined in an
15                  epidemiologic study and individuals with mothers who had lower self-esteem had greater
16                  Pb-related decreases in MDI and PDI. An epidemiologic study evaluated cognitive
17                  reserve as a modifier of the associations between Pb and cognitive and motor functions.
18                  Cognitive reserve was an effect measure modifier for the association between Pb and
19                  cognitive function but not motor function.
      November 2012                             6-44                  Draft - Do Not Cite or Quote

-------
References for Chapter 6
     Absalon. D; Slesak. B. (2010). The effects of changes in cadmium and lead air pollution on cancer
        incidence in children. Sci Total Environ 408: 4420-4428.
        http://dx.doi.0rg/10.1016/i.scitotenv.2010.06.030

     Alghazal. MA; Lenartova. V: Holovska. K; Sobekova. A; Palis. M; Legath. J. (2008a). Activities of
        antioxidant and detoxifying enzymes in rats after lead exposure. Acta Vet (Beogr) 77: 347-354.
        http://dx.doi.org/10.2754/avb200877030347

     Baghurst. PA; McMichael. AJ; Wigg. NR; Vimpani. GV: Robertson. EF; Roberts. RJ: Tong. SL. (1992).
        Environmental exposure to lead and children's intelligence at the age of seven years: The Port Pirie
        cohort study. N Engl J Med 327: 1279-1284. http://dx.doi.org/10.1056/NEJM199210293271805

     Barbosa. F. Jr; Sandrim. VC: Uzuelli. JA; Gerlach. RF; Tanus-Santos. JE. (2006c). eNOS genotype-
        dependent correlation between whole blood lead and plasma nitric oxide products concentrations.
        Nitric Oxide 14: 58-64. http://dx.doi.Org/10.1016/i.niox.2005.09.007

     Basha. MR: Murali. M: Siddiqi. HK: Ghosal. K: Siddiqi. OK: Lashuel. HA: Ge. YW: Lahiri. DK: Zawia.
        NH. (2005). Lead (Pb) exposure and its effect on APP proteolysis and A beta aggregation. FASEB J
        19: 2083-2084. http://dx.doi.org/10.1096/fi.05-4375fie

     Bellinger. D; Leviton. A; Sloman. J. (1990). Antecedents and correlates of improved cognitive
        performance in children exposed in utero to low levels of lead. Environ Health Perspect 89: 5-11.

     Berrahal. AA; Lasram. M; El Elj. N; Kerkeni. A; Gharbi. N; El-Fazaa. S. (2011). Effect of age-dependent
        exposure to lead on hepatotoxicity and nephrotoxicity in male rats. Environ Toxicol 26: 68-78.
        http://dx.doi.org/10.1002/tox.20530

     Bhatti. P: Stewart. PA; Hutchinson. A; Rothman. N; Linet. MS: Inskip. PD; Rajaraman. P. (2009). Lead
        exposure, polymorphisms in genes related to oxidative stress,  and risk of adult brain tumors. Cancer
        Epidemiol Biomarkers Prev 18: 1841-1848. http://dx.doi.org/10.1158/1055-9965.EPI-09-0197

     Bleecker. ML; Ford. DP; Celio. MA; Vaughan. CG: Lindgren. KN. (2007a). Impact of cognitive reserve
        on the relationship of lead exposure and neurobehavioral performance. Neurology 69: 470-476.
        http://dx.doi.org/10.1212/01.wnl.0000266628.43760.8c

     Bogden. JD; Kemp. FW: Han. S: Murphy. M; Fraiman. M; Czerniach. D; Flynn. CJ: Banua. ML;
        Scimone. A; Castrovilly. L; Gertner. SB. (1995). Dietary calcium and lead interact to modify maternal
        blood pressure, erythropoiesis, and fetal and neonatal growth in rats during pregnancy and lactation. J
        Nutr 125: 990-1002.

     Brubaker. CJ: Dietrich. KN; Lanphear. BP; Cecil. KM.  (2010). The influence of age of lead exposure on
        adult gray matter volume. Neurotoxicology 31: 259-266.
        http://dx.doi.0rg/10.1016/i.neuro.2010.03.004

     Campanella. R; Mielke. HW. (2008). Human geography of New Orleans' high-lead geochemical setting.
        Environ  Geochem Health 30: 531-540. http://dx.doi.org/10.1007/sl0653-008-9190-9

     Cantonwine. D: Hu. H; Tellez-Rojo. MM; Sanchez. BN; Lamadrid-Figueroa. H; Ettinger. AS: Mercado-
        Garcia. A; Hernandez-Avila. M; Wright. RO. (2010b). HFE gene variants modify the association
        between maternal lead burden and infant birthweight: A prospective birth cohort study in Mexico City,
        Mexico.  Environ Health Global Access Sci Source 9: 43. http://dx.doi.org/10.1186/1476-069X-9-43

     CDC (Centers for Disease Control and Prevention). (2009). Prevalence of autism spectrum disorders ~
        Autism and developmental disabilities monitoring network, United States, 2006. MMWR Surveill
        Summ 58: 1-20.
November 2012                              6-45                   Draft - Do Not Cite or Quote

-------
     Cecil. KM; Brubaker. CJ: Adler. CM; Dietrich. KN; Altave. M; Egelhoff. JC: Wessel S: Elangovan. I;
        Hornung. R; Jarvis. K; Lanphear. BP. (2008). Decreased brain volume in adults with childhood lead
        exposure. PLoS Med 5: el 12. http://dx.doi.org/10.1371/iournal.pmed.0050112

     Chen. S: Golemboski. KA; Piepenbrink. M; Dietert. RR. (2004). Developmental immunotoxicity of lead
        in the rat: Influence of maternal diet. J Toxicol Environ Health A 67: 495-511.
        http://dx.doi.org/10.1080/15287390490276520

     Chen. Y; Zhao. JX; Liu. JW: Cui. J: Li.  L; Tian. W. (2008c). Lack of association of delta-aminolevulinic
        acid dehydratase genotype with blood lead levels in environmentally exposed children of Uygur and
        Han populations. Acta Paediatr 97: 1717-1720. http://dx.doi.Org/10.llll/j.1651-2227.2008.01003.x

     Chia. SE; Zhou. HJ; Yap. E; Tham. MT; Dong. NV; Hong Tu. NT; Chia. KS. (2006). Association of renal
        function and delta-aminolevulinic acid dehydratase polymorphism among Vietnamese and Singapore
        workers exposed to inorganic lead. Occup Environ Med 63: 180-186.
        http://dx.doi.org/10.1136/oem.2005.021154

     Glaus Henn. B; Schnaas. L; Ettinger. AS: Schwartz. J: Lamadrid-Figueroa. H; Hernandez-Avila. M;
        Amarasiriwardena. C: Hu. H; Bellinger. DC: Wright. RO: Tellez-Rojo. MM. (2012). Associations of
        early childhood manganese  and lead co-exposure with  neurodevelopment. Environ Health Perspect
        120:  126-131. http://dx.doi.org/10.1289/ehp.1003300

     Corv-Slechta. DA; Virgolini. MB;  Rossi-George. A; Thiruchelvam. M; Lisek. R; Weston. D. (2008).
        Lifetime consequences of combined  maternal lead and stress  [Review]. Basic Clin Pharmacol Toxicol
        102: 218-227. http://dx.doi.Org/10.llll/i.1742-7843.2007.00189.x

     de Souza Lisboa. SF; Goncalves. G: Komatsu. F; Queiroz. CAS: Almeida. AA; Moreira. EG. (2005).
        Developmental lead exposure induces depressive-like behavior in female rats. Drug Chem Toxicol 28:
        67-77. http://dx.doi.org/10.1081/dct-200039696

     Di Lorenzo. L;  Silvestroni. A; Martino. MG: Gagliardi. T; Corfiati. M; Soleo. L. (2006). Evaluation of
        peripheral blood neutrophil  leucocytes  in lead-exposed workers. Int Arch Occup Environ Health 79:
        491-498. http://dx.doi.org/10.1007/s00420-005-0073-4

     Di Lorenzo. L;  Vacca. A; Corfiati.  M; Lovreglio. P; Soleo. L. (2007). Evaluation of tumor necrosis factor-
        alpha and granulocyte colony-stimulating factor serum levels in lead-exposed smoker workers.  Int J
        Immunopathol Pharmacol 20: 239-247.

     Dietrich. KN; Krafft. KM; Shukla. R: Bornschein. RL; Succop. PA. (1987b). The neurobehavioral effects
        of early lead exposure. In SR Schroeder (Ed.), Toxic substances and mental retardation:
        Neurobehavioral toxicology and teratology (pp. 71-95). Washington, DC: American Association on
        Mental Deficiency.

     Dixon. SL; Gaitens. JM; Jacobs. DE; Strauss. W: Nagaraia. J: Pivetz. T; Wilson. JW: Ashley. PJ. (2009).
        Exposure of US children to  residential dust lead, 1999-2004: II. The contribution of lead-contaminated
        dust to children's blood lead levels. Environ Health Perspect 117: 468-474.
        http://dx.doi.org/10.1289/ehp.11918

     Donald.  JM; Cutler. MG: Moore. MR. (1987). Effects of lead in the laboratory mouse. Development and
        social behaviour after lifelong exposure to 12 microM lead in drinking fluid. Neuropharmacology 26:
        391-399. http://dx.doi.org/10.1016/0028-3908(87)90194-8

     Donald.  JM: Cutler. MG: Moore. MR; Bardlev. M. (1986). Effects of lead in the laboratory mouse~2:
        Development and social behaviour after lifelong administration of a small dose of lead acetate in
        drinking fluid. Neuropharmacology 25: 151-160.

     Dyatlov. VA; Lawrence. DA. (2002). Neonatal lead exposure potentiates sickness behavior induced by
        Listeria monocytogenes infection of mice. Brain Behav Immun  16: 477-492.
        http://dx.doi.org/10.1006/brbi.2001.0641

     Elmarsafawv. SF; Jain. NB; Schwartz. J: Sparrow. D; Nie. HL; Hu. H. (2006). Dietary calcium as a
        potential modifier of the relationship of lead burden to  blood pressure. Epidemiology 17: 531-537.
        http://dx.doi.org/10.1097/01.ede.0000231285.86968.2b
November 2012                              6-46                   Draft - Do Not Cite or Quote

-------
     Fadrowski. JJ; Navas-Acien. A; Tellez-Plaza. M; Guallar. E; Weaver. VM; Furth. SL. (2010). Blood lead
        level and kidney function in US adolescents: The Third National Health and Nutrition Examination
        Survey. Arch Intern Med 170: 75-82. http://dx.doi.org/10.1001/archinternmed.2009.417

     Filippelli. GM; Laidlaw. MAS: Latimer. JC: Raftis. R. (2005). Urban lead poisoning and medical
        geology: An unfinished story. Geological Society of America Today 15: 4-11.
        httrj://dx.doi.org/10.1130/1052-5173(2005)015<4:ULPAMG>2.0.CO:2

     Franklin. CA; Inskip. MJ; Baccanale. CL; Edwards. CM; Manton. WI; Edwards. E; O'Flahertv. EJ.
        (1997). Use of sequentially administered stable lead isotopes to investigate changes in blood lead
        during pregnancy in a nonhuman primate (Macaca fascicularis). Toxicol Sci 39: 109-119.

     Froehlich. TE; Lanphear. BP; Auinger. P; Hornung. R; Epstein. JN; Braun. J; Kahn. RS. (2009).
        Association of tobacco and lead exposures with attention-deficit/hyperactivity disorder. Pediatrics
        124: E1054-E1063. http://dx.doi.org/10.1542/peds.2009-0738

     Froehlich. TE; Lanphear. BP; Dietrich. KN; Cory-Slechta. DA; Wang. N; Kahn. RS. (2007). Interactive
        effects of a DRD4 polymorphism, lead and sex on executive functions in children. Biol Psychiatry 62:
        243-249. http://dx.doi.0rg/10.1016/i.biopsvch.2006.09.039

     Fullmer. CS. (1992). Intestinal interactions of lead and calcium [Review]. Neurotoxicology 13: 799-807.

     Fullmer. CS: Edelstein. S: Wasserman. RH. (1985). Lead-binding properties of intestinal calcium-binding
        proteins. J Biol Chem 260: 6816-6819.

     Gaitens. JM; Dixon. SL; Jacobs. DE; Nagaraia. J: Strauss. W: Wilson. JW: Ashley. PJ. (2009). Exposure
        of US children to residential dust lead, 1999-2004:1.  Housing and demographic factors. Environ
        Health Perspect 117: 461-467. http://dx.doi.org/10.1289/ehp.11917

     Gao. A; Lu. XT; Li. QY; Tian. L. (2010a). Effect of the delta-aminolevulinic acid dehydratase gene
        polymorphism on renal and neurobehavioral function in workers exposed to lead in China. Sci Total
        Environ 408: 4052-4055. http://dx.doi.Org/10.1016/i.scitotenv.2010.04.024

     Garrido Latorre. F; Hernandez-Avila. M; Orozco. JT; Medina. CAA; Aro. A; Palazuelos. E; Hu. H.
        (2003). Relationship of blood and bone lead to menopause and bone mineral density among middle-
        age women in Mexico  City. Environ Health Perspect  111: 631-636.

     Glass. TA; Bandeen-Roche. K; McAtee. M; Bolla. K; Todd. AC: Schwartz. BS. (2009). Neighborhood
        psychosocial hazards and the association of cumulative lead dose with cognitive function in older
        adults. Am J Epidemiol 169: 683-692. http://dx.doi.org/10.1093/aie/kwn390

     Gollenberg. AL; Hediger. ML; Lee. PA; Himes. JH; Buck Louis. GM. (2010). Association between lead
        and cadmium and reproductive hormones in peripubertal U.S. girls. Environ Health Perspect 118:
        1782-1787.  http://dx.doi.org/10.1289/ehp.1001943

     Gulson. B; Mizon. K; Smith. H; Eisman. J: Palmer. J: Korsch. M; Donnelly. J: Waite. K. (2002). Skeletal
        lead release during bone resorption: Effect of bisphosphonate treatment in a pilot study. Environ
        Health Perspect 110: 1017-1023.

     Gulson. B; Mizon. KJ; Palmer. JM; Korsch. MJ: Taylor. AJ; Mahaffev. KR.  (2004a). Blood lead changes
        during pregnancy and postpartum with calcium supplementation.  Environ Health Perspect 112:
        1499-1507.  http://dx.doi.org/10.1289/ehp.6548

     Gulson. BL; Jameson. CW: Mahaffev. KR; Mizon. KJ: Korsch. MJ: Vimpani. G. (1997). Pregnancy
        increases mobilization of lead from maternal skeleton. Transl Res 130: 51-62.
        http://dx.doi.org/10.1016/S0022-2143(97)90058-5

     Han. S: Li. W: Jamil. U; Dargan. K; Orefice. M; Kemp. FW: Bogden. JD. (1999). Effects of weight loss
        and exercise on the distribution of lead and essential trace elements in rats with prior lead exposure.
        Environ Health Perspect 107: 657-662. http://dx.doi.org/10.1289/ehp.99107657

     Hertz-Picciotto. I; Schramm. M; Watt-Morse. M; Chantala. K; Anderson. J:  Osterloh. J. (2000). Patterns
        and determinants of blood lead during pregnancy. Am J Epidemiol 152: 829-837.
        http://dx.doi.0rg/10.1093/aie/152.9.829
November 2012                              6-47                   Draft - Do Not Cite or Quote

-------
     Jamieson. JA; Stringer. DM; Zahradka. P; Taylor. CG. (2008). Dietary zinc attenuates renal lead
        deposition but metallothionein is not directly involved. Biometals 21: 29-40.
        http://dx.doi.org/10.1007/sl0534-007-9090-v

     Jamieson. JA; Taylor. CG: Weiler. HA. (2006). Marginal zinc deficiency exacerbates bone lead
        accumulation and high dietary zinc attenuates lead accumulation at the expense of bone density in
        growing rats. Toxicol Sci 92: 286-294. http://dx.doi.org/10.1093/toxsci/kfi201

     Jedrychowski. W: Perera. F; Jankowski. J; Mrozek-Budzyn. D; Mroz. E; Flak. E; Edwards. S: Skarupa.
        A; Lisowska-Miszczyk. I. (2009a). Gender specific differences in neurodevelopmental effects of
        prenatal exposure to very low-lead levels: The prospective cohort study in three-year olds. Early Hum
        Dev 85: 503-510. http://dx.doi.Org/10.1016/j.earlhumdev.2009.04.006

     Jones. EA; Wright. JM; Rice. G: Buckley. BT; Magsumbol. MS: Barr. DB; Williams. BL. (2010). Metal
        exposures in an inner-city neonatal population. Environ Int 36: 649-654.
        http://dx.doi.0rg/10.1016/i.envint.2010.04.007

     Jones. RL; Homa. DM; Meyer. PA; Brodv. DJ: Caldwell. KL; Pirkle. JL; Brown. MJ. (2009a). Trends in
        blood lead levels and blood lead testing among US children aged 1 to 5 Years, 1988-2004. Pediatrics
        123: e376-e385. http://dx.doi.org/10.1542/peds.2007-3608

     Joseph. CLM; Havstad. S: Ownbv. PR; Peterson. EL; Maliarik. M; McCabe. MJ. Jr; Barone. C: Johnson.
        CC. (2005). Blood lead level and risk of asthma. Environ Health Perspect 113: 900-904.
        http://dx.doi.org/10.1289/ehp.7453

     Jurczuk. M; Moniuszko-Jakoniuk. J; Brzoska. MM. (2006). Involvement of some low-molecular thiols in
        the peroxidative mechanisms of lead and ethanol action on rat liver and kidney. Toxicology 219:
        11-21. http://dx.doi.0rg/10.1016/i.tox.2005.10.022

     Kim. JH: Lee. KH: Yoo. DH: Kang. D: Cho. SH: Hong. YC. (2007). GSTM1 and TNF-alpha gene
        polymorphisms and relations between blood lead and inflammatory markers in a non-occupational
        population. Mutat Res Genet Toxicol  Environ Mutagen 629: 32-39.
        http://dx.doi.0rg/10.1016/i.mrgentox.2007.01.004

     Kim. Y; Kim. BN; Hong. YC: Shin. MS: Yoo. HJ; Kim. JW: Bhang. SY; Cho. SC. (2009b). Co-exposure
        to environmental lead and manganese affects the intelligence of school-aged children.
        Neurotoxicology 30: 564-571. http://dx.doi.Org/10.1016/i.neuro.2009.03.012

     Kiziler. AR; Aydemir. B; Onaran. I; Alici. B; Ozkara. H; Gulyasar. T; Akyolcu. MC. (2007). High levels
        of cadmium and lead in seminal fluid and blood of smoking men are associated with high oxidative
        stress and damage in infertile  subjects. Biol Trace Elem Res 120: 82-91.
        http://dx.doi.org/10.1007/sl2011-007-8020-8

     Kordas. K; Ettinger. AS: Bellinger. DC: Schnaas. L; Tellez Rojo. MM; Hernandez-Avila. M; Hu. H;
        Wright. RO. (2011). A dopamine receptor (DRD2) but not dopamine transporter (DAT1) gene
        polymorphism is associated with neurocognitive development of Mexican preschool children with
        lead exposure. J Pediatr 159: 638-643. http://dx.doi.Org/10.1016/j.jpeds.2011.03.043

     Kordas. K; Ettinger. AS: Lamadrid-Figueroa. H; Tellez-Rojo. MM; Hernandez-Avila. M; Hu. H; Wright.
        RO. (2009). Methylenetetrahydrofolate reductase  (MTHFR) C677T, A1298C and Gl 793 A genotypes,
        and the relationship between maternal folate intake, tibia lead and infant size at birth. Br J Nutr 102:
        907-914. http://dx.doi.org/10.1017/s0007114509318280

     Kordas. K; Queirolo. El; Ettinger. AS: Wright. RO: Stoltzfus. RJ. (2010). Prevalence and predictors of
        exposure to multiple metals in preschool children  from Montevideo, Uruguay. Sci Total Environ 408:
        4488-4494. http://dx.doi.Org/10.1016/i.scitotenv.2010.06.041

     Krieg. EF. Jr; Butler. MA; Chang. MH; Liu. TB; Yesupriya. A; Lindegren. ML; Dowling. N. (2009). Lead
        and cognitive function in ALAD genotypes in the  Third National Health and Nutrition Examination
        Survey. Neurotoxicol Teratol 31: 364-371. http://dx.doi.Org/10.1016/i.ntt.2009.08.003
November 2012                              6-48                   Draft - Do Not Cite or Quote

-------
     Krieg. EF. Jr; Butler. MA; M-h. C: Liu. T; Yesupriya. A; Dowling. N; Lindegren. ML. (2010). Lead and
        cognitive function in VDR genotypes in the Third National Health and Nutrition Examination Survey.
        Neurotoxicol Teratol 32: 262-272. http://dx.doi.Org/10.1016/i.ntt.2009.12.004

     Lagerkvist. BJ; Ekesrydh. S: Englyst. V: Nordberg. GF; Soderberg. HA; Wiklund. DE. (1996). Increased
        blood lead and decreased calcium levels during pregnancy: A prospective study of Swedish women
        living near a smelter. Am J Public Health 86: 1247-1252.

     Laidlaw. MAS: Filippelli. GM. (2008). Resuspension of urban soils as a persistent source of lead
        poisoning in children: A review and new directions. Appl Geochem 23: 2021-2039.
        http://dx.doi.0rg/10.1016/i.apgeochem.2008.05.009

     Lamadrid-Figueroa. H; Tellez-Rojo. MM; Hernandez-Cadena. L; Mercado-Garcia. A; Smith. D; Solano-
        Gonzalez. M; Hernandez-Avila. M; Hu. H. (2006). Biological markers of fetal lead exposure at each
        stage of pregnancy. J Toxicol Environ Health A 69: 1781-1796.
        http://dx.doi.org/10.1080/15287390600630195

     Lanphear. BP; Matte. TD; Rogers. J: Clickner. RP; Dietz. B; Bornschein. RL; Succop. P; Mahaffev. KR;
        Dixon. S: Galke. W: Rabinowitz. M; Farfel. M; Rohde. C: Schwartz. J: Ashley. P; Jacobs. DE. (1998).
        The contribution of lead-contaminated house dust and residential soil to children's blood lead levels: A
        pooled analysis of 12 epidemiologic studies. Environ Res 79: 51-68.
        http://dx.doi.org/10.1006/enrs.1998.3859

     Laxen. DPH; Raab. GM; Fulton. M. (1987). Children's blood lead and exposure to lead in household dust
        and water ~ a basis for an environmental standard for lead in dust. Sci Total Environ 66: 235-244.

     Leasure. JL; Giddabasappa. A; Chanev. S: Johnson. JE; Pothakos. K; Lau. YS: Fox. DA. (2008). Low-
        level human equivalent gestational lead exposure produces sex-specific motor and coordination
        abnormalities and late-onset obesity in year-old mice. Environ Health Perspect 116: 355-361.
        http://dx.doi.org/10.1289/ehp.10862

     Lee. JE; Naqi. SA; Kao. E; Dietert. RR. (2002). Embryonic exposure to lead: Comparison of immune and
        cellular responses in unchallenged and virally stressed chickens. Arch Toxicol 75: 717-724.
        http://dx.doi.org/10.1007/s00204-001-0283-9

     Leggett. RW. (1993). An age-specific kinetic model of lead metabolism in humans. Environ Health
        Perspect 101: 598-616.

     Levin. R; Brown. MJ; Kashtock. ME; Jacobs. DE; Whelan. EA; Rodman. J: Schock. MR; Padilla. A;
        Sinks. T. (2008). Lead exposures in US children, 2008: Implications for prevention. Environ Health
        Perspect 116: 1285-1293. httrj://dx.doi.org/10.1289/ehp.H241

     Liu. J; McCaulev. L; Compher. C: Yan. C: Shen. X; Needleman. H; Pinto-Martin. JA.  (2011 a). Regular
        breakfast and blood lead levels among preschool children. Environ Health Global Access Sci Source
        10: 28. http://dx.doi.org/10.1186/1476-069X-10-28

     Mahaffev. KR; Annest. JL. (1986). Association of erythrocyte protoporphyrin with blood lead level and
        iron status in the second national health and nutrition examination survey, 1976-1980. Environ Res 41:
        327-338. http://dx.doi.org/10.1016/80013-9351(86)80194-3

     Marcus. AH; Schwartz. J. (1987). Dose-response curves for erythrocyte protoporphyrin vs blood lead:
        Effects of iron status. Environ Res 44: 221-227. http://dx.doi.org/10.1016/S0013-9351(87)80230-X

     Martuzevicius. D; Grinshpun. SA; Reponen. T; Gorny. RL; Shukla. R; Lockev. J: Hu.  S: McDonald. R;
        Biswas. P: Kliucininkas. L; LeMasters. G. (2004). Spatial and temporal variations of PM2.5
        concentration and composition throughout an urban area with  high freeway density: The Greater
        Cincinnati study. Atmos Environ 38: 1091-1105. http://dx.doi.Org/10.1016/j.atmosenv.2003.ll.015

     Mason. H; Gallagher. F; Sen. D. (2005). Window renovation and exposure to lead—an observational
        study. Occup Med (Lond) 55: 631-634. http://dx.doi.org/10.1093/occmed/kqil59
November 2012                              6-49                   Draft - Do Not Cite or Quote

-------
     Masso-Gonzalez. EL; Antonio-Garcia. MT. (2009). Natural antioxidants protect against lead-induced
        damage during pregnancy and lactation in rat's pups. Ecotoxicol Environ Saf 72: 2137-2142.
        http://dx.doi.0rg/10.1016/i.ecoenv.2009.03.013

     McGivern. RF; Sokol. RZ; Berman. NG. (1991). Prenatal lead exposure in the rat during the third week
        of gestation: Long-term behavioral, physiological and anatomical effects associated with reproduction.
        Toxicol Appl Pharmacol 110: 206-215. http://dx.doi.org/10.1016/S0041-008X(05)80003-l

     McMichael. AJ; Baghurst. PA; Vimpani. GV: Robertson. EF; Wigg. NR; Tong. SL. (1992).
        Sociodemographic factors modifying the effect of environmental lead on neuropsychological
        development in early childhood. Neurotoxicol Teratol 14: 321-327.
        http://dx.doi.org/10.1016/0892-0362(92)90038-C

     McNeill. FE; Stokes. L; Brito. JA; Chettle. PR; Kave. WE. (2000). 109Cd K x-ray fluorescence
        measurements of tibial lead content in young adults exposed to lead in early childhood. Occup
        Environ Med 57: 465-471. http://dx.doi.Org/10.1136/oem.57.7.465

     Menke. A; Muntner. P; Batuman. V; Silbergeld. EK; Guallar. E. (2006). Blood lead below 0.48
        micromol/L (10 microg/dL) and mortality among US adults. Circulation 114: 1388-1394.
        http://dx.doi.org/10.1161/circulationaha.106.628321

     Mielke. HW: Laidlaw. MA; Gonzales. CR. (2011c). Estimation of leaded (Pb) gasoline's continuing
        material and health impacts on  90 US urbanized areas [Review]. Environ Int 37: 248-257.
        http://dx.doi.0rg/10.1016/i.envint.2010.08.006

     Mielke. HW: Powell. ET; Shah. A; Gonzales. CR: Mielke. PW. (2001). Multiple metal contamination
        from house paints: Consequences of power sanding and paint scraping in New Orleans. Environ
        Health Perspect  109: 973-978. http://dx.doi.org/10.1289/ehp.01109973

     Miranda. ML; Edwards. SE; Swamy. GK; Paul. CJ: Neelon. B. (2010). Blood lead levels among pregnant
        women: Historical versus contemporaneous exposures. Int J Environ Res Public Health 7: 1508-1519.
        http://dx.doi.org/10.3390/iierph7041508

     Miranda. ML; Kim. D; Galeano. M; Paul. CJ: Hull. AP; Morgan. SP (2007a). The relationship between
        early childhood blood lead levels and performance on end of grade tests. Environ Health Perspect 115:
        1242-1247. http://dx.doi.org/10.1289/ehp.9994

     Miyaki. K; Lwin. H; Masaki. K; Song. YX; Takahashi. Y; Muramatsu. M; Nakavama. T. (2009).
        Association between a polymorphism of aminolevulinate dehydrogenase (ALAD) gene and blood lead
        levels  in Japanese subjects. Int  J Environ Res Public Health 6: 999-1009.
        http://dx.doi.org/10.3390/ijerph6030999

     Montenegro. MF; Barbosa. F. Jr; Sandrim. VC: Gerlach. RF; Tanus-Santos. JE. (2006). A polymorphism
        in the delta-aminolevulinic acid dehydratase gene modifies plasma/whole blood lead ratio. Arch
        Toxicol 80: 394-398. http://dx.doi.org/10.1007/s00204-005-0056-v

     Muntner. P; Menke. A; DeSalvo. KB; Rabito. FA; Batuman. V. (2005). Continued decline in blood lead
        levels  among adults in the United States - The National Health and Nutrition Examination Surveys.
        Arch Intern Med 165: 2155-2161. http://dx.doi.org/10.1001/archinte.165.18.2155

     Muwakkit. S: Nuwayhid. I; Nabulsi. M; al Haij. R; Khoury. R; Mikati. M; Abboud. MR. (2008). Iron
        deficiency in young Lebanese children: Association with elevated blood lead levels. J Pediatr Hematol
        Oncol 30: 382-386.  http://dx.doi.org/10.1097/MPH.Ob013e318165b283

     Mykkanen. HM; Wasserman. RH.  (1981). Gastrointestinal absorption of lead (203Pb)  in chicks: Influence
        of lead, calcium, and age. JNutr 111: 1757-1765.

     Navas-Acien. A; Tellez-Plaza. M; Guallar. E; Muntner. P; Silbergeld. E; Jaar. B; Weaver. V. (2009). Blood
        cadmium and lead and chronic  kidney disease in US adults: Ajoint analysis. Am J Epidemiol 170:
        1156-1164. http://dx.doi.org/10.1093/aje/kwp248

     NCHS (National Center for Health Statistics). (2008). 2007-2008 NHANES:  Demographics. Retrieved
        from ftp://ftp.cdc.gov/pub/Health  Statistics/NCHS/nhanes/2007-2008/DEMO E.xpt
November 2012                              6-50                  Draft - Do Not Cite or Quote

-------
     NCHS (National Center for Health Statistics). (2009). NHANES: 2007-2008 data documentation,
        codebook, and frequencies: Blood cadmium and lead. Atlanta, GA: Centers for Disease Control and
        Prevention, http://www.cdc.gov/nchs/nhanes/nhanes2007-2008/pbcd e.htm

     Nriagu. J: Burt. B; Linder. A; Ismail. A; Sohn. W. (2006). Lead levels in blood and saliva in a low-income
        population of Detroit, Michigan. Int J Hyg Environ Health 209: 109-121.
        http://dx.doi.0rg/10.1016/i.iiheh.2005.ll.005

     O'Flahertv. EJ. (1995). Physiologically based models for bone-seeking elements: V. Lead absorption and
        disposition in childhood [Review]. Toxicol Appl Pharmacol 131: 297-308.
        http://dx.doi.org/10.1006/taap. 1995.1072

     O'Flahertv. EJ. (1998). A physiologically based kinetic model for lead in children and adults [Review].
        Environ Health Perspect 106: 1495-1503. http://dx.doi.org/10.1289/ehp.98106s61495

     Olnev. JW. (1969). Brain lesions, obesity and others disturbances in mice treated with monosodium
        glutamate. Science 164:  719-721. http://dx.doi.org/10.1126/science.164.3880.719

     Park. SK; Schwartz. J: Weisskopf. M; Sparrow. D; Vokonas. PS: Wright. RO: Coull. B; Nie. H; Hu. H.
        (2006). Low-level lead exposure, metabolic syndrome, and heart rate variability: The VA Normative
        Aging Study. Environ Health Perspect 114:  1718-1724.

     Peters. JL; Kubzansky. L; McNeely. E; Schwartz. J: Spiro. A. Ill; Sparrow. D; Wright. RO: Nie. H; Hu. H.
        (2007). Stress as a potential modifier of the impact of lead levels on blood pressure: The Normative
        Aging Study. Environ Health Perspect 115:  1154-1159. http://dx.doi.org/10.1289/ehp.10002

     Peters. JL; Wright. RJ: Weisskopf. MG: Spiro. A. Ill; Schwartz. J: Sparrow.  D: Nie. H; Hu. H; Wright.
        RO. (2008). Interaction of stress and lead burden on cognition in older men: The VA Normative Aging
        Study [Abstract]. Epidemiology 19: S134. http://dx.doi.org/10.1097/01.ede.0000339926.08729.35

     Pillai. P; Patel. R; Pandva. C: Gupta. S. (2009). Sex-specific effects of gestational and lactational
        coexposure to lead and cadmium on hepatic phase I and phase II xenobiotic/steroid-metabolizing
        enzymes and antioxidant status. J Biochem  Mol Toxicol 23: 419-431.
        http://dx.doi.org/10.1002/ibt.20305

     Pilsner. JR; Hu. H; Wright. RO: Kordas. K; Ettinger. AS: Sanchez.  BN; Cantonwine. D: Lazarus. AL;
        Cantoral. A; Schnaas. L; Tellez-Rojo. MM; Hernandez-Avila. M. (2010). Maternal MTHFR genotype
        and haplotype predict deficits in early cognitive development in a lead-exposed birth cohort in Mexico
        City. Am J Clin Nutr 92: 226-234. http://dx.doi.org/10.3945/aicn.2009.28839

     Pineda-Zavaleta. AP;  Garcia-Vargas. G: Borja-Aburto. VH; Acosta-Saavedra. LC: Aguilar. EV; Gomez-
        Mufioz. A; Cebrian. ME; Calderon-Aranda. ES. (2004). Nitric oxide and superoxide anion production
        in monocytes from children exposed to arsenic and lead in region Lagunera, Mexico. Toxicol Appl
        Pharmacol 198:  283-290. http://dx.doi.Org/10.1016/i.taap.2003.10.034

     Pirkle. JL: Brodv. DJ: Gunter. EW: Kramer. RA; Paschal. DC: Flegal. KM; Matte. TD.  (1994). The
        decline in blood lead levels in the United States: The National Health and Nutrition Examination
        Surveys (NHANES). JAMA 272: 284-291.  http://dx.doi.org/10.1001/iama.1994.03520040046039

     Pizent. A; Macan. J: Jurasovic. J: Varnai. VM; Milkovic-Kraus. S: Kanceljak-Macan. B. (2008).
        Association of toxic and essential metals with atopy markers and ventilatory lung function in women
        and men. Sci Total Environ 390: 369-376. http://dx.doi.Org/10.1016/i.scitotenv.2007.10.049

     Pleis. JR: Lucas. JW:  Ward. BW. (2009). Summary health statistics for U.S.  adults: National Health
        Interview Survey, 2008.  (DHHS 2010-1570). Hyattsville, MD: National  Center for Health Statistics.
        http://www.cdc.gov/nchs/data/series/sr 10/srlO 242.pdf

     Popovic. M; McNeil!. FE; Chettle. PR; Webber. CE; Lee. CV: Kave. WE. (2005). Impact of occupational
        exposure on lead levels in women. Environ Health Perspect 113: 478-484.
        http://dx.doi.org/10.1289/ehp.7386
November 2012                              6-51                   Draft - Do Not Cite or Quote

-------
     Rabstein. S: Unfried. K; Ranft. U: Illig. T; Kolz. M; Mambetova. C: Vlad. M; Roman. C: Weiss. T;
        Becker. D; Bruning. T; Pesch. B. (2008). Lack of association of delta-aminolevulinate dehydratase
        polymorphisms with blood lead levels and hemoglobin in Romanian women from a lead-contaminated
        region. J Toxicol Environ Health A 71: 716-724. http://dx.doi.org/10.1080/15287390801985190

     Rajan. P; Kelsev. KT; Schwartz. JD; Bellinger. DC: Weuve. J: Spiro. A. Ill; Sparrow. D; Smith. TJ; Nie.
        H; Weisskopf. MG: Hu. H; Wright. RO. (2008). Interaction of the delta-aminolevulinic acid
        dehydratase polymorphism and lead burden on cognitive function: The VA Normative Aging Study. J
        Occup Environ Med 50: 1053-1061. http://dx.doi.org/10.1097/JOM.Ob013e3181792463

     Rajaraman. P; Stewart. PA; Samet. JM; Schwartz. BS: Linet. MS: Zahm. SH; Rothman. N; Yeager. M;
        Fine. HA; Black. PM; Loeffler. J: Shapiro. WR; Selker. RG: Inskip. PP. (2006). Lead, genetic
        susceptibility, and risk of adult brain tumors. Cancer Epidemiol Biomarkers Prev 15: 2514-2520.
        http://dx.doi.org/10.1158/1055-9965.EPI-06-0482

     Rezende. VB; Barbosa. F. Jr; Montenegro. MF;  Sandrim. VC: Gerlach. RF; Tanus-Santos. JE. (2008).
        Haplotypes of vitamin D receptor modulate the circulating levels of lead in exposed subjects. Arch
        Toxicol 82: 29-36. http://dx.doi.org/10.1007/s00204-007-0231-4

     Ris. MD; Dietrich. KN; Succop. PA; Berger. OG: Bornschein. RL. (2004). Early exposure to lead and
        neuropsychological outcome in adolescence. J Int Neuropsychol Soc 10: 261-270.
        http://dx.doi.org/10.1017/S1355617704102154

     Rossi-George. A; Virgolini. MB; Weston. D; Corv-Slechta. DA. (2009). Alterations in glucocorticoid
        negative feedback following maternal Pb, prenatal stress and the combination: A potential biological
        unifying mechanism for their corresponding disease profiles. Toxicol Appl Pharmacol 234:  117-127.
        http://dx.doi.0rg/10.1016/i.taap.2008.10.003

     Rothenberg. SJ: Karchmer. S: Schnaas. L; Perroni. E; Zea. F; Alba. JF. (1994a). Changes in serial blood
        lead levels during pregnancy. Environ Health Perspect 102: 876-880.
        http://dx.doi.org/10.1289/ehp.94102876

     Sarasua. SM; Vogt. RF; Henderson. LO; Jones. PA; Lybarger. JA. (2000). Serum immunoglobulins and
        lymphocyte subset distributions in children and adults living in communities assessed for lead and
        cadmium exposure. J Toxicol Environ Health A 60: 1-15. http://dx.doi.org/10.1080/009841000156556

     Saxena. DK; Singh. C: Murthy. RC: Chandra. SV. (1991). Lead distribution in iron-deficient rats: I. Lead
        exposure during the period of organogenesis. Nutr Res 11: 471-477.
        http://dx.doi.org/10.1016/80271-5317(05)80009-0

     Schell. LM; Denham. M; Stark. AD; Ravenscroft. J: Parsons. P: Schulte. E. (2004). Relationship between
        blood lead concentration and dietary intakes of infants from 3 to 12 months of age. Environ Res 96:
        264-273. http://dx.doi.0rg/10.1016/i.envres.2004.02.008

     Schober. SB; Mirel. LB; Graubard. BI; Brodv. DJ: Flegal. KM. (2006). Blood lead levels and death from
        all causes, cardiovascular disease, and cancer: Results from the NHANES III Mortality Study. Environ
        Health Perspect  114: 1538-1541. http://dx.doi.org/10.1289/ehp.9123

     Schuhmacher. M; Hernandez. M; Domingo. JL; Fernandez-Ballart JD: Llobet JM; Corbella. J. (1996). A
        longitudinal study of lead mobilization during pregnancy: Concentrations in maternal and umbilical
        cord blood. Trace Elem Electroly 13: 177-181.

     Schwartz. BS: Lee. BK; Lee. GS: Stewart. WF;  Simon. D; Kelsev. K;  Todd. AC. (2000a). Associations of
        blood lead, dimercaptosuccinic acid-chelatable lead, and tibia lead with polymorphisms in the vitamin
        D receptor and delta-aminolevulinic acid dehydratase genes. Environ Health Perspect 108: 949-954.
        http://dx.doi.org/10.2307/3435053

     Schwartz. J. (1994). Low-level lead exposure and children's IQ: A meta-analysis and search for a
        threshold. Environ Res 65: 42-55. http://dx.doi.org/10.1006/enrs.1994.1020

     Scinicariello. F; Murray. HE; Moffett. DB; Abadin. HG: Sexton. MJ; Fowler.  BA. (2007). Lead and delta-
        aminolevulinic acid dehydratase polymorphism: Where does  it lead? A meta-analysis. Environ Health
        Perspect 115: 35-41. http://dx.doi.org/10.1289/ehp.9448
November 2012                              6-52                   Draft - Do Not Cite or Quote

-------
     Scinicariello. F; Yesupriya. A; Chang. MH; Fowler. BA. (2010). Modification by ALAD of the
        association between blood lead and blood pressure in the U.S. population: Results from the Third
        National Health and Nutrition Examination Survey. Environ Health Perspect 118: 259-264.
        http://dx.doi.org/10.1289/ehp.0900866

     Selevan. SG: Rice. DC: Hogan. KA; Euling. SY; Pfahles-Hutchens. A; Bethel. J. (2003). Blood lead
        concentration and delayed puberty in girls. N Engl J Med 348: 1527-1536.
        http://dx.doi.org/10.1056/NEJMoa020880

     Shaik. AP; Jamil. K. (2009). Individual susceptibility and genotoxicity in workers exposed to hazardous
        materials like lead. J Hazard Mater 168: 918-924. http://dx.doi.Org/10.1016/i.jhazmat.2009.02.129

     Sherry. B; Blanck. HM; Galuska. DA; Pan. L; Dietz. WH; Balluz. L. (2010). Vital signs: State-specific
        obesity prevalence among adults - United States, 2009. MMWR Recomm Rep 59: 951-955.

     Singh. C: Saxena. DK; Murthy. RC: Chandra. SV. (1993b). Embryo-fetal development influenced by lead
        exposure in iron-deficient rats. Hum Exp Toxicol 12: 25-28.
        http://dx.doi.org/10.1177/096032719301200105

     Six. KM; Gover. RA. (1970). Experimental enhancement of lead toxicity by low dietary calcium. Transl
        Res 76: 933-942.

     Sobekova. A; Holovska. K; Lenartova. V: Legath. J: Javorsky. P. (2009). The alteration of glutathione
        peroxidase activity in rat organs after lead exposure. Acta Physiol Hung 96: 37-44.
        http://dx.doi.Org/10.1556/APhvsiol.96.2009.l.4

     Sobin. C: Gutierrez. M; Alterio. H. (2009). Polymorphisms of delta-aminolevulinic acid dehydratase
        (ALAD) and peptide transporter 2 (PEPT2) genes in children with low-level lead exposure.
        Neurotoxicology 30: 881-887. http://dx.doi.Org/10.1016/i.neuro.2009.08.006

     Songdej. N; Winters. PC: McCabe. MJ. Jr; van Wijngaarden. E. (2010). A population-based assessment of
        blood lead levels in relation to inflammation. Environ Res 110: 272-277.
        http://dx.doi.0rg/10.1016/i.envres.2009.12.008

     SSDAN CensusScope (Social Science Data Analysis Network, CensusScope). (2010a). United States:
        Age distribution [Database]. Ann Arbor, Michigan:  Social Science Data Analysis Network. Retrieved
        from http://www.censusscope.org/us/chart age.html

     SSDAN CensusScope (Social Science Data Analysis Network, CensusScope). (2010b). United States:
        Population by race [Database]. Ann Arbor, Michigan. Retrieved from
        http://www.censusscope.org/us/chart race.html

     SSDAN CensusScope (Social Science Data Analysis Network, CensusScope). (2010c). United States:
        Poverty by age [Database]. Ann Arbor, Michigan. Retrieved from
        http://www.censusscope.org/us/chart_povertv.html

     Surkan. PJ; Schnaas. L; Wright. RJ: Tellez-Rojo. MM; Lamadrid-Figueroa. H; Hu. H; Hernandez-Avila.
        M; Bellinger. DC: Schwartz. J: Perroni. E; Wright. RO.  (2008). Maternal self-esteem, exposure to
        lead, and child neurodevelopment. Neurotoxicology 29: 278-285.
        http://dx.doi.0rg/10.1016/i.neuro.2007.ll.006

     Theppeang. K;  Glass. TA; Bandeen-Roche. K; Todd. AC: Rohde. CA; Schwartz. BS. (2008b). Gender and
        race/ethnicity differences in lead dose biomarkers. Am J Public Health 98: 1248-1255.
        http://dx.doi.org/10.2105/aiph.2007.118505

     Theppeang. K;  Schwartz. BS: Lee. BK; Lustberg. ME; Silbergeld. EK; Kelsev. KT;  Parsons. PJ: Todd.
        AC. (2004). Associations of patella lead with polymorphisms in the vitamin D receptor, delta-
        aminolevulinic acid dehydratase and endothelial nitric oxide synthase genes. J Occup Environ Med
        46: 528-537.
November 2012                              6-53                   Draft - Do Not Cite or Quote

-------
     Tian. Y; Green. PG: Stamova. B; Hertz-Picciotto. I; Pessah. IN; Hansen. R; Yang. X; Gregg. JP;
        Ashwood. P; Jickling. G: Van de Water. J; Sharp. FR. (2011). Correlations of gene expression with
        blood lead levels in children with autism compared to typically developing controls. Neurotox Res 19:
        1-13. http://dx.doi.org/10.1007/sl2640-009-9126-x

     Tong. S: McMichael. AJ; Baghurst. PA. (2000). Interactions between environmental lead exposure and
        sociodemographic factors on cognitive development. Arch Environ Health 55: 330-335.
        http://dx.doi.org/10.1080/00039890009604025

     Tsaih. SW: Korrick. S: Schwartz. J; Amarasiriwardena. C: Aro. A; Sparrow. D; Hu. H. (2004). Lead,
        diabetes, hypertension, and renal function: The Normative Aging Study. Environ Health Perspect 112:
        1178-1182. http://dx.doi.org/10.1289/ehp.7024

     U.S. Census Bureau. (2010). U.S. population projections [Database]. Retrieved from
        http://www.census.gov/population/www/projections/projectionsagesex.html

     U.S. EPA (U.S. Environmental Protection Agency). (2006b). Air quality criteria for lead: Volume I of II
        [EPA Report]. (EPA/600/R-05/144aF). Research Triangle Park, NC.
        http ://cfpub. epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823

     U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment for paniculate
        matter [EPA Report]. (EPA/600/R-08/139F). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546

     U.S. EPA (U.S. Environmental Protection Agency). (2010).  Integrated science assessment for carbon
        monoxide [EPA Report]. (EPA/600/R-09/019F). Research Triangle Park, NC.
        http ://cfpub .epa. gov/ncea/cfm/recordisplav. cfm?deid=218686

     Uzbekov. MG: Bubnova. NI; Kulikova. GV. (2007). Effect of prenatal lead exposure on superoxide
        dismutase activity in the brain and liver of rat fetuses. Bull Exp Biol Med 144: 783-785.
        http://dx.doi.org/10.1007/sl0517-007-0431-l

     Virgolini. MB; Rossi-George. A; Lisek. R; Weston. DP; Thiruchelvam. M; Cory-Slechta. DA. (2008a).
        CNS effects of developmental Pb exposure are enhanced by combined maternal and offspring stress.
        Neurotoxicology 29: 812-827. http://dx.doi.Org/10.1016/i.neuro.2008.03.003

     Virgolini. MB; Rossi-George. A; Weston. D; Cory-Slechta. DA. (2008b). Influence of low level maternal
        Pb exposure and prenatal stress on offspring stress challenge responsivity. Neurotoxicology 29:
        928-939. http://dx.doi.0rg/10.1016/i.neuro.2008.09.010

     Wananukul. W: Sura. T; Salaitanawatwong. P. (2006). Polymorphism of delta-aminolevulinic acid
        dehydratase and its effect on blood lead levels in Thai workers. Arch Environ  Occup Health 61: 67-72.
        http://dx.doi.Org/10.3200/AEOH.61.2.67-72

     Wang. FT; Hu. H; Schwartz. J: Weuve. J: Spiro. AS. Ill; Sparrow. D: Nie. HL; Silverman. EK; Weiss. ST;
        Wright. RO. (2007a). Modifying effects of the HFE polymorphisms on the association between lead
        burden and cognitive decline. Environ Health Perspect 115: 1210-1215.
        http://dx.doi.org/10.1289/ehp.9855

     Wang. GS: Fowler. BA. (2008). Roles of biomarkers in evaluating interactions among mixtures of lead,
        cadmium and arsenic [Review]. Toxicol Appl Pharmacol 233: 92-99.
        http://dx.doi.0rg/10.1016/i.taap.2008.01.017

     Wang. L; Wang. H; Hu. MZ; Cao. J: Chen. DW: Liu. Z. (2009c). Oxidative stress and apoptotic changes
        in primary cultures of rat proximal tubular cells exposed to lead. Arch Toxicol 83: 417-427.
        http://dx.doi.org/10.1007/s00204-009-0425-z

     Wang. YY; Sui. KX; Hong. LI; Ma. HY. (2009e). The effects of lead exposure on placenta! NF-kappaB
        expression and the consequences for gestation. Reprod Toxicol 27: 190-195.
        http://dx.doi.0rg/10.1016/i.reprotox.2008.12.006
November 2012                              6-54                   Draft - Do Not Cite or Quote

-------
     Weaver. VM; Ellis. LR; Lee. BK; Todd. AC: Shi. W: Ahn. KD; Schwartz. BS. (2008). Associations
        between patella lead and blood pressure in lead workers. Am J Ind Med 51: 336-343.
        http://dx.doi.org/10.1002/aiim.20573

     Weaver. VM: Lee. BK: Todd. AC: Ahn. KD: Shi. W: Jaar. BG: Kelsev. KT: Lustberg. ME: Silbergeld.
        EK; Parsons. PJ: Wen. J; Schwartz. BS. (2006). Effect modification by delta-aminolevulinic acid
        dehydratase, vitamin D receptor, and nitric oxide synthase gene polymorphisms on associations
        between patella lead and renal function in lead workers. Environ Res 102: 61-69.
        http://dx.doi.0rg/10.1016/i.envres.2006.01.001
     Weaver. VM: Schwartz. BS: Ahn. KD: Stewart. WF; Kelsev. KT; Todd. AC: Wen. J: Simon. DJ; Lustberg.
        ME; Parsons. PJ: Silbergeld. EK; Lee.  BK.  (2003b). Associations of renal function with
        polymorphisms in the "delta"-aminolevulinic acid dehydratase, vitamin D receptor, and nitric oxide
        synthase genes in Korean lead workers. Environ Health Perspect 111: 1613-1619.

     Weaver. VM: Schwartz. BS: Jaar. BG: Ahn. KD: Todd. AC: Lee. SS: Kelsev. KT: Silbergeld. EK:
        Lustberg. ME; Parsons. PJ: Wen. J: Lee. BK. (2005b). Associations of uric acid with polymorphisms
        in the "delta"-aminolevulinic acid dehydratase, vitamin D receptor, and nitric oxide synthase genes in
        Korean lead workers. Environ Health Perspect 113: 1509-1515. http://dx.doi.org/10.1289/ehp.7927

     Weiss. AL; Caravanos. J: Blaise. MJ; Jaeger. RJ. (2006).  Distribution of lead in urban roadway grit and its
        association with elevated steel structures. Chemosphere 65: 1762-1771.
        http://dx.doi.0rg/10.1016/i.chemosphere.2006.04.079
     Weuve. J: Kelsev. KT; Schwartz. J: Bellinger. D; Wright. RO: Raian. P; Spiro. A. Ill; Sparrow. D: Aro. A;
        Hu. H. (2006). Delta-aminolevulinic acid dehydratase polymorphism and the relation between low
        level lead exposure and the Mini-Mental Status Examination in older men: The Normative Aging
        Study. Occup Environ Med 63: 746-753. http://dx.doi.org/10.1136/oem.2006.027417

     Wright. JP: Dietrich. KN; Ris. MD; Hornung. RW: Wessel. SD; Lanphear. BP; Ho. M; Rae. MN. (2008).
        Association of prenatal and childhood blood lead concentrations with criminal arrests in early
        adulthood. PLoS Med 5: 732-740. http://dx.doi.org/10.1371/iournal.pmed.0050101

     Wright. RO: Tsaih. SW: Schwartz. J: Spiro. A; McDonald. K; Weiss. ST; Hu. H.  (2003). Lead exposure
        biomarkers and mini-mental status exam scores in older men. Epidemiology 14: 713-718.
        htrp://dx.doi.org/10.1097/01.EDE.0000081988.85964.db

     Wu. J: Edwards. R; He. XE; Liu. Z; Kleinman. M. (2010). Spatial analysis of bioavailable soil lead
        concentrations in Los Angeles, California. Environ Res 110: 309-317.
        http://dx.doi.0rg/10.1016/i.envres.2010.02.004

     Yang. Y; Ma. Y; Ni. L; Zhao. S: Li. L; Zhang. J: Fan. M; Liang. C: Cao. J: Xu. L. (2003). Lead exposure
        through gestation-only caused long-term learning/memory deficits in young adult offspring. Exp
        Neural 184: 489-495. http://dx.doi.org/10.1016/S0014-4886(03)00272-3

     Yapici. G: Can. G: Kiziler. AR; Aydemir. B; Timur. IH; Kavpmaz. A. (2006). Lead and cadmium exposure
        in children living around a coal-mining area in Yatagan, Turkey. Toxicol Ind Health 22: 357-362.

     Zahran. S: Mielke. HW: Gonzales. CR; Powell. ET; Weiler. S. (2010). New Orleans before and after
        hurricanes Katrina/Rita: A quasi-experiment of the association between soil lead and children's blood
        lead. Environ Sci Technol 44: 4433-4440. http://dx.doi.org/10.1021/esl00572s

     Zawia. NH; Basha. MR. (2005). Environmental risk factors and the developmental basis for Alzheimer's
        disease [Review]. RevNeurosci 16: 325-337.
     Zhang. A; Park. SK; Wright. RO: Weisskopf. MG: Mukherjee. B; Nie. H; Sparrow. D: Hu. H.  (2010a).
        HFE H63D  polymorphism as a modifier of the effect  of cumulative lead exposure on pulse pressure:
        The Normative Aging Study. Environ Health Perspect 118: 1261-1266.
        http://dx.doi.org/10.1289/ehp. 1002251

     Zhao. Y; Wang. L; Shen. HB; Wang. ZX; Wei. QY; Chen. F. (2007). Association between delta-
        aminolevulinic acid dehydratase (ALAD) polymorphism and blood lead levels: A meta-regression
        analysis. J Toxicol Environ Health 70:  1986-1994. http://dx.doi.org/10.1080/15287390701550946
November 2012                              6-55                   Draft - Do Not Cite or Quote

-------
       7 ECOLOGICAL  EFFECTS  OF  LEAD
 1                  This chapter synthesizes and evaluates the most policy-relevant science to help form the
 2                  foundation for the review of the secondary (welfare-based) NAAQS for Pb. The Clean
 3                  Air Act definition of welfare effects includes, but is not limited to, effects on soils, water,
 4                  wildlife, vegetation, visibility, weather, and climate, as well as effects on materials,
 5                  economic values, and personal comfort and well-being. This chapter discusses the effects
 6                  of Pb on ecosystem components and processes and is organized into five sections. The
 7                  introduction (Section 7.1) presents the organizing principles of this chapter and several
 8                  important general ecology concepts. An overview of fate and transport of Pb in
 9                  ecosystems including measured concentrations of this metal in various environmental
10                  media (i.e., soil, water, sediment) is presented in Section 7.2. Section 7.3 reviews the
11                  effects of Pb on terrestrial ecosystems; how soil biogeochemistry affects Pb
12                  bioavailability, biological effects of Pb exposure and subsequent vulnerability of
13                  particular ecosystems. A similar discussion of the effects of Pb on freshwater and
14                  saltwater ecosystems is presented in Section 7.4. including water-only exposures and
15                  sediment-related effects. The terrestrial, freshwater and saltwater sections each conclude
16                  with an integrative synthesis of new evidence for Pb effects and causal determinations,
17                  based on the synthesis of new  evidence and findings from previous Pb AQCDs. Section
18                  7.5 summarizes the causal determinations. Areas not addressed here include literature
19                  related to ingestion of Pb shot or pellets and studies that examine human health-related
20                  endpoints which are described in other chapters of this document.
      7.1        Introduction to Ecological  Concepts

21                  Metals, including Pb, occur naturally in the environment at measurable concentrations in
22                  soils, sediments, and water. Organisms have developed adaptive mechanisms for living
23                  with metals, some of which are required micronutrients (but not Pb). However,
24                  anthropogenic enrichment can result in concentrations that exceed the capacity of
25                  organisms to regulate internal concentrations, causing a toxic response and potentially
26                  death. Differences in environmental chemistry may enhance or inhibit uptake of metal
27                  from the environment, thus creating a spatial patchwork of environments that are at
28                  greater risk than other environments. Similarly, organisms vary in their degree of
29                  adaptation to, or tolerance of, the presence of metals. These fundamental principles of
30                  how metals interact with organisms and ecosystems are described in detail in EPA's
31                  Framework for Metals Risk Assessment (U.S. EPA. 2007c). This section introduces
32                  critical concepts for understanding how Pb from atmospheric deposition may affect
      November 2012                             7-1                  Draft - Do Not Cite or Quote

-------
 1                  organisms, communities, and ecosystems. The sections that follow provide more detail
 2                  for how aquatic and terrestrial ecosystems respond to Pb and how environmental
 3                  chemistry interacts with organisms to affect exposure and uptake.
      7.1.1  Ecosystem Scale, Function, and Structure

 4                  For this assessment, an ecosystem is defined as the interactive system formed from all
 5                  living organisms (biota) and their abiotic (chemical and physical) environment within a
 6                  given area (IPCC. 2007).The boundaries of what could be called an ecosystem are
 7                  somewhat arbitrary, depending on the focus of interest or study. Thus, the extent of an
 8                  ecosystem may range from very small spatial scales to, ultimately, the entire Earth
 9                  (IPCC. 2007). Ecosystems cover a hierarchy of spatial scales and can comprise the entire
10                  globe, biomes at the continental scale, or small, well-circumscribed systems such as a
11                  small pond (U.S. EPA. 2008e). A pond may be a small but complex system with multiple
12                  trophic levels ranging  from phytoplankton to several feeding guilds offish plus fish-
13                  eating birds or mammals. A large lake, on the other hand, may be a very simple
14                  ecosystem, such as the Great Salt Lake in Utah that covers approximately 1,700 square
15                  miles but contains only bacteria, algae, diatoms, and two invertebrate species. All
16                  ecosystems, regardless of size or complexity, share the commonality of multiple
17                  interactions between biota and abiotic factors, and a reduction in entropy through energy
18                  flow from photosynthetic organisms to top predators.  This includes both structural
19                  (e-g-, soil type and food web trophic levels) and functional (e.g., energy flow,
20                  decomposition, nitrification) attributes. Changes are often considered undesirable if
21                  important structural or functional components of ecosystems are altered following
22                  pollutant exposure (U.S. EPA. 1998V

23                  Ecosystems are most often defined by their structure,  and are based on the number and
24                  type of species present. Structure may refer to a variety of measurements including the
25                  species richness, abundance, community composition and biodiversity as well as
26                  landscape attributes. Individual organisms of the same species are similar in appearance
27                  and genetics,  and can interbreed and produce fertile offspring. Interbreeding groups of
28                  individual organisms within the same species that occupy some defined geographic space
29                  form populations, and populations of different species form communities (Barnthouse et
30                  al.. 2008). The community composition may also define an ecosystem type, such as a
31                  pine forest or a tall grass prairie. Pollutants can affect the ecosystem structure at any of
32                  these levels of biological organization (Suter et al., 2005). Individual plants or animals
33                  may exhibit changes in metabolism, enzyme  activities, hormone function, or overall
34                  growth rates or may suffer gross lesions, tumors, deformities, or other pathologies.
35                  Effects on the nervous system of animals may cause behavioral changes that alter


      November 2012                             7-2                  Draft - Do Not Cite or Quote

-------
 1                  breeding behaviors or predator avoidance. However, only some organism-level endpoints
 2                  such as growth, survival and reproductive output have been definitively linked to effects
 3                  at the population level and above. Examples of organism-level endpoints with direct links
 4                  to population level effects include mass mortality, gross anomalies, survival, fecundity
 5                  and growth (Suteretal.. 2004). Population level effects of pollutants include changes
 6                  over time in abundance or density (number of individuals in a defined area), age or sex
 7                  structure, and production or sustainable rates of harvest (Barnthouse et al., 2008).
 8                  Community level attributes affected by pollutants include species richness and abundance
 9                  (also known as biodiversity), dominance of one species over another, or size (area) of the
10                  community. Pollutants may affect communities in ways that are not observable in
11                  organisms or populations (Bartell 2007). including: (1) effects resulting from interactions
12                  between species, such as altering predation rates or competitive advantage; (2) indirect
13                  effects, such as reducing or removing one species from the assemblage and allowing
14                  another to emerge (Petraitis and Latham. 1999); and (3) alterations in trophic structure.

15                  Alternatively, ecosystems may be defined on a functional basis. "Function" refers to the
16                  suite of processes and interactions among the ecosystem components and their
17                  environment that involve nutrient and energy flow as well as other attributes including
18                  water dynamics and the  flux  of trace gases such as rates of photosynthesis,
19                  decomposition, nitrification,  or carbon cycling. Pollutants may affect abiotic conditions
20                  (e.g., soil chemistry), which indirectly influences biotic structure and function (Bartell
21                  2007). Feedback loops or networks influence the stability of the system, and can be
22                  mathematically described through simplistic or complex process, or energy flow, models
23                  (Bartell.  2007). For example, the Comprehensive Aquatic Systems Model (CASM) is a
24                  bioenergetics-based multicompartment model that describes the daily production of
25                  biomass  (carbon)  by populations of aquatic plants and animals over an annual cycle
26                  (DeAngelis et al.. 1989). CASM, originally designed to examine theoretical relationships
27                  between food web structure,  nutrient cycling, and ecosystem stability, has since been
28                  adapted for risk assessments  and has been applied to numerous lakes with a variety of
29                  pollutants (Bartell 2007). Likewise, other theoretical ecosystem models are being
30                  modified for use in assessing ecological risks from pollutant exposures (Bartell. 2007).

31                  Some ecosystems, and some  aspects of particular ecosystems, are less vulnerable to long-
32                  term consequences of pollutant exposure. Other ecosystems may be profoundly altered if
33                  a single attribute is affected.  Thus, spatial and temporal definitions of ecosystem structure
34                  and function become an essential factor in defining impacted ecosystem services and
35                  critical loads of particular pollutants, either as single pollutants or in combination with
36                  other stressors. Both ecosystem services (Section 7.1.2) and critical loads (Section 7.1.3)
37                  serve as benchmarks or measures of the impacts of pollutants on ecosystems.
      November 2012                             7-3                   Draft - Do Not Cite or Quote

-------
      7.1.2  Ecosystem Services

 1                  Ecosystem structure and function may be translated into ecosystem services (Daily.
 2                  1997). Ecosystem services are the benefits people obtain from ecosystems (UNEP, 2003).
 3                  Ecosystem services are defined as the varied and numerous ways that ecosystems are
 4                  important to human welfare and how they provide many goods and services that are of
 5                  vital importance for the functioning of the biosphere. This concept has gained recent
 6                  interest and support because it recognizes that ecosystems are valuable to humans, and
 7                  are important in ways that are not generally appreciated (Daily. 1997). Ecosystem
 8                  services also provide a context for assessing the collective effects of human actions on a
 9                  broad range of the goods and services upon which humans rely.

10                  In general, both ecosystem structure and function play essential roles in providing goods
11                  and services. Ecosystem processes provide diverse benefits including absorption and
12                  breakdown of pollutants, cycling of nutrients, binding of soil, degradation of organic
13                  waste, maintenance of a balance of gases in the air, regulation of radiation balance and
14                  climate, and fixation  of solar energy fWRI. 2000: Daily. 1997: Westman. 1977). These
15                  ecological benefits, in turn, provide economic benefits and values to society (Costanza et
16                  al.. 1997: Pimentel etal..  1997). Goods such as food crops, timber, livestock, fish and
17                  clean drinking water  have market value. The values of ecosystem services such as flood
18                  control, wildlife habitat, cycling of nutrients and removal of air pollutants are more
19                  difficult to measure (Goulder and Kennedy. 1997).

20                  Particular concern has developed within the past decade regarding the consequences  of
21                  decreasing biological diversity (Tilman. 2000: Ayensuetal.. 1999: Wall. 1999: Chapin et
22                  al.. 1998: Hooper and Vitousek. 1997). Human activities that decrease biodiversity also
23                  alter the complexity and stability of ecosystems and change ecological processes. In
24                  response, ecosystem  structure and function can be affected (Daily and Ehrlich. 1999:
25                  Wall. 1999: Chapin etal. 1998: Levin. 1998: Peterson et al.. 1998: Tilman. 1996: Tilman
26                  and Downing. 1994:  Pimm. 1984). Biodiversity is an important consideration at all levels
27                  of biological organization, including species,  communities, populations, and ecosystems.
28                  Human-induced changes in biotic diversity and alterations in the structure and
29                  functioning of ecosystems are two of the most dramatic ecological trends of the past
30                  century (U.S. EPA. 2004: Vitousek et al.. 1997).
      November 2012                             7-4                   Draft - Do Not Cite or Quote

-------
 1                  Hassan et al. (2005) identified four broad categories of ecosystem services:

 2                     • Supporting services are necessary for the production of all other ecosystem
 3                      services. Some examples include biomass production, production of
 4                      atmospheric O2, soil formation and retention, nutrient cycling, water cycling and
 5                      provisioning of habitat. Biodiversity is a supporting service in that it is
 6                      increasingly recognized to sustain many of the goods and services that humans
 7                      enjoy from ecosystems. These supporting services provide a basis for  an
 8                      additional three higher-level categories of services.
 9                     • Provisioning services such as products (Gitav et al.. 2001) i.e., food (including
10                      game meat, roots, seeds, nuts, and other fruit, spices, fodder), water, fiber
11                      (including  wood, textiles) and medicinal and cosmetic products.
12                     • Regulating services that are of paramount importance for human society such as
13                      (1) carbon  sequestration, (2) climate and water regulation, (3) protection from
14                      natural hazards such as floods, avalanches, or rock-fall, (4) water and air
15                      purification, and (5) disease and pest regulation.
16                     • Cultural services that satisfy human spiritual and aesthetic appreciation of
17                      ecosystems and their components.
     November 2012                             7-5                  Draft - Do Not Cite or Quote

-------
      7.1.3  Critical Loads as an Organizing Principle for Ecological Effects of
             Atmospheric Deposition

 1                  A critical load is defined as, "a quantitative estimate of an exposure to one or more
 2                  pollutants below which significant harmful effects on specified sensitive elements of the
 3                  environment do not occur according to present knowledge" (Nilsson and Grennfelt.
 4                  1988). Critical loads are a powerful organizing principle for information that links
 5                  atmospheric deposition with ecological impairment. They allow for heterogeneity in
 6                  ecosystem sensitivity and exposure which often results in critical load values that vary by
 7                  ecosystem (e.g., aquatic-water; aquatic-sediment; terrestrial), and differ by endpoint of
 8                  concern. It is important to consider that critical loads are often calculated assuming
 9                  steady state conditions (i.e., how much input is required to balance the rate of output),
10                  and there may be time required to reach the critical load (i.e., the lag time between onset
11                  of exposure and induction of measurable effects). The following types of information are
12                  required to calculate a critical load, each of which is discussed in more detail in the
13                  subsequent sections of this chapter:

14                     • Ecosystem at risk;
15                     • Receptors of concern (plants, animals, etc.);
16                     • Endpoints of concern (organism, population or community responses, changes
17                       in ecosystem services or functions);
18                     • Dose (concentration) - response relationships and threshold levels of effects;
19                     • Bioavailability and bioaccumulation rates;
20                     • Naturally occurring (background) Pb (or other metal) concentrations; and
21                     • Biogeochemical modifiers of exposure.
22                  There is no single "definitive" critical load for a pollutant, partly because critical load
23                  estimates reflect the current state-of-knowledge and policy priorities, and also because of
24                  local or regional differences among ecosystems (U.S. EPA, 2008e). Changes in scientific
25                  understanding may include, for example, expanded information about dose-response
26                  relationships, better understanding of bioavailability factors, and improved quantitative
27                  models for effects predictions. Changes in policy may include new mandates for resource
28                  protection, inclusion of perceived new threats that may exacerbate the effects of the
29                  pollutant of concern (e.g., climate change), and a better understanding of the value of
30                  ecosystem services.
      November 2012                             7-6                  Draft - Do Not Cite or Quote

-------
     7.2       Fate and Transport of Pb in  Ecosystems
 i
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
Fate and transport of Pb in ecosystems are difficult to assess because Pb detected in the
environment could have multiple sources and passes through various environmental
media within a watershed. These issues are described in detail in Section 3.3. Pb can be
emitted to air, soil, or water and then cycle through any or all of these media. In addition
to primary emission of particle-bound or gaseous Pb to the atmosphere, Pb can be
resuspended to the air from soil or dust (Section 7.2.2). Additionally, Pb-bearing PM can
be deposited from the  air to soil or water through wet and dry deposition. The
complicated nature of Pb fate and transport in ecosystems is illustrated in Figure 7-1 in
which the Venn diagram depicts how Pb can cycle through multiple environmental media
that encompass both terrestrial and aquatic systems (see also Figure 3-9). The
"air/soil/water" arrows illustrate Pb exposures to plants and animals. Many of the studies
presented in the subsequent material focus on observations of Pb exposure via one
medium: air, soil, sediment, or water.
                                        Newly Emitted Pb
                                     Historically Emitted  Pb
              NATURAL WATERS
              AND SEDIMENTS
                                                  OUTDOOR SOIL
                                                     AND DUST
Non-air Pb
  eleases
                                   AIR
                                  SOIL
                               SEDIMENT
                                 WATER
                                        AIR
                                        SOIL
                                     SEDIMENT
                                       WATER
                                 PLANT
                              EXPOSURE
                                      ANIMAL
                                    EXPOSURE
     Figure 7-1
    Fate of atmospheric Pb in ecosystems.
     November 2012
                          7-7
                    Draft - Do Not Cite or Quote

-------
      7.2.1      Fate and Transport
 1                  This section provides a brief overview of the fate and transport of Pb in ecosystems. Fate
 2                  of Pb is determined by the chemical and physical properties of the medium in wet
 3                  deposition, bodies of water, or soil (e.g., pH, salinity, oxidation status, flow rate and the
 4                  suspended sediment load and its constituents). Desorption, dissolution, precipitation,
 5                  sorption and complexation processes can all occur concurrently and continuously, leading
 6                  to transformations and redistribution of Pb within a watershed. The pH of water is of
 7                  primary importance in determining the likely chemical fate of Pb in terms of solubility,
 8                  precipitation, or organic complexation. For more detailed information about the fate and
 9                  transport of Pb, please see Section 3.3.

10                  Soluble Pb in air is mostly removed by wet deposition, and most of the insoluble Pb is
11                  removed by dry deposition. As a result, dry deposition is the major removal mechanism
12                  for Pb in coarse PM (which is mainly insoluble) and wet deposition is the most important
13                  removal mechanism for fine PM and Pb halides (which were more soluble)
14                  (Section 3.3.1). Recent research provides considerable evidence that appreciable amounts
15                  of Pb can accumulate on coarse PM during transport, and that the physical and chemical
16                  characteristics of Pb can be altered by this process due to accompanying transformations
17                  (Section 3.3.1.1). Atmospheric removal of metals by wet or dry deposition  is largely
18                  controlled by solubility of Pb in rain water. The relative importance of wet  and dry
19                  deposition is highly variable with  respect to location  and  season, probably reflecting both
20                  variations in Pb speciation and variations in external  factors such as pH and rain water
21                  composition (Section 3.3.1.2).

22                  Pb deposited to terrestrial ecosystems may remain in soils or eventually be  transported in
23                  runoff to streams, lakes or rivers in the watershed. Pb has a relatively long retention time
24                  in the organic soil horizon, although its movement through the soil column also suggests
25                  potential for contamination of groundwater (Section 3.3.3). Pb deposition to soils has
26                  decreased since the  phase-out of leaded on-road gasoline  (Section 3.3.3.1).  Recent studies
27                  of metal concentrations in leaf litter and organic roadside debris suggest that the litter can
28                  act as a temporary sink for metals  from the soil around and below leaves on the ground
29                  (Section 3.3.3.2). Leaching has been consistently observed to be a slower process for Pb
30                  than for other contaminants because Pb is only weakly soluble in pore  water, but
31                  anthropogenic Pb is more available for leaching than naturally occurring Pb in soil
32                  (Section 3.3.3.3). Overall, recent research confirms the generally  low mobility of Pb in
33                  soil. This  limited mobility is strongly dependent on colloid amount and composition, as
34                  well as pH, and may be greater in  some contaminated soils. Low mobility allows soils to
35                  act as a sink for atmospheric Pb potentially for decades or longer. Hence, atmospheric Pb
36                  concentrations that peaked several decades ago may still be present in  soil in the absence
37                  of remediation.

      November 2012                              7-8                  Draft - Do Not Cite or Quote

-------
 1                   Sources of Pb to surface waters include direct atmospheric deposition and indirect
 2                   deposition via runoff and industrial discharge (Section 3.3.2). Because dispersal in
 3                   waterways is a relatively rapid process, concentrations in surface waters are highest near
 4                   sources of pollution before substantial Pb removal by flushing, evaporation, and
 5                   sedimentation occurs. Transport in surface water is largely controlled by exchange with
 6                   sediments, and the cycling of Pb between water and sediments is governed by chemical,
 7                   biological, and mechanical processes that are affected by many factors, including salinity,
 8                   organic complexation, oxidation-reduction potential, and pH. Metals in waterways are
 9                   transported primarily as soluble chelates and ions, or adsorbed on colloidal surfaces,
10                   including secondary clay minerals, Fe and Mn oxides or hydroxides, and organic matter,
11                   and adsorption on organic or inorganic colloids is particularly important for Pb. The
12                   extent of sorption strongly depends on  particle size as smaller particles have larger
13                   collective surface  areas. Pb is relatively stable in sediments, with long residence times
14                   and limited mobility (Section 3.3.2.1). As described in previous sections, Pb enters and is
15                   distributed in bodies of water largely in PM form. In rivers, particle-bound metals can
16                   often account for > 75% of the total load (Section 3.3.2.2). The flux of Pb in aquatic
17                   ecosystems is therefore influenced by the dynamic physical and chemical interactions
18                   within a watershed.

19                   Particles associated with runoff are mostly PM, with a relatively small dissolved fraction,
20                   and dissolution of carbonate and related compounds are important contributors to Pb
21                   pollution in runoff waters. Pb release into runoff is dependent on storm intensity and
22                   length of dry periods between rain events. A "first flush effect" occurs with highest
23                   runoff concentrations observed at the beginning of a rain event.  Most recent studies have
24                   concluded that, during storm events, Pb is transported together with large PM. Some
25                   studies, however,  found that Pb was concentrated in the fine  PM fraction and,
26                   occasionally, Pb was found predominantly in the dissolved fraction. Since the
27                   2006 Pb AQCD, snowmelt and rain-on-snow events are better understood, and it has
28                   been observed that greater runoff occurs from snowmelt and  in rain on snow events than
29                   when snow is not  present, and that metals, including Pb, are often associated with coarse
30                   PM under these circumstances. Runoff in rural areas is strongly controlled by soil type
31                   and the presence of vegetation, with less runoff and  greater retention in mineral soils or
32                   when grass is present, and more runoff for soils high in organic matter (OM).

33                   Sediments can be  either a source or a sink for metals in the aquatic environment
34                   (Section 3.3.2). Release can be via re-suspension of the sediment bed via wind,  wave, and
35                   tidal action or by dissolution from sediment to the water column. Sediment resuspension
36                   from marine environments is important, with disturbance of bed sediments by tidal action
37                   in estuarine areas  resulting in a general greater capacity for re-suspension of PM. Recent
38                   research on Pb flux from sediments in natural waters has demonstrated that resuspended
      November 2012                              7-9                   Draft - Do Not Cite or Quote

-------
 1                  Pb is largely associated with OM or Fe and Mn particles, but that anoxic or depleted
 2                  oxygen environments in sediments play an important role in Pb cycling. This newer
 3                  research indicated that resuspension and release from sediments largely occurs during
 4                  discrete events related to storms. It has also confirmed that resuspension is an important
 5                  process that strongly influences the lifetime of Pb in bodies of water. Finally, there have
 6                  been important advances in understanding and modeling of Pb partitioning in complex
 7                  aquatic environments.
      7.2.2  Ecosystem Exposure, Lag Time and Re-entrainment of Historically
             Deposited Pb

 8                  Ecosystem exposure from atmospheric emissions of Pb depends upon the amount of Pb
 9                  deposited per unit time. Ecosystem response will also depend upon the form in which the
10                  Pb is deposited, the areal extent of such deposition, and modifying factors that affect Pb
11                  bioavailability in soil, sediments, and water (e.g., pH, organic matter) (Sections 7.3.2.
12                  7.4.2 and 7.4.3). However, there is frequently a lag in time between when metals are
13                  emitted and when an effect is seen, particularly in terrestrial ecosystems and, to a lesser
14                  extent, in aquatic sediments. This is because the buffering capacity of soils and sediments
15                  permits Pb to become sequestered into organic matter, reducing its availability for uptake
16                  by organisms. The lag time from start of emissions to achieving a critical load can be
17                  calculated as the time to reach steady state after Pb was initially added to the system.
18                  Excluding erosion processes, the time required to achieve 95% of steady state is about 4
19                  half-lives (ti^)1 (Smolders  et al.. 2007). Conversely, once emissions cease, the same
20                  amount of time is required to reduce metal concentrations to background levels.
      1 Time required to reduce the initial concentration by 50% if metal input is zero.

      November 2012                             7-10                  Draft - Do Not Cite or Quote

-------
 1                   Time to steady state for metals in soils depends upon rates of erosion, uptake by plants,
 2                   and leaching or drainage from soils. Ignoring erosion, half-life of metals can be predicted
 3                   (Smolders et al.. 2007) for a soil as:

                                                       _ 0.69 xdx 10,000
                                                   1/2      y x TF + -§-
                                                                   PKd

                                                                                            Equation 7-1

 4                   where:
 5                   d is the soil depth in meters (m)
 6                   y is the annual crop yield (tons/ha-yr)
 7                   TF is the ratio of the metal concentration in plant to that in soil
 8                   R is the net drainage loss from the soil depth of concern (m3/ha-yr)
 9                   Pis the bulk density of soil [kg(dry weight)/L]
10                   Kd is the ratio of the metal concentration in soil to that in soil pore solution (L/kg)
11                   Metals removed by crops (or plants in general) comprise a very small fraction of the total
12                   soil metal and can be ignored for the purpose of estimating time to steady state. Thus,
13                   equation 7-1 is simplified to:

                                                           0.69 xdx 10,000
                                                                                            Equation 7-2
14                   and becomes a function of soil depth, the amount of rainfall, soil density, and soil
15                   properties that affect Kd. Pb has a relatively long time to steady state compared to other
16                   metals, as shown in Table 7-1.
      November 2012                              7-11                  Draft - Do Not Cite or Quote

-------
      Table 7-1    Comparison among several metals: Time to achieve 95% of steady
                    state metal concentration in soil; example in a temperate system.
Metal
Se
Cu
Cd
Pb
Cr
Loading rate (g/ha-yr)
100
100
100
100
100
Kd (L/kg)
0.3
480a
690a
19,000a
16,700a
Time (years)
1.3
1 ,860a
2,670a
73,300a
64,400a
      aMean Kd (ratio of total metal concentrations in soils to that in soil pore water); and Time to achieve 95% of steady-state
      concentration in soil. (49 Dutch soils) (de Groot et al.. 1998).
      Note: Based on a soil depth of 25 cm, a rain infiltration rate of 3,000 m3/ha-yr, and the assumption that background was zero at the
      start of loading.
      Source: Reprinted with permission of CRC Press, Smolders et al. (2007)

 1                  In aquatic systems, ti/2 for Pb in the water column depends on the ratio of the magnitudes
 2                  of the fluxes coming from and going into the sediment, the ratio of the depths of the
 3                  water column and sediment,  and the sediment ti/2. Sediment tU2 is dependent upon the
 4                  particulate and dissolved fractions and is calculated as for soils (Equation 7-2).

 5                  Re-entrainment of Pb particles via windblown dust from surface  soils or dry sediments
 6                  may occur. Amount and distance of re-entrained particles and deposition rates are
 7                  dependent upon wind velocity and frequency; size, density, shape, and roughness of the
 8                  particle; soil or sediment moisture; and terrain features including openness (including
 9                  amount of vegetation), aspect relative to wind direction, and surface roughness.
10                  Resuspension is defined in terms of a resuspension factor, K, with units of m"1, or a
11                  resuspension rate (A), with units of sec"1 (Equation 7-3). The resuspension rate, A, is the
12                  fraction of a surface contaminant that is released per time and is defined by:
                                                               R
                                                          A=c

                                                                                           Equation 7-3
13                  where:
14                  R is the upward resuspension flux (|ig/m2/sec)
15                  C is the soil (or dry sediment) Pb concentration ((ig/m2)
16                  Such emissions may have local impacts, but are not likely to have long-range effects, as
17                  particles generally remain low to the ground and are not lifted into the atmosphere.
18                  Although re-entrainment may alter the particle size distribution in a local area, it
19                  generally does not alter the bioavailable fraction, and deposited particles will be subject

      November 2012                             7-12                  Draft - Do Not Cite or Quote

-------
 1                  to the same biogeochemical forces affecting bioavailability. Therefore, exposure via
 2                  re-entrainment should be considered additive to exposure from atmospheric particulate
 3                  deposition in terrestrial and aquatic ecosystems.
      7.2.3  Concentrations in Non-Air Media

 4                  Pb from multiple sources moves through environmental media as described in Section
 5                  7.2.1 and Figure 7-1 and has led to measurable Pb concentrations in soil, water, sediment
 6                  and biota in terrestrial and aquatic ecosystems (Table 7-2). The highest concentrations of
 7                  Pb in the environment are currently found near Pb sources, such as metal smelters and
 8                  industrial processing. After phase-out of Pb from on-road gasoline, Pb concentrations
 9                  have decreased considerably in rain, snowpack and surface waters. Declining Pb
10                  concentrations in tree foliage, trunk sections, and grasses, as well as surface sediments
11                  and soils in some locations, have also been observed (U.S. EPA. 2006b). In contrast, Pb
12                  is retained  in soils and sediments, where it may provide a historical record of deposition
13                  and associated concentrations. In remote lakes, sediment profiles indicate higher Pb
14                  concentrations in near surface sediment as compared to pre-industrial era sediment from
15                  greater depth, with peak concentrations between 1960 and 1980 (when leaded on-road
16                  gasoline was at peak use).

17                  Atmospheric deposition has led to measurable Pb concentrations observed in rain,
18                  snowpack,  soil, surface waters, sediments, agricultural plants, livestock and wildlife.
19                  Concentrations of Pb in moss, lichens, peat, and  aquatic bivalves have been used to
20                  understand spatial and temporal distribution patterns of air Pb concentrations. The
21                  amount of  Pb in ecosystems is influenced by numerous factors, however, and it is not
22                  currently possible to determine the contribution of atmospherically-derived Pb from total
23                  Pb. Food, drinking water, and inhalation are major routes of exposure for livestock and
24                  terrestrial wildlife. Ingestion and water intake are the major routes of Pb exposure for
25                  aquatic organisms. In these  exposure pathways, the bioavailable Pb may be from multiple
26                  sources. Information on ambient Pb concentrations in non-air media and biota is reported
27                  in Section  3.6. and concentrations considered in the  interpretation of the ecological
28                  evidence are tabulated in Table 7-2.
      November 2012                             7-13                  Draft - Do Not Cite or Quote

-------
     Table 7-2     Ambient Pb Concentrations in  Non-Air Media and Biota Considered
                     for Ecological Assessment.
Media
Soil
Freshwater
Sediment
Saltwater Sediment
Fresh Surface
Water
(Dissolved Pb)b
Pb Concentration
National Average:
18.9 mg Pb/kg (dry weight)
Range of state averages:
5-38.6 mg Pb/Kg (dry weight)
Median:
73 mg Pb/kg (dry weight)
Median:
28 mg Pb/kgb (dry weight)
Range:
0.6 to 1 ,050 mg Pb/kga
Median: 0.50 ug Pb/Lb; Max:
30 ug Pb/L, 95th percentile
1.1 ug Pb/L
Years Data
Obtained
1961-1997
1996-2001
1991-2003
Dates not available
1991-2003
References
U.S. EPA (2007d. 2006b.
Mahler etal. (2006)
U.S. EPA (2006b)
Sadiq (1992)
U.S. EPA (2006b)
2003b)




                       Range: 0.0003-0.075 ug Pb/L
                       (Set of National Parks in western
                       U.S.)
                                                     2002-2007
                                                    Field and Sherrell (2003).
                                                    U.S. National Park Service (2011)
Saltwater0
Vegetation
Vertebrate
Range: 0.01 - 27 ug Pb/L
Lichens: 0.3-5 mg Pb/kg (dry
weight) (Set of National Parks in
western U.S.)
Grasses: 31% (percent of soil Pb in
grass)
Fish:
Dates not available
2002-2007
1980s-2000s
1991-2001
Sadiq (1992)
U.S. National Park Service
Vandenhove et al. (2009)
U.S. EPA (2006b)

(2011)


                       Geometric Mean:
                       0.59 mg Pb/kg
                       (dry weight) (whole fish)
                       0.15 mg Pb/kg
                       (dry weight) (liver)
                       Range:
                       0.08-22.6 mg Pb/kg
                       (dry weight) (whole fish)
                       0.01-12.7 mg Pb/kg
                       (dry weight) (liver)
                       Fish (from a set of national parks
                       in western U.S.):
                       0.0033 (fillet) to 0.97 (liver)
                       mg Pb/kg (dry weight)
                       Moosed:
                       0.008-0.029 mg Pb/kg
                       (dry weight) (meat)
                       0.012-0.023 mg Pb/kg
                       (dry weight) (liver)
                                                     2002-2007
                                                    U.S. National Park Service (2011)
     aNo information available regarding wet or dry weight
     "Based on synthesis of NAWQA data reported in 2006 Pb AQCD (U.S. EPA. 2006b)
     °Data from a combination of brackish and marine saltwater samples. In general, Pb in seawater is higher in coastal areas and
     estuaries since these locations are closer to sources of Pb contamination and loading from terrestrial systems.
     dThree moose in one Alaskan park
1

2

o
5

4
The most extensive survey of background soil Pb concentration in the contiguous U.S.

was conducted between 1961 and 1976, and comprised 1,319 non-urban, undisturbed

sampling locations, where 250 cm3  of soil was collected at a depth of 20 cm (Shacklette

and Boerngen. 1984). The lower detection limit was 10 mg Pb/kg, and 14% of the 1,319
     November 2012
                              7-14
Draft - Do Not Cite or Quote

-------
 1                  samples were below it. The mean Pb concentration was 19.3 mg Pb/kg, the median 15 mg
 2                  Pb/kg, and the 95th percentile 50 mg/kg. Sixteen locations had Pb concentrations
 3                  between 100 and 700 mg Pb/kg. These results were in agreement with 3 previous
 4                  surveys. When creating the Ecological Soil Screening Level (Eco-SSL) guidance
 5                  document, the U.S. EPA (2007d, 2003b) augmented these data with observations from an
 6                  additional 13 studies conducted between 1982 and 1997, most of them limited to one
 7                  state. The resulting data were summarized using state means for each of the fifty states.
 8                  Those state means ranged between 5 and 38.6 mg Pb/kg, with an overall national mean of
 9                  18.9 mg Pb/kg. No new data on background concentrations of Pb in U.S. soils have been
10                  published since 2005.

11                  The 2006 Pb AQCD reported representative median and range of Pb concentrations in
12                  surface waters (median 0.50  (ig Pb/L, range 0.04 to 30 (ig Pb/L) and sediments (median
13                  28 mg Pb/kg dry weight, range 0.5 to 12,000 mg Pb/kg dry weight) in the U.S. based on a
14                  synthesis of National Water Quality Assessment (NAWQA) data (U.S. EPA. 2006c). In
15                  an additional study using data collected from 1996-2001 the median Pb concentration in
16                  sediment was reported to be  73 mg Pb/kg dry weight (Mahler et al.. 2006). A range of
17                  0.01 to 27 (ig Pb/L for saltwater was reported by Sadiq although the values are not
18                  specific to the U.S. and include open sea areas as well as estuarine and coastal waters
19                  (Sadiq, 1992). In general, Pb in seawater is higher in coastal areas and estuaries  since
20                  these locations are closer to sources of Pb contamination and loading from terrestrial
21                  systems (Sadiq.  1992).

22                  Measured concentrations of Pb in soils, sediment and water are not necessarily
23                  representative of the amount of Pb that is bioavailable to plants, invertebrates and
24                  vertebrates. Both bioaccessibility and bioavailability (Sections 7.3.3. 7.4.3, and 7.4.11) of
25                  Pb are dependent upon the physical, chemical, and biological conditions under which an
26                  organism is exposed at a particular geographic location. Experimental exposures may be
27                  difficult to compare with exposures under natural field conditions in terrestrial and
28                  aquatic systems where a variety of abiotic and biotic modifying factors affect Pb toxicity.
      7.3       Terrestrial Ecosystem Effects

      7.3.1  Introduction to Effects of Pb on Terrestrial Ecosystems

29                  Numerous studies of the effects of Pb on components of terrestrial systems were
30                  reviewed in the 1977 Pb AQCD, the 1986 Pb AQCD and the 2006 Pb AQCD. The focus
31                  of the present document is on studies published since the  last AQCD. Many of those
32                  studies were conducted near stationary sources of atmospheric Pb such as metal

      November 2012                             7-15                  Draft - Do Not Cite or Quote

-------
 1                   industries and mines, or using soil collected near those sources. Increasing proximity to
 2                   the source was often used to generate a gradient of increasing exposure. As may be
 3                   expected, concentrations found in close proximity to those sources are many times
 4                   greater than those found at most locations around the country (data on concentrations of
 5                   Pb in U.S. soils are reviewed in Section 7.2.3 and summarized in Table 7-2). and as
 6                   indicated in the present document's Preamble, concentrations within one to two orders of
 7                   magnitude of current conditions were considered. In addition, it is important to note that
 8                   in all studies where a gradient of multiple concentrations was used, effects increased with
 9                   increasing concentration. This is an important aspect in determining causality (see
10                   Preamble), and therefore justifies inclusion of some studies with very high exposures.
11                   Inclusion of those studies also provides potential data for establishing dose-response
12                   relationships, and predicting effects at all concentrations, including those found away
13                   from stationary sources. Finally, some studies at very high concentrations were used to
14                   provide mechanistic information on Pb toxicity, allow for comparison of Pb uptake
15                   across taxa, or demonstrate the wide range of sensitivity among closely-related species.

16                   Concentrations used in studies where Pb was added to soil experimentally are difficult to
17                   relate to concentrations found in natural environments that have been exposed to Pb
18                   pollution. As reviewed in the following sections, there is ample evidence that multiple
19                   factors, many of them known but not quantified, interact with Pb concentration to
20                   produce responses of widely varying magnitude for similar concentrations, or similar
21                   responses for varying concentrations of Pb. Thus, experimental concentrations that
22                   appear relatively low may be most comparable to relatively high concentrations in natural
23                   soils, and vice-versa. The various factors  that interact with Pb concentration, and the
24                   evidence for those interactions, are discussed in the following sections. However, the
25                   same justifications for inclusion apply to  added-Pb experiments as they do to studies
26                   where proximity to sources is used to vary exposure: gradients of Pb concentrations
27                   create gradients of response, and they often provide information on underlying
28                   mechanisms of toxicity even if the concentrations cannot be easily compared to natural
29                   ones.

30                   The literature on terrestrial ecosystem effects of Pb published since the 2006 Pb AQCD,
31                   is considered with brief summaries from the  1977 Pb AQCD, the 1986 Pb AQCD and the
32                   2006 Pb AQCD, where relevant. Section  7.3 is organized to consider uptake of Pb and
33                   effects at the species level, followed by community and ecosystem level effects. Recent
34                   evidence for Pb effects on reproduction, growth and survival in terrestrial plants,
35                   invertebrates and vertebrates is  summarized in Table 7-4. Alterations to reproduction,
36                   growth and survival of terrestrial organisms can lead to changes at the community and
37                   ecosystem levels of biological organization such as decreased abundance, reduced taxa
38                   richness, and shifts in species composition (Section 7.1). Soil biogeochemistry of Pb is
      November 2012                              7-16                   Draft - Do Not Cite or Quote

-------
 1                   reviewed in Section 7.3.2. Section 7.3.3 considers the bioavailability and uptake of Pb by
 2                   plants, invertebrates, and vertebrates in terrestrial systems. Biological effects of Pb on
 3                   terrestrial ecosystem components including plants and lichen, invertebrates, and
 4                   vertebrates (Section 7.3.4) are followed by data on exposure and response of terrestrial
 5                   species (Section 7.3.5). Effects of Pb at the ecosystem level of biological organization are
 6                   discussed in Section 7.3.6. Section 7.3 concludes with a discussion of critical loads in
 7                   terrestrial systems (Section 7.3.7). soil screening levels (Section 7.3.8), characterization
 8                   of sensitivity and vulnerability of ecosystem components (Section 7.3.9). and effects on
 9                   ecosystem services (Section 7.3.10). Concentration of Pb in soil is expressed in mg Pb/kg
10                   soil, and concentration in solutions applied to soil or extracted from soil is expressed in
11                   mg Pb/L  solution.
      7.3.2  Soil Biogeochemistry and its Influence on Bioavailability

12                   According to data presented in the 2006 Pb AQCD (U.S. EPA. 2006b). the fraction of
13                   soil metal that is directly available to plants is the fraction found in soil pore water, even
14                   though the concentration of metals in pore water is generally small relative to bulk soil
15                   concentration. At any given bulk soil concentration, the amount of Pb dissolved in soil
16                   solution is controlled by at least six variables: (1) solubility equilibria; (2) adsorption-
17                   desorption relationship of total Pb with inorganic compounds (e.g., oxides of Al, Fe, Si,
18                   Mn; clay minerals); (3) adsorption-desorption reactions of dissolved Pb phases on soil
19                   organic matter; (4) pH; (5) cation exchange capacity (CEC); and (6)  aging. Adsorption-
20                   desorption of Pb to soil solid phases is largely controlled by total metal loading.
21                   Therefore, areas with high Pb deposition will exhibit a lower fraction of total Pb
22                   partitioned to inorganic and organic matter. Decreasing soil pH, CEC, and organic matter
23                   have been strongly correlated to increases in the concentration of dissolved Pb species.
24                   Aging of metals in soils results in decreased amounts of labile metal as the Pb  becomes
25                   incorporated into the soil  solid phase (McLaughlin et al.. 2010). Data from recent studies
26                   have further defined the impact of pH, CEC, organic matter (OM), and aging on Pb
27                   mobilization and subsequent bioavailability in soils.
              7.3.2.1  pH, CEC and Salinity

28                   Models of metal bioavailability calibrated from 500+ soil toxicity tests on plants,
29                   invertebrates, and microbial communities indicated that soil pH and CEC are the most
30                   important factors governing both metal solubility and toxicity (Smolders et al.. 2009).
31                   The variability of derived EC50 values was most closely associated with CEC. Smolders
32                   et al. (2007) determined that 12 to 18 months of artificial aging of soils amended with

      November 2012                              7-17                  Draft - Do Not Cite or Quote

-------
 1                   metal decreased the soluble metal fraction by approximately one order of magnitude.
 2                   Relatedly, lower soil pH in forest environments relative to adjacent agricultural land
 3                   resulted in higher solubility, and the mobility of smelter-produced metals was found to be
 4                   greater in forest than in agricultural lands (Douav et al.. 2009). Further, decreasing the
 5                   soil pH via simulated acid rain events increased naturally occurring Pb bioavailability in
 6                   field tests (Hu et al.. 2009b). Miretzky et al. (2007) also showed that the concentration of
 7                   mobile Pb was increased in acidic soils, and discovered that Pb adsorption to sandy loam
 8                   clay was a function of weak electrostatic bonds with charged soil surfaces and was
 9                   influenced by Fe and Mn oxide. Dayton et al. (2006) and Bradham et al. (2006) used path
10                   analysis to help identify the main determinants of both organism Pb content and
11                   responses from among multiple soil characteristics. In parallel studies with lettuce and
12                   earthworms, they amended an array of 21 soils with varying characteristics with the same
13                   amount of Pb (2,000 mg /kg as Pb nitrate), and found that in lettuce, the main
14                   determinant of both accumulation and biological responses was OC, with contribution
15                   from pH and Fe/Al oxides. These later characteristics only influenced accumulation and
16                   responses through their own impact on CEC. In earthworms, the main determinant of
17                   accumulation was pH, with contribution from CEC, but only through its association with
18                   other variables including OC, Fe/Al oxides, and pH. The main determinant of
19                   reproductive effects in earthworms was Fe/Al oxides, while pH drove differential
20                   mortality between the various  soils.

21                   Salinity can also alter Pb mobility and bioavailability in soils. Application of CaCl2,
22                   MgCl, or NaCl salts to field-collected soils containing 31 to 2,764 mg Pb/kg increased
23                   the proportion of mobile metal. As the strength of the salt application was increased from
24                   0.006 to 0.3 M, the proportion of released Pb increased from less than 0.5% to over 2%
25                   for CaCl2 and from less than 0.5% to over 1% for MgCl (Acostaet al.. 2011). However,
26                   the majority of salinity-induced effects occurred in soils containing less than 500 mg
27                   Pb/kg, and the proportion of released Pb decreased with increasing total soil Pb
28                   concentrations. In addition, the authors noted that Pb release from soils under increasing
29                   salinity was reduced at higher carbonate concentrations, indicating that the effect of soil
30                   salinity on Pb release is dependent on still other soil factors. A sequential extraction
31                   procedure was employed by Ettler et al. (2005) to determine the relative bioavailability of
32                   different Pb fractions present in soils collected from a mining and smelting area in the
33                   Czech Republic. Five Pb fraction categories were identified: (Fraction A) exchangeable,
34                   (Fraction B) acid extractable (bound to carbonates), (Fraction C) reducible (bound to Fe
35                   and Mn oxides), (Fraction D) oxidizable (complexed with organic carbon), and  (Fraction
36                   E) residual (silicates). Tilled agricultural soils were found to have decreased Pb, likely as
37                   a result of repeated cultivation, with the majority of Pb represented  as the reducible
38                   Fraction C. Pb concentration in undisturbed forest soils, however, was largely present as
39                   the exchangeable fraction (A), weakly bound to soil OM. However, the validity of

      November 2012                              7-18                   Draft - Do Not Cite or Quote

-------
 1                   associating sequentially extracted fractions with discrete geochemical components has
 2                   not been definitively established, and as a consequence, the association between
 3                   fractionation and bioavailability remains uncertain.
              7.3.2.2 Organic Matter

 4                   Organic matter decreases bioavailability of Pb, but as it is turned over and broken down,
 5                   pedogenic minerals become more important in Pb sequestration (Schroth et al.. 2008).
 6                   Shaheen and Tsadilas (2009) noted that soils with higher clay content, organic matter,
 7                   total calcium carbonate equivalent, and total free sesquioxides also exhibited higher total
 8                   Pb concentration, indicating that less Pb had been taken up by resident plant species.
 9                   Huang et al. (2008) examined the re-mobilization potential of Pb in forest soils, and
10                   determined that mobilization of total Pb  was strongly associated with dissolved organic
11                   matter (DOM). Groenenberg et al. (2010) used a non-ideal competitive adsorption
12                   Donnan model to explain the variability  of organic matter binding affinity and
13                   uncertainties associated with metal speciation. They found that natural variations in fulvic
14                   acid binding properties were the most important variable in predicting Pb speciation. Guo
15                   et al. (2006b) determined that the -COOH and -OH groups associated with soil OM were
16                   important factors in Pb sequestration in soil, and Pb sorption was increased as pH was
17                   raised from 2 to 8. Because organic content increased the Pb sequestration efficiency of
18                   soils, OM content had  an inhibitory effect on Pb uptake by woodlouse species Oniscus
19                   asellus and Porcellio scaber (Gal et al., 2008). Vermeulen et al. (2009) demonstrated that
20                   invertebrate bioaccumulation of Pb from contaminated soils was dependent on pH and
21                   OM, but that other unidentified habitat-dependent factors also contributed. The
22                   relationship of bioaccumulation and soil concentration was modified by pH and OM, and
23                   also by habitat type. Kobler et al.  (2010) showed that the migration of atmospherically
24                   deposited Pb in soil matrices was strongly influenced by soil type, indicating that certain
25                   soil types may retain Pb for longer periods of time than others. In soils characterized by
26                   well-drained substrate  and limestone bedrock, Pb concentration decreased over time,
27                   likely as a result of water drainage and percolation. The authors contrasted this
28                   observation with reports of prolonged residence time in humic soils, particularly at the
29                   lower depths of the humus layer. They theorized that the most significant Pb migration
30                   route was transportation of particulate-bound Pb along with precipitation-related  flow
31                   through large soil pores.

32                   A number of recent laboratory studies have further defined the relationship of soil
33                   biogeochemical characteristics and Pb uptake by plants. As noted above, Dayton  et al.
34                   (2006) found through path analysis that the main determinant of both accumulation and
35                   biological responses in lettuce grown on amended soil was OC. As part of a metal


      November 2012                             7-19                   Draft - Do Not Cite or Quote

-------
 1                   partitioning study, Kalis et al. (2007) determined that not only did metal concentration in
 2                   the soil solution decrease as pH increased, but pH-mediated metal adsorption at the root
 3                   surface ofLolium perenne determined root Pb concentration, with concentration in the
 4                   shoot correlated with root concentration. Interestingly, Kalis et al. (2007) and Lock et al.
 5                   (2006) also observed that the influx of Pb in the water-soluble fraction had an impact on
 6                   soil pH. In addition, 1 (iM humic acid decreased root Pb concentration in L. perenne
 1                   plants grown in 0.1 and 1 (iM Pb solution, likely as a result of Pb complexation and
 8                   sequestration with the added OM (Kalis et al.. 2006). Ma et al.  (2010) also reported that
 9                   long-term agricultural cultivation can decrease the rate of Pb desorption in soil through a
10                   gradual OM-enrichment. Phosphorous soil amendments equivalent to 35 mg P/kg soil
11                   were observed to reduce the quantity of DPTA-extractable Pb from an average of 19 and
12                   24 mg Pb/kg in unamended soils to 12 to 15 mg Pb/kg in P-amended soils. As a result,
13                   maize and soybean seedlings accumulated significantly less Pb: average concentrations in
14                   soybean shoot and root ranged from 4.4 to 5.2 mg Pb/kg with P addition (versus 9.21 mg
15                   Pb/kg without), while maize shoot concentrations average between 4.8 to 5.3 mg Pb/kg in
16                   P-amended soils (as compared with 10.16 mg Pb/kg in controls) (Xie etal., 2011).
              7.3.2.3 Aging

17                   Smolders et al. (2007) defined aging as the process responsible for decreasing the
18                   bioavailability of metals in soils independently of their persistence.  Smolders et al. (2009)
19                   reviewed the effects of aging of Pb in soils on the toxicity of Pb to plants and soil
20                   invertebrates, with aging achieved in most studies primarily by leaching amended soil,
21                   but also through natural binding and complexation. In nearly half of the Pb soil studies
22                   reviewed, responses that were observed with freshly amended soil could no longer be
23                   detected following soil leaching, indicating that aged soils likely contain less bioavailable
24                   Pb. The authors concluded that competitive binding between soil ligands and biotic
25                   ligands on plant roots or invertebrate guts can be used to model the  relationship of
26                   observed availability and toxicity of metals in soils. Because this concept is the basis of
27                   the Biotic Ligand Model (BLM) (Section 7.3.3). the authors proposed a terrestrial BLM
28                   approach to estimate the risk of metals to terrestrial organisms. However, Antunes et al.
29                   (2006) noted that there were several key challenges involved in development of a
30                   terrestrial BLM applicable to plants, particularly the reliable measurement of free ion
31                   activities and ligand concentration in the rhizosphere, the identification of the organisms'
32                   ligands associated with toxicity, and the possible need to incorporate kinetic dissolution
33                   of metal-ligand complexes as sources of free ion. Further, Pb in aged field soils has been
34                   observed to be less available for uptake into terrestrial organisms, likely as  a result of
35                   increased sequestration within the soil particles (Antunes et al.. 2006).  Magrisso et al.
      November 2012                              7-20                   Draft - Do Not Cite or Quote

-------
 1                   (2009) used a bioluminescent strain of the bacterium Cupriavidus metallidurans to detect
 2                   and quantify Pb bioavailability in soils collected adjacent to industrial and highway areas
 3                   in Jerusalem, Israel, and in individual simulated soil components freshly spiked with Pb.
 4                   The bacterium was genetically engineered to give off the bioluminescent reaction as a
 5                   dose-dependent response, and was inoculated in soil slurries for three hours prior to
 6                   response evaluation. Spiked soil components induced the bioluminescent response, and
 7                   field-collected components did not. However, the comparability of the simulated soils
 8                   and their Pb concentration with the field-collected samples was not entirely clear. Lock et
 9                   al.  (2006) compared the Pb toxicity to springtails (Folsomia Candida) from both
10                   laboratory-spiked soils and field-collected Pb-contaminated soils of similar Pb
11                   concentrations. Total Pb concentrations of 3,877 mg Pb/kg dry weight and higher always
12                   caused significant effects on F. Candida reproduction in the spiked soils. In field soils,
13                   only the soil  with the highest Pb concentration of 14,436 mg Pb/kg dry weight
14                   significantly affected reproduction. When expressed as soil pore-water concentrations,
15                   reproduction was never significantly affected at Pb concentrations of 0.5 mg Pb/L,
16                   whereas reproduction was always significantly affected at Pb concentrations of 0.7 mg
17                   Pb/L and higher, independent of the soil treatment. Leaching soils prior to use in
18                   bioassays had only a slight effect on Pb toxicity to resident springtails, suggesting that
19                   among the processes that constitute aging of Pb in field soils, leaching is not particularly
20                   important with respect to bioavailability.

21                   Red-backed salamanders (Plethodon cinereus) exposed to Pb-amended soils (553 mg
22                   Pb/kg, 1,700 mg Pb/kg, 4,700 mg Pb/kg,  and 9,167 mg Pb/kg) exhibited lowered appetite
23                   and decreased white blood cell counts at the two highest concentrations, as compared to
24                   controls (Bazar etal. 2010). However, salamanders tolerated field-collected, aged soils
25                   containing Pb concentration of up to 16,967 mg Pb/kg with no significant deleterious
26                   effects.

27                   In summary, studies published during the past 5 years continue to substantiate the
28                   important role that soil geochemistry plays in sequestration or release of Pb. Soil pH and
29                   CEC have long been known to be the primary controlling factors of the amount of
30                   bioavailable  Pb in soils, and a recent review of more than 500 studies corroborates these
31                   findings (Smolders et al.. 2009). Fe and Mn oxides are now known to also play an
32                   important role in Pb sequestration in soils. Pb binds to OM, although relatively weakly,
33                   and as the  OM  is broken down the Pb may be released into soil solution. Leaching of
34                   metal through soil pores may be the primary route for loss of bioavailable soil Pb; OM
35                   may reduce leaching and thus appear to be associated with Pb sequestration. Aging of Pb
36                   in soils (through incorporation of the metal into the particulate solid-phase of the soil)
37                   results in long term binding of the metal,  and reduced bioavailability of Pb to plants and
38                   soil organisms.
      November 2012                             7-21                   Draft - Do Not Cite or Quote

-------
      7.3.3  Bioavailability in Terrestrial Systems

 1                   Bioavailability was defined in the 2006 Pb AQCD as "the proportion of a toxin that
 2                   passes a physiological membrane (the plasma membrane in plants or the gut wall in
 3                   animals) and reaches a target receptor (cytosol or blood)" (U.S. EPA. 2006c). In 2007,
 4                   EPA took cases of bioactive adsorption into consideration and revised the definition of
 5                   bioavailability as "the extent to which bioaccessible metals absorb onto, or into, and
 6                   across biological membranes of organisms, expressed as a fraction of the total amount of
 7                   metal the organism is proximately exposed to (at the sorption surface) during  a given
 8                   time and under defined conditions" (U.S. EPA. 2007c). The bioavailability of metals
 9                   varies widely depending on the physical, chemical, and biological conditions under
10                   which an organism is exposed (U.S. EPA. 2007c). Characteristics of the toxicant itself
11                   that affect bioavailability are: (1) chemical form or species, (2) particle size, (3) lability,
12                   and (4) source. The bioavailability of a metal is also dependent upon the fraction of metal
13                   that is bioaccessible. As stated in the Framework for Metals Risk Assessment (U.S. EPA.
14                   2007c). the bioaccessible fraction of a metal is the portion (fraction or percentage) of
15                   environmentally available metal that actually interacts  at the organism's contact surface
16                   and is potentially available for absorption or adsorption by the organism. The Framework
17                   states that "the bioaccessibility, bioavailability, and bioaccumulation properties of
18                   inorganic metals in soil, sediments, and aquatic systems are interrelated and abiotic
19                   (e.g., organic carbon) and biotic (e.g., uptake and metabolism). Modifying factors
20                   determine the amount of an inorganic metal that interacts at biological surfaces (e.g., at
21                   the gill, gut, or root tip epithelium) and that binds to and is absorbed across these
22                   membranes. A major challenge is to consistently and accurately measure quantitative
23                   differences in bioavailability between multiple forms of organic metals in the
24                   environment." A conceptual diagram presented in the Framework for Metals Risk
25                   Assessment (U.S. EPA. 2007c) summarizes metals bioavailability and bioaccumulation
26                   in aquatic, sediment and soil media (Figure 7-2).
      November 2012                             7-22                   Draft - Do Not Cite or Quote

-------
Bioaccessible Fraction (BF) ":
Percent soluble metal ion
concentration relative to total
metal concentration (measured in
solution near biomembrane)

Relative Bioavai lability (RBA) b:
Percent adsorbed or absorbed
compared to reference material
(measure of membrane dynamics)

Absolute Bioavailability (ABA) c:
Percent of metal mass absorbed
internally corrpared to external
exposure (measures systerric
uptake/accumulation)
                            Bioaccessibility
                    X
io\vailability
                         Environmental availability \
                         	Exposure ==
                                          Bioaccumulation  of  metal
                                    ==============-» Efects
                                                      lembrane
                                                      uptake
                                       \
                                                    Physiological
                                                   V^rnembrane
                        Total Metal Concentration
                              Predation
                              Foraging
                           Toxicoloical
                           accumulation
                                                             Detoxification
                                                              and Storage
      Benign  /
   accumulation
  Internal  r
 Transport
    and
Distribution
      aBF is most often measured using in vitro methods (e.g., artificial stomach), but it should be validated by in vivo methods.
      bRBA is most often estimated as the relative absorption factor, compared to a reference metal salt (usually calculated on the basis of
      dose and often used for human risk, but it can be based on concentrations).
      °ABA is more difficult to measure and used less in human risk; it is often used in ecological risk when estimating bioaccumulation or
      trophic transfer.
      Source: ERG (2004) and U.S. EPA (2007c).

      Figure 7-2       Conceptual diagram for evaluating bioavailability processes and
                         bioaccessibility for metals in soil, sediment, or aquatic systems.
 i
 2
 3
 4
 5
 6
 7
 8
 9
10
The BLM attempts to integrate the principal physical and chemical variables that
influence Pb bioavailability. The model considers the reactions of Pb with biological
surfaces and membranes (the site of action) to predict the bioavailability and uptake of
the metal (Figure 7-3). and integrates the binding affinities of various natural ligands and
the biological uptake rates of organisms to predict both the bioaccessible and bioavailable
fraction of Pb in the environment, and to determine the site-specific toxicity of the
bioavailable fraction. In principle, the BLM can be used for determining toxicity in water,
sediment, and soil media, however, the parameter values that influence BLM are, in
general, characterized to a greater extent in aquatic systems than in terrestrial systems
(Section 7.4.4).
      November 2012
                             7-23
                  Draft - Do Not Cite or Quote

-------
                             Organic Matter
                             Complexation
                                              Inorganic Ligand
                                                Complexation
      Source: Reprinted with permission of John Wiley and Sons ; from Di Toro et al. (2001)

      Figure 7-3       Schematic diagram of the biotic ligand model.

 1                  New information on sources of Pb in terrestrial ecosystems, and their influence on
 2                  subsequent bioavailability, was reviewed in  Chapter_3, while new information on the
 3                  influence of soil biogeochemistry on speciation and chemical lability was presented in
 4                  Section 7.3.2. This section summarizes recent literature on uptake and subsequent
 5                  presence of Pb in tissues. The 2006 Pb AQCD (U.S. EPA. 2006b) extensively reviewed
 6                  the methods available for quantitative determination of the mobility, distribution, uptake,
 7                  and fluxes of atmospherically delivered Pb in ecosystems, and they are not reviewed in
 8                  this section. The 2006 Pb AQCD also reported bioaccumulation factors (BAF) and
 9                  bioconcentration factors (BCF). BAF is defined as the field measurement of metal
10                  concentration in tissues, including dietary exposures, divided by metal concentration in
11                  environmental media (Smolders et al., 2007). BCF is defined as the same measurement
12                  carried out in artificial media in the laboratory that does not include dietary exposure
13                  (Smolders et al., 2007). The EPA Framework for Metals Risk Assessment states that the
14                  latest scientific data on bioaccumulation do not currently support the use of BCFs and
15                  BAFs when applied as generic threshold criteria  for the hazard potential of metals (U.S.
16                  EPA. 2007c).
     November 2012
7-24
Draft - Do Not Cite or Quote

-------
              7.3.3.1  Terrestrial Plants

 1                  At the time of the 1977 Pb AQCD, it was understood that Pb uptake in plants was
 2                  influenced by plant species and by the available Pb pool in the soils (U.S. EPA. 1977).
 3                  The role of humic substances in binding Pb was better characterized by the 1986 Pb
 4                  AQCD where it was stated that most plants cannot survive in soil containing
 5                  10,000 (ig Pb/g (mg Pb/kg) dry weight if the pH is below 4.5 and the organic content is
 6                  below 5% (U.S. EPA. 1986b). At the time of the 1986 AQCD , it was thought that Pb can
 7                  be absorbed across the leaf surface into internal plant tissues, but that the vast majority of
 8                  uptake is via roots (U.S. EPA.  1986b). The 2006 Pb AQCD (U.S. EPA. 2006b) noted that
 9                  terrestrial plants accumulate  atmospheric Pb primarily via two routes: direct stomatal
10                  uptake into foliage, and incorporation of atmospherically deposited Pb from soil into root
11                  tissue, followed by variable translocation to other tissues. Foliar Pb may include both
12                  incorporated Pb (i.e., from atmospheric gases or particles) and surficial particulate Pb
13                  deposition. Although the plant may eventually absorb the surficial component, its main
14                  importance is its likely contribution to the exposure  of plant consumers. This section will
15                  first review recent studies on uptake of Pb by plants through foliar and soil routes, and
16                  their relative contribution, followed by the consideration of translocation of Pb from roots
17                  to shoots, including a discussion of variability in translocation among species. Data on
18                  ambient Pb levels associated with vegetation are summarized in Section 3.6.6.


                    Leaf and Root Uptake

19                  Although Pb is not an essential metal, it is taken up from soils through the symplastic
20                  route, the same active ion transport mechanism used by plants to take up water and
21                  nutrients and move them across root cell membranes (U.S. EPA. 2006c). As with all
22                  nutrients, only the proportion of a metal present in soil pore water is directly available for
23                  uptake by plants. In addition, soil-to-plant transfer factors in soils enriched with Pb have
24                  been found to better correlate with bioavailable Pb soil concentration, defined as DTPA-
25                  extractable Pb, than with total Pb concentration (U.S. EPA. 2006c). Since the publication
26                  of the 2006 Pb AQCD, suggestive evidence has become available that a substantial
27                  proportion of Pb accumulated in shoots of some species of trees originates in direct leaf
28                  uptake of atmospheric Pb. Evidence  for such direct uptake is weaker in herbaceous
29                  plants, and all data came from near stationary sources.

30                  Field studies carried out in the vicinity of Pb smelters have determined the relative
31                  importance of direct foliar uptake  and root uptake of atmospheric Pb deposited in soils.
32                  Hu and Ding (2009) analyzed ratios  of Pb isotopes in the shoots of commonly grown
33                  vegetables and in soil at three distances from a smelter (0.1, 0.2, 5.0 km). Pb isotope
34                  ratios in plants and soil were different at two of those locations, leading the authors to the

      November 2012                             7-25                  Draft - Do Not Cite or Quote

-------
 1                   conclusion that airborne Pb was being assimilated via direct leaf uptake. Soil Pb
 2                   concentration in the rhizosphere at the three sites ranged between 287 and 379 mg Pb/kg
 3                   (Site I), 155 and 159 mg Pb/kg (Site II), and 58 and 79 mg Pb/kg (Site III, selected as the
 4                   control site). The median shoot and root Pb concentrations at each site were 36 and
 5                   47 mg Pb/kg, 176 and 97 mg Pb/kg, and  1.3 and 7 mg Pb/kg, respectively, resulting in
 6                   shootroot Pb ratios exceeding 1.0 in Site I (for Malabar spinach [Basella alba],
 1                   ratio = 1.6, and amaranth \Amaranthus spinosus], ratio =  1.1), and in Site II (for the
 8                   weeds Taraxacum mongolicum, ratio = 1.9, and Rostellariaprocumbens, ratio = 1.7).
 9                   However, the two species studied at Site II were not studied at Site I or Site III. In the
10                   control site (Site III), no plant was found with a Pb shootroot ratio greater than 1.0. Hu
11                   and Ding (2009) concluded that metal  accumulation was greater in shoot than in root
12                   tissue, which suggested both high atmospheric Pb concentration and direct  stomatal
13                   uptake into the shoot tissue.

14                   Cui et al. (2007) studied seven weed species growing in the vicinity of an old smelter
15                   (average soil Pb concentration of 4,020 mg Pb/kg) in Liaoning, China, to measure Pb
16                   accumulation rates in roots and shoots. Cutleaf groundcherry (Physalis angulatd)
17                   accumulated the most Pb, with root and shoot concentration of 527 and 331 mg Pb/kg,
18                   respectively, and velvetleaf (Abutilon theophrasti) was the poorest absorber of Pb (root
19                   and shoot concentration of 39 and 61 mg Pb/kg, respectively). In all  cases, weed species
20                   near the smelter accumulated more Pb than plants from non-polluted environments (5 mg
21                   Pb/kg), indicating that aerially deposited Pb produced by smelting is bioavailable to
22                   plants. However, the ratio of rootshoot Pb concentration varied by species, and the
23                   authors presented no data to differentiate Pb taken up from soil from Pb incorporated via
24                   foliar uptake. Angelova et al. (2010) examined Pb uptake by rapeseed plants (Brassica
25                   napus) grown in heavy metal contaminated soils 0.5 km and 15 km from the Non-Ferrous
26                   Metal Works, in Bulgaria. Average surface soil Pb concentration decreased with distance
27                   from the plant (200.3 and 24.6 mg Pb/kg, respectively), as did average DTPA-extractable
28                   Pb (69.7 and 4.9 mg Pb/kg, respectively). Pb content in stems and leaves in rapeseed
29                   grown at 0.5 km from the plant averaged 1.73 and 8.69 mg Pb/kg ; average stem and leaf
30                   Pb concentrations in rapeseed grown at the more distant location were  reported as 0.72
31                   and 1.42 mg Pb/kg, respectively (Angelova et al.. 2010).

32                   Pb plant BAFs for plants grown in 70 actively cropped fields in California averaged
33                   0.052 for vegetable crops and 0.084 for grains; the highest reported Pb BAF (0.577) was
34                   found in onions. Authors compared the BAFs based on total Pb and Pb in solution and
35                   determined that both were accurate predictors of plant uptake (Chen et al.. 2009b).
36                   Likewise, Zhang et al. (20lib) compiled Pb uptake  data for several crop species in
37                   China, and reported an average BAF for grains (rice) of 0.009 (0.0009-0.03) and
38                   0.41(0.0007-0.17) for leafy vegetables, such as spinach, Chinese cabbage and celery
      November 2012                             7-26                  Draft - Do Not Cite or Quote

-------
 1                   (Zhang et al.. 20 lib). Chrastny et al. (2010) characterized the Pb contamination of an
 2                   agricultural soil in the vicinity of a shooting range. Pb was predominantly in the form of
 3                   PbO and PbCO3, and Pb was taken up by plants through both atmospheric deposition
 4                   onto the plant and by root uptake.

 5                   The Pb content of ripe date palm (Phoenix dactyliferd) fruit collected in Riyadh, Saudi
 6                   Arabia was determined to be indicative of areas of heavy industrialization and
 7                   urbanization; Pb concentrations in fruit flesh ranged from 0.34 to 8.87 mg Pb/kg dry
 8                   weight, with the highest Pb date concentrations detected near freeways and industrial
 9                   areas (Aldjain et al.. 2011). Likewise, Pb concentrations in rosemary (Rosmarinus
10                   officinalis) flowers, stems, and leaves were significantly higher in the urban areas of Al-
11                   Mafraq and Irbid, Jordan than in the smaller town of Ma'an, Jordan (53.6 to 86.5 mg
12                   Pb/kg versus 16.2 to 16.7 mg Pb/kg). Authors noted a significant difference between Pb
13                   concentrations in washed and unwashed rosemary samples, indicating that aerial
14                   deposition and surface dust is likely a significant source of plant-associated Pb (El-Rjoob
15                   et al.. 2008).

16                   Bilberry (Vaccinium myrtillus), accumulated the highest amount of Pb out of four total
17                   herbaceous species growing in Slovakian spruce ecosystems with variable soil Pb
18                   concentrations, giving BAFs of 0.09 to 0.44, depending on location (Kuklova et al..
19                   2010). Because of their long life  spans, trees can provide essential information regarding
20                   the sources of bioavailable Pb. A Scots pine forest in northern Sweden was found to
21                   incorporate atmospherically derived Pb pollution directly from ambient air, accumulating
22                   this Pb in bark, needles, and shoots (Klaminder et al.. 2005). Nearly 50% of total tree
23                   uptake was estimated to be from direct adsorption from the atmosphere, as determined
24                   using isotopic ratios and a binary mixing model. Further, Aznar et al. (2009a) found that
25                   the Pb content of black spruce (Picea mariand) needles collected along a metal
26                   contamination gradient emanating from a Canadian smelter in Murdochville, Quebec,
27                   showed a significant decrease in  Pb concentration with increasing distance from the
28                   smelter. Interestingly, older needles were determined to accumulate larger quantities of
29                   Pb than younger ones. Foliar damage and growth reduction were  also observed in the
30                   trees (Aznar et al.. 2009a). They  were significantly correlated with Pb concentration in
31                   the litter layer. In addition, there  was no correlation between diminished tree growth and
32                   Pb concentration in the deeper mineral soil layers, strongly suggesting that only current
33                   atmospheric Pb was affecting trees (Aznar et al.. 2009b). Similarly, Kuang et al. (2007)
34                   noted that the Pb concentration in the inner bark ofPinus massoniana trees growing
35                   adjacent to a Pb-Zn smelter in the Guangdong province of China was much higher
36                   (1.87 mg Pb/kg dry weight) than in reference-area trees. Because concentration in the
37                   inner bark was strongly correlated with concentration in the outer bark, they concluded
38                   that the origin of the Pb was atmospheric.
      November 2012                             7-27                  Draft - Do Not Cite or Quote

-------
 1                  Dendrochronology (tree ring analysis) has become an increasingly important tool for
 2                  measuring the response of trees to Pb exposure (Watmough. 1999). Tree ring studies
 3                  reviewed in the 1977 Pb AQCD showed that trees could be used as indicators of
 4                  increasing environmental Pb concentrations with time. Additional studies in the 1986 Pb
 5                  AQCD indicated that Pb could be translocated from roots to the upper portions of the
 6                  plant and that the amounts translocated are in proportion to concentrations of Pb in soil
 7                  (U.S. EPA. 1986b). The advent of laser ablation inductively coupled plasma mass
 8                  spectrometry has made measurement of Pb concentration in individual tree rings possible
 9                  (Witte et al., 2004; Watmough. 1999). This allows for close analysis of the timing of Pb
10                  uptake relative to smelter activity and/or changes in soil chemistry. For example, Aznar
11                  et al. (2008a) measured Pb concentration in black spruce tree rings to determine the
12                  extent and timing of atmospheric deposition near the Murdochville smelter. Variability in
13                  tree-ring Pb content seemed to indicate that trees accumulated and sequestered
14                  atmospheric Pb in close correlation with the rates of smelter emission, but that
15                  sequestration lagged about 15  years behind exposure. However, the ability to determine
16                  time of uptake from the location in growth rings is weakened in species that transfer Pb
17                  readily from outer bark to inner bark. Cutter and Guyette (1993) identified species with
18                  minimal radial translocation from among a large number of tree species, and
19                  recommended the following temperate zone North American species as suitable for metal
20                  dendrochronology studies: white oak (Quercus alba), post oak (Q. stellatd), eastern red
21                  cedar (Juniperus virginiana), old-growth Douglas fir (Pseudotsuga menziesii), and big
22                  sagebrush (Artemisia tridentata). In addition, species such as bristlecone pine (Pinus
23                  aristata), old-growth redwood  (Sequoia sempervirens), and giant sequoia (S. giganted)
24                  were deemed suitable for local purposes. Patrick and Farmer (2006) determined that
25                  European sycamore (Acer pseudoplatanus) are not suitable for this type of
26                  dendrochronological analysis because of the formation of multiple annual  rings.

27                  Pb in sapwood and heartwood is more likely a result of soil uptake than of direct
28                  atmospheric exposure (Guvette et al.. 1991). Differentiation of geogenic soil Pb in tree
29                  tissue from Pb that originated in the atmosphere requires measurement of stable Pb
30                  isotope ratios (Patrick. 2006). Tree bark samples collected from several areas of the
31                  Czech Republic were subjected to stable Pb isotope analysis to determine the source and
32                  uptake of atmospheric Pb  (Conkova and Kubiznakova. 2008). Results indicated that
33                  beech bark is a more efficient accumulator of atmospheric Pb than spruce bark. A
34                  decrease in the 206Pb/207Pb ratio was measured in bark and attributed to increased usage of
35                  leaded gasoline between 1955 and 1990; an increased 206Pb/207Pb ratio was ascribed to
36                  coal combustion (Conkova and Kubiznakova. 2008). Similarly, Savard et al. (2006)
37                  compared isotope ratios of 206Pb/207Pb and 208Pb/206Pb in tree rings from spruce trees
38                  sampled at a control site near Hudson Bay, with those sampled near the Home smelter
39                  active since 1928, in Rouyn-Noranda, Canada.  The concentration of total Pb showed a

      November 2012                             7-28                  Draft - Do Not Cite or Quote

-------
 1                   major increase in 1944 and a corresponding decrease of the 206Pb/207Pb ratios, suggesting
 2                   that the smelter was responsible for the increased Pb uptake (Savard et al.. 2006). The
 3                   authors suggested that the apparent delay of 14 years may have been attributable to the
 4                   residence time of metals in airborne particles the buffering effect of the soils and, to a
 5                   lesser extent, mobility of heavy metals in tree stems. Furthermore, through the use of the
 6                   two different isotope ratios, Savard et al. (2006) were able to differentiate three types of
 7                   Pb in tree rings: natural (derived from the mineral soil horizons), industrial (from coal
 8                   burning urban pollution), and mining (typical of the volcanogenic massive sulfide ore
 9                   deposits treated at the Home smelter).

10                   Devall et al. (2006) measured Pb uptake by bald-cypress trees (Taxodium distichum)
11                   growing in a swamp near a petroleum refinery and along a bank containing
12                   Pb-contaminated dredge spoils. They measured Pb in tree cores and showed
13                   greater uptake of Pb by trees in the swamp than by trees growing on the dredge spoil
14                   bank, attributing the difference to exposure source (refinery versus dredge spoils) and
15                   differences in soil chemistry between the swamp and the dredge spoil bank (Devall et al..
16                   2006). Similarly, Gebologlu et al. (2005) found no correlation between proximity to
17                   roadway and accumulated Pb in tomato and bean plants at sites adjacent to two state
18                   roads in Turkey (average Pb concentration 5.4 and 6.0 mg Pb/kg), indicating that uptake
19                   may be influenced by multiple factors, including wind direction, geography, and soil
20                   chemistry. Average Pb levels in leaves were 0.6 and 0.5 mg Pb/kg for tomato and bean
21                   plants, respectively, while fruit concentration averaged 0.4 mg Pb/kg for both species.
22                   Conversely, if foliar contamination is due primarily to dust  deposition, distance from a
23                   source such as a road  may be easily correlated with Pb concentration on the plants. For
24                   example, Ai-Khlaifat  and Al-Khashman (2007) collected unwashed date palm (Phoenix
25                   dactyliferd) leaves at 3-meter trunk height from trees in Jordan to assess the extent of Pb
26                   contamination from the city of Aqaba. Whereas relatively low levels of Pb were detected
27                   in leaves collected at background sites (41 mg Pb/kg), leaves collected adjacent to
28                   highway sites exhibited the highest levels of Pb (177 mg Pb/kg). The authors determined
29                   that Pb levels in date palm leaves correlated with industrial  and human activities
30                   (e-g-, traffic density) (Ai-Khlaifat and Al-Khashman. 2007). Likewise, Pb concentrations
31                   were significantly enriched in tree bark samples and road dust collected in highly
32                   urbanized areas of Buenos Aires, Argentina (approximate average enrichment factors of
33                   30 and 15 versus reference samples) (Fujiwara et al.. 2011). However, decreases in tissue
34                   Pb concentration with increasing distance from stationary sources can also follow from
35                   decreasing Pb in soil.  Bindler et al. (2008) used Pb isotopes to assess the relative
36                   importance of pollutant Pb versus natural Pb for plant uptake and cycling in Swedish
37                   forested soils. The Pb isotopic composition of needles/leaves and stemwood of different
38                   tree species and ground-cover plants indicated that the majority of Pb present in these
39                   plant components was derived from the atmosphere, either through aerial interception or

      November 2012                              7-29                  Draft - Do Not Cite or Quote

-------
 1                  actual uptake through the roots. For the ground-cover plants and the needles/leaves, the
 2                  206Pb/207Pb isotopic ratios (1.12 to 1.20) showed that the maj ority of Pb was of
 3                  anthropogenic origin. Stemwood and roots have higher 206Pb/207Pb ratio values (1.12 to
 4                  1.30) which showed the incorporation of some natural Pb as well as anthropogenic Pb.
 5                  For pine trees, the isotopic ratio decreased between the roots and the apical stemwood
 6                  suggesting that much of the uptake of Pb by trees is via aerial exposure. Overall, it was
 7                  estimated that 60-80% of the  Pb in boreal forest vegetation originated from pollution; the
 8                  Pb concentrations were, however, quite low - not higher than 1  mg Pb/kg plant material,
 9                  and usually in the range of 0.01-0.1 mg Pb/kg plant material (while soils had a range of 5
10                  to 10 mg Pb/kg in the mineral horizons and 50 to 150 mg Pb/kg in the O horizons).
11                  Overall, the  forest vegetation recycles very little of the Pb present in soils (and thus does
12                  not play a direct role in the Pb biogeochemical cycle in boreal forest soils).

13                  Fungal species, as represented by mushrooms, accumulate Pb from soils to varying
14                  degrees. Based on the uptake of naturally occurring 210Pb, Guillen et al. (2009)
15                  established that soil-associated Pb was bioavailable for uptake by mushrooms, and that
16                  the highest 210Pb accumulation was observed in Fomes fomentarius mushrooms, followed
17                  by Lycoperdon perlatum, Boletus aereus, and Macrolepiota procera, indicating some
18                  species  differences. Benbrahim et al. (2006) also showed species differences in uptake of
19                  Pb by wild edible mushrooms, although they found no significant correlations between
20                  Pb content of mushrooms and soil Pb concentration. Pb concentrations in mushroom
21                  carpophores ranged from 0.4  to 2.7 mg Pb/kg from sites with  soil concentrations ranging
22                  from 3.6 and 7.6 mg Pb/kg dry soil. Likewise, Semreen and Aboul-Enein (2011).
23                  reported the heavy metal uptake of wild edible mushrooms collected in various
24                  mountainous regions of Jordan. Pb BCFs ranged between 0.05 (Russula delicd) and 0.33
25                  (Bovistaplumbea) for six mushroom species.  Pb BAFs for edible mushrooms collected
26                  from quartzite acidic soils in central Spain (containing  19.2 mg Pb/kg) ranged from 0.07
27                  (Macrolepiotaprocera) to 0.45 (Lepista nuda) (Campos and Tejera. 2011).


                    Translocation and Sequestration of Pb in Plants

28                  In the 1977 Pb AQCD it was  recognized that most Pb taken up from soil remains in the
29                  roots and that distribution to other portions of the plant is variable among species (U.S.
30                  EPA. 1977). The 2006 Pb AQCD (U.S. EPA. 2006b) stated that most of the Pb absorbed
31                  from soil remains bound in plant root tissues either because (1) Pb may be deposited
32                  within root cell wall material, or (2) Pb may be sequestered within root cell organelles.
33                  More recent research largely confirms that Pb taken up from soil largely remains in roots,
34                  but suggests that some species translocate meaningful amounts into shoot tissue.
35                  Sequestration of Pb may be a protective mechanism for the plant. Recent findings have
36                  been consistent with this hypothesis: Han et al. (2008) observed Pb deposits in the cell

      November 2012                            7-30                  Draft - Do Not Cite or Quote

-------
 1                  walls and cytoplasm of malformed cells of Iris lactea exposed to 0 to 10 mM Pb (0 to
 2                  2,072 mg Pb/L) solution in sand culture for 28 days. They hypothesized that preferential
 3                  sequestration of Pb in a few cells, which results in damage to those cells, helps in
 4                  maintaining normal overall plant activities through the sacrifice of a small number of
 5                  active cells. Similarly, macroscopic analysis of the roots of broad bean (Viciafabd)
 6                  cultivated in mine tailings (average Pb concentration of 7,772 mg Pb/kg) by Probst et al.
 7                  (2009) revealed dark ultrastructural abnormalities that were demonstrated to be metal-
 8                  rich particles located in or on root cell walls. It is unclear whether the presence of these
 9                  structures had any effect on overall plant health.

10                  Clark et al. (2006) investigated Pb bioavailability in garden soils in Roxbury and
11                  Dorchester, MA. The sources of Pb were considered to be Pb from paints and from
12                  leaded gasoline additives, with 40 to 80% coming from paint. The average Pb
13                  concentration in foliar tissue of bean plants was  14 ± 5 mg Pb/kg while the concentration
14                  in the bean pod was only 20.6 mg Pb/kg. For mustard plants, there was a linear
15                  relationship (R2=0.85) between Pb concentration in plant tissues and Pb concentration in
16                  the soil (both for plants grown in situ and those grown under greenhouse conditions).

17                  Murray et al. (2009) investigated the uptake and accumulation of Pb in several vegetable
18                  species (carrot [Daucus carota], radish [Raphanus sativus], lettuce [Lactuca sativa],
19                  soybean [Glycine max], and wheat \Triticum aestivum]) from metal-contaminated soils,
20                  containing 10 to 40 mg Pb/kg and demonstrated that most Pb remained in the roots. No
21                  Pb was measured in the above-ground edible soybean and wheat tissues, while carrots,
22                  the most efficient accumulator of Pb, contained a maximum  Pb tissue concentration of
23                  12 mg Pb/kg dry mass. Similarly, (Cho et al.. 2009) showed  that green onion (Allium
24                  fistulosum) plants also take up little Pb when planted in soil spiked with Pb nitrate. No
25                  plant tissues contained a Pb concentration greater than 24 mg Pb/kg when grown for
26                  14 weeks in soils  of up to 3,560 mg Pb/kg, and the majority of bioavailable Pb was
27                  determined to be contained within the roots. Chinese spinach (Amaranthus dublus) also
28                  translocates very little Pb to stem and leaf tissue, and uptake from Pb-containing soils (28
29                  to 52 mg Pb/kg) is minimal (Mellem et al.. 2009). Wang et al. (20lie) determined tissue-
30                  specific BCFs for wheat grown in soils containing 93 to 1,548 mg Pb/kg. Although the
31                  average calculated root BCF was 0.3, very little Pb was translocated to shoots (average
32                  BCF=0.02), shells (0.006), and kernels (0.0007) (Wang et al.. 20lie). Sonmez et al.
33                  (2008) reported that Pb accumulated by three weed species (Avena sterilis, Isatis
34                  tinctoria, Xanthium strumarium) grown in Pb-spiked soils was largely concentrated in the
35                  root tissues, and little was translocated to the shoots (Sonmez et al.. 2008).
      November 2012                             7-31                  Draft - Do Not Cite or Quote

-------
 1                   The Pb BCFs for alfalfa (Medicago sativd) and crimson clover (Trifolium incarnatum)
 1                   grown in mixtures of heavy metals (Pb concentrations of 10 to 500 (ig Pb/kg) were
 3                   reportedly low. For alfalfa, BCFs ranged from 0.02 to 0.12, while for crimson clover,
 4                   these values were between 0.04 and 0.06 (Comino et al.. 2011). The low shoot-root
 5                   translocation factors reported for alfalfa (0.17 to 0.43) indicated that plant Pb content was
 6                   largely contained in root tissue. Businelli et al. (2011) calculated whole-plant Pb BAFs
 7                   for lettuce, radish, tomato and Italian ryegrass using Pb-spiked soils (average values of
 8                   0.025, 0.021, 0.032, and 0.65, respectively). Again, the majority of accumulated Pb was
 9                   stored in root tissue, with comparatively little translocated to above-ground tissues
10                   (Businelli etal.. 2011).

11                   Recent research has shown that Pb translocation to stem and leaf tissues does occur at
12                   significant rates in some species, including the legume Sesbania drummondii (Peralta-
13                   Videa et al., 2009) and buckwheat (Fagopyrum esculentum) (Tamura et al., 2005). Wang
14                   et al. (2006b) noted that Pb soil-to-plant transfer factors were higher for leafy vegetables
15                   (Chinese cabbage, pak-choi, and water spinach) than for the non-leafy vegetables tested
16                   (towel gourd, eggplant, and cowpea). Tamura et al. (2005) demonstrated that buckwheat
17                   is an efficient translocator of Pb. Buckwheat grown in Pb-containing soils collected from
18                   a shooting range site (average 1M HC1 extractable Pb= 6,643 mg Pb/kg) preferentially
19                   accumulated Pb in leaves (8,000 mg Pb/kg) and shoots (4,200 mg Pb/kg), over root
20                   tissues (3,300 mg Pb/kg). Although plant growth was unaffected, this level of leaf and
21                   shoot accumulation is likely to have significant implications for exposure of herbivores.
22                   Similarly, Shaheen and Tsadilas (2009) reported that vegetables (pepper, okra, and
23                   eggplant) grown in soils containing 24 to 30 mg Pb/kg total Pb were more likely to
24                   accumulate Pb in leaves (range: undetected to 25 mg Pb/kg) rather than in fruits (range:
25                   undetected to 19 mg Pb/kg); however, no significant correlation between soil Pb
26                   concentration and plant tissue Pb concentration could be established (Shaheen and
27                   Tsadilas. 2009). Tobacco plants were also observed to take up significant amounts of Pb
28                   into leaf tissue. Field-grown plants in soils containing an average of 19.8 mg Pb/kg
29                   contained average lower, middle and upper leaf Pb concentrations of 11.9, 13.3, and
30                   11.6 mg Pb/kg respectively (Zaprianova et al.. 2010). Uptake by tobacco plants was
31                   correlated with both total soil Pb concentrations and the mobile Pb fraction (average
32                   3.8 mg Pb/kg soil).

33                   There is broad variability in uptake and translocation among plant species, and
34                   interspecies variability has been shown to interact  with other factors such as soil type. By
35                   studying multiple species in four Pb-Zn mining sites  in Yunnan, China, Li et al. (2009d)
36                   demonstrated not only significant differences in uptake and translocation among the
37                   species studied, but also modification of the effect on species by type of soil. Plants
38                   sampled represented nine species from four families—Caryophyllaceae, Compositae,
      November 2012                             7-32                  Draft - Do Not Cite or Quote

-------
 1                  Cruciferae, and Pteridaceae. Overall, soil Pb concentration averaged 3,772 mg Pb/kg dry
 2                  weight, with the highest site average measured at the Minbingying site (5,330 mg Pb/kg),
 3                  followed by Paomaping (2,409 mg Pb/kg), Jinding (1,786 mg Pb/kg),  and Qilinkeng
 4                  (978 mg Pb/kg). The highest average shoot Pb concentration (3,142 mg Pb/kg) was
 5                  detected in Stellaria vestita (Caryophyllaceae) collected at Paomaping, while Sinopteris
 6                  grevilloides (Pteridaceae) collected from Minbingying exhibited the lowest shoot Pb
 7                  concentration (69 mg Pb/kg). A similar trend was detected in root tissues. S. vestita root
 8                  collected from the Paomaping area contained the maximum Pb concentration measured
 9                  (7,457 mg Pb/kg), while the minimum root Pb levels were measured in Picris
10                  hieracioides (Pteridaceae) tissues collected from Jinding. These results indicate
11                  significant interspecies differences in Pb uptake, as well as potential soil-specific
12                  differences in Pb bioavailability. S. vestita, in particular, was determined to be an
13                  efficient accumulator of Pb, with a maximum enrichment coefficient of 1.3. Significant
14                  correlations between soil Pb concentration and average shoot and root Pb levels were also
15                  established (Li et al.. 2009d). Within plant species, the variability in uptake and
16                  translocation of Pb may extend to the varietal level. Antonious and Kochhar (2009)
17                  determined uptake of soil-associated Pb for 23 unique genotypes from four species of
18                  pepper plants (Capsicum chinense, C.frutescens, C. baccatum, and C. annum). Soil Pb
19                  concentration averaged approximately 0.6 mg Pb/kg dry soil. No Pb was detected in the
20                  fruits of any of the 23 genotypes, except two out of seven genotypes of C. baccatum,
21                  which had 0.9 and 0.8 mg Pb/kg dry weight Pb in fruit.

22                  Recent studies substantiated findings from the 2006 Pb AQCD that plants store a large
23                  portion of Pb in root tissue. Pb soil-to-plant transfer factors are higher for leafy
24                  vegetables than for the non-leafy vegetables (Wang et al.. 2006b) and  buckwheat has
25                  recently been shown to be an efficient translocator of Pb from soil to above-ground
26                  shoots (Tamura et al.. 2005).

27                  Field studies carried out in the vicinity of Pb smelters (Hu et al.. 2009b)  show that Pb
28                  may accumulate in shoot tissue through direct stomatal uptake rather than by soil-root-
29                  shoot translocation. For instance, Hovmand and Johnsen (2009) determined that about
30                  98% of Pb sequestered in Norway spruce needles and twigs was derived from
31                  atmospheric sources, and that less  than 2% of Pb was translocated from the roots
32                  (Hovmand et al.. 2009). Dendrochronology has become more advanced in recent years
33                  and is a useful tool for monitoring historical uptake of Pb  into trees exposed to
34                  atmospheric or soil Pb. Trees accumulate and sequester atmospheric Pb in close
35                  correlation with the rate of smelter emissions, although one study indicated that
36                  sequestration can lag behind exposure from emissions by  15 years. Pb in the outer woody
37                  portion of the tree is more likely the result of direct atmospheric exposure, while Pb in
      November 2012                             7-33                   Draft - Do Not Cite or Quote

-------
 1                  sapwood is more likely a result of soil uptake. This difference provides an important tool
 2                  for analyzing source apportionment of Pb accumulation in plants (Guvette et al.. 1991).
              7.3.3.2  Terrestrial Invertebrates

 3                  At the time of publication of the 2006 Pb AQCD (U.S. EPA. 2006b). little information
 4                  was available regarding the uptake of atmospheric Pb pollution (direct or deposited) by
 5                  terrestrial invertebrate species. Consequently, few conclusions could be drawn
 6                  concerning the  Pb uptake rate of particular species although there was some evidence that
 7                  dietary or habitat preferences may influence exposure and uptake. Recent literature
 8                  indicates that invertebrates can accumulate Pb from consuming a Pb-contaminated diet
 9                  and from exposure via soil, and that uptake and bioaccumulation of Pb by invertebrates is
10                  lower than that observed for other metals.
                    Snails

11                  Pauget et al. (2011) reported that uptake of Pb from soil by the land snail (Cantareus
12                  asperses) was most significantly influenced by soil pH and organic matter, as increases in
13                  these variables were correlated to decreased Pb bioavailability. Cantareus asperses snails
14                  exposed to dietary Pb at 3.3, 86, and 154 mg/kg of diet (spiked with Pb sulfate) for up to
15                  64 days were found to assimilate a significant proportion of Pb, and feeding rates were
16                  unaffected by the presence of the metal (Beebv and Richmond. 2010). While BCFs for
17                  Cd were observed to increase over the 64-day study period, the rate of Pb assimilation
18                  remained consistent over time and the authors inferred the absence of a regulatory
19                  mechanism for uptake of Pb. The authors speculated that uptake is a function of growth
20                  or cell turnover instead. Helix aspersa snails rapidly accumulated Pb from contaminated
21                  soil (1,212 mg Pb/kg) and from eating contaminated lettuce (approximately 90 mg Pb/kg
22                  after  16 weeks' growth on Pb-contaminated soil) during the first 2 weeks of exposure, at
23                  which point snail body burdens reached a plateau (Scheifler et al., 2006b). There were no
24                  observed effects of Pb exposure or accumulation on survival or growth in C. asperses or
25                  H. aspersa. In another study (Ebenso and Ologhobo. 2009b). juvenile Achatina achatina
26                  snails confined in cages on former Pb-battery waste dump sites were found to accumulate
27                  Pb from both plant and soil sources. Soil Pb concentration averaged 20, 200, and
28                  1,200 mg Pb/kg at the three main waste sites, while leaf tissues of radish  (Raphanus
29                  sativus) grown at these sites averaged 7, 30, and 68 mg Pb/kg  dry weight, respectively.
30                  Concentration of Pb in  snail tissues rose with concentration in both soil and plants, and
31                  the authors found that for both sources, a log-log relationship could be estimated with a
32                  very close  fit (r2 =0.94 and 0.95, respectively). Pb concentration in snail tissues averaged
      November 2012                             7-34                  Draft - Do Not Cite or Quote

-------
 1                   12, 91, and 468 mg Pb/kg, respectively, at the three sites, which the authors stipulated
 2                   were above the maximum permissible concentration of Pb for human consumption of
 3                   mollusks, mussels, and clams (1.5 mg Pb/kg tissue) as determined by the U.K. Food
 4                   Standards Agency. Pb concentration in snail tissues generally is much lower than that of
 5                   the soil substrates upon which they were reared, but higher than in other soil-dwelling
 6                   organisms. De Vaufleury et al. (2006) exposed Helix aspersa snails to standardized
 7                   (International Organization for Standardization methodology [ISO 11267:1999])
 8                   artificial-substrate soils amended with sewer sludge containing 13, 26, 39, or 52 mg
 9                   Pb/kg  for 28  days without supplemental food. After the exposure period, snail foot tissue
10                   contained increased levels of Pb—1.9, 1.7, and 1.5 mg Pb/kg dry weight versus
11                   concentration averaging 0.4 mg Pb/kg in control organisms. Viscera also exhibited
12                   increased Pb levels at the two highest exposures, with measured tissue concentration of
13                   1.2 and 1.1 mg Pb/kg, respectively, as compared with control tissue Pb levels of 0.4 mg
14                   Pb/kg. However, there was no significant increase in snail-tissue Pb concentration when
15                   natural soil was used in place of ISO  medium, and there was no relationship between soil
16                   Pb concentration and snail tissue concentration, strongly suggesting the presence of soil
17                   variables that modify bioavailability.  Notten et al. (2008) investigated the origin of Pb
18                   pollution in soil, plants, and snails by means of Pb isotope ratios. They found that a
19                   substantial proportion of Pb in both plants and snails was from current atmospheric
20                   exposure.

21                   Finally, a study by Coeurdassier et al. (2007) found that the presence  of snails was
22                   associated with higher Pb content in earthworms, suggesting that snails themselves may
23                   have an effect on bioavailability.


                     Earthworms

24                   Accumulation studies conducted with Eisenia sp. earthworms documented the difficulty
25                   of extrapolating accumulation kinetic constants from one soil type to  another, and
26                   showed that many soil physiochemical properties, including pH, organic matter, and
27                   CEC, among others, affect metal bioavailability (Nahmani et al.. 2009). Source of Pb,
28                   and  proportion of soil:leaf litter also affect Pb bioavailability. Bradham et al.  (2006)
29                   examined the effect of soil chemical and physical properties on Pb bioavailability.
30                   Eisenia andrei earthworms were  exposed to 21  soils with varying chemical and physical
31                   properties that were freshly spiked with Pb to give a standard concentration of 2,000 mg
32                   Pb/kg  dry weight. At equivalent Pb exposure, the main determinants of both internal
33                   earthworm Pb concentration and mortality were pH first (with lower pH resulting in
34                   higher concentration and mortality), then CEC. However, the apparent importance of
35                   CEC was due to its correlation with several other less important soil characteristics.
36                   These  data corroborate that Pb bioavailability and toxicity are increased in acidic soils

      November 2012                             7-35                  Draft - Do Not  Cite or Quote

-------
 1                  and in soils with a low CEC (Section 7.3.2). This finding was confirmed by Gandois et al.
 2                  (2010). who determined that the free-metal-ion fraction of total Pb concentration in field-
 3                  collected soils was largely predicted by pH and soil Fe content.

 4                  The role of soil profile and preferred depth was studied using eight species of earthworms
 5                  from 27 locations in Switzerland, representing three ecophysiological groups (Ernst et al..
 6                  2008): epigeic (surface-dwelling worms), endogeic (laterally burrowing worms that
 7                  inhabit the upper soil layers), and anecic (vertically burrowing worms that reach depths
 8                  of 6 inches). For epigeic and anecic earthworms, the total concentration of Pb in leaf litter
 9                  and in soil, respectively, were the most important drivers of Pb body burdens. By
10                  contrast, the level of Pb in endogeic  earthworms  was largely determined by soil pH and
11                  CEC. As a result of these differences, the authors suggested that atmosphere-sourced Pb
12                  may be more bioavailable to epigeic than endogeic species, because it is less dependent
13                  on modifying factors.  Suthar et al. (2008). on the other hand, found higher Pb
14                  bioaccumulation in the endogeic earthworm Metaphire posthuma than in the anecic
15                  earthworm species Lampito mauritii, and speculated that differences in Pb tissue level
16                  arose from differing life-history strategies,  such as feeding behaviors, niche preferences,
17                  and burrowing patterns, all of which exposed the endogeic species to greater Pb
18                  concentration. Garg et al. (2009) reported that the smaller native earthworm
19                  Allolobophoraparva accumulated significantly greater Pb concentrations than E. fetida.
20                  Subsequently, it was concluded that  native  earthworm species may exhibit a higher Pb
21                  accumulation potential as a result of increased tolerance  to the heavy metal (Garg et al..
22                  2009).

23                  Earthworm activity can alter Pb bioavailability and subsequent uptake by earthworms
24                  themselves and other organisms. Sizmur and Hodson (2009) speculated that earthworms
25                  affect Pb mobility by modifying the  availability of cations or anions. The concentration
26                  of water-soluble Pb was observed to increase following earthworm (Lumbricus terrestris)
27                  feeding activity in field-collected soils containing 132.7, 814.9, and 821.4 mg total Pb/kg
28                  (calculated BAFs of 0.27, 0.33, and 0.13, respectively) (Alonso-Azcarate et al.. 2011).
29                  However, Coeurdassier et al. (2007) found  that snails did not have a higher Pb content
30                  when earthworms were present, and  that unexpectedly, Pb was higher in earthworm
31                  tissue when snails were present.

32                  Despite significant Pb uptake by earthworms, Pb in earthworm tissue may not be
33                  bioavailable to predators. Pb in the earthworm (Aporrectodea caliginosa) was determined
34                  to be contained largely in the granular fraction (approximately 60% of total Pb), while the
35                  remaining Pb body burden was in the tissue, cell membrane, and intact cell fractions
36                  (Vijver et al.. 2006). However, this may vary by  species, as (Li et al.. 2008b) found that
37                  more than half of the Pb accumulated by E. fetida was contained within earthworm tissue
      November 2012                             7-36                  Draft - Do Not Cite or Quote

-------
 1                   and cell membranes. Regardless, Vijver et al. (2006) concluded that only a minority of
 2                   earthworm-absorbed Pb would be lexicologically available to cause effects in the
 3                   earthworms or in their predators.


                     Arthropods

 4                   Pb and other metals were analyzed in honeybees (Apis mellifera) foraging in sampling
 5                   sites that included both urban areas and wildlife reserves in central Italy. (Perugini et al..
 6                   2011). Pb in whole bees ranged from 0.28 to 0.52 mg Pb/kg with the highest
 7                   concentration in honeybees caught in hives near an airport. Cicadas pupating in
 8                   historically Pb-arsenate-treated soils accumulated Pb at concentrations similar to those
 9                   reported previously for earthworms (Robinson et al.. 2007). Likewise, tissue Pb levels
10                   measured in Coleoptera specimens collected from areas  containing average soil
11                   concentration of 45 and 71 mg Pb/kg exhibited a positive relationship with soil Pb
12                   content, although abundance was unaffected (Schipper et al.. 2008). By contrast, two
13                   grasshopper species inhabiting Pb and Cd-contaminated areas near Zn smelting facilities
14                   exhibited different Pb accumulation rates. Locust (Locusta migratoria) collected from
15                   areas with an average Pb soil concentration of 540mg Pb/kg contained 47 mg Pb/kg,
16                   while grasshoppers (Acrida chinensis) inhabiting the same area accumulated 93.9 mg
17                   Pb/kg (Zhang etal.. 2012). This gives respective BAFs of 0.09 and 0.17. Similarly, the
18                   Pb sequestration rates that were observed in two woodlouse species, O. asellus and
19                   P. scaber, were  species-dependent (Gal et al.. 2008). Both species were field collected at
20                   Pb-contaminated sites (average concentration, 245 mg Pb/kg dry weight; range,
21                   21-638 mg Pb/kg dry weight), with O. asellus Pb levels  averaging 43  mg Pb/kg over all
22                   sites, while P. scaber contained no detectable Pb residues. Pb concentration measured in
23                   granivorous rough harvester ants (Pogonomyrmex rugosus), in the seeds of some plant
24                   species they consume, and in surface soil, were all shown to decline with increasing
25                   distance from a former Pb smelter near El Paso, Texas, where soil leachable Pb at the
26                   three sites of ant collection ranged from 0.003 to 0.117 mg Pb/kg (Del Toro et al.. 2010).
27                   Ants accumulated approximately twice as much Pb as was measured in  seeds, but the
28                   study did not separate the effects of dietary exposure from those of direct contact with
29                   soil or respiratory intake.
              7.3.3.3  Terrestrial Vertebrates

30                  At the time of the 1977 Pb AQCD few studies of Pb exposure and effects in wild animals
31                  other than birds had been conducted. A limited number of rodent trapping studies near
32                  roadsides indicated general trends of species differences in Pb uptake and higher
      November 2012                             7-37                  Draft - Do Not Cite or Quote

-------
 1                  concentrations of Pb in habitats adjacent to high-traffic areas (U.S. EPA. 1977). In the
 2                  1986 Pb AQCD concentration of Pb in bone tissue was reported for selected herbivore,
 3                  omnivore and carnivore species [Table 8-2 in (U.S. EPA, 1986b)1.

 4                  Tissue Pb residues in birds and mammals associated with adverse toxicological effects
 5                  were presented in the 2006 Pb AQCD. In general, avian blood, liver, and kidney Pb
 6                  concentrations of 0.2-3 (ig Pb/dL, 2-6 mg Pb/kg wet weight, and 2-20 mg Pb/kg wet
 7                  weight, respectively, were linked to adverse effects. A few additional studies of Pb
 8                  uptake and tissue residues in birds and mammals conducted since 2006 are reviewed
 9                  here.

10                  In a study of blood Pb levels in wild Steller's eiders (Polysticta stelleri) and black scoters
11                  (Melanitta nigra) in Alaska, the authors compiled avian blood Pb data from available
12                  literature to develop reference values for sea ducks (Brown et al., 2006). The background
13                  exposure reference value  of blood Pb was <20 (ig Pb/dL, with levels between 20 and
14                  59 (ig Pb/dL as indicative of Pb exposure. Clinical toxicity was in the range of
15                  60-99 (ig Pb/dL in birds while >100 (ig Pb/dL results in acute, severe toxicity. In
16                  measurement of blood Pb with a portable blood Pb analyzer, only 3% of birds had values
17                  indicating exposure and none of the birds had higher blood Pb levels or clinical signs of
18                  toxicity. Tissue distribution of Pb in liver, kidney, ovary and testes of rain quail (Coturnix
19                  coramandelica) following oral dosing of 0.5 mg Pb/kg, 1.25 mg Pb/kg or 2.5 mg Pb/kg
20                  Pb acetate for 21 days indicated that Pb uptake was highest in liver and kidney and low in
21                  ovary and testes (Mehrotra et al.. 2008). Resident feral pigeons (Columba livid) captured
22                  in the urban and industrial areas of Korea exhibited increased lung Pb concentration,
23                  ranging from 1.6 to 1.9 mg Pb/kg wet weight (Nam and Lee. 2006). However, tissue
24                  concentration did not correlate with atmospheric Pb concentration, so the authors
25                  concluded that ingestion of particulate Pb (paint chips, cement, etc.) in the urban and
26                  industrial areas was responsible for the pigeons' body burden. Similarly, 70% of
27                  American woodcock (Scolopax minor) chicks  and 43% of American woodcock young-of-
28                  year collected in Wisconsin, U.S., exhibited high bone Pb levels of 9.6-93 mg Pb/kg dry
29                  weight and 1.5-220 mg Pb/kg, respectively, even though radiographs of birds'
30                  gastrointestinal tracts revealed no evidence of shot ingestion (Strom et al., 2005). Authors
31                  hypothesized that unidentified anthropogenic sources may have caused the observed
32                  elevated Pb levels.

33                  In addition to birds, soil-dwelling mammals can also bioaccumulate atmospherically-
34                  sourced Pb. Northern pocket gophers (Thomomys talpoides) trapped within the Anaconda
35                  Smelter Superfund Site were shown to accumulate atmospherically deposited Pb. Gopher
36                  liver and carcass Pb concentration averaged 0.3 and 0.4 mg Pb/kg wet weight on low Pb
37                  soils (47 mg Pb/kg),  0.4 and 0.9 mg Pb/kg wet weight in medium Pb soils (95 mg Pb/kg)
      November 2012                             7-38                  Draft - Do Not Cite or Quote

-------
 1                  and 1.6 and 3.8 mg Pb/kg wet weight in high Pb soils (776.5 mg Pb/kg) (Reynolds et al..
 2                  2006). Likewise, rats trapped in the vicinity of a Kabwe, Zambia Pb-Zn mine had
 3                  significantly elevated liver and kidney Pb concentrations. Soil Pb concentrations were
 4                  measured between 9 and 51,188 mg Pb/kg (approximate average of 200 mg Pb/kg dry
 5                  weight), while rat liver and kidney Pb concentrations ranged between 0.009 and 7.3 mg
 6                  Pb/kg dry weight and 0.3 and 22.1 mg Pb/kg dry weight, respectively. Consequently,
 7                  residence in the mining region was correlated to significantly increased Pb body burdens
 8                  for rats (Nakavama et al.. 2011). Angelova et al. (2010) reared rabbits on a fodder
 9                  mixture containing Pb-contaminated rapeseed grown adjacent to a metal works plant.
10                  Following a four-week exposure, Pb was most heavily concentrated in rabbit kidney
11                  tissue (3.9 mg Pb/kg and 1.9 mg Pb/kg, for high and low diet respectively), bone (1.0 and
12                  0.3 mg Pb/kg, respectively), and liver (0.6 and 0.4 mg Pb/kg, respectively). Yucatan
13                  micropigs (Sus scrofd) and Sprague-Dawley rats (Rattus norvegicus) reared on
14                  Pb-contaminated soil (5% of 1,000 mg Pb/kg soil as dietary component) consumed
15                  significantly different amounts of Pb. Over a 30-day period, rats consumed an average of
16                  19.4 mg Pb, while  micropig intake averaged 948 mg Pb (Smith et al.. 2009a). This
17                  resulted in significantly higher Pb accumulation in micropigs, based on liver, blood,
18                  kidney and bone Pb concentrations (average concentrations of 1.2, 25, 0.9, and 9 mg
19                  Pb/kg for micropigs, and 0.2, 7, 0.5, and 1.5 mg Pb/kg for rats, respectively).

20                  Casteel et al.  (2006) found that bioavailability of Pb from environmental soil samples in
21                  swine (Sus domesticd) depended on Pb form or type, with high absorption of cerussite
22                  and Mn-Pb oxides  and poor absorption of galena and anglesite. Juvenile swine
23                  (approximately 5-6 weeks old and weighing 8-11  kg) were fed Pb-contaminated soils
24                  collected from multiple sources for 15 days (concentration range of 1,270 to 14,200 mg
25                  Pb/kg) to determine the relative bioavailability. While Pb concentrations were roughly
26                  equivalent in blood, liver, kidney, and bone tissues, individual swine exhibited different
27                  uptake abilities (Casteel et al.. 2006).

28                  Consistent with observations in humans, dietary Ca2+ deficiency (0.45 mg Ca2+ daily
29                  versus 4 mg under normal conditions) was linked to increased accumulation of Pb in
30                  zebra finches (Taeniopygia guttatd) that were provided with drinking water containing
31                  20 mg Pb/L (Dauwe et al.. 2006). Liver and bone Pb concentration were increased by an
32                  approximate factor of three, while Pb concentration in kidney, muscle, and brain tissues
33                  were roughly doubled by a Ca2+-deficient diet. However, it is not known whether this
34                  level of dietary Ca2+ deficiency is common in wild populations of birds.
      November 2012                             7-39                  Draft - Do Not Cite or Quote

-------
              7.3.3.4  Food Web

 1                  In addition to the organism-level factors reviewed above, understanding the
 2                  bioavailability of Pb along a simple food chain is essential for determining risk to
 3                  terrestrial animals. While the bioavailability of ingested soil or particles is relatively
 4                  simple to measure  and model, the bioavailability to secondary consumers of Pb ingested
 5                  and sequestered by primary producers and primary consumers is more complex. Kaufman
 6                  et al. (2007) caution that the use of total Pb concentration in risk assessments can result in
 7                  overestimation of risk to ecological receptors, and they suggest that the bioaccessible
 8                  fraction may provide a more realistic approximation of receptor exposure and effects.
 9                  This section reviews recent literature that estimates the bioaccessible fraction of Pb in
10                  dietary items of higher order consumers, and various studies suggesting that Pb may be
11                  transferred through the food chain but that trophic transfer of Pb results in gradual
12                  attenuation, i.e., lower concentration at  each successive trophic level.

13                  Earthworm and plant vegetative tissue collected from a rifle and pistol range that
14                  contained average  soil Pb concentration of 5,044 mg Pb/kg were analyzed for Pb content
15                  and used to model  secondary bioavailability to mammals (Kaufman et al.. 2007).
16                  Earthworms were determined to contain an average of 727 mg Pb/kg, and the Pb content
17                  of unwashed leaf tissues averaged 2,945 mg Pb/kg. Canonical correspondence  analysis
18                  detected no relationship  between earthworm and soil Pb concentration, but did show
19                  correlation between unwashed vegetation and soil concentration. The  authors noted that
20                  the relatively high  Pb  concentration of unwashed as opposed to washed vegetation
21                  indicated the potential importance of aerial deposition (or dust resuspension) in
22                  determining total vegetative Pb concentration. Based on the mammalian gastric model,
23                  they noted that 50% of vegetation tissue Pb and 77% of earthworm tissue Pb was
24                  expected to be bioavailable to consumers. The avian gizzard model indicated that 53% of
25                  soil Pb and 73% of earthworm Pb was bioaccessible to birds; and, for both mammals and
26                  birds, the bioaccessible fraction of Pb was a function of total Pb concentration.

27                  The transfer of Pb  from  soils contaminated by a Pb-Zn mine was limited along a soil-
28                  plant-insect-chicken food chain (Zhuang et al.. 2009). In soils averaging 991 mg Pb/kg,
29                  plants of the fodder plant Rumex patientia X tianschanicus sequestered an average of
30                  1.6 mg Pb/kg wet weight in the shoot tissue, while larvae of the leafworm Spodoptera
31                  litura accumulated an average Pb concentration of 3.3 mg Pb/kg wet weight S.  litura-fed
32                  chickens (Gallus gallus  domesticus) accumulated 0.58 mg Pb/kg and 3.6 mg Pb/kg in
33                  muscle and liver tissue,  respectively, but only liver Pb burden was increased significantly
34                  relative to controls. A large proportion of ingested Pb was excreted with the feces.
35                  Likewise, an insectivorous bird species, the black-tailed godwit (Limosa limosa) was
36                  shown to accumulate Pb from earthworms residing in Pb-contaminated soils (Roodbergen
      November 2012                             7-40                  Draft - Do Not Cite or Quote

-------
 1                   et al.. 2008). Pb concentration in eggs and feathers was increased in areas with high soil
 2                   and earthworm Pb concentration (336 and 34 mg Pb/kg, respectively): egg Pb
 3                   concentration averaged 0.17 mg Pb/kg and feather concentration averaged 2.8 mg Pb/kg.
 4                   This suggests that despite a residence breeding time of only a few months, this bird
 5                   species could accumulate Pb when breeding areas are contaminated.

 6                   Rogival et al. (2007) showed significant positive correlations between soil Pb
 7                   concentration along a gradient (approximately 50 to 275 mg Pb/kg) at a metallurgical
 8                   plant, and Pb concentration in both acorns (from Quercus robur) and earthworms
 9                   (primarily Dendrodrilus rubidus and Lumbricus rubellus) collected on site. Acorn and
10                   earthworm Pb contents were, in turn, positively correlated with the Pb concentration in
11                   the liver,  kidney, and bone tissues of locally trapped wood mice (Apodemus sylvaticus).

12                   The uptake and transfer of Pb from soil to native plants and to red deer (Cervus elaphus)
13                   was investigated in mining areas of the Sierra Madrona Mountains in Spain (Reglero et
14                   al.. 2008). The authors reported a clear pattern between plant Pb concentration and the Pb
15                   content of red deertissues with attenuation (i.e., decreasing concentration) of Pb up the
16                   food chain. Interestingly, soil geochemistry likely was affected by mining activity as
17                   Holm oak (Quercus ilex), gum rockrose (Cistus ladanifef), elm leaf blackberry (Rubus
18                   ulmifolius), and grass (Graminae) tissues collected from mining areas exhibited increased
19                   Pb levels  (up to 98 mg Pb/kg in grasses and 21 mg Pb/kg in oak) despite the fact that total
20                   soil Pb concentrations were not significantly greater than those of the non-mining areas.

21                   Positive relationships were observed between Cepaea nemoralis snail tissue Pb levels
22                   and Pb concentration measured in Urtica dioica leaves in field-collected samples from
23                   areas characterized by metal soil contamination (approximately 200 to 400 mg Pb/kg)
24                   (Notten et al.. 2005). Inouye et al. (2007) found that several invertebrate prey offence
25                   lizards, including Acheta domestica crickets, Tenebrio molitor beetles, and P. scaber
26                   isopods, accumulate Pb from dietary exposures (10, 50, 100, 250, 500, 750, and 1,000 mg
27                   Pb/kg) lasting between 44 and 72 days. By day 44, Pb  body burdens of crickets were 31,
28                   50 and 68 mg Pb/kg (wet weight) at the three highest dietary exposures, respectively.
29                   Isopods and beetle larvae accumulated significantly less Pb, with average body burdens
30                   of 10,  15, and 14 mg Pb/kg following 56 days of exposure; and 12, 14, and 31 mg Pb/kg
31                   following 77 days of exposure, respectively. For all invertebrates tested, Pb was
32                   sequestered partly in the exoskeleton, and partly in granules. Exoskeleton Pb may be
33                   available  to predators, but returns to background level  with each shedding, while granular
34                   Pb is likely unavailable, at least to other invertebrates (Vijver et al.. 2004).

35                   In a comparison of rural and urban blackbirds (Turdus meruld), Sheifler et al.(2006a)
36                   found that while Pb concentration in unwashed tail  feathers was equivalent in both
37                   populations,  urban birds had higher tissue concentrations. Pb content of urban


      November 2012                              7-41                   Draft - Do Not Cite or Quote

-------
 1                   earthworms was also higher than that of rural earthworms. Hypothesizing that tail feather
 2                   Pb reflected deposition from air and resuspended dust, the authors suggested that elevated
 3                   Pb in the urban birds was mostly dietary in origin.

 4                   Overall, studies of Pb transfer in food webs have established the existence of pervasive
 5                   trophic transfer of the metal, but no consistent evidence of trophic magnification. It
 6                   appears that on the contrary, attenuation is common as Pb is transferred to higher trophic
 7                   levels. However, many individual transfer steps, as from particular plants to particular
 8                   invertebrates, result in concentration, which may then be undone when stepping to the
 9                   next trophic level. It is possible that whether trophic transfer is magnifying or attenuating
10                   depends on Pb concentration itself. Kaufman et al. (2007) determined that, at low
11                   concentrations of soil Pb, risk to secondary consumers (birds and mammals) was driven
12                   by the bioavailability of Pb in worm tissues, while at high soil concentrations,
13                   bioavailability of soil-associated Pb was more critical. The authors concluded that
14                   incorporation of bioavailability/bioaccessibility measurements in terrestrial risk
15                   assessments could lead to more accurate estimates of critical Pb  levels in soil and biota.
16                   Finally, while trophic magnification does greatly increase exposure of organisms at the
17                   higher levels of the food web, these studies establish that atmospherically deposited Pb
18                   reaches species that have little direct exposure to it. For those species, detrimental effects
19                   are not a function of whether they accumulate more Pb than the species they consume,
20                   but of the absolute amount they are exposed to, and their sensitivity to it.
      7.3.4  Biological Effects of Pb in Terrestrial Systems

21                   Various effects can be observed in exposed terrestrial species following uptake and
22                   accumulation of Pb. While many of the responses are specific to organism type, induction
23                   of antioxidant activities in response to Pb exposure has been reported in plants,
24                   invertebrates, and vertebrates. In this section, the observed biological effects caused by
25                   exposure to atmosphere-derived Pb will be discussed, while the results of dose-response
26                   experimentation will be addressed in Section 7.3.5. Because environmental releases of Pb
27                   often include simultaneous release of other metals, it can be difficult to identify
28                   Pb-specific effects in field studies, with the exception of effects from leaded gasoline and
29                   some Pb smelter deposition. Many laboratory studies that expose organisms to natural
30                   soils (or to biosolids-amended soils) also include exposure to multiple metals. There is
31                   some information about mechanisms of metal interactions, such as through competition
32                   for binding locations on specific enzymes or on cellular receptors, but generally such
33                   interactions (particularly of multiple metals) are not well understood (ATSDR. 2004).
34                   Despite a few well-known examples of metal antagonism (e.g., Cu and Mo or Cd and
35                   Zn), it is common practice to assume additivity of effects  (Fairbrother et al.. 2007).


      November 2012                             7-42                   Draft - Do Not Cite or Quote

-------
 1                   Because this review is focused on effects of Pb, studies reviewed for this section and the
 2                   following include only those for which Pb was the only, or primary, metal to which the
 3                   organism was exposed. All reported values are from exposures in which concentrations
 4                   of Pb were analytically verified unless nominal concentrations are stated.
              7.3.4.1  Terrestrial Plants and Lichen

 5                   Pb exposure has been linked to decreased photosynthesis in affected plants, significant
 6                   induction of antioxidant activities, genetic abnormalities, and decreased growth.
 7                   Induction of antioxidant responses in plants has been shown to increase tolerance to
 8                   metal soil contamination, but at sufficiently high levels, antioxidant capacity is exceeded,
 9                   and metal exposure causes peroxidation of lipids and DNA damage, eventually leading to
10                   accelerated senescence and potentially death (Stobrawa and Lorenc-Plucinska. 2008).


                     Effects on Photosystem and Chlorophyll

11                   Photosynthesis and mitosis were recognized as targets of Pb toxicity in plants  in the 1977
12                   Pb AQCD and additional effects of Pb on these processes were reported in subsequent Pb
13                   AQCDs (U.S. EPA. 2006c. 1986b. 1977). The effect of Pb exposure on the structure and
14                   function of plant photosystem II was recently studied in giant duckweed, Spirodela
15                   polyrrhiza (Ling and Hong. 2009). Although this is an aquatic plant, photosystem II is
16                   present in all plants. This finding thus provides support for effects on photosystem II
17                   being the cellular-level mechanism that leads to decreases photosynthesis observed in
18                   other plants. The Pb concentration of extracted photosystem II particles was found to
19                   increase with increasing environmental Pb concentration, and increased Pb concentration
20                   was shown to decrease emission peak intensity at 340 nm, amino acid excitation peaks at
21                   230 nm, tyrosine residues, and absorption intensities. This results in decreased efficiency
22                   of visible light absorption by affected plants. The authors theorized that Pb2+ may replace
23                   either Mg2+ or Ca2+ in chlorophyll or the oxygen-evolving center, inhibiting photosystem
24                   II function through an alteration of chlorophyll structure. Consistently with these  results,
25                   Wu et al. (2008c) demonstrated that Pb exposure interfered with and decreased light
26                   absorption by spinach (Spinacia oleracea) plants. Spinach seeds were soaked in 5, 12, or
27                   25 mM Pb chloride (1036, 2486, or 5180 mg Pb/L) for 48 hours prior to germination, and
28                   following 42 days of growth, plants were sprayed with Pb chloride solutions. Chloroplast
29                   absorption peak intensity, fluorescence quantum yield at 680 nm, and whole-chain
30                   electron transport rate all decreased with Pb exposure, as did photosystem II
31                   photoreduction and oxygen evolution. Similarly, the photosynthetic rate of maize (Zea
32                   mays) seedlings decreased over 21 days exposure to Pb, and measured leaf Pb
      November 2012                             7-43                  Draft - Do Not Cite or Quote

-------
 1                  concentrations in photosynthetically-depressed seedlings ranged from approximately 100
 2                  to 300 mg Pb/kg dry weight (Ahmad et al.. 2011). Liu et al. (2010a) observed that
 3                  chlorophyll a and b content in wheat grown in soils spiked with Pb nitrate rose with
 4                  length of exposure until 14 days, at which point chlorophyll decreased. At nominal
 5                  exposures of 0.1 and 0.5 mM Pb (20.72 and 103.6 mg Pb/L) in hydroponic solution for
 6                  50 days, concentration of chlorophyll a and b was decreased in radish (R. sativus)
 1                  (Kumar and Tripathi. 2008). Changes in chlorophyll content in response to Pb were also
 8                  observed in lichen and moss species following exposures intended to simulate
 9                  atmospheric deposition (Carreras and Pignata. 2007). Usnae ambfyoclada lichen was
10                  exposed to aqueous Pb solutions of 0.5, 1, 5, and 10 mM Pb nitrate (103.6, 207.2, 1,036,
11                  and 10,360 mg Pb/L); chlorophyll a concentration was shown to decrease with increasing
12                  Pb exposure. However, the ratio of lichen dry weight to fresh weight increased following
13                  Pb exposures. It should be noted that highly productive Sphagnum mosses accumulated
14                  atmospheric Pb at the  same rate as slower growing mosses, indicating that moss growth
15                  allowed for further Pb uptake, rather than a "dilution" effect (Kempter et al.. 2010). As
16                  compared to other metals, however, Pb caused less physiological damage, which the
17                  authors attributed to the metal's high affinity for binding to and sequestration within cell
18                  walls (Carreras and Pignata. 2007).

19                  The effect of Pb exposure on chlorophyll content of the moss and liverwort species
20                  Thuidium delicatulum, T. sparsifolium, and Ptychanthus striatus was investigated
21                  following immersion in six solutions of Pb nitrate containing from 10"10 to 10"2 M Pb
22                  (0.00002 to 2,072 mg  Pb/L) (Shakva et al.. 2008). Both chlorophyll a and total
23                  chlorophyll content of the mosses (T. delicatulum and T. sparsifolium) decreased with
24                  increasing Pb exposure. For the liverwort, increasing  Pb exposure resulted in decreases in
25                  content of chlorophyll a, chlorophyll b, and total chlorophyll. Further, the total
26                  chlorophyll content of Hypnumplumaeforme mosses  was decreased by 5.8% following
27                  exposure to the highest concentration, while lower exposures slightly elevated
28                  chlorophyll content.

29                  These studies suggest that exposure to Pb has an impact on photosynthetic pigments, but
30                  the exposure methods (seed soaking, spraying of Pb chloride solutions, hydroponic
31                  growth systems) make it difficult to compare these effects to what might occur under the
32                  uncontrolled conditions encountered in natural environments. These experiments bring to
33                  light the presence of effects, and the underlying mechanisms, but strong uncertainties
34                  remain regarding the natural concentrations at which  theses effects would be observed.
      November 2012                             7-44                  Draft - Do Not Cite or Quote

-------
                     Response of Antioxidants

 1                   Increased antioxidant activity is a common response to Pb exposure, although this
 2                   endpoint may not necessarily be an indication of deleterious effects on plant vitality.
 3                   Increases in reactive oxygen species with increasing exposure to Pb from 20 mg Pb/kg
 4                   soil to 2,000 mg Pb/kg have been demonstrated in broad bean (Viciafabd) (Wang et al..
 5                   2010c; Wang et al., 2010a; Wang et al., 2008b) and tomato (Lycopersicon esculentum)
 6                   (Wang et al.. 2008a). where they were accompanied up to approximately 500mg Pb/kg by
 7                   proportional increases in superoxide dismutase (SOD), glutathione, guaiacol peroxidase,
 8                   and lipid peroxidation, as well as decreases in catalase. Spinach seedlings grown in soil
 9                   containing six increasing concentrations of Pb from 20 to 520 mg Pb/kg exhibited higher
10                   production of reactive oxygen species, increased rates of lipid peroxidation and increased
11                   SOD concentrations. Many of these responses persisted for 50 days after germination and
12                   growth in the Pb-contaminated soil (Wang etal. 201 la). Similarly, the bryophyte mosses
13                   Hypnum plumaeforme, Thuidium cymbifolium, and Brachythecium piligerum exposed to
14                   Pb solutions of greater than 0.1 mM Pb  for 48 hours exhibited increased production of
15                   «O2 and H2O2, although no single moss species could be identified as most sensitive to
16                   Pb exposure (Sun et al..  2011). Increased rates of lipid peroxidation were also observed in
17                   Pb-exposed mosses; however, SOD and catalase activity was suppressed or unaffected by
18                   Pb.

19                   Reddy et al. (2005) found that horsegram (Macrotyloma uniflorum) and bengalgram
20                   (Cicer arietinum) plants watered with Pb solutions containing 200, 500, and 800 mg Pb/L
21                   exhibited increased antioxidant activity: at exposures of 800 mg Pb/L, root and shoot
22                   SOD activity increased to 2-3 times that of controls, and induction was slightly higher in
23                   M. uniflorum. Similarly, catalase, peroxidase, and glutathione-S-transferase activities
24                   were elevated in Pb-stressed plants, but were again higher forM uniflorum. Antioxidant
25                   activities were also markedly greater in  the root tissues than the shoot tissues of the two
26                   plants, and were very likely related to the higher Pb accumulation rate of the roots.  The
27                   effectiveness of the up-regulation of antioxidant systems in preventing damage from Pb
28                   uptake was evidenced by lower malondialdehyde (MDA) (a chemical marker of lipid
29                   peroxidation) concentration inM uniflorum versus C.  arietinum, indicating a lower rate
30                   of lipid peroxidation in response to M. uniflorum's higher antioxidant activity.

31                   Gupta et al. (2010) contrasted responses of two ecotypes ofSedum alfredii  (an Asian
32                   perennial herb), one an accumulator of Pb collected from a Pb and Zn mining area,  and
33                   the other not. Glutathione level was increased in both, and root and shoot lengths were
34                   decreased following long-term exposures to Pb up to 200 (iM (41.4 mg Pb/L) in
35                   hydroponic solution. However, the  accumulator plants exhibited greater SOD and
36                   ascorbate peroxidase activity, likely as a result of greater Pb uptake and a concurrent
37                   increased detoxification  capacity. Similar results were reported by Islam et al. (2008):

      November 2012                             7-45                   Draft - Do Not Cite or Quote

-------
 1                   following Pb exposures of 200 (iM (41.4 mg Pb/L), catalase, ascorbic acid, and
 2                   glutathione levels of another Chinese herb, Elsholtzia argyi, were increased, while SOD
 3                   and guaiacol peroxidase activities decreased. Microscopic analysis also showed that
 4                   affected plants exhibited abnormal chloroplast structures. The response of glutathione
 5                   was further confirmed in wheat (Liu et al., 2010a) grown in soils spiked with Pb nitrate.
 6                   Evidence of increasing lipid peroxidation (MDA accumulation) with increasing Pb
 7                   exposure was also found in mosses (Sun et al., 2009) and lichens. Lichens field-collected
 8                   from the trunks of poplar (Populus tremula) trees in eastern Slovakia were chemically
 9                   analyzed for metal concentration arising from exposure to smelter pollution (Dzubaj et
10                   al., 2008).  These concentrations (ranging from 13 to 1,523 mg Pb/kg dry weight) were
11                   assessed in relation to physiological variables, including chlorophyll a and b, carotenoids,
12                   photosystem II activity, CO2 gas exchange (respiration), and MDA content. Lichen Pb
13                   levels were significantly correlated only with MDA content. Determination of plant
14                   chitinase content following exposure to As, Cd and Pb indicated that while levels of these
15                   defense proteins were elevated by As and Cd, chitinase levels were not increased
16                   following exposure to Pb (Bekesiova et al., 2008). As in studies of effects on
17                   photosynthesis, the methods used for exposure make it difficult to compare these effects
18                   to what might occur under the uncontrolled conditions encountered in natural
19                   environments.
                     Growth

20                   Evidence of effects of Pb on higher growth processes in terrestrial plants was reported in
21                   early NAAQS reviews. Impacts to growth can lead to effects at the population-level of
22                   biological organization and higher (Section 7.1.1). Growth effects of Pb on plants in the
23                   1977 Pb AQCD primarily included visible growth responses observed in laboratory
24                   studies with plants grown in artificial nutrient culture (U.S. EPA. 1977). No Pb toxicity
25                   was observed in plants growing under field conditions at the time of the 1977 Pb AQCD.
26                   Indirect effects of Pb on plant growth (i.e., inhibition of uptake of other nutrients when
27                   Pb is present in the plant) were also reported in the 1977 Pb AQCD. In the 1986 Pb
28                   AQCD mechanisms of Pb effects on growth included reduction of photosynthetic rate,
29                   inhibition of respiration, cell elongation, root development or premature senescence (U.S.
30                   EPA. 1986b). All of these effects were observed to occur in isolated cells or in plants
31                   grown hydroponically in solutions comparable to 1 to 2 mg Pb/kg soil or in soils with
32                   10,000 mg Pb/kg or greater (U.S. EPA. 1986R Pb effects on other plant processes,
33                   especially maintenance, flowering and hormone development had not been studied at the
34                   time of the  1986 Pb AQCD and remain poorly characterized.
      November 2012                             7-46                  Draft - Do Not Cite or Quote

-------
 1                   Recent evidence for growth effects in terrestrial plants available since the
 2                   2006 Pb AQCD is reviewed below and summarized in Table 7-4. Both growth and
 3                   carotenoid and chlorophyll content of Brassica juncea (mustard) plants were negatively
 4                   affected by Pb exposure (John et al.. 2009). Nominal Pb treatments of 1,500 (iM (311 mg
 5                   Pb/L) as Pb acetate solution decreased root lengths and stem heights by 50% after 60
 6                   days. Exposure to 600 (iM Pb (124 mg Pb/L) and greater decreased carotenoid content,
 7                   while chlorophyll a was decreased at Pb exposures of 450 (iM (93 mg Pb/L) and higher.
 8                   However, when smelter ash-spiked soils containing 1,466 mg Pb/kg (and 18.6 mg Cd/kg)
 9                   or 7,331 mg Pb/kg (98.0 mg/kg Cd) were used to grow maize (Zea mays), as well as other
10                   metals in high concentrations, effects were seen in growth or chlorophyll production  only
11                   at the higher concentration (Komarek et al.. 2009). Given the  low solubility of smelter
12                   ash, these observations are consistent with solubility being a key determinant of
13                   bioavailability. Similarly, wheat seedling growth was unaffected when exposed to soil
14                   leachate containing up to 0.7 mg Pb/L for six weeks. Lettuce seedling root growth was
15                   negatively correlated to leachate Pb concentration, but this correlation was only
16                   significant for week 3 and week 6 measurements. Authors concluded that although the
17                   total concentrations of multiple metals in tested soils and leachates exceeded Canadian
18                   Environmental Quality Guidelines, no toxic or only slightly toxic effects occurred
19                   following exposure to the metal mixture (Chapman et al.. 2010).

20                   Chinese cabbage  (Brassica pekinensis) exposed to Pb-containing soils exhibited
21                   depressed nitrogen assimilation as measured by shoot nitrite content, nitrate reductase
22                   activity, and free  amino acid concentration (Xiong et al.. 2006). The authors planted
23                   germinated cabbage seeds in soils spiked with Pb acetate to give final soil concentrations
24                   of 0.2, 4, and 8 mM Pb/kg dry weight total Pb (41.4, 828.8 and 1,657.6 mg Pb/kg ) and
25                   collected leaf samples for 11 days. At exposures of 4 and 8 mM Pb/kg (828.8 and 1,657.6
26                   mg Pb/kg), leaf nitrite content was decreased by 29% and 20%, while nitrate content was
27                   affected only at the highest Pb exposure (70% of control levels). Free amino acid content
28                   in exposed plants was 81% and 82% of control levels, respectively. B. pekinensis shoot
29                   biomass was observed to decrease with increasing Pb exposures, with biomass at the two
30                   highest Pb exposures representing 91% and 84% of control growth, respectively.

31                   Nitrogen, potassium, and phosphorus concentrations in the shoot and root tissues of four
32                   canola cultivars (Brassica napus) also decreased as spiked soil Pb concentrations
33                   increased from 0 to 90 mg/kg. At the highest soil Pb concentration, nitrogen
34                   concentrations were reduced 56% in roots and 58% in shoots  versus control levels, while
35                   phosphorous concentrations were reduced 37% and 45%, respectively, and potassium
36                   content decreased by 42% in both tissues (Ashraf et al.. 2011). Cultivation in Pb-spiked
37                   soils was also linked to decreased shoot and root biomass (32% and 62%, respectively at
38                   90 mg Pb/kg).
      November 2012                             7-47                  Draft - Do Not Cite or Quote

-------
                    Genetic and Reproductive Effects

 1                  Exposure to Pb also resulted in genetic abnormalities, including bridges, condensed
 2                  bivalents, and laggards, in the meiotic cells of pea plants (Lathyrus sativus) (Kumar and
 3                  Tripathi, 2008). Seeds were germinated in soils amended with Pb nitrate at concentrations
 4                  of 25, 50, 100, 200, and 300 mg Pb/kg, and concentrations of 100 mg Pb/kg and greater
 5                  were found to be genotoxic or detrimental to pea viability. Cenkci et al. (2010) exposed
 6                  fodder turnip (B. rapa) to 0.5 to 5 mM of Pb nitrate (103.6 to 1036 mg Pb/L) for 6 days
 7                  and showed decreased genetic template stability (as quantified by random amplified
 8                  polymorphic DNA profiles) and decreased photosynthetic pigments.

 9                  Two genotypes of maize seedlings exhibited a significant and concentration-dependent
10                  reduction in seed germination following 7 days of Pb treatment in nutrient solution of
11                  0.01, 0.1 and 1 mg Pb/L as Pb sulfate (Ahmad et al.. 2011). Pb exposure also decreased
12                  germination rate and growth, and increased pollen sterility in radish grown for 50 days in
13                  hydroponic solutions containing 0.5 mM Pb (104 mg Pb/L) (Kumar and Tripathi.  2008).
14                  Plants exposed to Pb exhibited decreased growth, curling and chlorosis of young leaves,
15                  and decreased root growth. In addition, Gopal and Rizvi (2008)  showed that Pb exposure
16                  increased uptake of phosphorus and iron and decreased sulfur concentration in radish
17                  tops.

18                  Interestingly, as in zebra finch (Section 7.3.3.3)  Ca2+ was found to moderate the effects of
19                  Pb in both monocotyledon and dicotyledon plant seedlings, with tomato (Lycopersicon
20                  esculentum), rye (Lolium sp.), mustard, and maize plants exhibiting increased tolerance
21                  to Pb exposures of 5, 10, and 20 mg Pb/L in the  presence of Ca2+ concentration of 1.2
22                  mM (249 mg Pb/L) and higher (Antosiewicz. 2005).
              7.3.4.2  Terrestrial Invertebrates

23                  Exposure to Pb also causes antioxidant effects, reductions in survival and growth, as well
24                  as decreased fecundity in terrestrial invertebrates as summarized in the 2006 Pb AQCD
25                  (U.S. EPA. 2006b). Alterations in reproduction, growth and survival at the species level
26                  can lead to effects at the population-level of biological organization and higher
27                  (Section 7.1.1). In addition to these endpoints, recent literature also indicates that Pb
28                  exposure can cause significant neurobehavioral aberrations, and in some cases,
29                  endocrine-level impacts. Second-generation effects have been observed in some
30                  invertebrate species.

31                  The morphology of y-aminobutyric acid (GABA) motor neurons in Caenorhabditis
32                  elegans nematodes was affected following exposure  to Pb nitrate for 24 hours  (Du and
      November 2012                             7-48                  Draft - Do Not Cite or Quote

-------
 1                  Wang. 2009). The authors determined that exposures as low as 2.5 (iM Pb nitrate (0.5 mg
 2                  Pb/L) could cause moderate axonal discontinuities, and observed a significant increase in
 3                  the number of formed gaps and ventral cord gaps at Pb nitrate exposures of 75 and
 4                  200 (iM (6 and 41 mg Pb/L). Younger C. elegans larvae were more likely to exhibit
 5                  neurobehavioral toxicity symptoms in response to Pb exposure at 2.5 (iM (0.5 mg Pb/L)
 6                  (Xing et al.. 2009b). Neural degeneration, as demonstrated by dorsal and ventral cord
 7                  gaps and neuronal loss was also more pronounced in young larval C. elegans than in
 8                  older larvae and adults (Xing et al.. 2009c). C. elegans nematodes exposed to Pb
 9                  concentration as low as 2.5 (iM (0.5 mg Pb/L) for 24 hours also exhibited significantly
10                  altered behavior characterized by decreased head thrashes and body bends. Exposures of
11                  50  (iM Pb (10 mg Pb/L) and greater decreased the number of nematode forward turns
12                  (Wang and Xing.  2008). Chemotaxis toward NaCl, cAMP, and biotin was also decreased
13                  in C. elegans nematodes exposed to Pb concentration greater than 2.5 (iM (0.5 mg Pb/L)
14                  (Xing et al., 2009a). This evidence suggests that Pb may exert neurotoxic action in
15                  invertebrates as it does in vertebrates. However, it is unclear how these behavioral
16                  aberrations would affect fitness or survival  (Wang and Xing.  2008).

17                  In a study of C. elegans exposed to 4 sub-lethal concentrations of Pb nitrate between 25
18                  and 100 (iM (5 and 21 mg Pb/L), Vigneshkumar et al. (In Press) observed upregulation of
19                  both catalase and  antimicrobial response-related genes. When challenged with addition of
20                  a pathogenic strain ofPseudomonas aeruginosa, exposed C. elegans showed greater
21                  resistance to microbial colonization than controls.

22                  Younger individuals also appear to be more sensitive to the reproductive effects of Pb
23                  exposure. Guo et  al. (2009) showed that concentrations of 2.5, 50, and 100 (iM Pb (0.5,
24                  10, and 21 mg Pb/L) had greater significant adverse effects on reproductive output when
25                  early-stage larval  C. elegans were exposed. Adult C. elegans exhibited decreased brood
26                  size only when exposed to the highest Pb concentration.
27                  The progeny of C. elegans nematodes exposed nominally to 2.5, 75, and 200
28                  Pb nitrate (0.5, 16, and 41 mg Pb/L) exhibited significant indications of multi-
29                  generational toxicity (Wang and Peng. 2007). Life spans of offspring were decreased by
30                  increasing parental Pb exposure, and were comparable to the reductions in parental life-
31                  spans. Similarly, diminished fecundity was observed in the progeny of C elegans
32                  exposed to Pb (9%,  19%, and 31% reductions of control  fecundity, respectively),
33                  although the decrease was smaller than in the exposed parental generation (reductions of
34                  52%, 58%, and 65%, respectively). Significant behavioral defects affecting locomotion
35                  were also observed in the offspring, but these impacts were not determined to be
36                  concentration-dependent. Reproductive effects of Pb exposure were also observed in
37                  springtails F. Candida following 10-day exposure to Pb-spiked soils. Egg hatch
      November 2012                             7-49                   Draft - Do Not Cite or Quote

-------
 1                   significantly decreased at nominal concentrations of 1,600 mg Pb/kg dry soil and higher
 2                   and the EC50 for hatching was 2,361 mg Pb/kg dry soils (Xu et al. 2009b).

 3                   E. andrei earthworms exposed to 21 different soils, each containing 2,000 mg Pb/kg
 4                   freshly added Pb, for 28 days exhibited highly variable mortality, ranging from 0% to
 5                   100%, (Bradham et al.. 2006).  Pb body burden of exposed worms ranged from 29 to
 6                   782 mg Pb/kg. Internal Pb concentration was also negatively correlated to reproductive
 7                   output. CEC and pH were found to be the principal soil characteristics determining the
 8                   differences in those effects, although the apparent role of CEC may only have been due to
 9                   its correlation with other soil characteristics. Low soil Pb concentration (5 mg Pb/kg) also
10                   decreased the protein content of E. fetida earthworms during a 7-day exposure (Li  et al..
11                   2009b). Higher Pb concentration had no  effect on protein production. However, cellulase
12                   activity was increased by the 7-day exposures to Pb at all exposure concentrations  (31%,
13                   13%, and 23% of control activity at exposures of 5, 50, and 500 mg Pb/kg, respectively),
14                   which the authors reported as an indication of detrimental effects on worm metabolism.
15                   By contrast, Svendsen et al.  (2007) found thatZ. rubellus earthworms exposed for 42
16                   days to field-collected smelter-polluted soils containing average Pb concentration of 106,
17                   309, and 514 mg Pb/kg dry weight exhibited normal survival and cocoon production
18                   rates, even though they accumulated more Pb with increased environmental
19                   concentration. The much smaller effect may be explained by the  increased aging time
20                   undergone by field soil, causing a larger fraction of the total Pb to be complexed and
21                   sequestered by organic and inorganic compounds.  Similarly, earthworms (E. fetida)
22                   exposed to field-collected soils with concentrations of Pb and As up to 390 mg/kg  and
23                   128 mg/kg, respectively, due to historical treatments of Pb-arsenate pesticides, exhibited
24                   no change in survival, behavior or morphology (Delistraty and Yokel. In Press). Soil
25                   aging (e.g., from of the time of Pb-arsenate applications in 1942 to  soil collection in
26                   approximately 2009) likely reduced Pb bioavailability to earthworms.

27                   As in plants, induction of metal chelating proteins  and antioxidant activity in
28                   invertebrates is affected by exposure to Pb. Metallothionein production in earthworms
29                   (Lampito mauritif) was significantly induced following exposure to Pb-contaminated soil.
30                   Tissue metallothionein levels increased after a two-week exposure to 75 to 300 mg Pb/kg
31                   soil, although by 28 days levels had begun to decrease, perhaps as a result of Pb toxicity
32                   (Maity et al., 2011). Further, the induction of antioxidant activity was correlated to
33                   standard toxicity measurements in Thebapisana snails (Radwan et al.. 2010). Topical
34                   application of Pb solutions (estimated to be  500 to 2,000 (ig Pb per animal) to snails
35                   resulted in decreased survival,  increased catalase and glutathione peroxidase activities,
36                   and decreased glutathione concentration. The 48-hour LD50 concentration was
37                   determined to be 653 (ig per snail, as measured in digestive gland tissue. Snail
38                   glutathione content was decreased at exposures of 72.2% of the 48-hour LD50 value,
      November 2012                             7-50                  Draft - Do Not Cite or Quote

-------
 1                   while Pb exposure at 40% of the 48-hour LD50 value induced catalase and glutathione
 2                   peroxidase activities.

 3                   Dietary exposure to Pb also affected T. pisana snail growth. After three weeks on
 4                   Pb-contaminated diet, snail feeding rates were depressed by all Pb exposures (nominal
 5                   concentration of 50 to 15,000 mg Pb/kg  diet dry weight) (El-Gendv et al.. 2011). A five
 6                   week dietary exposure to 1,000 mg Pb/kg and greater resulted  in reduced snail growth.
 7                   Food consumption, growth, and shell thickness were also observed to decrease with
 8                   increasing diet Pb in juvenile A. achatina snails (7 levels between 0 and 1,344 mg Pb/kg
 9                   feed, for 12 weeks) (Ebenso and Ologhobo. 2009a). A similar  depression of growth was
10                   observed in sentinel juvenile A. achatina snails deployed at Pb-polluted sites in the Niger
11                   Delta region of Nigeria. Although snail mortality was not increased significantly by
12                   exposure to soil Pb up to 1,200 mg Pb/kg, a concentration-dependent relationship was
13                   established for growth, with significant reduction observed at  12-week exposures to
14                   20 mg Pb/kg (Ebenso and Ologhobo. 2009b).  However, consumption of field-collected
15                   Pb-polluted U. dioica leaves containing  3 mg  Pb/kg stopped all reproductive output in
16                   C. nemoralis. Snails also exhibited diminished food consumption rates when offered
17                   leaves with both low (1.5 mg Pb/kg) and high Pb content, but the mechanism of the
18                   dietary aversion was not defined (Notten et al.. 2006).

19                   Chronic dietary exposure to Pb was also examined in post-embryonic oribatid mites
20                   (Archegozetes longisetosus) (Kohler et al., 2005). Both algae and bark samples were
21                   soaked in 100 mg/L Pb as Pb nitrate and provided as diet and substrate, respectively, to
22                   larval mites. In addition to elevated heat shock proteins  (hsp70),  90.8% of the
23                   protonymphs exhibited significant  leg deformities, including abnormal claws, shortened
24                   and thickened legs, and translocated setae. Although not specifically discussed, it is very
25                   likely that these deformities would decrease mite mobility, prey capture, and reproductive
26                   viability. While there is some evidence that oribatid mites exhibit Pb avoidance behavior,
27                   this response may not significantly reduce Pb  exposure and effects. Although soil-
28                   inhabiting mites (Oppia nitens) were observed to avoid high Pb concentrations, the EC50
29                   for this behavior was approximately five times higher than the chronic EC50 for
30                   reproduction (8,317 and 1,678 mg Pb/kg, respectively) (Owojori et al., 2011).
31                   Consequently, it is unlikely that oribatid mites will avoid soils containing toxic Pb
32                   concentrations.

33                   Lock et al. (2006) compared the toxicity of both laboratory-spiked soils and field-
34                   collected Pb-contaminated soils to  springtails  (F. Candida). The 28-day EC50 values
35                   derived for F. Candida ranged  from 2,060 to 3,210 mg Pb/kg in leached and unleached
36                   Pb-spiked soils, respectively, whereas field-collected soils had no significant effect on
37                   springtail reproduction up to (but not including) 14,436  mg Pb/kg (Lock et al., 2006).
      November 2012                             7-51                   Draft - Do Not Cite or Quote

-------
 1                   Consequently, leaching soils prior to use in bioassays had only a slight effect on Pb
 2                   toxicity to resident springtails, and did not provide an appropriate model for field-
 3                   weathered, Pb-contaminated soils. This indicates that physiochemical factors other than
 4                   leaching may be more important determinants of Pb bioavailability. A 4-week exposure
 5                   to Pb-amended soils containing up to 3,200 mg Pb/kg (nominal concentration) had no
 6                   significant effect on Sinella curviseta springtail survival or reproduction (Xu et al..
 7                   2009a).

 8                   Carabid beetles (Pterostichus oblongopunctatus) inhabiting soils contaminated by
 9                   pollution from a Pb-Zn smelter (containing 136 to 2,635 mg Pb/kg) were field-collected
10                   and then laboratory-reared for two generations (Lagisz and Laskowski. 2008). While
11                   fecundity was positively correlated to soil metal concentration (e.g., more eggs were
12                   produced by females collected from contaminated areas), the hatching rate of eggs
13                   diminished with increasing soil metal contamination. For the Fl generation, females
14                   produced by parents inhabiting highly polluted areas exhibited decreased body mass. The
15                   authors stated that these results indicate that invertebrates inhabiting metal- (or Pb-)
16                   contaminated soils could face "significantly altered life-history parameters." Similarly,
17                   aphids (Brevicoryne brassicae) reared on cabbage and radish plants exposed to 0.068 mg
18                   Pb daily exhibited altered development and reproduction when compared to those reared
19                   on non-exposed plants. Development time was increased by approximately two days,
20                   which led to a reduction in relative fecundity (Gorur. 2007). Although the authors noted
21                   that study exposures were greater than what would be expected in naturally polluted
22                   areas, Pb exposure under field conditions could alter invertebrate life history patterns.

23                   Several studies suggest that Pb may  disrupt hormonal homeostasis in invertebrates. Shu
24                   et al.  (2009) reported that vitellogenin production in both male and female S.  litura moths
25                   was disrupted following chronic dietary exposure to Pb. Adult females reared on diets
26                   containing 25, 50, 100, or 200 mg Pb/kg exhibited decreased vitellogenin mRNA
27                   induction, and vitellogenin levels decreased with increasing Pb exposure. In addition,
28                   vitellogenin mRNA induction was detected in males  exposed to 12 and 25 mg Pb/kg, and
29                   low levels of vitellogenin were found at those lower Pb exposures, when males normally
30                   do not produce any. In the Asian earthworm (Pheretima guillelmi), sperm morphology
31                   was found to be altered significantly following 2-week exposure to soils containing
32                   nominal concentration of 1,000, 1,400,  1,800, and 2,500 mg Pb/kg (Zheng and Li. 2009).
33                   Common deformities were swollen head and head helices, while head bending was also
34                   recorded in  some cases. These deformities were observed following exposures to
35                   concentration below the 14-day LC50 (3,207 mg Pb/kg) and below the concentration at
36                   which weight was diminished (2,800 mg Pb/kg). Experimentation with the model
37                   organism Drosophila indicates that Pb exposure may increase time to pupation and
      November 2012                             7-52                  Draft - Do Not Cite or Quote

-------
 1                  decrease pre-adult development, both of which are endocrine-regulated (Hirsch et al.,
 2                  2010).
              7.3.4.3  Terrestrial Vertebrates

 3                  Pb poisoning is one of the earliest recognized toxicoses of terrestrial vertebrates,
 4                  occurring primarily through the ingestion of spent shot by birds. While the focus of the
 5                  ISA is on more environmentally relevant exposures, studies of Pb poisoning provide
 6                  historical context for the review. The widespread nature of this toxicosis was first noticed
 7                  in American waterfowl around the turn of the last century (see (Jones. 1939) for an
 8                  historical summary). Wetmore (1919) demonstrated that Pb shot caused the observed
 9                  effects and described in detail the species affected, associated symptoms, and additional
10                  factors involved. By 1959, the estimated annual loss of waterfowl to Pb poisoning was
11                  2-3 percent of the fall population (Bellrose. 1959).  Smaller numbers of shorebirds and
12                  upland game birds were also found poisoned by Pb  (Locke and Thomas. 1996).

13                  The first reported Pb poisoning of a bald eagle (Haliaeetus leucocephalus) was described
14                  by Mulhern et al. (1970). and subsequently several hundred bald eagle Pb poisonings
15                  were diagnosed throughout the U.S. prior to the ban on use of Pb shot for waterfowl
16                  hunting (Kramer and Redig, 1997). Eagles and other raptors are poisoned by consuming
17                  Pb pellets imbedded in the flesh of ducks or upland prey species and may also be exposed
18                  to other sources of Pb, such as fishing sinkers and weights (Kramer and Redig. 1997).
19                  The use of Pb shot for waterfowl hunting was banned in 1991 due to the poisoning of
20                  bald eagles, which had been previously added to the endangered species list and were
21                  specially protected under the Bald Eagle Protection Act  of 1940.

22                  Anderson et al. (2000) reported that by 1997, mallard (Anas platyrhynchus) deaths from
23                  Pb poisoning in the Mississippi flyway were reduced by 64 percent, and ingestion of
24                  toxic pellets had declined by 78 percent. They estimated the ban prevented approximately
25                  1.4 million duck deaths in the  first 6-year period. However, Pb exposure remains
26                  widespread in bald eagles, although blood Pb concentrations have significantly decreased
27                  (Kramer and Redig. 1997). The endangered California condor (Gymnogyps
28                  californianus)  also continues to have significantly elevated blood Pb levels as well as
29                  Pb-associated mortality resulting from exposure to ammunition fragments contained in
30                  food items (Cade. 2007; Church et al.. 2006). Although there is a significant amount of
31                  information on Pb tissue residues of mammals, there are very few reports of Pb
32                  poisoning; exceptions are reports of Pb poisoned bats in a cave in the southern U.S. and
33                  small mammals in the vicinity of several smelters (Shore and Rattner. 2001).
      November 2012                             7-53                  Draft - Do Not Cite or Quote

-------
 1                  At the time of the 1977 Pb AQCD few studies of the effects of exposure to Pb had been
 2                  conducted in wild animals other than birds, and the majority of those studies were of
 3                  direct poisoning (U.S. EPA. 1977). Several studies of domestic animals grazing near Pb
 4                  smelters indicated that horses are more susceptible than cattle to chronic Pb exposure
 5                  although the findings were not conclusive due to the presence of other metals. Delta-
 6                  aminolevulinic acid dehydratase (ALAD) was recognized as a sensitive indicator of Pb
 7                  exposure in rats and waterfowl. In the 1986 Pb AQCD, additional effects of Pb on small
 8                  mammals and birds were reported. According to the 2006 Pb AQCD (U.S. EPA. 2006b).
 9                  commonly observed effects of Pb on avian and mammalian wildlife include decreased
10                  survival, reproduction, and growth, as well as effects on development and behavior. More
11                  recent experimental data presented here expand and support these conclusions, and also
12                  indicate that Pb can exert other effects on exposed terrestrial vertebrates, including
13                  alteration of hormones and other biochemical variables.

14                  Since the 2006 Pb AQCD, there is additional evidence for hematological effects of Pb
15                  exposure in terrestrial vertebrates. Red-backed salamanders (Plethodon cinereus)
16                  exposed to Pb-amended soils (553, 1,700, 4,700, and 9,167 mg Pb/kg) by Bazar et al.
17                  (2010) exhibited lowered appetite and decreased white blood cell counts at the two
18                  highest concentrations, but tolerated field-collected, aged soils containing Pb
19                  concentrations  of up to 16,967 mg Pb/kg with no significant deleterious effects. The
20                  white blood cell count of adult South American toads, (Bufo arenarum) was also
21                  decreased by weekly sublethal i.p. injections of Pb acetate at 50 mg Pb/kg body weight,
22                  (Chiesa et al.. 2006). The toads also showed altered serum profiles and increased number
23                  of circulating blast cells. Final toad blood Pb levels were determined to be 8.6 mg Pb/dL,
24                  although it is unclear whether this is representative of Pb concentrations observed in field
25                  B. arenarum populations exposed to Pb. The authors suggested that, based on these
26                  findings, long-term environmental exposure to Pb could affect toad immune response. In
27                  western fence lizards (S. occidentalis), sub-chronic (60-day) dietary exposure to 10 to
28                  20 mg Pb/kg per day resulted in significant sublethal effects, including decreased cricket
29                  consumption, decreased testis weight, decreased body fat, and abnormal posturing and
30                  coloration (Salice et al.. 2009). Long-term dietary Pb exposures are thus likely to
31                  decrease lizard fitness.

32                  Even in cases of high environmental Pb exposures, however, linking Pb body burdens to
33                  biological effects can be difficult. Pb concentration in the breast feathers, washed tail
34                  feathers, and blood of field-collected blackbirds (Turdus meruld) were determined to be
35                  3.2 mg  Pb/kg, 4.9 mg Pb/kg, and 0.2 mg Pb/kg wet mass in urban birds, as opposed to
36                  1.4 mg  Pb/kg, 1 mg Pb/kg, and 0.05 mg Pb/kg in rural birds (Scheifler et al.. 2006a).
37                  However, the elevated Pb tissue concentrations in urban birds were not significantly
38                  correlated to any index of body condition.
      November 2012                             7-54                  Draft - Do Not Cite or Quote

-------
 1                  The long-term effect of atmospheric Pb deposition on pied flycatcher (Ficedula
 1                  hypoleuca) nestlings was determined in native communities residing in the Laisvall
 3                  mining region of Sweden (Berglund et al.. 2010). Moss samples indicated that Pb
 4                  deposition in study areas ranged between 100 and 2,000 mg Pb/kg dry weight during
 5                  operations and 200 and 750 mg Pb/kg when operations ceased. A simultaneous slight
 6                  reduction was observed in pied flycatcher blood Pb levels, from 0.4 to 0.3 mg Pb/kg.
 7                  However, clutch size was decreased in pied flycatchers inhabiting the mining area both
 8                  during and after mining operations, and mean nestling mortality was 2.5 times higher in
 9                  the mining region than in reference areas during mining operations, and 1.7 higher five
10                  years after cessation of mining operations. The authors noted that Pb deposition in the
11                  mining region remained elevated even after mining operations ceased, and that stable Pb
12                  isotope analysis suggested that smelter Pb remained available to pied flycatcher through
13                  the transfer of historically deposited Pb in soil to prey items.

14                  Berglund et al. (2010) also analyzed ALAD activity in pied flycatchers at the later period,
15                  and found that it was 46% lower at the mine site. Beyer et al. (2004) observed that
16                  elevated blood Pb levels in several types of birds inhabiting the Tri-State Mining District
17                  (Oklahoma, Kansas, Missouri) were correlated with decreases in ALAD activity. Based
18                  on reduction in ALAD activity, robins (Turdus migratorius) were most sensitive to Pb
19                  exposure (35% reduction), followed by cardinals (Cardinalis cardinalis), waterfowl, and
20                  bobwhite quail (Colinus virginianus) (40%, 41%, and 56% reductions, respectively).
21                  Eagle owl (Bubo bubo) nestlings  living in a historical mining area in Spain also exhibited
22                  elevated blood Pb levels (average 8.61 (ig/dL as compared to an average reference area
23                  value of 3.18 (ig/dL), and this was correlated to an approximate 60% reduction in ALAD
24                  activity (Gomez-Ramirez et al.. 2011). Hansen et al. (2011 a) determined that ground-
25                  feeding songbirds were frequently exposed to Pb within the Coeur d'Alene, ID mining
26                  region. Robins, in particular, were significantly likely to exhibit blood Pb levels in the
27                  clinical and severe clinical poisoning ranges (50 to 100 (ig/dL and >100 (ig/dL,
28                  respectively). Ingested soil Pb accounted for almost all of the songbirds' exposure to Pb,
29                  with Pb exposure correlated with estimated soil ingestion rates (20% for robins, 17% for
30                  song sparrows, and 0.7% for Swainson's thrushes, Catharus ustulatus). More than half of
31                  the robins and song  sparrows from all contaminated sites and more than half of the
32                  Swainson's thrushes from highly contaminated sites showed at least 50% inhibition of
33                  ALAD. The highest hepatic Pb concentration of 61 mg/kg (dry weight) was detected in a
34                  song sparrow (Hansen et al.. 201 la).

35                  Blood Pb was significantly elevated in waterfowl in the Lake Coeur d'Alene areas of
36                  Blackwell Island and Harrison Slough (mean sediment concentrations of 679 and
37                  3,507 mg Pb/kg dry weight, respectively). Twenty-seven percent of the waterfowl
38                  sampled in the Blackwell Island region had blood Pb concentrations suggestive of severe
      November 2012                             7-55                  Draft - Do Not Cite or Quote

-------
 1                   clinical poisoning (average concentration=0.17 mg Pb/kg); in the Harrison Slough, 60%
 2                   of sampled waterfowl had highly elevated blood Pb levels that exceeded the severe
 3                   clinical poisoning threshold (average concentration=2,2 mg Pb/kg) (Spears  et al., 2007).
 4                   The level of corticosteroid hormones in field populations of white stork nestlings
 5                   (Ciconia ciconid) in a mining area affected by Pb and other metals was positively
 6                   correlated with blood Pb levels (Baos et al.. 2006). The effect was more pronounced for
 7                   single nestlings than for multiple-chick broods. Surprisingly, average blood Pb levels in
 8                   chicks inhabiting reference areas was 910 (ig Pb/dL (± 51), which was higherthan blood
 9                   Pb levels from the mining area (440 ± 340 (ig Pb/dL). However, the correlation between
10                   blood Pb levels and the corticosteroid stress response in white stork nestlings was
11                   observed in both groups of birds. Burger and Gochfeld (2005) exposed herring gull
12                   (Lams argentatus) chicks to Pb acetate via an i.p.  injection of 100 mg Pb/kg body
13                   weight, to produce feather Pb concentration approximately equivalent to those observed
14                   in wild gulls. Pb-exposed gulls exhibited  abnormal behaviors, including decreased
15                   walking and food begging, erratic behavioral thermoregulation, and diminished
16                   recognition of caretakers. Interestingly, subchronic exposure of Japanese quail (Coturnix
17                   coturnixjaponica) to 5 and 50 mg Pb/L in drinking water caused an increase in their
18                   immune response. Exposed quail exhibited significantly lower rates of death or health
19                   effects (including septicemia, perihepatitis, and pericarditis among others) than control
20                   animals  following infection with Escherichia coli, and the incidence of infection-related
21                   effects was dependent on Pb exposure (Nain and Smits. 2011). These observations
22                   contrast with immunotoxicology results in mice reported in Section 5.6.5.1.

23                   Again, dietary or other health deficiencies unrelated to Pb exposure  are likely to
24                   exacerbate the effects of Pb. Ca2+-deficient female zebra finches (T.  guttatd) had a
25                   suppressed secondary humoral immune response following 28-day exposures to 20 mg
26                   Pb/L in drinking water (Snoeijs et al.. 2005). This  response, however, was not observed
27                   in birds fed sufficient Ca2+. Although a significant finding, these data are difficult to
28                   interpret under field conditions where the overall health of avian wildlife may not be
29                   easily determined.

30                   Chronic Pb exposures were also demonstrated to affect several mammalian  species.
31                   Young adult rats reared on a diet containing 1,500 mg Pb/kg Pb acetate for 50 days
32                   demonstrated less plasticity in learning than non-exposed rats (McGlothan et al.. 2008).
33                   indicating that Pb exposure caused significant alteration in neurological function. Yu et
34                   al. (2005) showed that dietary Pb exposure affected both the growth and endocrine
35                   function of gilts (S.  domestica). Consumption of 10 mg Pb/kg diet resulted in lower body
36                   weight and food intake after 120 days of dietary exposure; Pb exposure decreased final
37                   weight by 8.2%, and average daily food intake of Pb-exposed pigs was decreased by
38                   6.8% compared to control intake. Additionally, concentration of estradiol, luteinizing
      November 2012                             7-56                   Draft - Do Not Cite or Quote

-------
 1                   hormone, and pituitary growth hormone were decreased (by 12%, 14%, and 27% versus
 2                   controls, respectively), while blood Pb level was increased by 44% to an average
 3                   2.1 (ig/dL. In cattle grazing near Pb-Zn smelters in India, blood Pb levels were positively
 4                   correlated with plasma levels of the thyroid hormones thyroxine (T4) and tri-
 5                   iodothyronine (T3) and the hepatic biomarkers alanine transaminase and aspartate
 6                   transaminase (Swarup et al.. 2007). Total lipids, total protein and albumin levels were
 7                   decreased in the same animals. Rodriguez-Estival et al. (2011) determined that red deer
 8                   (Cervus elaphus) and wild boar (Sus scrofd) inhabiting a Pb-contaminated mining area in
 9                   Spain exhibited increased liver and bone Pb concentrations (geometric means  of 0.35 and
10                   0.46 mg Pb/kg for red deer, and 0.81 and 7.36 mg Pb/kg for wild boar, respectively).
11                   These tissue concentrations were correlated to a significant decrease in red deer
12                   glutathione production, but corresponded to an increase in wild boar glutathione
13                   (Rodriguez-Estival et al.. 2011). Authors proposed that the different antioxidant
14                   responses may be indicative of different Pb susceptibilities in the two species.

15                   Following previous reports of in vivo follicle and oocyte damage in animals with
16                   low-level Pb accumulation, Nandi et al. (2010) treated oocytes of buffalo (Bubalus
17                   bubalis) in vitro with Pb at concentrations ranging from 0.005 to 10 mg/L in one-day
18                   cultures indicated a significant decline in viability of oocytes at 1 mg/L. Dose-dependent
19                   effects on oocyte viability, morphological abnormalities, cleavage, blastocyst yield and
20                   blastocyst hatching were observed in Pb-treated oocytes with maturation significantly
21                   reduced at 2.5 mg/L and 100% oocyte death at 32 mg/L. These results appear to confirm
22                   previous reports, but the in vitro concentrations of Pb are difficult to relate to in vivo
23                   exposures. On the other hand, the reproductive viability of wild red deer from the
24                   Pb-contaminated mining area of Spain studied by Rodriguez-Estival et al. (2011) was
25                   shown to be altered, with 11% and 15% reductions in spermatozoa and acrosome
26                   integrity observed in male deer from the mining area compared with those residing in
27                   reference  areas (Reglero et al., 2009a).
      7.3.5  Exposure and Response of Terrestrial Species

28                   Evidence regarding exposure-response relationships and potential thresholds for Pb
29                   effects on terrestrial populations can inform determination of standard levels that are
30                   protective of terrestrial ecosystems. Given that exposure to Pb may affect plants,
31                   invertebrates and vertebrates at the organism, population, or community level,
32                   determining the rate and concentration at which these effects occur is essential in
33                   predicting the overall risk to terrestrial organisms. This section updates available
34                   information derived since the 2006 Pb AQCD, summarizing several dose-response
35                   studies with soil invertebrates. As shown in the studies summarized in Table 7-4. several
      November 2012                             7-57                  Draft - Do Not Cite or Quote

-------
 1                   experiments have been published that used multiple levels of Pb under controlled
 2                   conditions. However, none of them treated Pb concentration as a continuous variable,
 3                   i.e., none attempted to analyze results as a concentration-response relationship. In
 4                   addition, given the well-established presence of strong interactions with variables such as
 5                   pH, CEC, OC, or aging, applying exposure-response relationships from those
 6                   experiments to natural conditions with different values of those interacting variables
 7                   could be difficult.

 8                   Dose-dependent responses in antioxidant enzymes were observed in adult L. mauritii
 9                   earthworms exposed to soil-associated Pb contamination (75, 150, 300 mg Pb/kg) (Maitv
10                   et al.. 2008). By day seven of exposure, glutathione-S-transferase activity and glutathione
11                   disulfide concentration were positively correlated with increasing Pb exposures, while
12                   glutathione concentration exhibited a negative dose-response relationship with soil Pb
13                   concentration. However, these trends had become insignificant by the end of the total
14                   exposure period (28 days), as a result of normalization of antioxidant systems following
15                   chronic exposure. This strongly suggests that changes to earthworm antioxidant activity
16                   are an adaptive response to Pb exposures.

17                   Both survival and reproductive success of E. fetida earthworms showed concentration-
18                   dependent relationships with  soil Pb concentration during the course of standard 14- and
19                   56-day toxicity tests (Jones et al.. 2009b). Five levels of Pb soil concentration were
20                   prepared for the acute 14-day study via spiking with Pb  nitrate—0, 300, 711, 1,687, and
21                   2,249 mg Pb/kg, while soil concentration of 0, 355, 593, 989, and  1,650 mg Pb/kg were
22                   used in chronic (56-day) earthworm bioassays. A 14-day acute LC50 of 2,490 mg Pb/kg
23                   was determined from the dose-response relationship, while the approximate 56-day
24                   NOEC (no observed effect concentration) and EC50 values were about 400 mg Pb/kg and
25                   1,000 mg Pb/kg, respectively. Jones et al. (2009b)made  use of continuous (regressional)
26                   models to characterize the relationship between Pb soil concentration and Pb
27                   accumulation in earthworms, but did not use continuous models for the relationship of
28                   exposure and other responses. Currie et al. (2005) observed mortality of E. fetida after 7
29                   and 14 days in spiked field soil at seven levels of Pb (0 to 10,000 mg Pb/kg).  They
30                   reported LC50  values of 2,662 mg Pb/kg at 7 days and 2,589 mg Pb/kg at 14 days or
31                   2,827 mg Pb/kg at both 7 and 14 days, depending on the number of worms in the
32                   experimental enclosure.

33                   Other studies have shown no  correlation between Pb concentration in either earthworm
34                   tissue or soil, and earthworm  survival rate. Although the Pb content  of E. fetida held in
35                   metal-contaminated soils containing between 9.7 and 8,600 mg Pb/kg was positively
36                   correlated with Pb concentration of fully aged soil collected from disused mines, there
37                   was no statistical relationship with earthworm survival during a 42-day exposure period
      November 2012                             7-58                  Draft - Do Not Cite or Quote

-------
 1                   (Nahmani et al., 2007). However, Pb concentration in soil leachate solution was
 2                   significantly correlated with decreased earthworm survival and growth (linear regression:
 3                   R2= 0.64, p<0.0001). The 42-day Pb EC50 for E. fetida growth was 6,670 mg Pb/kg.

 4                   Langdon et al. (2005) exposed three earthworm species (E. andrei, L. rubellus, and
 5                   A. caliginosd) to Pb nitrate-amended soils at concentrations of 1,000 to  10,000 mg Pb/kg
 6                   to determine species variability in uptake and sensitivity. Twenty-eight-day LC50 values
 7                   for the three species were 5,824 mg Pb/kg, 2,867 mg Pb/kg, and 2,747 mg Pb/kg,
 8                   respectively, indicating thatZ. rubellus and A. caliginosa are significantly more
 9                   vulnerable to Pb contamination than E. andrei, a common laboratory species. This is
10                   comparable to previous findings by Spurgeon et al. (1994) who reported 14-day LC50 of
11                   4,480 mg Pb/kg and 50-day LC50 of 3,760 mg Pb/kg for E. fetida, another standard
12                   laboratory test species. In the more recent study of E. fetida sensitivity summarized
13                   above, Jones (2009b)  reported LC50 values for E. fetida that are similar to those for
14                   L. rubellus and A. caliginosa. It is likely that these apparent species differences are a
15                   result of differential bioavailability of the Pb in test soils. However, the Pb body burden
16                   of all three species in the study by Langdon et al. (2005) increased with increasing
17                   environmental concentration, and there were no species differences in Pb tissue content.
18                   When given a choice between treated and untreated soils, all worm species exhibited
19                   significant avoidance  of Pb-contaminated soils, and altering pH (and, consequently, Pb
20                   bioavailability) had no impact on avoidance (Langdon et al.. 2005). Field earthworms
21                   may thus be able to reduce their exposure to Pb through behavior.

22                   Reproductive success of other soil invertebrates is impacted by Pb. The organismal and
23                   population-level responses of the springtail Paronychiurus kimi to Pb were determined by
24                   Son et al. (2007) using artificial soils, following the 1999 ISO methodology. The 7-day
25                   Pb LC50 was determined to be 1,322 mg  Pb/kg dry weight, while the 28-day reproduction
26                   EC50 was established as 428 mg Pb/kg. The intrinsic rate of population increase was
27                   lower at a Pb soil concentration of  1,312 mg Pb/kg, and the authors estimated that, at this
28                   level, P. kimi populations would be extirpated. The authors noted that, in this case, the
29                   reproductive endpoint overestimated the population-level risk for P. kimi springtails
30                   exposed to Pb, and proposed that more specific measures of population-level endpoints
31                   (such as the reduction in intrinsic rate of increase)  be used to determine risk to
32                   populations. Menta et al. (2006) showed that a nominal soil concentration of 1,000 mg
33                   Pb/kg decreased the reproductive output of two collembolans, Sinella coeca and
34                   F. Candida. Pb concentration of 50, 100, and 500 mg Pb/kg slightly but significantly
35                   depressed S. coeca adult survival, while F.  Candida survival was  statistically unaffected
36                   by Pb exposure. The hatching success ofF. Candida eggs was diminished by 10-day
37                   exposure to Pb-spiked soils; the 10-day EC50 for hatching success was reported as
      November 2012                             7-59                  Draft - Do Not Cite or Quote

-------
 1                  2,361 mg/kg Pb (Xu et al.. 2009b). However, authors noted that egg development was
 2                  more sensitive to Cu and Zn exposure, and by comparison, was less susceptible to Pb.

 3                  In addition to species variability, physical and chemical factors affecting Pb
 4                  bioavailability were also demonstrated to significantly influence the toxicity of Pb to
 5                  terrestrial species. As noted previously in Section 7.3.2. laboratory-amended artificial
 6                  soils provide a poor model for predicting the toxicity of Pb-contaminated field soils,
 7                  because aging and leaching processes, along with variations in physiochemical properties
 8                  (pH, CEC, OM), influence metal bioavailability. Consequently, toxicity values derived
 9                  from exposure-response experimentation with laboratory-spiked soils probably
10                  overestimate true environmental risk, with the possible exception of highly acidic sandy
11                  soils. Because toxicity is influenced by bioavailability of soil biogeological and chemical
12                  characteristics,  extrapolation of toxic concentrations between different field-collected
13                  soils will be difficult. Models that account for those modifiers of bioavailability, such as
14                  the terrestrial BLM proposed by Smolders et al. (2009). have proven difficult to develop
15                  due to active physiological properties of soil organisms affecting either uptake (such as
16                  root phytochelatins) or sequestration of Pb (such as granule formation in root tissues and
17                  earthworms, or substitution of Pb for calcium in bones).
      7.3.6  Terrestrial Community and Ecosystem Effects

18                  A study reviewed in the 1977 Pb AQCD provided evidence for Pb effects on forest-
19                  nutrient cycling and shifts in community composition. Reduced arthropod density,
20                  biomass and richness were observed in the vicinity of a smelting complex in southeastern
21                  Missouri where Pb, Cd, Zn and Cu were measured in the litter layer and soil (U.S. EPA.
22                  1977; Watson et al.. 1976). In the 1986 Pb AQCD it was reported that Pb at
23                  environmental concentrations occasionally found near roadsides and smelters (10,000 to
24                  40,000 mg Pb/kg dry weight) can eliminate populations of bacteria and fungi on leaf
25                  surfaces and in soil. At soil concentrations of 500 to 1,000 mg Pb/kg or higher,
26                  populations of plants, microorganisms, and invertebrates may shift toward Pb-tolerant
27                  populations of the same or different species (U.S. EPA. 1986b).

28                  According to the 2006 Pb AQCD (U.S. EPA. 2006b). natural terrestrial ecosystems near
29                  significant Pb stationary sources (such as smelters and mines) exhibited a number of
30                  ecosystem-level effects, including decreased species diversity, changes in floral and
31                  faunal community composition, and  decreasing vigor of terrestrial vegetation. These
32                  findings are summarized in Table AX7-2.5.2 of the Annex to the 2006 Pb AQCD (U.S.
33                  EPA. 2006c). More recent literature explored the interconnected effects  of Pb
      November 2012                             7-60                   Draft - Do Not Cite or Quote

-------
 1                  contamination on soil bacterial and fungal community structure, earthworms, and plant
 2                  growth, in addition to impacts on soil microbial community function.

 3                  Inoculation of maize plants with Glomus intraradices arbuscular mycorrhizal fungi
 4                  isolates decreased Pb uptake from soil, resulting in lower shoot Pb concentration and
 5                  increased plant growth and biomass (Sudova and Vosatka. 2007). Similarly, Wong et al.
 6                  (2007) showed that the presence of arbuscular mycorrhizal fungi improved vetiver grass
 7                  (Vetiveria zizanioides) growth, and while Pb uptake was stimulated at low soil
 8                  concentration (10 mg Pb/kg), it was depressed at higher concentration (100 and 1,000 mg
 9                  Pb/kg). Bojarczuk and Kieliszewska-Rokicka (2010) found that the abundance of
10                  ectomycorrhizal fungi was negatively correlated with the concentration of metals,
11                  including Pb, in the  leaves of silver birch seedlings. Arbuscular mycorrhizal fungi may
12                  thus protect plants growing in Pb-contaminated soils. Microbes too may dampen Pb
13                  uptake and ameliorate its deleterious effects: biomass of plants grown in metal-
14                  contaminated soils (average Pb concentration 24,175 mg Pb/kg dry weight) increased
15                  with increasing soil  microbial biomass and enzymatic activity (Epelde et al.. 2010).
16                  However, above certain Pb concentration, toxic effects on both plants and microbial
17                  communities may prevent these ameliorating effects. R.Y. Yang et al. (2008b) found that
18                  both the mycorrhizal colonization and the growth ofSolidago canadensis were negatively
19                  affected by soil Pb contamination. They suggested that, more generally, Pb-mediated
20                  alterations in plant-fungal dynamics may be the cause of ecological instability in
21                  terrestrial vegetative communities exposed to metals.

22                  The presence of both earthworms and arbuscular mycorrhizal fungi decreased the
23                  mobility of Pb in mining soils undergoing phytoremediation (Maet al.. 2006).
24                  Inoculation with both earthworms and fungi increased plant growth at sites contaminated
25                  with mine tailings compared to that observed at sites with 75% less Pb contamination.
26                  Most likely, this was a result of the decrease in bioavailable (DTPA-extractable and
27                  ammonium acetate-extractable) Pb to 17% to 25% of levels in areas without the
28                  earthworm and arbuscular mycorrhizal fungi amendments. The presence of earthworms
29                  in metal-contaminated soils decreased the amount of water-soluble Pb (Sizmur and
30                  Hodson. 2008). but despite this decrease,  ryegrass accumulated more Pb from
31                  earthworm-worked soils than soils without worms present. Sizmur and Hodson
32                  speculated that increased root dry biomass may explain the increased uptake of Pb in the
33                  presence of earthworms. However, Sizmur et al. (2011) found that the presence of anecic
34                  (deep-burrowing) earthworms (L. terrestris) increased soil leachate Pb concentrations by
35                  190%. The authors observed that worms promoted a faster breakdown of organic matter,
36                  which caused a decrease in soil pH and a concurrent increase in Pb solubility. As a result,
37                  ryegrass (L. perenne) accumulated a greater amount of Pb in systems with earthworms
38                  (Sizmur et al.. 2011). Further, the presence of earthworms (Lumbricus terrestris) was
      November 2012                             7-61                  Draft - Do Not Cite or Quote

-------
 1                   found to increase Pb concentrations in both maize and barley, although growth of these
 2                   species was unaffected (Ruiz et al.. 2011). Authors noted that worm activity increased Pb
 3                   extraction yields by factors of 4.4 and 7.6, for barley and maize. By contrast,
 4                   Coeurdassier et al. (2007) found that Pb was higher in earthworm tissue when snails were
 5                   present, but that snails did not have a higher Pb content when earthworms were present.

 6                   Microbial communities of industrial soils containing Pb concentrations of 61, 456, 849,
 7                   1,086, and 1,267 mg Pb/kg dry weight were also improved via revegetation with native
 8                   plants, as indicated by increased abundances of fungi, actinomycetes, gram-negative
 9                   bacteria, and protozoa, as well as by enhanced fatty acid concentration (Zhang et al..
10                   2006). Increased plant diversity ameliorated the effects of soil Pb contamination (300 and
11                   600 mg Pb/kg) on the soil microbial community (Yang et al.. 2007).

12                   The effect of Pb on microbial community function has been quantified previously using
13                   functional endpoints such as respiration rates, fatty acid production, and soil acid
14                   phosphatase and urease activities, which may provide an estimate of ecological impacts
15                   separate from  microbial diversity and abundance measurements.  Most studies of metal-
16                   induced changes in microbial communities have been conducted using mixtures of
17                   metals. However, Akerblom et al. (2007) tested the effects of six metals (Cr, Zn, Mo, Ni,
18                   Cd, and Pb) individually. All tested metals had a similar effect on the species
19                   composition of the microbial community. Exposure to a high Pb  concentration (52 mg
20                   Pb/kg) also negatively affected respiration rates. Total phospholipid fatty acid content
21                   was determined to negatively correlate with increasing Pb exposure, indicating alteration
22                   of the microbial community. When Yang et al. (2006) compared the microbial properties
23                   of metal-contaminated urban soils to those of rural soils, significant differences were
24                   detected in basal community respiration rates and microbial abundance. The urban soils
25                   studied contained multiple metal contaminants, but microbial biomass was the only
26                   measured endpoint to be significantly and negatively correlated to Pb concentration.
27                   Similarly, the  fungal community in a naturally Pb-enriched forest in Norway exhibited
28                   differences in  community composition and abundance when compared with other, low Pb
29                   sites. The number of colony-forming fungal units was diminished by soil Pb, and was
30                   approximately 10 times lower in the highest Pb soil concentration (-4300 mg Pb/kg).
31                   Further, only one fungus species was isolated from both high Pb  and control soils,
32                   indicating highly  divergent communities; species diversity was also reduced by high soil
33                   Pb concentrations (Baath et al.. 2005). These studies suggest that anthropogenic Pb
34                   contamination may affect soil microbial communities, and alter their ecological function.
35                   However, (Khan et al.. 2010c) reported that it is possible for indicators of microbial
36                   activity to recover after an initial period depression. (Khan et al.. 2010c) found that
37                   following a 2-week exposure to three levels of Pb (150, 300, and 500 mg Pb/kg), the
38                   number of culturable bacteria at the highest exposure concentration tested was decreased.
      November 2012                             7-62                  Draft - Do Not Cite or Quote

-------
 1                   Acid phosphatase and urease levels (measures of soil microbial activity) decreased
 2                   significantly, but they had recovered by the ninth week. Another study (Bamborough and
 3                   Cummings. 2009) reported that no changes in bacterial and actinobacterial diversity in
 4                   metallophytic soils containing 909 to 5,280 mg Pb/kg (43 to 147 mg Pb/kg bioavailable
 5                   Pb (as defined by the study authors)). Soil bacteria community structure and basal
 6                   respiration rates were examined in natural soils with pH values ranging from 3.7 to 6.8
 7                   (Lazzaro et al., 2006). Six soil types of differing pH were treated with Pb nitrate
 8                   concentrations of 0.5, 2, 8, and 32 mM (104, 414,  1,658, and 6,630 mg Pb/L). Basal
 9                   respiration was decreased in two soil types tested at the highest Pb treatment (32 mM,
10                   =6,630 mg Pb/L), and in a third at the two highest Pb treatments (8 and 32 mM, =1658
11                   and 6,630 mg Pb/L). Terminal Restriction Fragment Length Polymorphism analysis
12                   indicated that bacterial community structure was only slightly altered by Pb treatments.
13                   While pH was correlated with the amount of water-soluble Pb, these increases were
14                   apparently not significant enough to affect bacterial communities, because there were no
15                   consistent relationships between soil pH and respiration rate or microbial community
16                   structure at equivalent soil Pb concentration. Pb contamination was also demonstrated to
17                   reduce phenol oxidase activity in several type of soils; concentrations between 5 and 50
18                   nM Pb (0.001 and 0.01 mg Pb/L) significantly decreased phenol oxidase activity in all
19                   soils tested, while 400 nM (0.08 mg Pb/L) and greater completely arrested phenol
20                   oxidase activity in one soil tested (a high pH sandy loam) (Carine et al., 2009). Carine et
21                   al. (2009) suggested that the decreased soil enzymatic activity resulted from changes in
22                   the microbial  community following Pb exposure. Pb concentrations between 50 and
23                   500 mg Pb/kg significantly reduced microbial abundance and diversity, and also resulted
24                   in lower soil phosphatase, urease, and dehydrogenase activities  (Gao etal.. 201 Ob).
25                   Further, the weekly soil carbon dioxide evolution rate was significantly reduced by
26                   concentrations of 5,  10, and 50 mg Pb/g, which also indicated decreased microbial
27                   respiration and adverse effects on the microbial community (Nwachukwu and Pulford.
28                   2011). Gai et al. (2011) examined the microbial activity of three soils via
29                   microcalorimetric methods following Pb exposure. They noted an increase in activity
30                   immediately following Pb application (giving 10, 20, 40, 80, and 160 mg Pb/kg), and
31                   theorized that this was a result of rapid mortality of sensitive microbial species, followed
32                   by a concurrent proliferation of Pb-tolerant microorganisms. As Pb concentrations
33                   increased, however, the calculated  microbial growth rate constant decreased, indicating a
34                   suppression of microbial activity (Gai et al.. 2011). Authors  also noted a strong
35                   correlation between microcalorimetry estimates and the number of colony forming units
36                   isolated from  soil samples.

37                   Pb exposure negatively affected the prey capture ability of certain fungal species.
38                   Nematophagous fungi are important predators of soil-dwelling nematodes, collecting
39                   their prey with sticky nets, branches, and rings. The densities of traps they constructed

      November 2012                              7-63                  Draft - Do Not Cite or Quote

-------
 1                   decreased in soils treated with 0.15 mM Pb chloride (31 mg Pb/L) (Mo et al., 2008). This
 2                   suppression caused a subsequent reduction in fungal nematode capturing capacity, and
 3                   could result in increased nematode abundance.

 4                   In a study of microbial communities and enzyme activity, Vaisvalavicius et al. (2006)
 5                   observed that high concentration of soil metals were linked to a significant reduction in
 6                   soil microorganism abundance and diversity. Soil columns spiked with Cu, Zn, and
 7                   Pb acetate (total Pb concentration of 278 to 838 mg Pb/kg, depending on depth) exhibited
 8                   a 10- to 100-fold decrease in microbial abundance, with specific microbe classes
 9                   (e.g., actinomycetes) seemingly more affected than others (Vaisvalavicius et al.. 2006).
10                   Concurrently, decreases in soil enzymatic activity were also observed, with saccharase
11                   activity decreased by 57-77%, dehydrogenase activity by 95-98%, and urease activity
12                   65-97%. Although this suggests that Pb contamination may alter the nutrient cycling
13                   capacity of affected soil communities, it is difficult to separate the impact of Pb in this
14                   study from the contributions of Cu and Zn that were also added. In contrast, Zeng et al.
15                   (2007) reported that soil concentrations of 300 mg Pb/kg and less stimulated soil
16                   enzymatic activity. Both urease and dehydrogenase levels were increased and rice dry
17                   weight was unaffected by concentrations of 100  and 300 mg Pb/kg. However, at 500 mg
18                   Pb/kg, both rice  and soil enzyme activities and microbial biomass were decreased
19                   suggesting impacts at the community level for the soil-rice system. The authors proposed
20                   that these concentrations could be considered the critical Pb concentration in  rice paddy
21                   systems (Zeng et al.. 2007).

22                   The microbial communities of soils collected from a Pb-Zn mine and a Pb-Zn smelter
23                   were significantly affected by Pb and other metals (e.g., Cd) (Hu et al.. 2007b). At a mine
24                   site, Pb concentration of 57 to 204 mg Pb/kg and Cd concentration of 2.4 to 227 mg
25                   Cd/kg decreased the number of bacteria-forming colonies extracted from soils. Principal
26                   component analysis of microbial community structure demonstrated that different
27                   communities were associated with different metal soil concentration. Similarly, soil
28                   microbial communities exposed to metal contamination from a smelter site (soil Pb
29                   concentration ranging from 30 to 25,583 mg Pb/kg dry weight) showed decreased
30                   bacterial functional diversity (although fungal functional diversity increased) and  no
31                   effects on soil respiration rates were observed (Stefanowicz et al.. 2008). This led the
32                   authors to conclude that bacterial diversity is a more sensitive endpoint and a better
33                   indicator of metal exposure than fungal diversity or microorganism activity. In a similar
34                   study, Kools et al.  (2009) showed that soil ecosystem variables measured after a 6-month
35                   exposure to metal-contaminated soil indicated that Pb concentration (536 or 745 mg
36                   Pb/kg) was an important driver of soil microbial species biomass and diversity.
      November 2012                             7-64                  Draft - Do Not Cite or Quote

-------
 1                   Pb-resistant bacterial and fungal communities were extracted regularly from soil samples
 2                   at a shooting range site in southern Finland (Hui et al.. 2009). While bioavailable Pb
 3                   concentration averaged 100 to 200 mg Pb/kg as determined by water extraction, the total
 4                   Pb concentrations measured on  site were 30,000 to 40,000 mg Pb/kg. To determine Pb
 5                   tolerance, bacterial colonies extracted and cultured from shooting range and control soils
 6                   were grown on media containing either 0.4 or 1.8 mM Pb (83 or 373 mg Pb/L). While
 7                   bacteria isolated from control soil did not proliferate on high-Pb media, shooting-range
 8                   soil microbe isolates grew on high-Pb media and were deemed Pb tolerant. The authors
 9                   noted that bacterial species common in control samples were not detected among the
10                   Pb-tolerant species isolated from shooting-range soils. They speculated that if long-term
11                   exposure to minimally bioavailable Pb can alter the structure of soil decomposer
12                   communities, decomposition rates could be altered. However, this would require that the
13                   microbial ecosystem decomposing function be altered along with structure, and the
14                   authors provided no evidence for alteration of function.

15                   Microbial communities associated with habitats other than soils are also affected by
16                   exposure to atmospherically deposited Pb. Alder (Alnus nepalensis) leaf microorganism
17                   populations were greater in number at non-affected sites than at sites adjacent to a major
18                   Indian highway with increased Pb pollution (Joshi. 2008). The density, species richness,
19                   and biomass of testate amoebae communities grown on Sphagnum fallax mosses were
20                   significantly decreased following moss incubation in Pb solutions of either 0.6 or 2.5 mg
21                   Pb/L (Nguyen-Viet et al.. 2008). More importantly, species richness and density were
22                   negatively correlated with Pb concentration accumulated within the moss tissue. The
23                   structure of microbial communities associated with lichen surfaces was affected by lichen
24                   trace-element accumulation, including Pb content. Lichens collected from industrial areas
25                   had elevated Pb concentration (10 to 20 mg Pb/kg versus 5 to 7 mg Pb/kg in urban and
26                   rural areas, respectively) and housed bacterial communities characterized by increased
27                   cyanobacteria biomass (Meyer et al..  2010).

28                   Following a 28-day exposure to field-collected soils contaminated with metals (including
29                   Pb at 426 mg Pb/kg), both population growth and individual growth of the earthworm
30                   L. rubellus were diminished (Klok et al.. 2006). The authors proposed that, although
31                   these reductions were unlikely to result in extirpation, avian predators such as the godwit
32                   (Limosa limosa) that feed heavily on  earthworms may be affected by a reduction of
33                   available earthworm biomass.

34                   During the past 5 years, there has been increasing interest in the effects of Pb and other
35                   metals on the functional aspects of soil microbial communities. Most studies show that
36                   Pb decreases diversity and function of soil microorganisms. However, in an example of
37                   ecological mutualism, plant-associated arbuscular mycorrhizal fungi were found to
      November 2012                             7-65                  Draft - Do Not Cite or Quote

-------
 1                  protect the host plant from Pb uptake, while fungal viability is protected by the host
 2                  plants. Similarly, soil microbial communities (bacterial species as well as fungi) in
 3                  Pb-contaminated soils are improved by revegetation. A few studies have  reported on
 4                  effects of Pb to populations of soil invertebrates. They demonstrated that Pb can decrease
 5                  earthworm population density, although not to levels that would result in local extinction.
 6                  There have been no recently reported studies on the potential effects of Pb on terrestrial
 7                  vertebrate populations or communities, or possible indirect effects through reduction of
 8                  prey items such as earthworms.
      7.3.7  Critical Loads in Terrestrial Systems

 9                  The general concept and definition of critical loads is introduced in Section 7.1.3 of this
10                  chapter [also see Section 7.4 of the 2006 Pb AQCD (U.S. EPA. 2006c)1. An international
11                  workshop was conducted in 2005 on the development of critical loads for metals and
12                  other trace elements (Lofts et al., 2007). Among the findings of the workshop it was
13                  reported that soil transport and transformation processes are key in controlling the fate of
14                  metals and trace elements, thus their importance in the input-output mass balance needs
15                  to be considered. The degree to which these processes are understood and can be
16                  quantified varies. Complexation, sorption, ion exchange and precipitation are well
17                  understood under laboratory conditions, but to a lesser extent in the field (Lofts et al..
18                  2007). Slower processes of weathering and fixation are less well understood or studied
19                  than leaching (Lofts et al., 2007).

20                  As noted in previous sections, soil pH and organic matter influence Pb availability. De
21                  Vries et al. (2007) demonstrate that critical limits, measured as critical  reactive metal
22                  content, can significantly vary between soil types that differ in pH and  organic matter.
23                  Critical limits of Pb increased from 30 to 64 (mg Pb/kg) over a pH range of 4-7 when soil
24                  organic matter content was 5%, while these limits increased from 187 to 400 (mg Pb/kg)
25                  over the same pH range when organic content was 80%. These implications suggest that
26                  critical limits increase with increasing soil organic matter. This has important
27                  consequences for forest soils because many are covered by an organic layer where roots,
28                  fungi and other microorganisms are located. Baath (1989) evaluated the effects of organic
29                  matter on critical limits for microorganisms, measured via enzyme synthesis, litter
30                  decomposition and soil respiration. Results indicate critical limits are up to four times
31                  higher in the organic (135 to 976 mg Pb/kg) than the mineral soil layer (32 to 690 mg
32                  Pb/kg) at hazardous concentration ranging from 5-50% of species. In general, De Vries et
33                  al.  (2007) found support that ecotoxicological critical limits in European soils for Pb
34                  decrease with increasing pH.
      November 2012                             7-66                  Draft - Do Not Cite or Quote

-------
 1                   Several methods are routinely used for Pb risk assessment of terrestrial animals. Buekers
 2                   et al. (2009) proposed the use of a Tissue Residue Approach as a risk estimation method
 3                   for terrestrial vertebrates that eliminates the need for quantitative estimation of food
 4                   intake or Pb species bioavailability. Blood Pb no observed effect concentration (NOEC)
 5                   and lowest observed effects concentration (LOEC) data derived from 25 studies
 6                   examining the effects of Pb exposure on growth, reproduction, and hematological
 7                   endpoints were used to construct a series of species sensitivity distributions for mammals
 8                   and birds. They also used the HC5 criterion (5th percentile of species NOEC values for
 9                   collection of species) proposed by Aldenberg and Slob (1993). For mammals, the HC5
10                   values obtained ranged from 11 to 18 (ig Pb/dL  blood; HC5 values for birds ranged from
11                   65 to 71 (ig Pb/dL. The authors proposed the use of 18 and 71 (ig Pb/dL as critical
12                   threshold values for mammals and birds respectively, which are below the lowest NOEC
13                   for both data sets used, and are above typical background Pb values. It is difficult to
14                   determine environmental Pb toxicity given the variation of physiochemical  and soil
15                   properties that alter bioavailability and toxicity.  This variability makes it difficult to
16                   extrapolate between areas.  Furman et al. (2006)  proposed the use of a physiologically
17                   based extraction test to predict risks posed to waterfowl from environmental Pb
18                   contamination. The extraction process was modeled after gastric and intestinal conditions
19                   of waterfowl, and was used to gauge the bioavailability of Pb from freshly amended and
20                   aged contaminated soils. The concentration of Pb extracted through the use  of the
21                   physiologically based extraction test was demonstrated to be significantly correlated to
22                   Pb tissue concentration in waterfowl exposed via in vivo studies of the same soils.

23                   There are few critical loads for Pb reported for terrestrial ecosystems in the  U.S.;
24                   however, work has been conducted in Europe. Given that local conditions (including
25                   historic loading, soil transport and transformation processes) are key elements to critical
26                   load calculation the utility of critical loads that are developed from other countries for
27                   application to U.S. ecosystems is unclear. The most recent European publications on Pb
28                   critical loads include assessments of the U.K., Netherlands and Italy. Hall et al. (2006)
29                   used the critical load approach to conduct a national risk assessment of atmospheric Pb
30                   deposition for the U.K. While specific regions were determined to have low critical load
31                   values for Pb (central England, the Pennines, and southern Wales), the authors noted that
32                   this approach can be significantly biased, as available ecotoxicological data used in the
33                   modeling were from studies that were not conducted in  soils representative  of all U.K.
34                   soils. De Vries et al. (2009) similarly observed that the uncertainty inherent in a critical
35                   load approach to Pb risk assessment is influenced by the critical concentration of
36                   dissolved metal and the absorption coefficients of exposed soils. However, this approach
37                   did indicate that for forest soils in the Netherlands, 29% of the areas would  be expected
38                   to exceed the critical load, based on currently available toxicity data and Pb pollution
39                   data (de Vries and Groenenberg, 2009).  Similarly, although Pb soil concentrations in the

      November 2012                             7-67                  Draft - Do Not Cite or Quote

-------
 1                   Bologna Province of Italy were far below concentrations harmful to soil organisms,
 2                   current atmospheric Pb deposition rates suggest that critical load exceedances are likely
 3                   in the future, unless annual Pb emissions are decreased (Morselli et al., 2006).

 4                   Given the heterogeneity of ecosystems affected by Pb, and the differences in expectations
 5                   for ecosystem services attached to different land uses, it is expected that there will be a
 6                   range of critical load values for Pb for soils within the U.S. In the short term, metal
 7                   emissions generally have greater effects on biota in freshwater systems than in terrestrial
 8                   systems because metals are more readily immobilized in soils than in sediment. However,
 9                   over the longer term, terrestrial systems may be more affected particularly by those
10                   metals with a long soil residence time, such as Pb.
      7.3.8  Soil Screening Levels

11                   Developed by EPA, ecological soil screening levels (Eco-SSLs) are maximum
12                   contaminant concentrations in soils that are predicted to result in little or no quantifiable
13                   effect on terrestrial receptors. These conservative values were developed so that
14                   contaminants that could potentially present an unacceptable hazard to terrestrial
15                   ecological receptors are reviewed during the risk evaluation process while removing from
16                   consideration those that are highly unlikely to cause significant effects. The studies
17                   considered for the Eco-SSLs for Pb and detailed consideration of the criteria for
18                   developing the Eco-SSLs are provided in the 2006 Pb AQCD (U.S. EPA. 2006c).
19                   Preference is given to studies using the most bioavailable form of Pb, to derive
20                   conservative values. Soil concentration protective of avian and mammalian diets are
21                   calculated by first converting dietary concentration to dose (mg/kg body weight per day)
22                   for the critical study, then using food (and soil) ingestion rates and conservatively derived
23                   uptake factors to calculate soil concentration that would result in unacceptable dietary
24                   doses. This frequently results in Eco-SSL values below the average background soil
25                   concentration [19 mg Pb/kg dry weight (U.S. EPA. 2005b. 2003b)1. as is the case with Pb
26                   for birds. The Pb Eco-SSL  was completed in March 2005 and has not been updated since.
27                   Values for terrestrial birds, mammals, plants, and soil invertebrates are 11, 56, 120, and
28                   1,700 mg Pb/kg soil (dry weight), respectively.
      November 2012                             7-68                   Draft - Do Not Cite or Quote

-------
      7.3.9  Characterization of Sensitivity and Vulnerability

 1                  Research has long demonstrated that Pb affects survival, reproduction, growth,
 2                  metabolism, and development in a wide range of species. The varying severity of these
 3                  effects depends in part upon species differences in metabolism, sequestration, and
 4                  elimination rates. Dietary factors also influence species sensitivity to Pb. Because of
 5                  effects of soil aging and other bioavailability factors discussed above (Section 7.3.2). in
 6                  combination with differing species assemblages and biological accessibility within prey
 7                  items, ecosystems may also differ in their sensitivity and vulnerability to Pb. The
 8                  2006 Pb AQCD reviewed many of these factors which are updated herein by reference to
 9                  recent literature.
              7.3.9.1  Species Sensitivity

10                  There is wide variation in sensitivity of terrestrial species to Pb exposure, even among
11                  closely related organisms. Langdon et al. (2005) showed a two-fold difference in LC50
12                  values among three common earthworm species, with the standard laboratory species,
13                  E. andrei, being the least sensitive. Mammalian NOEC values expressed as blood Pb
14                  levels were shown to vary by a factor of 8, while avian blood NOECs varied by a factor
15                  of 50 (Buekers et al.. 2009). Age at exposure, in particular, may affect sensitivity to Pb.
16                  For instance, earlier instar C. elegans were more likely than older individuals to exhibit
17                  neurobehavioral toxicity following Pb exposure (Xing et al.. 2009b). and also
18                  demonstrated more pronounced neural degeneration than  older larvae and adults (Xing et
19                  al.. 2009c).
              7.3.9.2  Nutritional Factors

20                  Dietary factors can exert significant influence on the uptake and toxicity of Pb in many
21                  species of birds and mammals. The 2006 Pb AQCD (U.S. EPA. 2006b) describes how
22                  Ca2+, Zn, Fe, vitamin E, Cu, thiamin, P, Mg, fat, protein, minerals, and ascorbic acid
23                  dietary deficiencies increase Pb absorption and its toxicity. For example, vitamin E
24                  content was demonstrated to protect against Pb-induced lipid peroxidation in mallard
25                  ducks. Generally, Pb exposure is more likely to produce behavioral effects in conjunction
26                  with a nutrient-deficient diet. As previously reported in the 2006 Pb AQCD, Ca2+
27                  deficiencies may increase the  susceptibility of different terrestrial species to Pb, including
28                  plant (Antosiewicz. 2005). avian (Dauwe et al.. 2006; Snoeijs et al.. 2005) and
29                  invertebrate species. Antosiewicz (2005) determined that, for plants, Ca2+ deficiency
30                  decreased the sequestration capacity of several species (tomato, mustard, rye, and maize),

      November 2012                             7-69                  Draft - Do Not Cite or Quote

-------
 1                   and that this likely resulted in an increased proportion of Pb at sites of toxic action.
 2                   Because Pb ions can interact with plant Ca2+ channel pores, in the presence of low Ca2+
 3                   and high Pb concentration, a higher proportion of Pb can interact with these channels and
 4                   be taken up by plants. A similar phenomenon has been observed in invertebrates, where
 5                   the metabolic pathway of metals mimics the metabolic pathway of Ca2+ [Simkiss et al.
 6                   (1982). as cited in Jordaens et al. (2006)1. Hence, in environments with
 7                   disproportionately high Pb versus Ca2+ concentration, accumulation of Pb may be
 8                   accelerated, as in plants. Ca2+ deficiency in birds was demonstrated to stimulate the
 9                   production of Ca2+-binding proteins in the intestinal tract, which extract more Ca2+ from
10                   available diet; however, this response also enhances the uptake and accumulation of Pb
11                   from diet and drinking water [Fullmer (1997). as cited in Dauwe et al. (2006)].
              7.3.9.3 Soil Aging and Site-Specific Bioavailability

12                   Total soil Pb concentration is a poor predictor of hazards to avian or mammalian wildlife,
13                   because site-specific biogeochemical and physical properties (e.g., pH, OM, metal oxide
14                   concentration) can affect the sequestration capacity of soils. Additionally, soil aging
15                   processes have been demonstrated to decrease the bioavailable Pb fraction;  as such,
16                   laboratory toxicity data derived from spiked soils often overestimate the environmental
17                   risk of Pb. Smolders et al. (2009) compared the toxicity of freshly Pb-spiked soils to
18                   experimentally aged spiked soils and field-collected Pb-contaminated soils. Experimental
19                   leaching and aging was demonstrated to increase invertebrate Pb EC50 values by factors
20                   of 0.4 to greater than 8; in approximately half the cases, the proportionality  of toxicity to
21                   Pb content disappeared following experimental aging of freshly spiked soils through
22                   leaching. The leaching-aging factor for Pb was determined to be 4.2,  and represented the
23                   ratio of EDi0 values derived in aged soils to freshly spiked soils (factors greater than one
24                   indicate decreased toxicity in aged field soils relative to laboratory spiked soils).
25                   Consequently, the sensitivity of terrestrial vertebrates to environmental Pb exposures will
26                   be heavily dependent on the relative rate of aging and site-specific bioavailability.
              7.3.9.4 Ecosystem Vulnerability

27                   Relative vulnerability of different terrestrial ecosystems to effects of Pb can be inferred
28                   from the information discussed above on species sensitivity and how soil geochemistry
29                   influences the bioavailability and toxicity of Pb. Soil ecosystems with low pH,
30                   particularly those with sandy soils, are likely to be the most sensitive to the effects of Pb.
31                   Examples of such systems are forest soils, including oak, beech, and conifer forests.
32                   The Pine Barrens in southern New  Jersey (also known as the Pinelands) is an example  of

      November 2012                             7-70                  Draft - Do Not Cite or Quote

-------
 1                   a highly vulnerable ecosystem: it is a dense coniferous (pine) forest with acidic, sandy,
 2                   nutrient poor soil. As agricultural areas are taken out of production and revert to old
 3                   fields and eventually forests, their vulnerability to Pb is likely to increase as a result of
 4                   decreasing OM and acidification of soils (from discontinuation of fertilizing and liming).
 5                   On the other hand, increasing density of native or invasive plants with associated
 6                   arbuscular mycorrhizal fungi will likely act to ameliorate some of the effects of Pb (see
 7                   previous  discussion of studies by Sudova and Vostka (2007) and Wong et al. (2007). It is,
 8                   however, difficult to categorically state that certain plant or soil invertebrate communities
 9                   are more vulnerable to Pb than others, as the available toxicity data have not yet been
10                   standardized for differences in bioavailability (because of use of different Pb salts,
11                   different soil properties, and different lengths of aging of soil prior to testing), nutritional
12                   state, or organism age, or other interacting factors. Data from field studies are
13                   complicated by the co-occurrence of other metals and alterations of pH, such as
14                   acidification from  SO2 in smelter emissions, which are almost universal at sites of high
15                   Pb exposure, especially at mine or smelter sites. However, because plants primarily
16                   sequester Pb in the roots, uptake by soil invertebrates is the  most likely pathway for Pb
17                   exposure of higher trophic level organisms. Invertivores are likely at higher risk than
18                   herbivores. In fact, estimations of Pb risk at a former Pb smelter in northern France
19                   indicated that area Pb concentration presented the greatest threat to insectivorous bird and
20                   mammal species, but only minimal risk to soil invertebrate and herbivorous mammals
21                   (Fritsch et al.. 2010). By extension, birds and mammals in ecosystems with a richer
22                   biodiversity of soil invertebrates may be more vulnerable to Pb than those in ecosystems
23                   with fewer invertebrates (e.g., arid locations). Regardless, the primary determinant of
24                   terrestrial ecosystem vulnerability is soil geochemistry, notably pH, CEC, and amount of
25                   OM.
      November 2012                              7-71                   Draft - Do Not Cite or Quote

-------
      7.3.10 Ecosystem Services Associated with Terrestrial Systems

 1                  Pb deposited on the surface of, or taken up by organisms has the potential to alter the
 2                  services provided by terrestrial biota to humans. There are no publications at this time
 3                  that specifically focus on the ecosystem services affected by Pb in terrestrial systems and
 4                  the directionality of impacts is not always clear. For example, terrestrial soils provide a
 5                  service to aquatic ecosystems by sequestering Pb through sorption and precipitation. At
 6                  the same time, the sequestration of Pb by soils may result in a degredation in the quality
 7                  of soil and may result in decreased crop productivity. The evidence reviewed in the
 8                  present document illustrates that Pb can cause ecological effects in each of the four main
 9                  categories of ecosystem services (Section 7.1.2) as defined by Hassan et al. (2005). These
10                  effects are sorted into ecosystem services categories and summarized here:

11                     • Supporting: altered nutrient cycling, decreased biodiversity, decline of
12                       productivity, food production for higher trophic levels
13                     • Provisioning: plant yields
14                     • Regulating: decline in soil quality, detritus production
15                     • Cultural: ecotourism and cultural heritage values related to ecosystem integrity
16                       and biodiversity, impacts to terrestrial vertebrates.
17                  A few studies since the 2006 Pb AQCD, consider the impact of metals in general on
18                  ecosystem services. Honeybees are important for provisioning services such as
19                  pollination and production of honey. They can be exposed to atmospheric Pb by direct
20                  deposition or through Pb associated with plants, water or soil. In a study of heavy metals
21                  in honeybees in central Italy, there was a statistically significant difference in Pb between
22                  bees collected in wildlife reserves compared to bees collected in urban areas with the
23                  highest concentration of Pb detected from bees caught in hives near an airport (Perugini
24                  et al.. 2011). In a review of the effects of metals on insect behavior, ecosystem services
25                  provided by insects such as detritus reduction and food production for higher trophic
26                  levels were evaluated by considering  changes in ingestion behavior and taxis (Mogren
27                  and Trumble. 2010). Pb was shown in a limited number of studies to affect ingestion by
28                  insects. Crickets (Chorthippus spp) in heavily contaminated sites reduced their
29                  consumption of leaves in the presence of increasing Cd and Pb concentrations (Migula
30                  and Binkowska. 1993). Decreased feeding activity in larval and adult Colorado potato
31                  beetle (Leptinotarsa decemlineata) were observed as a result of dietary exposures of Pb
32                  and Cu (Kwartirnikov et al.. 1999). while no  effects were found in ingestion studies of Pb
33                  with willow leaf beetle, Lochmaea caprae (Rokvtova et al.. 2004) mottled water hyacinth
      November 2012                             7-72                  Draft - Do Not Cite or Quote

-------
 1                  weevil, Neochetina eichhorniae (Kay and Haller. 1986) and hairy springtail, Orchesella
 1                  cincta (Van Capelleveen et al.. 1986J.

 3                  Soil health for agricultural production and other soil-associated ecosystem services is
 4                  dependent upon the maintenance of four major functions: carbon transformations,
 5                  nutrient cycles, soil structure maintenance, and the regulation of diseases and pests and
 6                  these parameters may be altered by metal deposition (Kibblewhite et al., 2008). Pb
 7                  impacts to terrestrial  systems reviewed in the previous sections provide evidence for
 8                  impacts to supporting, provisioning, and regulating ecosystem services provided by soils.
 9                  For example, earthworms were shown to impact soil metal mobility and availability,
10                  which in turn resulted in changes to microbial populations (biodiversity), pH, dissolved
11                  organic carbon, and metal speciation (Sizmur and Hodson, 2009). all of which  may
12                  directly affect soil fertility.

13                  Pb is bioaccumulated in plants, invertebrates and vertebrates inhabiting terrestrial and
14                  aquatic systems that receive Pb from atmospheric deposition. This represents a potential
15                  route for Pb mobilization into the food web or into food products. For example, Pb
16                  bioaccumulation in leaves and roots of an edible plant may represent an adverse impact to
17                  the provisioning of food, an essential ecosystem service. Although there is no consistent
18                  evidence of trophic magnification there is substantial evidence of trophic transfer. It is
19                  through consumption of Pb-exposed prey or Pb-contaminated food that atmospherically
20                  deposited Pb reaches species that may have very little direct exposure to it.

21                  There is limited evidence of Pb impacts to plant productivity. Productivity of gray birch
22                  (Betula populifolid) was impaired in soils with elevated As, Cr, Pb, Zn and V (Gallagher
23                  et al.. 2008). Tree growth measured in both individuals and at the assemblage level using
24                  satellite imagery and field spectrometry was significantly decreased with increasing metal
25                  load in soil.
      7.3.11 Synthesis of New Evidence for Pb Effects in Terrestrial Systems

26                  This synthesis of the effects of Pb on terrestrial ecosystems covers information from the
27                  publication of the 2006 Pb AQCD to present. It is followed in Section 7.5 by
28                  determinations of causality that take into account evidence dating back to the  1977 Pb
29                  AQCD.

                    High concentrations of Pb
30                  The state-level mean concentration of Pb in U.S. soils ranges from 5 to 39 mg Pb/kg.
31                  Studies of the effects of Pb use much higher concentrations, whether they use soils that

      November 2012                             7-73                   Draft - Do Not Cite or Quote

-------
 1                  have been exposed to Pb pollution, or experimental amendment with salts of Pb (Table
 2                  7-4). Studies that were conducted in situ or used soil collected from natural environments
 3                  all took place near stationary sources, i.e., in highly contaminated areas. All of them took
 4                  advantage of gradients of exposure produced by distance from the source to create
 5                  several levels of Pb, sometimes with only two sampling locations-control and elevated-
 6                  rather a larger set of levels representative of the whole gradient. In most cases, the
 7                  highest concentration of Pb in the study is very high relative to those found anywhere
 8                  except heavily exposed sites. In amendment experiments, variation in Pb was generated
 9                  by addition of Pb salts to either natural or artificial soils. These experiments often
10                  included concentrations that were even higher than those found in heavily exposed
11                  natural environments. In either type of study, however, effects gradually increased with
12                  increasing exposure, and studies that include very high exposures were thus informative
13                  for lower exposures as well. This would not be true if there was clear evidence of the
14                  presence of discontinuity (breakpoint) in the relationship of exposure and effects, but
15                  without evidence of discontinuity, the presence of effects at elevated exposures implies
16                  effects at lower exposures. Using concentration-response models where concentration is
17                  taken as a continuous variable to analyze data with multiple values of Pb concentration
18                  would allow  better estimation of the size of effects at any value of exposure, including
19                  low ones, and also better estimation of uncertainties around the  size of effects. However,
20                  none of studies with multiple Pb concentrations used a continuous model to characterize
21                  the relationship between concentration and effects.


                    Comparability of effect concentrations

22                  Strong interactions of Pb concentration and  several other soil variables, including pH,
23                  CEC, OC, and Fe/Al oxides have been amply demonstrated with respect to various
24                  biological responses. For example, Dayton et al.(2006) and Bradham et al.(2006) tested
25                  an array of different soils to which the same amount of Pb was added, using lettuce and
26                  earthworms,  respectively. They found differences in biological effects that were as large
27                  as 27, 35, or  even 72-fold between soils.

28                  In studies where Pb was introduced through amendment, those interacting variables can
29                  be changed experimentally in a controlled way, or held constant. In studies where natural
30                  soils were used in which Pb originated from pollution, they are left to vary freely. In
31                  either case, the presence and magnitude of those interactions make calculations of
32                  expected responses under other sets of conditions particularly difficult, as well as
33                  comparisons between studies conducted under different conditions.

34                  In addition, the amount of Pb dissolved in soil pore water determines the impact of soil
35                  Pb on terrestrial ecosystems to a much greater extent than the total amount present. It has
      November 2012                             7-74                  Draft - Do Not Cite or Quote

-------
 1                   long been established that the amount of Pb dissolved in soil solution is controlled by at
 2                   least six variables: (1) solubility equilibria; (2) adsorption-desorption relationship of total
 3                   Pb with inorganic compounds; (3) adsorption-desorption reactions of dissolved Pb phases
 4                   on soil OM; (4) pH; (5) CEC; and (6) aging. Since 2006, further details have been
 5                   contributed to the understanding of the role of pH, CEC, OM, and aging. Smolders et al.
 6                   (2009) demonstrated that the two most important determinants of solubility (and also
 7                   toxicity) in soils are pH and CEC. However, they had previously shown that aging,
 8                   primarily in the form  of initial leaching following deposition, decreases soluble metal
 9                   fraction by approximately one order of magnitude (Smolders et al.. 2007).  Since 2006,
10                   OM has been confirmed as an important influence on Pb sequestration, leading to longer-
11                   term retention in soils with higher OM content, and also creating the potential for later
12                   release of deposited Pb. Aging, both under natural conditions and simulated through
13                   leaching , was shown to substantially decrease bioavailability to plants, microbes, and
14                   vertebrates. However, most studies report some measure of total extracted  Pb, or total
15                   added Pb, rather than pore water or soluble fraction.


                     Plants

16                   Recent studies with herbaceous species growing at various distances from  smelters added
17                   to the  existing strong  evidence that atmospherically transported Pb is taken up by plants.
18                   These studies did not establish the relative proportion that originated from  atmospheric
19                   Pb deposited in the soil, as opposed to that taken up directly from the atmosphere through
20                   the leaves. Studies found that in trees, Pb that is taken up through the roots is then
21                   generally translocated from the roots to other parts. However, multiple recent studies
22                   showed that in trees, the proportion of Pb that is taken up through the leaves is likely to
23                   be very substantial. One study attempted to quantify it, and suggested that  50% of the Pb
24                   contained in Scots Pine in Sweden is taken up directly from the atmosphere
25                   (Section 7.3.3.1). Studies with herbaceous plants  found  that in most species tested, soil
26                   Pb taken up by the roots is not translocated into the stem and leaves, but when growth
27                   and survival were reported, growth of the whole plant decreased with increasing Pb, and
28                   mortality increased (Table  7-4). Experimental studies have added to the existing evidence
29                   of photosynthesis impairment in plants exposed to Pb, and have found damage to
30                   photosystem II due to alteration of chlorophyll structure, as well as decreases in
31                   chlorophyll content in diverse taxa, including lichens and mosses. A substantial amount
32                   of evidence of oxidative stress in response to Pb exposure has also been produced.
33                   Reactive oxygen species were found to increase in broad bean and tomato  plants exposed
34                   to increasing concentrations of soil Pb, and a concomitant increase in superoxide
35                   dismutase, glutathione,  peroxidases, and lipid peroxidation, as well as decreases in
36                   catalase were observed  in the same plants. Monocot, dicot, and bryophytic taxa grown in
      November 2012                             7-75                   Draft - Do Not Cite or Quote

-------
 1                  Pb-contaminated soil or in experimentally spiked soil all responded to increasing
 2                  exposure with increased antioxidant activity. In addition, genotoxicity, decreased
 3                  germination, and pollen sterility were observed in some experiments. All effects were
 4                  small outside of contaminated areas (Section 7.3.4.1).

 5                  Invertebrates

 6                  Since the 2006 Pb AQCD, various species of terrestrial snails have been found to
 7                  accumulate Pb from both diet and soil, although effects on growth, survival and
 8                  reproduction are inconsistent. Recent studies with earthworms have found that both
 9                  internal concentration of Pb and mortality increase with decreasing soil pH and CEC. In
10                  addition, tissue concentration differences have been found in species of earthworms that
11                  burrow in different soil layers. The  rate of accumulation in each of these species could
12                  result from layer differences in interacting factors such as pH and CEC (Section 7.3.3.2).
13                  Because earthworms often sequester Pb in granules, some authors have suggested that
14                  earthworm Pb is not bioavailable to their predators. There is some evidence that
15                  earthworm activity increases Pb availability in soil, but it is inconsistent. In arthropods
16                  collected at contaminated sites, recent studies found gradients in accumulated Pb that
17                  corresponded to  gradients in soil with increasing distance from stationary sources.
18                  Recently published studies have shown neuronal damage in nematodes exposed to low
19                  concentrations of Pb (2.5  uM = 0.5  mg Pb/L), accompanied by behavioral abnormalities.
20                  Reproductive effects were found at lower exposure in younger nematodes, and effects on
21                  longevity and fecundity were shown to persist for several generations. Increased
22                  mortality was found in earthworms, and was strongly dependent on soil characteristics
23                  including pH,  CEC, and aging. Snails exposed to Pb through either topical application or
24                  through consumption of Pb-exposed plants had increased antioxidant activity and
25                  decreased food consumption, but effects on growth and survival were inconsistent.
26                  Effects on arthropods exposed through soil or diet varied with species and exposure
27                  conditions, and included diminished growth and fecundity in springtails, endocrine and
28                  reproductive anomalies, and body deformities. Increasing concentration of Pb in the
29                  exposure medium generally resulted in increased effects within each study, but the
30                  relationship between concentration  and effects varied between studies, even when the
31                  same medium, e.g., soil, was used (Section 7.3.4.2). Evidence suggested that aging and
32                  pH are important modifiers.

33                  Vertebrates

34                  There were few recent studies of Pb bioavailability and uptake in vertebrates since the
35                  2006 Pb AQCD. A study  of two species of sea ducks in Alaska found that 3% of the birds
36                  had tissue levels of Pb that indicated exposure above background. Urban pigeons in
37                  Korea were found to accumulate 1.6 to 1.9 mg Pb/kg wet weight Pb in the lungs, while in
      November 2012                             7-76                   Draft - Do Not Cite or Quote

-------
 1                  Wisconsin 70% of American woodcock chicks and 43% of young-of-year had elevated
 2                  bone Pb (9.6 to 93 mg Pb/kg dry weight in chicks, 1.5 to 220 mg Pb/kg dry weight in
 3                  young-of-year). None of the locations for these studies was in proximity to stationary
 4                  sources of heavy contamination, and none was able to identify the origin of the Pb
 5                  (Section 7.3.3.3).  Effects on amphibians and reptiles included decreased white blood cell
 6                  counts, decreased testis weight, and behavioral anomalies. However, large differences in
 7                  effects were observed at the same concentration of Pb in soil, depending on whether the
 8                  soil was freshly amended, or field-collected from contaminated areas. As in most studies
 9                  where the comparison was made, effects were smaller when  field-collected soils were
10                  used.  A study at the Anaconda Smelter Superfund site found increasing Pb accumulation
11                  in gophers with increasing soil Pb around the location of capture. Effects of dietary
12                  exposure were studied in several mammalian species, and cognitive, endocrine,
13                  immunological, and growth effects were observed. Pigs fed various Pb-contaminated
14                  soils showed that the  form of Pb determined accumulation, and another study showed
15                  lower feed efficiency and weight in pigs with 2,08 versus 1.44 ug Pb/dl in blood,
16                  originating in Pb-sulfate feed supplement. In some birds, maternal elevated blood Pb
17                  level was associated in recent studies with decreased hatching success, smaller clutch
18                  size, high corticosteroid level, and abnormal behavior.  Some species show little or no
19                  effect of elevated  blood Pb level. A study of Japanese quail found that Pb added to the
20                  diet could improve survival and incidence of several pathologies, and a long term study
21                  of pied flycatchers at a mine site produced mixed evidence for the effects of Pb
22                  (Section 7.3.4.3).

23                  Food web

24                  Recent studies were able to measure Pb in the components of various food chains that
25                  included soil, plants, invertebrates, arthropods and vertebrates. They confirmed that
26                  trophic transfer of Pb is pervasive, but no consistent evidence of trophic magnification
27                  was found (Section 7.3.3.4).

28                  Community and Ecosystem Effects

29                  New evidence of effects of Pb at the community and ecosystem levels of biological
30                  organization include several studies of the ameliorative effects of mycorrhizal fungi on
31                  plant growth, attributed to decreased uptake of Pb by plants,  although both mycorrhizal
32                  fungus and plant were negatively affected. The presence of both earthworms and
33                  mycorrhizal fungi decreased solubility and mobility of Pb in soil in one study, but the
34                  presence of earthworms was associated with higher uptake of Pb by plants in another.
35                  The presence of snails increased uptake of Pb by earthworms, but not vice-versa. Most
36                  recently published research on community and ecosystem effects of Pb has focused on
37                  soil microbial communities, which have been shown to be impacted in both composition
      November 2012                             7-77                  Draft - Do Not Cite or Quote

-------
 1                  and activity. Many recent studies have been conducted using mixtures of metals, but have
 2                  tried to separate the effects of individual metals when possible. One study compared the
 3                  effects of 6 metals individually (Akerblom et al.. 2007). and found that their effects on
 4                  community composition were similar. In studies that included only Pb, or where effects
 5                  of Pb could be separated,  soil microbial activity was generally diminished, but in some
 6                  cases recovered overtime. Species and genotype composition were consistently altered,
 7                  and those changes were long-lasting or permanent (Section 7.3.6) .

 8                  Exposure-Response

 9                  Several studies with various organisms have included gradients of Pb exposure. None has
10                  characterized the exposure-relationship using a continuous model of exposure-response.
11                  However, evidence indicates clearly that increased exposure to Pb is associated with
12                  increases in observed effects in terrestrial ecosystems. Evidence also demonstrates that
13                  many factors, including species and various soil physiochemical properties, interact
14                  strongly with Pb concentration to modify those effects. In terrestrial ecosystems, where
15                  soil is generally the main  component of the exposure route, Pb aging is a particularly
16                  important factor, and one  that may be difficult to reproduce experimentally. Without
17                  adequate quantification of those interactions, characterizations of exposure-response
18                  relationships may be difficult to transfer outside of experimental settings.
      7.3.12 Causal Determinations for Pb in Terrestrial Systems

19                  In the following sections, organism-level effects on reproduction and development,
20                  growth and survival are considered first since these endpoints can lead to effects at the
21                  population level or above and are important in ecological risk assessment.
22                  Neurobehavioral effects are considered next followed by sub-organismal responses
23                  (hematological effects, physiological stress) for which Pb has been shown to have an
24                  impact in multiple species and across taxa, including humans. Causal determinations for
25                  terrestrial, freshwater and saltwater ecological effects are summarized in Table 7-3.
              7.3.12.1      Reproductive and Developmental Effects-Terrestrial Biota

26                  In terrestrial invertebrates and vertebrates, evidence assessed for the present document
27                  and in Pb AQCDs indicates an association between reproductive effects and Pb exposure.
28                  Impaired fecundity at the organism level of biological organization can result in a decline
29                  in abundance and/or extirpation of populations, decreased taxa richness, and decreased
30                  relative or absolute abundance at the community level (Suter et al.. 2005; U.S. EPA.
31                  2003a). Evaluation of the literature on Pb effects in terrestrial species indicates that

      November 2012                             7-78                  Draft - Do Not Cite or Quote

-------
 1                  exposure to Pb is associated with reproductive effects. Various endpoints have been
 2                  measured in various taxa of terrestrial organisms to assess the effect of Pb on fecundity,
 3                  development, and hormone homeostasis. Although reproductive effects were
 4                  demonstrated, no single endpoint in a single taxon has been extensively studied. Recent
 5                  evidence available since the 2006 Pb AQCD for effects of Pb on reproductive endpoints
 6                  in terrestrial species is summarized in Table 7-4.

 7                  In terrestrial plants, few studies were available to the 2006 Pb AQCD (U.S. EPA. 2006b).
 8                  and few are available more recently that specifically address reproductive effects of Pb
 9                  exposure. Two genotypes of maize seedlings exhibited a significant and concentration-
10                  dependent reduction in seed germination following 7 days of Pb treatment in nutrient
11                  solution of nominal concentration of 0, 0.007, 0.7 and  7 mg Pb/L as Pb sulfate (Ahmad et
12                  al.. 2011). Germination inhibition and chromosomal abnormalities also increased in a
13                  concentration-dependent manner in Grass pea grown in soil irrigated with solutions
14                  containing nominal concentration of 0 to 188 mg Pb/L (Kumar and Tripathi. 2008).
15                  However, germination increased in a broad sample of soils when amended with 2,000 mg
16                  Pb/kg (Davton et al.. 2006).

17                  In terrestrial invertebrates, Pb can alter developmental timing,  hatching success, sperm
18                  morphology and hormone homeostasis. The number of species studied has been small,
19                  but reproductive effects consistently increase with increasing exposure. The
20                  2006 Pb AQCD reported effects on reproduction in collembolans and earthworms, with
21                  LOECs and NOECs typically well above Pb soil concentrations observed away from
22                  stationary sources of contamination, more recently, an increase in development time
23                  (approximately two days) and a  reduction in relative fecundity were observed in aphids
24                  feeding on plants exposed to high concentrations of Pb (Goriir. 2007). Hatching success
25                  of the collembolan F. Candida was decreased following 10 day nominal exposure to
26                  Pb-spiked soils (EC50 2,361 mg Pb/kg dry soils) (Xu et al., 2009b). Sperm morphology in
27                  Asian earthworms was significantly altered following 2-week exposures to soils
28                  containing nominal concentration of 1,000, 1,400, 1,800 and 2,500 mg Pb/kg soil (Zheng
29                  and Li. 2009). Pb may also disrupt hormonal homeostasis in invertebrates as studies with
30                  moths have suggested (Shu et al.. 2009). Adult female moths reared on diets containing
31                  25, 50, 100, or 200 mg Pb/kg exhibited decreased vitellogenin mRNA induction, and
32                  vitellogenin levels were demonstrated to decrease with increasing Pb exposure. Evidence
33                  of multi-generational toxicity effects of Pb is also present in terrestrial invertebrates,
34                  specifically springtails, mosquitoes, carabid beetles and nematodes where decreased
35                  fecundity in progeny of Pb-exposed individuals was observed. The magnitude of effects
36                  is variable, but they are present in multiple phyla, and increase with increasing exposure
37                  within studies. Reproductive effects in terrestrial invertebrates are also coherent with
38                  similar effects observed in aquatic invertebrates.
      November 2012                             7-79                   Draft - Do Not Cite or Quote

-------
 1                   In terrestrial vertebrates, there is evidence for reproductive effects associated with Pb
 2                   exposure in recent evidence and Pb AQCDs. The 2006 Pb AQCD (U.S. EPA. 2006c)
 3                   concluded that exposure to affects reproduction and development in terrestrial
 4                   vertebrates. Effects reported in that document included declines in clutch size, number of
 5                   young hatched, number of young fledged, decreased fertility, and decreased eggshell
 6                   thickness observed in birds near areas of Pb contamination and in birds with elevated Pb
 7                   tissue concentration regardless of location. More recently, decreased testis weight was
 8                   observed in lizards administered a sublethal dose of 10 or 20 mg Pb/kg day by oral
 9                   gavage for 60 days (Salice et al. 2009). Few studies in the field have addressed
10                   reproductive effects of Pb specifically in mammals, due to most available data in wild or
11                   grazing animals being from near smelters, where animals are co-exposed to  other metals.
12                   For example, the reproductive viability of red deer (C. elaphus) inhabiting a
13                   Pb-contaminated mining area of Spain was shown to be altered, with 11% and 15%
14                   reductions in spermatozoa and acrosome integrity observed in male deer from the mining
15                   area compared with those  residing in reference areas (Reglero et al.. 2009aX but multiple
16                   other metals were present  at high concentrations.  Evidence from AQCDs and the present
17                   document for terrestrial vertebrates is coherent with evidence from freshwater
18                   amphibians, and fish (Section 7.4.12.1). However, experimental evidence obtained using
19                   mammals in the context of human health research demonstrates a consistency of adverse
20                   effects of Pb on sperm  (Section 5.8.4.1) and the onset of puberty in males and females
21                   (Sections 5.8.1.1 and 5.8.1.2) with strong evidence from both toxicology and
22                   epidemiology studies. Other reproductive endpoints including spontaneous abortions,
23                   pre-term birth, embryo development, placental development, low birth weight,
24                   subfecundity, hormonal changes, and teratology were also affected, but less consistently
25                   (Section 5.8).

26                   For reproductive and developmental effects in terrestrial ecosystems, the current body of
27                   evidence is inadequate  to conclude that exposure to Pb is causal in plants, and is
28                   sufficient to conclude that there is a causal relationship in invertebrates and  vertebrates.
              7.3.12.2     Growth Effects-Terrestrial Biota

29                  Alterations in growth at the organism level of biological organization can have impacts at
30                  the population, community and ecosystem levels. In terrestrial ecosystems, evidence for
31                  effects of Pb on growth is strongest in terrestrial plants, although these effects are
32                  typically observed in laboratory studies with high exposure concentrations or in field
33                  studies near stationary  sources. In terrestrial plants, there is evidence over several decades
34                  of research that Pb inhibits photosynthesis and respiration, all of which can reduce the
35                  growth of the plant (U.S. EPA. 2006c. 1986a. 1977).Decreases in chlorophyll a and b


      November 2012                              7-80                  Draft - Do Not Cite or Quote

-------
 1                  content have been observed in various algal and plant species. Many laboratory toxicity
 2                  studies report effects on the growth of plants in synthetic growing media, but observed
 3                  effects typically occur at concentrations higher than the average background
 4                  concentrations in U.S. soils (19 mg Pb/kg dry weight) (U.S. EPA. 2005b) and there are
 5                  few field studies. Effects on plant growth can result in reduced productivity and
 6                  decreased biomass. The 2006 Pb AQCD relied principally on evidence assembled in the
 7                  Ecological Soil Screening Levels for Lead document (U.S. EPA. 2005b), which
 8                  concluded that growth (biomass) was the most sensitive and ecologically relevant
 9                  endpoint for plants.

10                  Evidence for growth effects in terrestrial fauna is sparse. In the 1986 Pb AQCD, a study
11                  was reviewed in which the Fl and F2 generations of the springtail Onychiurus armatus
12                  fed a diet of Pb-exposed fungi (0.008 to 3.1 mg Pb/g) experienced a delay in achieving
13                  maximum length (Bengtsson et al., 1983). The authors suggested that the reduced growth
14                  may be accompanied by delayed sexual maturity. The 2006 Pb AQCD (U.S. EPA.
15                  2006b) reported that growth effects observed in both terrestrial invertebrates and
16                  vertebrates were more pronounced in juvenile organisms, underscoring the importance of
17                  lifestage to overall Pb  susceptibility. Recent evidence available since the 2006 Pb AQCD
18                  for effects of Pb on growth endpoints in terrestrial species is summarized in Table 7-4:
19                  reduced growth of garden snail T. pisana, increasing with increasing exposure, was
20                  observed following a five week dietary exposure to eight nominal concentrations of Pb
21                  (El-Gendy et al., 2011). Studies also show concentration-dependent inhibition of growth
22                  in earthworms raised in Pb-amended soil (Zheng and Li. 2009; Currie et al.. 2005;
23                  Langdon et al., 2005).  In AQCDs, growth effects of Pb have been reported in birds
24                  (changes in juvenile weight gain), at concentrations typically  higher than currently found
25                  in the environment away from heavily exposed sites. The current body of evidence is
26                  sufficient to conclude that there is a causal relationship between Pb exposures and growth
27                  effects in terrestrial plants, and that a causal relationship is likely to exist between Pb
28                  exposure and growth effects in terrestrial invertebrates. Evidence is inadequate to
29                  establish causal relationship between Pb exposures and growth effects in terrestrial
30                  vertebrates.
              7.3.12.3     Survival-Terrestrial Biota

31                  The relationship between Pb exposure and survival has been well demonstrated in
32                  terrestrial species as presented in the Pb AQCDs and in Section 7.3.5 of the present
33                  document. Exposure can be either lethal, or produce sublethal effects that diminish
34                  survival probabilities. In the 1977 Pb AQCD, deaths from Pb poisoning in domestic
35                  animals caused by emissions from stationary sources were reported (U.S. EPA. 1977).


      November 2012                             7-81                  Draft - Do Not Cite or Quote

-------
 1                  Additional studies in the 1986 and 2006 Pb AQCDs and current ISA provide evidence for
 2                  a concentration-dependent response of mortality in terrestrial biota. Recent evidence
 3                  available since the 2006 Pb AQCD for effects of Pb on survival in terrestrial species is
 4                  summarized in Table 7-4.

 5                  Survival is a biologically important response that can have direct impact on population
 6                  size. Survival is often quantified using LC50 (the concentration of toxicant where 50%
 7                  mortality is observed or modeled), which may be a better metric for acute toxicity than
 8                  for typical environmental exposure, which is more often comparatively low, and
 9                  cumulative or chronic. From the LC50 data on Pb in this review and previous Pb AQCDs,
10                  a wide range of sensitivity to Pb is evident across taxa and within genera. As expected,
11                  reported LC50 are usually much higher than current environmental levels of Pb in the U.S
12                  away from heavily exposed sites, even though physiological dysfunction that adversely
13                  impacts the fitness of an organism often occurs well below concentrations that result in
14                  mortality. When available, LCio, NOEC or LOEC have been reported in the present
15                  document.

16                  Pb is generally not phytotoxic to plants at concentrations found in the environment away
17                  from heavily exposed sites, probably due to the fact that plants often sequester large
18                  amounts of Pb in roots, and that translocation to other parts of the plant is limited. No
19                  data have become available to change this assessment since the 2006 Pb AQCD.

20                  Survival of soil-associated organisms is adversely affected by Pb exposure. In the 1986
21                  Pb AQCD it was reported that Pb at the high extreme of concentrations found near
22                  roadsides and smelters at the time (10,000 to 40,000 mg Pb/kg dry weight) can eliminate
23                  populations  of bacteria and fungi on leaf surfaces and in soil. Severe impairment of
24                  decomposition has long been accepted to be one of the most apparent results of soil
25                  contamination with Pb and other metals. In nematodes, the 2006 Pb AQCD reported LC50
26                  values varying from 10 to 1,550 mg Pb/kg dry weight dependent upon soil OM content
27                  and soil pH (U.S. EPA. 2006c). In earthworms, 14 and 28 day LC50 values typically fell
28                  in the range  of 2,400-5,800 mg Pb/kg depending upon the species tested. More recent
29                  evidence has been consistent with these values, and also showed concentration-dependent
30                  decreases in survival in collembolans and earthworms under various experimental
31                  conditions. The evidence in terrestrial invertebrates is coherent with evidence in
3 2                  fire shwater invertebrate s.

33                  In terrestrial avian and mammalian species, toxicity is observed in laboratory studies over
34                  a wide range of doses (<1 to > 1,000 mg Pb/kg body weight-day) as reviewed for the
35                  development of Eco-SSLs (U.S. EPA. 2005b). and subsequently reported in the
36                  2006 Pb AQCD. The NOAELs for survival ranged from 3.5 to 3,200 mg Pb/kg-day.
37                  Surprisingly, the only study to have reported survival data following exposure to Pb in an
      November 2012                             7-82                  Draft - Do Not Cite or Quote

-------
 1                  avion species since the 2006 Pb AQCD, found that survival was greater than in controls
 2                  in quail exposed to 50 mg Pb/L in drinking water for 7 weeks (Nain and Smits. 2011).
 3                  Evidence for association of Pb exposure with mortality in terrestrial vertebrates is
 4                  coherent with observations in freshwater vertebrates (Section 7.4.12.3). The evidence is,
 5                  therefore, sufficient to conclude that a causal relationship is likely to exist between Pb
 6                  exposures and survival in terrestrial vertebrates and that there is a causal relationship
 7                  between Pb exposures and survival in terrestrial invertebrates. The evidence is inadequate
 8                  to conclude that there is a causal relationship between Pb exposures and survival in
 9                  terrestrial plants.
              7.3.12.4     Neurobehavioral Effects-Terrestrial Biota

10                  The central nervous system of animals was recognized as a target of Pb toxicity in the
11                  1977 Pb AQCD (U.S. EPA. 1977). and subsequent Pb reviews have provided additional
12                  supporting evidence of Pb as a neurotoxicant in terrestrial invertebrates and vertebrates.
13                  Effects of Pb on neurological endpoints in terrestrial animal taxa include changes in
14                  behaviors that may decrease the overall fitness of the organism such as food
15                  consumption, prey capture ability and avoidance behaviors.

16                  Some organisms exhibit behavioral avoidance while others do not seem to detect the
17                  presence of Pb (U.S. EPA. 2006c). Decreased food consumption of Pb-contaminated diet
18                  has been demonstrated in some invertebrates (snails) and vertebrates (lizards, pigs).
19                  Decreased food consumption were observed in juvenile A. achatina snails exposed to
20                  Pb-contaminated (concentration greater than 134 mg Pb/kg) diet for 12 weeks (Ebenso
21                  and Ologhobo. 2009a). Similarly, feeding rate in  T. pisana  snails was depressed  in 3
22                  week dietary nominal exposures of 50 to  15,000 mg Pb/kg (El-Gendy et al.. 2011). while
23                  other snails exposed to Pb at similar concentrations have shown no effects on feeding rate
24                  (Beeby and Richmond. 2010). Consumption of 10 mg/Pb kg diet resulted in lower food
25                  intake  after 120 days of dietary exposure  in pigs (S. domestied) (Yu et al.. 2005).

26                  In limited studies available on nematodes there is evidence  that Pb may affect the ability
27                  to escape or avoid predation (Wang and Xing. 2008). Additional new evidence of
28                  changes in the morphology of GABA motor neurons was also found in nematodes
29                  (C. elegans) (Du and Wang. 2009).

30                  Gull chicks experimentally exposed to Pb exhibited abnormal behaviors such as
31                  decreased walking, erratic behavioral thermoregulation and food begging that could make
32                  them more vulnerable in the wild (Burger and Gochfeld. 2005). Pb was administered via
33                  injection to reach a Pb concentration in feathers equivalent  to Pb levels in feathers of wild
      November 2012                             7-83                   Draft - Do Not Cite or Quote

-------
 1                  gull populations. Lizards exposed to Pb through diet of 10 to 20 mg Pb/kg per day for 60
 2                  days in the laboratory exhibited abnormal coloration and posturing behaviors.

 3                  These findings in terrestrial invertebrates and vertebrates are coherent with findings from
 4                  studies in aquatic biota that showed neurobehavioral alterations in various species offish,
 5                  and also in some aquatic invertebrates (Section 7.4.12.4).  They are also coherent with
 6                  findings in laboratory animals that show that Pb induces changes in learning and memory
 7                  (Section 5.3.2.3). as well as attention and motor skills (Section 5.3.3.1). New behaviors
 8                  induced by exposure to Pb reviewed in Chapter_5 that are  relevant to effects of Pb
 9                  observed in terrestrial systems include hyperactivity and mood disorders, effects on
10                  visual and auditory sensory systems, changes in structure  and function of neurons and
11                  supporting cells in the brain, and effects on the blood brain barrier. Mechanisms that
12                  include the displacement of physiological cations, oxidative stress and changes in
13                  neurotransmitters and receptors are also reviewed. Data from ecological studies  are
14                  highly coherent with these data from animal experiments, especially neurobehavioral
15                  findings and evidence of structural changes. Overall, the evidence from aquatic  and
16                  terrestrial systems  is sufficient to conclude that a causal relationship is likely to  exist
17                  between Pb exposures and neurobehavioral effects in invertebrates and vertebrates in
18                  terrestrial ecosystems.
              7.3.12.5     Hematological Effects-Terrestrial Biota

19                  Hematological responses are commonly reported effects of Pb exposure in vertebrates in
20                  terrestrial systems. In the 1977 Pb AQCD, ALAD was recognized as the most sensitive
21                  indicator of Pb exposure in rats (U.S. EPA. 1977). Furthermore, inhibition of ALAD was
22                  associated with death of waterfowl following ingestion of Pb shot. In the 1986 Pb AQCD,
23                  decreases in red blood cell ALAD activity were documented in birds and mammals near a
24                  smelter (Beyer et al.. 1985). Additional evidence for effects on blood parameters and
25                  their applicability as biomarkers of Pb  exposure in terrestrial birds and mammals were
26                  presented in the 2005 Ecological Soil Screening Levels for Lead, the 2006 Pb AQCD  and
27                  the current ISA (U.S. EPA. 2006c. 2005b). Field studies available since the
28                  2006 Pb AQCD, include evidence for elevated blood Pb levels correlated with decreased
29                  ALAD activity in songbirds and owls living in historical mining areas (Gomez-Ramirez
30                  etal., 2011; Hansen et al., 201 la).

31                  This evidence is strongly coherent with evidence from freshwater invertebrates and
32                  vertebrates (Section 7.4.12.5) and observations from human epidemiologic and animal
33                  toxicology studies showing that exposure to Pb induces effects on hematological
34                  endpoints, including altered heme synthesis mediated through decreased ALAD and
      November 2012                             7-84                  Draft - Do Not Cite or Quote

-------
 1                  ferrochelatase activities, decreased red blood cell survival and function, and increased red
 2                  blood cell oxidative stress. Taken together, the overall weight of human epidemiologic
 3                  and animal toxicological evidence is sufficient to conclude that a causal relationship
 4                  exists between Pb exposure and decreased RBC survival and function, and altered heme
 5                  synthesis in humans and in laboratory animals (Section 5.7). Based on observations in
 6                  terrestrial birds and mammals and additionally supported by observations in aquatic
 7                  organisms, and toxicological and epidemiological findings in laboratory animals and
 8                  humans evidence is sufficient to conclude that there is a causal relationship between Pb
 9                  exposures and hematological effects in terrestrial vertebrates. The evidence is inadequate
10                  to conclude that there is a causal relationship between Pb exposures and hematological
11                  effects in terrestrial invertebrates.
              7.3.12.6     Physiological Stress-Terrestrial Biota

12                  Induction of enzymes associated with oxidative stress response in terrestrial plants,
13                  invertebrates and vertebrates is a recognized effect of Pb exposure (U.S. EPA. 2006c).
14                  Several studies from the 2006 Pb AQCD in birds and plants provide evidence that Pb
15                  induces lipid peroxidation, however, exposures in these studies were higher than would
16                  be found generally in the environment (U.S. EPA. 2006c). Building on the body of
17                  evidence presented in the 2006 Pb AQCD, recent studies provide evidence of
18                  upregulation of antioxidant enzymes and increased lipid peroxidation associated with Pb
19                  exposure in many species of plants, invertebrates and vertebrates. In plants, increases of
20                  antioxidant enzymes with Pb exposure occur in some terrestrial species at concentrations
21                  approaching the average Pb concentrations in U.S. soils (18.9 mg Pb/kg). For example, in
22                  a series of studies Wang et al. observed increases in reactive oxygen species with
23                  increasing exposure to Pb from 20 mg Pb/kg soil to 2,000 mg Pb/kg in broad bean (V.
24                  fabd) (Wangetal..2010c: Wang etal.. 201 Oa: Wang et al.. 2008b) and tomato (L.
25                  esculentum) (Wang et al., 2008a). where they were accompanied up to approximately
26                  500mg Pb/kg by proportional increases in SOD, glutathione, guaiacol peroxidase, and
27                  lipid peroxidation, as well as decreases in catalase. Spinach seedlings grown in soil
28                  containing six increasing concentrations of Pb from 20 to 520 mg Pb/kg exhibited higher
29                  production of reactive oxygen species, increased rates of lipid peroxidation and increased
30                  SOD concentrations. (Wang et al.. 201 la). Markers of oxidative damage are also
31                  observed in terrestrial invertebrates, including snails and earthworms, and in terrestrial
32                  mammals. Across these biota, there are differences in the induction of antioxidant
33                  enzymes that appear to be species-dependent.
      November 2012                             7-85                   Draft - Do Not Cite or Quote

-------
 1                   The oxidative stress responses associated with Pb exposure in terrestrial plants,
 2                   invertebrates and vertebrates are consistent with responses in freshwater
 3                   (Section 7.4.12.6) and saltwater organisms (Section 7.4.21.6). and in humans
 4                   (Section 5.2.4). This oxidative stress is characterized by increased presence of reactive
 5                   oxygen species and membrane and lipid peroxidation that can promote tissue damage,
 6                   cytotoxicity, and dysfunction. Increases in reactive oxygen species are often followed by
 7                   a compensatory and protective upregulation in antioxidant enzymes, such that this
 8                   upregulation is itself indicative of oxidative stress conditions. Continuous production of
 9                   reactive oxygen species may overwhelm this defensive process, leading to further
10                   oxidative stress and injury.

11                   Upregulation of antioxidant enzymes and increased lipid peroxidation are considered
12                   reliable biomarkers of stress, and provide evidence that Pb exposure induces a stress
13                   response in those organisms which may itself increase susceptibility to other stressors and
14                   reduce individual fitness. Evidence is sufficient to conclude that there is a causal
15                   relationship between Pb exposures and physiological stress in terrestrial plants, and that a
16                   causal relationship is likely to exist between Pb exposure and physiological stress in
17                   terrestrial invertebrates and vertebrates.
              7.3.12.7     Community and Ecosystem Level Effects-Terrestrial Biota

18                   Most direct evidence of community and ecosystem level effects is from near stationary
19                   sources where Pb concentrations are higher than typically observed environmental
20                   concentrations for this metal. Impacts of Pb on terrestrial ecosystems near smelters,
21                   mines, and other industrial sources have been studied for several decades. Emissions of
22                   Pb from smelting and other industrial activities are accompanied by other trace metals
23                   (e-g-, Zn, Cu, Cd) and SO2 that may cause toxic effects independently or in concert with
24                   Pb. Those impacts include decreases in species diversity and changes in floral and faunal
25                   community composition. Ecosystem-level field studies are complicated by the
26                   confounding of Pb exposure with other factors such as the presence of trace metals and
27                   acidic deposition and the inherent variability in natural systems. In natural systems, Pb is
28                   often found co-existing with other stressors, and observed effects may be due to
29                   cumulative toxicity.

30                   In laboratory and microcosm studies where it is possible to isolate the effect of Pb, this
31                   metal has been shown to  alter competitive behavior of species, predator-prey interactions
32                   and contaminant avoidance. These dynamics may change species abundance and
33                   community structure at higher levels of ecological organization. Uptake of Pb into
34                   aquatic and terrestrial organisms and subsequent effects on mortality, growth,
      November 2012                             7-86                  Draft - Do Not Cite or Quote

-------
 1                  physiological stress, blood, neurobehavioral and developmental and reproductive
 2                  endpoints at the organism level are expected to have ecosystem-level consequences, and
 3                  thus provide consistency and plausibility for causality in ecosystem-level effects.

 4                  In the 1977 Pb AQCD the potential for Pb to interfere with ecosystem level processes
 5                  was explored in a detailed review of a study on the effects of Pb on relationships between
 6                  arthropods and leaf litter decomposition (U.S. EPA.  1977). Reduced arthropod density,
 7                  biomass and richness were observed in the vicinity of a Pb smelting complex in  Missouri.
 8                  There were also several studies correlating feeding habits, habitat, and Pb concentrations
 9                  in body tissues reported in the 1977 Pb AQCD, specifically in insects and small
10                  mammals indicating that species differences in Pb concentrations are determined in part
11                  by trophic position and habitat preference.

12                  In the 1986 Pb AQCD it was reported that Pb at environmental concentrations
13                  occasionally found near roadsides and smelters (10,000 to 40,000 mg Pb/kg dry  weight
14                  [mg Pb/kg]) can eliminate populations of bacteria and fungi on leaf surfaces and in soil
15                  (U.S. EPA. 1986b). Some key populations of soil microorganisms and invertebrates die
16                  off at 1,000 mg Pb/kg soil interrupting the flow of energy through decomposition
17                  processes and altering community structure. At soil concentrations of 500 to 1,000 mg
18                  Pb/kg or higher, populations of plants, microorganisms, and invertebrates may shift
19                  toward Pb-tolerant populations of the same or different species (U.S. EPA. 1986b).

20                  The 2006 Pb AQCD reported that decreased species diversity, changes in floral and
21                  faunal community composition and decreased vigor of terrestrial vegetation were
22                  observed in ecosystems surrounding former smelters including the Anaconda smelter in
23                  southwestern Montana (U.S. EPA. 2006c). Several studies in the 2006 Pb AQCD
24                  documented reduced organic matter decomposition rates and decreased microbial
25                  biomass in areas heavily polluted by metals. Lower abundance and reduced biodiversity
26                  of soil invertebrate communities were observed in field surveys in proximity to Pb
27                  stationary sources.

28                  Recent evidence published since the 2006  Pb AQCD (summarized in Table 7-4) supports
29                  previous findings of a link between high concentration of soil metals and substantial
30                  changes in soil microorganism community composition, as well as decreased abundance
31                  and diversity. In a naturally Pb-enriched forest in Norway, The number of fungal colony
32                  forming units was approximately 10 times lower in the highest Pb soil concentration
33                  (~4.5 mg Pb/g dry weight) than in control  soils (Baath et al.. 2005). The composition of
34                  the fungal community was drastically altered, with only one species common to  both
35                  soils, and the number of species present was substantially lower.
      November 2012                             7-87                   Draft - Do Not Cite or Quote

-------
 1                   The effect of Pb on microbial community function has been quantified previously using
 2                   functional endpoints such as respiration rates, fatty acid production, and soil acid
 3                   phosphatase and urease activities. These may provide estimate of ecological impacts that
 4                   emphasize functionality irrespective of microbial diversity or abundance measurements.
 5                   Studies available since the 2006 Pb AQCD provide further evidence of Pb effects on
 6                   microbial processes. Pb contamination reduced phenol oxidase activity in several types of
 7                   soils; concentrations between 5 and 50 nM Pb 0.001  and 0.01 mg Pb/L significantly
 8                   decreased phenol oxidase activity in all soils tested, while 400 nM (0.08 mg Pb/L) and
 9                   greater completely arrested phenol oxidase activity in one soil tested (a high pH sandy
10                   loam) (Carine et al., 2009). Pb concentrations between 50 and 500 mg Pb/kg significantly
11                   reduced microbial abundance and diversity, and also resulted in lower soil phosphatase,
12                   urease, and dehydrogenase activities (Gao et al., 2010b). When the microbial properties
13                   of metal-contaminated urban soils were compared to those of rural  soils, significant
14                   differences (Sudova and Vosatka. 2007) were detected in basal community respiration
15                   rates and microbial abundance (Yang et al.. 2006). Gai et al. (2011) examined the
16                   microbial activity of three soils via microcalorimetric methods following Pb exposure.
17                   They noted an increase in activity immediately following Pb application (giving 10, 20,
18                   40, 80, and 160 mg Pb/kg), and theorized that this was a result of rapid mortality of
19                   sensitive microbial  species, followed by a concurrent proliferation of Pb-tolerant
20                   microorganisms. As Pb concentrations increased, however, the calculated microbial
21                   growth rate constant decreased, indicating a suppression of microbial activity (Gai et al..
22                   2011). Akerblom et al. (2007) tested the effects of six metals (Cr, Zn, Mo, Ni, Cd,  and
23                   Pb) individually. All tested metals had a similar effect on the species composition of the
24                   microbial community. Exposure to a high Pb concentration (52 mg Pb/kg) negatively
25                   affected respiration rates.

26                   In addition to  microbial communities, there is new evidence for effects of Pb on other
27                   terrestrial ecosystem components. Increased plant diversity was shown to ameliorate
28                   effects of Pb contamination on a microbial community (Yang et al.. 2007). The presence
29                   of arbuscular mycorrhizal fungi may protect plants growing in Pb-contaminated soils
30                   (Bojarczuk and Kieliszewska-Rokicka. 2010; Sudova and Vosatka. 2007). Invertebrates
31                   affected by Pb in terrestrial systems may be altering community structure. Recent
32                   evidence since the 2006 Pb AQCD, indicates that some  species of worms avoid
33                   Pb-contaminated soils (Langdon et al., 2005). Reductions in microbial and detritivorous
34                   populations can affect the success of their predators (U.S. EPA. 2006c). Following a
35                   28-day exposure to field-collected soils contaminated with metals (including Pb at
36                   426 mg Pb/kg), both population growth and individual growth of the earthworm
37                   L. rubellus were diminished (Klok et al.. 2006). The  authors proposed that, although
38                   these reductions were unlikely to result in extirpation, avian predators such as the godwit
39                   (Limosa limosa) that feed heavily on earthworms may be affected by a reduction of

      November 2012                             7-88                  Draft - Do Not Cite or Quote

-------
 1                  available earthworm biomass. Furthermore, the presence of earthworms increased Pb
 2                  uptake by plants (Ruiz etal.. 2011; Sizmur et al.. 2011).

 3                  In terrestrial ecosystems, most studies show decreases in microorganism abundance,
 4                  diversity, and function with increasing soil Pb concentrations in areas near point-sources.
 5                  Specifically,  shifts in nematode communities, bacterial species, and fungal diversity have
 6                  been observed. Most evidence for Pb toxicity to terrestrial plants, invertebrates and
 7                  vertebrates is from single-species assays in laboratory studies. Although the evidence is
 8                  strong for effects of Pb on growth (Section 7.3.12.2), reproduction (Section 7.3.12.1) and
 9                  survival (Section 7.3.12.3) in certain species, considerable uncertainties exist in
10                  generalizing effects observed under small-scale, particular conditions up to predicted
11                  effects at the  ecosystem level of biological organization. In many cases it is difficult to
12                  characterize the nature and magnitude of effects and to quantify relationships between
13                  ambient concentrations of Pb and ecosystem response due to existence of multiple
14                  stressors, variability in field conditions, and to differences in Pb bioavailability at that
15                  level of organization. However, the cumulative evidence for Pb effects at higher levels of
16                  ecological organization is sufficient to conclude that a causal relationship is likely to exist
17                  between Pb exposures and the alteration of species richness, species composition and
18                  biodiversity in terrestrial ecosystems.
      7.4       Aquatic Ecosystem Effects


      7.4.1  Introduction to Effects of Pb on Aquatic Ecosystems

19                  This section of the Pb ISA reviews the recent literature published since the
20                  2006 Pb AQCD (U.S. EPA. 2006c). on the effects of Pb on freshwater and saltwater
21                  ecosystems. Freshwater and marine/estuarine systems  are considered separately due to
22                  differences in Pb speciation, bioavailability of Pb, salinity, and physiological adaptations
23                  of organisms in freshwater versus  saltwater environments, as modifying factors for Pb
24                  toxicity. The focus is on the effects of Pb to aquatic organisms including algae, aquatic
25                  plants, invertebrates, vertebrates, and other biota with  an aquatic lifestage
26                  (e.g., amphibians). In the freshwater and saltwater sections, aqueous concentrations of Pb
27                  are reported as (ig Pb/L and sediment concentrations are in mg Pb/kg.
28                  In the present document, studies in some freshwater and saltwater organisms are included
29                  where responses are observed at very high Pb concentrations that might not be expected
30                  in most environmental scenarios or where the relevance of the exposure method to
31                  atmospherically-deposited Pb is unknown. These studies can provide mechanistic
32                  information on Pb toxicity, allow for comparison of Pb uptake across taxa, or

      November 2012                             7-89                  Draft - Do Not Cite or Quote

-------
 1                   demonstrate the wide range of sensitivity among closely-related species. Furthermore,
 2                   although exposure to Pb in natural systems is likely characterized as a chronic, low dose
 3                   exposure, it is not always feasible to conduct long-term experiments under natural
 4                   conditions. Observations from short-term experiments in which high concentrations are
 5                   used can help to elucidate the shape of concentration-response relationships and provide
 6                   evidence for a gradient of response to Pb exposure but the extent to which effects would
 7                   be observed at concentrations of Pb typically found in the environment is uncertain.

 8                   There are a few studies in the following sections for which effects are reported at very
 9                   low concentrations of Pb that appear to be below analytical detection limits. These
10                   studies are included to the extent that they provide information on responses to Pb.
11                   However, the difficulty in maintaining low concentrations of Pb and the potential for
12                   contamination limits the interpretation of the reported observations and consideration of
13                   the observed effects in the absence of analytical verification. In these cases, less weight is
14                   placed on study findings in drawing conclusions regarding the effects of Pb exposure.

15                   In the following sections, the literature on aquatic ecosystem effects of Pb, published
16                   since the 2006 Pb AQCD, is considered with brief summaries from the 1977 Pb  AQCD,
17                   the 1986 Pb AQCD and the 2006 Pb AQCD where relevant. Biogeochemistry of Pb in
18                   aquatic systems is reviewed in Section 7.4.2. Sections 7.4.3 and 7.4.4 consider the
19                   bioavailability and uptake of Pb by freshwater plants, invertebrates, and vertebrates.
20                   Biological effects of Pb on freshwater ecosystem components (plants, invertebrates, and
21                   vertebrates) are discussed in Section 7.4.5. In this section,  effects are generally presented
22                   from sub-organismal responses (i.e., enzymatic activities, changes in blood parameters)
23                   to endpoints relevant to the population-level and higher (growth, reproduction and
24                   survival; summarized in Table 7-5). Biological effects are  followed by data on exposure
25                   and response of freshwater species (Section 7.4.6). Effects of Pb at the freshwater
26                   ecosystem level of biological organization are discussed in Section 7.4.7. Section 7.4
27                   includes a discussion of critical loads in freshwater systems (Section 7.4.8).
28                   characterization of sensitivity and vulnerability of freshwater ecosystem  components
29                   (Section 7.4.9) and a discussion of Pb effects on ecosystem services (Section 7.4.10). A
30                   synthesis of the new evidence for Pb effects on freshwater organisms (Section 7.4.11) is
31                   followed by causal determinations based on evidence dating back to the  1977 Pb AQCD
32                   (Section 7.4.12). Corresponding sections on saltwater systems introduced in
33                   Section 7.4.13 include bioavailability of Pb in saltwater (Section 7.4.14). biological
34                   effects of Pb in saltwater (Section 7.4.15). exposure and response of saltwater species
35                   (Section 7.4.16). community and ecosystem level effects (Section 7.4.17) and
36                   characterization of sensitivity and vulnerability in saltwater species (Section 7.4.18) and
37                   ecosystem services (Section 7.4.19). The saltwater ecosystem section concludes with a
38                   synthesis of new evidence for Pb effects in marine/estuarine systems (Section 7.4.20) and
      November 2012                              7-90                  Draft - Do Not Cite or Quote

-------
 1                  causal determinations based on evidence dating back to earlier AQCDs when available
 2                  (Section 7.4.21V
      7.4.2  Biogeochemistry and Chemical Effects of Pb in Freshwater and Saltwater
             Systems

 3                  Quantifying Pb speciation in aquatic environments is critical for determining the toxicity
 4                  of the metal to aquatic organisms. As reviewed in the 2006 Pb AQCD (U.S. EPA. 2006b)
 5                  and discussed in detail in Sections 3.3 and 7.2 of this assessment (Fate and Transport),
 6                  the speciation process is controlled by many environmental factors. Although aerially
 7                  deposited Pb largely consists of the labile Pb fraction, once the atmospherically-derived
 8                  Pb enters surface waters its fate and bioavailability are influenced by Ca2+ concentration,
 9                  pH, alkalinity, total suspended solids, and dissolved organic carbon (DOC), including
10                  humic acids. In sediments, Pb is further influenced by the presence of sulfides and Fe and
11                  Mn oxides. For instance, in neutral to acidic aquatic environments, Pb is typically present
12                  as PbSO4, PbCl4, Pb2+, cationic forms of Pb hydroxide, and ordinary hydroxide
13                  [Pb(OH)2],  while in alkaline waters, common forms of Pb include Pb carbonates
14                  [Pb(CO3)] and hydroxides [Pb(OH)2]. In addition to these inorganic forms, Pb humate is
15                  present in the solid phase and Pb fulvate is present in solution. In freshwater systems, Pb
16                  complexes with inorganic OH" and CO32 and forms weak complexes with Cl;
17                  conversely, Pb speciation in seawater is a function of chloride concentration and the
18                  primary species are PbCl3, PbCO3, PbCl2, and PbCl+. In many, but not all aquatic
19                  organisms,  Pb dissolved in water can be the primary exposure route to gills or other biotic
20                  ligands. The toxicity associated with Pb in the water column or sediment pore waters is
21                  directly affected by the competitive binding of Pb to the anions listed above.

22                  Currently, national and state ambient water quality criteria for Pb attempt to adjust
23                  measured concentrations to better represent the bioavailable free ions, and express the
24                  criteria value as a function of the hardness (i.e., amount of Ca2+ and Mg ions) of the water
25                  in a specific aquatic system. Models such as the BLM (Figure 7-3) (Paquin et al., 2002;
26                  Pi Toro etal.. 2001) include an aquatic speciation model (WHAM V; see below)
27                  combined with a model of competitive binding to gill surfaces, and provides a more
28                  comprehensive method for expressing Pb concentrations at specific locations in terms of
29                  the bioavailable metal. Sediment quality criteria have not been established, although the
30                  EPA has developed methods based on equilibrium partitioning theory to estimate
31                  sediment benchmarks for Pb and a few other metals (U.S. EPA. 2005d).  The approach is
32                  based on the ratio of the sum of simultaneously extracted metals and amount of AVS,
33                  adjusted for the fraction of organic carbon present in the sediments, and is reviewed in
34                  detail in the 2006 Pb AQCD (U.S. EPA. 2006c). It is important to note that this method
      November 2012                            7-91                  Draft - Do Not Cite or Quote

-------
 1                  cannot accurately predict which sediments are toxic or which metal is the primary risk
 2                  driver.

 3                  A more detailed understanding of the biogeochemistry of Pb in aquatic systems (both the
 4                  water column and sediments) is critical to accurately predicting toxic effects of Pb to
 5                  aquatic organisms. It should be recognized, however, that in addition to exposure via
 6                  sediment and water, chronic exposures to Pb also include dietary uptake, even though the
 7                  toxicokinetics of this exposure pathway are not yet well understood in aquatic organisms
 8                  and the influence of the bioavailability factors described above is unknown. Furthermore,
 9                  changes in environmental factors that reduce the bioaccessible Pb fraction can result in
10                  either sequestration in sediments or subsequent release as mobile, bioaccessible forms.
11                  This section provides updated information about the influence of chemical parameters
12                  that affect Pb bioaccessibility in the aquatic environment (in sediments and the water
13                  column).

14                  Several models are available for estimating the speciation of dissolved Pb. These models
15                  were tested by Balistrieri and Blank (2008) by comparing the speciation of dissolved Pb
16                  in aquatic systems affected by historical mining activities with that predicted by  several
17                  models, including Windermere humic aqueous model (WHAM VI), non-ideal
18                  competitive absorption Donnan-type model (NICA-Donnan), and Stockholm humic
19                  model (SHM). Accurate prediction of labile Pb concentrations was achieved only with
20                  SHM, although other metal concentrations were better described by the WHAM model.
21                  Whereas both WHAM VI and NICA-Donnan predicted that the bulk of Pb contamination
22                  would be complexed with Fe,  SHM predicted Pb speciation predominantly characterized
23                  by Fe and inorganic Pb complexes. Predicted dynamic Pb concentrations developed with
24                  the WHAM VI and NICA-Donnan methods overestimated Pb concentrations measured
25                  using diffusive gradients in thin-films in Lake Greifen (Switzerland), but underestimated
26                  concentrations in Furbach stream (located in both the Coeur d'Alene and Spokane River
27                  Basins in Idaho), indicating that such models may not be able to accurately describe
28                  metal speciation under all environmental  conditions (Balistrieri and Blank. 2008).

29                  Quantification of different sediment metal-binding phases, including sulfide, organic
30                  carbon (OC), Fe, and Mn phases, is important to fully understand the bioaccessible
31                  fraction of Pb and the toxicity to benthic organisms (Simpson and Batley. 2007).
32                  However, physical disturbance, pH change, and even the biota themselves also alter
33                  sediment binding or release of Pb. Atkinson et al. (2007) studied the effects of pH on
34                  sequestration or release of Pb from sediments. Although high and circumneutral  water pH
35                  (8.1 and 7.2) did not affect the release of sequestered Pb from sediments, lowering the pH
36                  to 6 increased the concentration of Pb in overlying waters from less than 100 (ig  Pb/L to
37                  200-300 (ig Pb/L. Physical sediment disturbance also increased the amount of sediment-
      November 2012                             7-92                  Draft - Do Not Cite or Quote

-------
 1                  bound Pb released into the aqueous phase. When Pb-contaminated sediment was
 2                  physically disturbed, the dissolved oxygen content of the overlying water was observed
 3                  to significantly impact Pb mobilization, with greater Pb mobilization at lower dissolved
 4                  oxygen levels (3 to 9 mg/L O2) (Atkinson et al. 2007). In addition, although Pb
 5                  concentrations in the sediments of a mine-impacted wetland in Hezhang, China, were
 6                  determined to be strongly associated with organic/sulfide and residual fractions (e.g., 34
 7                  to 82% of total Pb), the presence of aquatic macrophytes altered the Pb speciation,
 8                  increasing the fraction of Pb bound to Fe-Mn oxides (42% to 47% of total Pb) (Bi et al..
 9                  2007). This phenomenon was investigated in greater depth by  Sundby et al. (2005). who
10                  determined that release of oxygen from macrophyte roots resulted in the oxidation of
11                  sediment-bound Pb, leading to the release of bioaccessible Pb  fractions (Sundby et al..
12                  2005).
              7.4.2.1  Other Metals

13                  Multiple metals are present simultaneously in many aquatic environments and may
14                  interact with one another influencing Pb uptake and toxicity. Interactions of Pb with other
15                  metals were reviewed in the 2006 Pb AQCD, and more recent evidence supports previous
16                  findings of altered bioavailability associated with metal mixtures. Komjarova and Blust
17                  (2008) looked at the effect of the presence of Cd2+ on the uptake of Pb by the freshwater
18                  cladoceran Daphnia magna. While Pb uptake rates were not affected by Cu, Ni or Zn,
19                  enhanced Pb accumulation was observed in the presence of 0.2 (iM Cd. The highest Pb
20                  concentration, 0.25 (iM (51.8 (ig Pb/L) in turn facilitated Cu uptake. Area-specific and
21                  whole organism Pb transport rates were greatest in the mid-intestine. It was concluded
22                  that Pb-induced disruptions of ion homeostasis and metal absorption processes might be a
23                  possible explanation of stimulated Pb uptake in the presence of Cd, as well as the
24                  increase in Cu uptake rates provoked by presence of Pb at its highest studied
25                  concentration. Komjarova and Blust (2009b) then considered the effect of Na, Ca2+ and
26                  pH on simultaneous uptake of Cd, Cu, Ni, Pb and Zn. Cd and Pb showed increased
27                  uptake rates at high Na concentration. It was thought that increased Na uptake rates
28                  promoted Pb entrance to the cell. With respect to the effect of pH, reduced proton
29                  competition begins to influence Pb uptake in waters with high pH. A clear suppression of
30                  Cd, Ni, Pb and Zn uptake was observed in the presence of Ca2+ (2.5 mM). Ca2+ has been
31                  reported to have a protective effect in other studies (involving  other organisms). The
32                  presence of other metals may also affect the uptake of Pb by fish. At low concentrations,
33                  Cd in a Pb-Cd mixture out-competed Pb at gill tissue binding sites in rainbow trout
34                  (Oncorhynchus mykiss), resulting in a less-than additive toxicity when fish were exposed
35                  to both metals in tandem (Birceanu et al.. 2008). Evidence for the presence of Pb
      November 2012                             7-93                   Draft - Do Not Cite or Quote

-------
 1                  influencing the uptake of other metals was observed in the marine bivalves Macomona
 1                  hliana and Austrovenus stutchburyi. Significantly, more Zn bioaccumulated in the
 3                  presence of Pb in these mussels than with Zn alone following a 10-day exposure to
 4                  spiked sediments (Fukunaga and Anderson. 2011).
              7.4.2.2  Biofilm

 5                  Farag et al. (2007) measured Pb concentrations in various media (water, colloids,
 6                  sediment, biofilm) as well as invertebrates and fish collected within the Boulder River
 7                  watershed, MT, U.S. They concluded that the fraction of Pb associated with Fe-oxides
 8                  was most frequently transferred to biofilms and the other biological components of the
 9                  sampled systems (Farag et al.. 2007). Consequently, an increase in the Pb Fe-oxide
10                  fraction could signify a potential increase in the bioaccessible pool of Pb. The authors
11                  also noted that this fraction may promote downstream transport of Pb contamination.
12                  Ancion et al. (2010) investigated whether urban runoff metal contaminants could modify
13                  biofilm bacterial community structure and diversity and therefore potentially  alter the
14                  function of biofilms in stream ecosystems. They found that accumulation rates for metals
15                  in biofilm were maximal during the first day of exposure and then decreased with time.
16                  Equilibrium between metal concentrations in the water and in the biofilm was reached for
17                  all metals after 7-14 days of exposure. The affinity of the biofilm for Pb was,  however,
18                  much greater than for Cu and Zn. With respect to recovery, the release of metals was
19                  slow and after 14 days in clean water 35% of Pb remained in the biofilm. By retaining
20                  and releasing such metal pollutants, biofilms may play a key role in determining both the
21                  concentration of the dissolved metals in the water column and the transfer of the metals
22                  to invertebrates and fish grazing on them. An enrichment factor of 6,000:1 for Pb
23                  between the biofilm and the water was measured  after 21 days exposure to synthetic
24                  urban runoff. The relatively slow  release of such metal may greatly influence the transfer
25                  of Pb to organisms feeding on the biofilms. This may be of particular importance during
26                  storm events when large amounts of Pb are present in the urban runoff. It was suggested
27                  that biofilms constitute an integrative indicator of metal exposure over a period of days to
28                  weeks.
      November 2012                             7-94                  Draft - Do Not Cite or Quote

-------
              7.4.2.3  Carbonate

 1                  An investigation of heavy metal concentrations in an industrially impacted French canal
 2                  (Deule canal) indicated that total extractable Pb in sediments ranged from 27 to
 3                  10,079 mg Pb/kg, with 52.3% present in Fe-Mn oxide fractions, 26.9% as organic sulfide
 4                  fraction, 10.7% in carbonates, and 10.1% in the residual fraction (Boughriet et al.. 2007).
 5                  The relatively high fraction of Pb associated with carbonates was not observed at other
 6                  sites, as sediments in these areas contained low proportions of carbonates. Hence,
 7                  addition of carbonates (either from anthropogenic or natural sources) can significantly
 8                  impact Pb speciation in sediments, and potential bioavailability to resident organisms. In
 9                  addition, increased surface water carbonate concentrations also reduced the bioaccessible
10                  Pb fraction as measured by chronic Pb accumulation in the fathead minnow, (Pimephales
11                  promelas) (Mager etal.. 2010). and by Pb toxicity to fathead minnow and the cladoceran
12                  (Ceriodaphnia dubid) (Mager et al.. 201 Ib).
              7.4.2.4  Dissolved Organic Matter (DOM)

13                  Uptake of Pb by water-column organisms is affected by the concentration of DOM
14                  (Mager etal.. 201 la; Mager etal.. 2010). In a 7-day chronic study with C. dubia, DOM
15                  protected against toxicity while water hardness was not protective (Mager et al., 201 la).
16                  The specific composition of DOM has been shown to affect the bioaccessibility of
17                  environmental Pb. Humic acid-rich DOM resulted in decreased free Pb ion concentration
18                  when compared to systems containing DOM with high concentrations of polysaccharides
19                  (Lamelas and Slavevkova. 2008). When the sequestering abilities of various components
20                  of DOM were compared, humic acid again was shown to be most efficient at reducing the
21                  Pb free ion concentration, followed by fulvic acid, alginic acid, polygalacturonic acid,
22                  succinoglycan, and xanthan (Lamelas et al., 2005). Lamelas et al. (2009) considered the
23                  effect of humic acid on Pb(II) uptake by freshwater algae taking account of kinetics and
24                  cell wall speciation. The uptake flux was described by a Michaelis-Menten type equation.
25                  Comparison of Cu(II), Cd(II) and Pb(II) uptake by green freshwater algae, (Chlorella
26                  Kessleri), in the presence of either citric acid or humic acid was made. The uptake fluxes,
27                  percentage adsorbed and percentage internalized for Cu and Cd were identical in the
28                  presence of either citric or humic acid. In contrast, however, there was a ten-fold increase
29                  in the respective values for Pb. The increase in adsorbed Pb was attributed to the increase
30                  in adsorption sites from the adsorbed humic acid on the surface of the algae. Two
31                  hypotheses were considered to explain the increase in internalized Pb and the
32                  internalization flux: (1) direct interaction of Pb-humic acid complexes with  the
33                  internalization sites, and (2) uptake of Pb(II) after dissociation from the Pb-humic acid
34                  complex. The authors favor the former hypothesis but no evidence is presented for the

      November 2012                             7-95                   Draft - Do Not Cite or Quote

-------
 1                  proposed ternary Pb-humic acid-internalized site complexes, nor is there an explanation
 2                  as to why this behavior is not observed for Cd or Cu.

 3                  There is evidence, however, that DOC/DOM does not have the same effect on free Pb ion
 4                  concentration in marine systems as in freshwater systems. No correlation was observed
 5                  between DOM concentration or composition and Pb toxicity when examined using the
 6                  sea urchin (Paracentrotus lividus) embryo-larval bioassay (Sanchez-Marin et al.. 2010a).
 7                  For marine invertebrates, the presence of humic acid increased both the uptake and
 8                  toxicity of Pb, despite the fact that a larger fraction of Pb is complexed with humic acid
 9                  (25 to 75%). Although the authors could not provide a precise explanation for this, they
10                  theorized that in marine environments,  addition of humic acid could induce and enhance
11                  uptake of Pb via membrane Ca2+ channels (Sanchez-Marin et al., 2010b). This
12                  mechanism was observed in the marine diatom (Thalassiosira weissflogii), in that humic
13                  acids absorbed to cell surfaces increased metal uptake; however, water column Pb-humic
14                  acid associations did appear to reduce free Pb ion concentrations (Sanchez-Marin et al..
15                  201 Ob). Formation of a ternary complex that is better absorbed by biological membranes
16                  was another proposed mechanism that could describe the increased bioaccessibility to
17                  marine invertebrates of Pb bound to humic acid (Sanchez-Marin et al.. 2007).

18                  Sanchez-Marin et al. (2011) subsequently have shown that different components of DOM
19                  have different effects on Pb bioavailability in marine systems. Their initial research using
20                  Aldrich humic acid found that increasing humic acid concentrations increased Pb uptake
21                  by mussel gills and increased toxicity to sea urchin larvae in marine environments
22                  (Sanchez-Marin et  al., 2007). In contrast, a subsequent investigation found that fulvic
23                  acid reduced Pb bioavailability in marine water (Sanchez-Marin et al.. 2011). The
24                  contradictory effects of different components of DOM on marine bioavailability likely
25                  reflect their distinct physico-chemical characteristics. More hydrophobic than fulvic acid,
26                  humic acid may adsorb directly with cell membranes and enhance Pb uptake through
27                  some (still unidentified) mechanism (Sanchez-Marin et al.. 2011).

28                  As little as 1 (iM of humic acid introduced into surface waters was sufficient to reduce Pb
29                  uptake by perennial ryegrass, Lolium perenne, grown in nutrient solution. This resulted
30                  from a decrease in the concentration of the free Pb fraction by several orders of
31                  magnitude following complexation with the OM. Pb content on the root surface was
32                  reduced to 1,658 mg Pb/kg from 4,144  mg Pb/kg following humic acid addition, and
33                  relative Pb absorption (absorption in the presence of humic acid divided by absorption in
34                  the absence of humic acid) was determined to be approximately 0.2 (Kalis et al.. 2006).
35                  Conversely, humic acid may increase the bioaccessible Pb fraction for green algae
36                  through formation of a ternary complex that promotes algal uptake of the metal. Lamelas
37                  and Slaveykova (2007) found that aqueous Pb formed complexes with humic acid, which
      November 2012                             7-96                  Draft - Do Not Cite or Quote

-------
 1                  in turn would become adsorbed to C. kesslerii algal surfaces, and that the presence of Pb
 2                  sorbed to humic acid did not interfere with humic acid-algae complexation. The authors
 3                  concluded that humic acids bound to algae acted as additional binding sites for Pb, thus
 4                  increasing the concentrations associated with the algal fraction (Lamelas and Slavevkova.
 5                  2007).
 6                  Based on the above, the recent literature indicates the existence of a number of deviations
 7                  from current models used to predict bioaccessibility of Pb. In marine aquatic systems, for
 8                  instance, surface water DOM was found to increase (rather than decrease) uptake of Pb
 9                  by fish gill structures, potentially through the alteration of membrane Ca2+-channel
10                  permeability. This phenomenon would not be accurately predicted by a BLM developed
11                  using data from freshwater organisms. Further, in both freshwater and marine
12                  environments, algal biosorption of labile Pb fraction was also increased by humic acid
13                  and DOM, likely through the formation of ternary complexes that increase Pb binding
14                  sites on the algal surface. Although it is unclear whether Pb in this form is available for
15                  toxic action  on algae, it is likely to comprise a significant source of dietary Pb for
16                  primary consumers. Moreover, the attempted field verification of freshwater
17                  bioaccessibility models was conducted at sites with distinct point-sources of Pb
18                  contamination, and only  one model (SHM) adequately predicted Pb bioaccessibility.
              7.4.2.5  Sulfides

19                  In sediments, Pb bioavailability is further influenced by sulfides. In the presence of
20                  sulfides, most of the reactive metal in sediments will form insoluble metal sulfide that is
21                  not bioavailable for uptake by benthic organisms. Acid volatile sulfide (AVS) has been
22                  used to predict the toxicity of Pb and other metals in sediments (Ankley etal. 1996; Di
23                  Toro et al.. 1992) and in the development of sediment quality criteria (Section 7.4.3). The
24                  role of sulfides in the flux of Pb from sediments is discussed further in Section 3.3.2.3.
      7.4.3  Introduction to Bioavailability and Biological Effects of Pb in Freshwater
             Ecosystems

25                  Freshwater ecosystems across the U.S. encompass many habitats including ponds,
26                  streams, rivers, wetlands and lakes. Concentrations of Pb available for fresh surface-
27                  water and freshwater sediments are reported in Section 7.2.3 and Table 7-2 and are
28                  summarized here. Representative median and range of Pb concentrations in surface
29                  waters (median 0.50 (ig Pb/L, range 0.04 to 30 (ig Pb/L), sediments (median 28 mg Pb/kg
30                  dry weight, range 0.5 to 12,000 mg Pb/kg dry weight) and fish tissues (geometric mean
      November 2012                             7-97                  Draft - Do Not Cite or Quote

-------
 1                  0.54 mg Pb/kg dry weight, range 0.08 to 23 mg Pb/kg dry weight [whole body]) in the
 2                  U.S. based on a synthesis of NAWQA data reported in the previous 2006 Pb AQCD
 3                  (U.S. EPA. 2006c). Additional information on ambient Pb levels in waters, sediments and
 4                  biota is presented in Section 3.6.5 and Table 7-2 including new data from the Western
 5                  Airborne Contaminants Assessment Project (WACAP) on Pb in environmental media and
 6                  biota from remote ecosystems in the western U.S. WACAP assessed concentrations of
 7                  semi-volatile organic compounds and metals in up to seven ecosystem components (air,
 8                  snow, water, sediment, lichen, conifer needles and fish) in watersheds of eight core
 9                  national parks during a multi-year project conducted from 2002-2007 (Landers et al.,
10                  2008). The goals of the study were to assess where these contaminants were
11                  accumulating in remote ecosystems in the western U.S., identify ecological receptors for
12                  the pollutants, and to determine the source of the air masses most likely to have
13                  transported the contaminants to the parks.

14                  The 2006 Pb AQCD (U.S. EPA. 2006b) provided an overview of regulatory
15                  considerations  for water and sediments in addition to consideration of biological effects
16                  and major environmental factors that modify the response of aquatic organisms to Pb
17                  exposure. Regulatory guidelines for Pb in water and sediments have not changed since
18                  the 2006 Pb AQCD, and are summarized below with consideration of limited new
19                  information on these criteria since the last review. This section is followed by new
20                  information on biogeochemistry, bioavailability and biological effects of Pb since the
21                  2006 Pb AQCD.

22                  The most recent ambient water quality criteria for Pb in freshwater were released in 1985
23                  (U.S. EPA. 1985) by the EPA Office of Water which employed empirical regressions
24                  between observed toxicity and water hardness to develop hardness-dependent equations
25                  for acute and chronic criterion. These criteria are published pursuant to Section 304(a) of
26                  the Clean Water Act and provide guidance to states and tribes to use in adopting water
27                  quality standards for the protection of aquatic life  and human health in surface water. The
28                  ambient water quality criteria for Pb are expressed as a criteria maximum concentration
29                  (CMC) for acute toxicity and criterion continuous concentration  (CCC) for chronic
30                  toxicity (U.S. EPA. 2009b). In freshwater, the CMC is 65 (ig Pb/L  and the CCC is
31                  2.5 (ig Pb/L at  a hardness of 100 mg/L.

32                  The 2006 Pb AQCD summarized two approaches for establishing sediment criteria for Pb
33                  based on either bulk sediment or equilibrium partitioning (Section 7.2.1 and
34                  Section AX7.2.1.4). The first approach is based on empirical correlations between metal
35                  concentrations in bulk sediment and associated biological effects to derive threshold
36                  effect concentrations (TEC) and probable effects concentrations (PEC) (MacDonald et
37                  al., 2000). The TEC/PEC approach derives numeric  guidelines to compare against bulk
      November 2012                             7-98                  Draft - Do Not Cite or Quote

-------
 1                  sediment concentrations of Pb. The other approach in the 2006 Pb AQCD was the
 2                  equilibrium partitioning procedure published by the EPA for developing sediment criteria
 3                  for metals (U.S. EPA. 2005d). The equilibrium partitioning approach considers
 4                  bioavailability by relating sediment toxicity to pore water concentration of metals. The
 5                  amount of simultaneously extracted metal (SEM) is compared with the metals extracted
 6                  via AVS since metals that bind to AVS (such as Pb) should not be toxic in sediments
 7                  where AVS occurs in greater quantities than SEM.

 8                  Since the 2006 Pb AQCD, both of these methods for estimating sediment criteria for
 9                  metals, have continued to be used and refined. The SEM approach was further refined in
10                  the development of the sediment BLM (Di Toro et al.. 2005). The BLM is discussed
11                  further in Sections 7.3.3 and 7.4.4. Comparison of empirical approaches with AVS-SEM
12                  in metal contaminated field sediments shows that samples where either method predicted
13                  there should be no toxicity due to metals, no toxicity was observed in chronic amphipod
14                  exposures (Besser et al.. 2009; MacDonald et al.. 2009). However, when the relationship
15                  between invertebrate habitat (epibenthic and benthic) and environmental Pb
16                  bioaccumulation was investigated, De Jonge et al. (2010) determined that different
17                  environmental fractions of Pb were responsible for invertebrate uptake and exposure. Pb
18                  uptake by benthic invertebrate taxa was not significantly correlated to AVS Pb levels, but
19                  rather to total sediment concentrations (De Jonge et al.. 2009). Conversely, epibenthic
20                  invertebrate Pb body burdens were better correlated to AVS concentrations, rather than
21                  total Pb sediment concentrations (De Jonge et al.. 2010).

22                  In the following sections, recent information since the 2006 Pb AQCD on Pb in
23                  freshwater ecosystems will be presented. Throughout the sections, brief summaries of
24                  conclusions from the 1977 Pb AQCD, the 1986 Pb AQCD and 2006 Pb AQCD are
25                  included where appropriate. The sections are organized to consider uptake of Pb and
26                  effects at the species level, followed by community and ecosystem level effects. New
27                  research on the bioavailability and uptake of Pb into freshwater organisms including
28                  plants, invertebrates and vertebrates is presented in Section 7.4.4. Effects of Pb on the
29                  physiology of freshwater flora and fauna (Section 7.4.5) are followed with data on
30                  exposure and response of freshwater organisms (Section 7.4.6). Responses at the
31                  community and ecosystem levels of biological organization are reviewed in Section 7.4.7
32                  followed by a brief consideration of critical loads in freshwater systems (Section 7.4.8),
33                  characterization of sensitivity and vulnerability of ecosystem components (Section 7.4.9)
34                  and a discussion of ecosystem services (Section 7.4.10). The freshwater ecosystem
35                  section concludes with a synthesis of new evidence (Section 7.4.11) and causal
36                  determinations based on evidence dating back to the 1977 Pb AQCD (Section 7.4.12).
      November 2012                             7-99                   Draft - Do Not Cite or Quote

-------
      7.4.4  Bioavailability in Freshwater Systems

 1                   Bioavailability was defined in the 2006 Pb AQCD as "the proportion of a toxin that
 2                   passes a physiological membrane (the plasma membrane in plants or the gut wall in
 3                   animals) and reaches a target receptor (cytosol or blood)." In 2007, EPA took cases of
 4                   bioactive adsorption into consideration and revised the definition of bioavailability as
 5                   "the extent to which bioaccessible metals absorb onto, or into, and across biological
 6                   membranes of organisms, expressed as a fraction of the total amount of metal the
 7                   organism is proximately exposed to (at the sorption surface) during a given time and
 8                   under defined conditions" (U.S. EPA. 2007c). See Section 7.3.3 for additional discussion
 9                   of bioavailability.

10                   The bioavailability of metals varies widely depending on the physical, chemical, and
11                   biological conditions under which an organism is exposed (U.S. EPA. 2007c). The
12                   bioavailability of a metal is also dependent upon the bioaccessible fraction of metal. The
13                   bioaccessible fraction of a metal is the portion (fraction or percentage) of
14                   environmentally available metal that actually interacts at the organism's contact surface
15                   and is potentially available for absorption or adsorption by the organism (U.S. EPA.
16                   2007c). The processes for evaluating bioavailability and bioaccessibility are presented in
17                   Figure 7-2 and in Section 7.3.3. In brief, trace metals, and their complexes, must first
18                   diffuse from the external medium to the surface of the organism (mass transport). Metal
19                   complexes may dissociate and re-associate in the time that it takes to diffuse to the
20                   biological surface. These processes are considered further in Chapter_3. To have an effect
21                   on the organism, metals must then react with a sensitive site on the biological membrane
22                   (adsorption/desorption processes), often but not necessarily followed by biological
23                   transport (internalization). Any of these processes may be the rate limiting step for the
24                   overall biouptake process. Internalization is, however, the key step in the overall
25                   biouptake process. Although the transport sites often have a high affinity for required
26                   metals they do not always have high selectivity and so a toxic metal may bind to the site
27                   of an essential metal with a similar ionic radius or co-ordination geometry, e.g., Pb2+,
28                   Cd2+ and Zn2+ are similar to Ca2+. At the molecular level, there are three major classes of
29                   transition metal transporter: P-type ATPases, Zn regulated transporter/iron-regulated
30                   transporter, and natural resistance associated macrophage proteins (Worms et al.. 2006).
31                   Of these, natural resistance associated macrophage proteins have been shown to promote
32                   the uptake of various metals including Pb. This type of trace metal transport can be
33                   described by Michaelis-Menten uptake kinetics and equilibrium considerations.
      November 2012                             7-100                  Draft - Do Not Cite or Quote

-------
                    Routes of Exposure

 1                  According to the 2006 Pb AQCD (U.S. EPA. 2006b). Pb adsorption, complexation,
 2                  chelation, etc., are processes that alter its bioavailability to different aquatic species, and
 3                  it was suggested that multiple exposure routes may be important in determining overall
 4                  bioavailability of Pb. Given its low solubility in water, bioaccumulation of Pb by aquatic
 5                  organisms may preferentially occur via exposure routes other than direct absorption  from
 6                  the water column, including ingestion of contaminated food and water, uptake from
 7                  sediment pore waters, or incidental ingestion of sediment. If uptake and accumulation are
 8                  sufficiently faster than depuration and excretion, Pb tissue levels may become sufficiently
 9                  high to result in physiological effects (Luoma and Rainbow. 2005). Pb accumulation rates
10                  are controlled, in part, by metabolic rate. Other factors that influence bioavailability of Pb
11                  to organisms in aquatic systems are reviewed in Section 7.4.2. As summarized in the
12                  2006 Pb AQCD, organisms exhibit three Pb accumulation strategies: (1) accumulation of
13                  significant Pb concentrations with low rate of loss resulting in substantial accumulation;
14                  (2) balance between excretion and bioavailable metal in the environment; and (3) very
15                  low metal uptake rate without significant excretion, resulting in weak net accumulation
16                  (Rainbow. 1996). Uptake experiments with aquatic plants, invertebrates and vertebrates
17                  reviewed in the 2006 Pb  AQCD showed increases in Pb uptake with increasing Pb in
18                  solution. The 2006 Pb AQCD findings included consideration of bioaccumulation in
19                  different trophic levels. Pb concentrations were found to be typically higher in algae and
20                  benthic organisms and lower in higher trophic-level consumers.

21                  In this section:

22                     1)  Recent information on bioavailability and uptake in algae, plants,
23                         invertebrates and vertebrates from freshwater systems are reviewed with
24                         summary material from the 2006 Pb AQCD and earlier Pb AQCDs where
25                         appropriate.
26                     2)  An overview of the BLM is presented as the most widely used method for
27                         predicting both the bioaccessible and bioavailable fractions of Pb in the
28                         aquatic environment. This is followed by a discussion of
29                     3)  Bioavailability in algae, plants, invertebrates and vertebrates. As reviewed by
30                         Wang and Rainbow (2008), aquatic organisms exhibit distinct patterns of
31                         metal bioaccumulation. The authors suggest that the observed differences in
32                         accumulation, body burden, and elimination between species are due to metal
33                         biogeochemistry and physiological and biological responses of the organism.
34                         The studies presented below generally support the observations of Wang and
      November 2012                            7-101                  Draft - Do Not Cite or Quote

-------
 1                         Rainbow (2008) that closely related species can vary greatly in
 2                         bioaccumulation of Pb and other non-essential metals.
 3                  The bioaccumulation and toxicity of Pb to aquatic organisms are closely linked to the
 4                  environmental fate of the metal under variable environmental conditions (Section 3.3) as
 5                  they are highly dependent upon the relative proportion of free metal ions in the water
 6                  column. However, information is lacking on the uptake of Pb through ingestion of
 7                  Pb-sorbed particles or dietary exposure to biologically-incorporated Pb. Such routes of
 8                  exposure are not included in models such as the BLM that predict toxicity as a function
 9                  of Pb concentration in the water column. This uncertainty may be greater for Pb than for
10                  other more soluble metals (such as Cu) as a greater proportion of the total mass of Pb in
11                  an aquatic ecosystem is likely to be bound to particulate matter. Therefore, estimating
12                  chronic toxicity of Pb to aquatic receptors may have greater uncertainty than predicting
13                  acute effects.

14                  BLM Models

15                  In addition to the biogeochemical effects that govern the environmental pool of
16                  accessible Pb, reactions of Pb with biological surfaces and membranes determines the
17                  bioavailability and uptake of the metal by aquatic organisms. The BLM (Figure 7-3)
18                  predicts both the bioaccessible and bioavailable fraction of Pb in the aquatic
19                  environment, and can be used to estimate the importance of environmental variables such
20                  as DOC in limiting uptake by aquatic organisms (Alonso-Castro et al.. 2009). The BLM
21                  integrates the binding affinities of various natural ligands in surface waters and the
22                  biological uptake rates of aquatic organisms to determine the site-specific toxicity of the
23                  bioavailable fraction.

24                  In the 2006 Pb AQCD, limitations of the use of BLM in developing air quality criteria
25                  were  recognized including the focus of this model on acute endpoints and the absence of
26                  consideration of dietary uptake as a route of exposure. Atmospheric deposition of Pb to
27                  aquatic systems and subsequent effects on ecosystem receptors is likely characterized as a
28                  chronic, cumulative exposure rather than an acute exposure. Recommendations from the
29                  2006 Pb AQCD included developing both chronic toxicity BLMs and BLMs that
30                  consider the dietary route of Pb uptake. The EPA recently incorporated the BLM into the
31                  Framework for Metal Risk Assessment (U.S. EPA. 2007c) and has published an ambient
32                  freshwater criteria document for Cu based on the BLM model (U.S. EPA. 2007a). This
33                  section reviews the literature from the past 5 years on applications of the BLM to
34                  predicting bioavailability of Pb to aquatic organisms. However, the primary focus of
35                  initial BLMs has been acute toxicity endpoints for fish and invertebrates following gill or
36                  cuticular uptake of metals.

      November 2012                             7-102                  Draft - Do Not Cite or Quote

-------
 1                   Di Toro et al. (2005) constructed BLMs for metals exposure in sediments, surface water,
 2                   and sediment pore water to determine how to most accurately predict the toxicity of
 3                   metals-contaminated sediments. Results from models were compared with literature-
 4                   derived acute toxicity values for benthic and epibenthic invertebrates to establish the
 5                   accuracy of the developed models. Although the models tended to overestimate the
 6                   toxicity of aqueous and sediment-bound Pb in freshwater environments, it was
 7                   determined that the model significantly underestimated Pb toxicity to marine
 8                   invertebrates (Di Toro et al.. 2005). This may be because pore water metal concentrations
 9                   were not modeled. Consequently, these results may suggest that either  1) mobilization of
10                   Pb concentrations from sediments into pore water is greater in marine environments, or 2)
11                   marine invertebrates are significantly more sensitive to Pb exposures than are freshwater
12                   species.

13                   A number of deviations from results predicted by Pb exposure models (such as the BLM)
14                   were documented by Ahlf et al. (2009). They highlighted that uptake of metals by
15                   sediment-dwelling bivalves was significantly greater than predicted, because bivalves
16                   accumulate Pb from multiple sources not included in the model, such as ingestion of
17                   algae, bacteria, and colloidal matter.  Species-specific dietary assimilation of ingested
18                   particulate-bound metals is also likely to play a role in the toxicity of Pb to aquatic
19                   organisms, yet insufficient data are available to permit modeling of this additional factor
20                   (Ahlf etal. 2009). The authors outlined the need for additional data in  developing
21                   bioavailability models for chronic metal exposures. As recent evidence suggests that the
22                   hydrophobic DOC fraction (e.g., humic and fulvic acids) sequesters the greatest fraction
23                   of Pb in aquatic systems (Pernet-Coudrier et al.. 2011), understanding the influence of
24                   this adsorption on Pb toxicity is critical for the prediction of chronic aquatic Pb toxicity.
25                   For instance, although the presence of humic acid is considered to reduce the bioavailable
26                   fraction of metals in surface water, green algae uptake and biosorption  of metals,
27                   including Pb, was actually increased  by humic acid. The authors determined that humic
28                   acid bound to algal surfaces served to increase the total number of metal binding sites
29                   over those afforded solely by the algal surface (Lamelas and Slavevkova. 2007). This
30                   highlights the complexity of modeling chronic metals bioavailability through multiple
31                   exposure routes, as humic acid would decrease gill or cuticular uptake of metals from the
32                   water column, but could potentially enhance dietary exposure by increasing algal metal
33                   content. Slaveykova and Wilkinson (2005) also noted that humic acid is likely to interact
34                   with other biological membranes and alter their permeability to metals, especially in
35                   acidic environments. Further, they observed that increased surface water temperatures
36                   can not only increase membrane permeability but also change metabolic rates, both of
37                   which can enhance metals uptake and assimilation; however, this factor is not included  in
38                   bioavailability models such as the BLM (Slaveykova and Wilkinson. 2005). Despite this,
      November 2012                             7-103                  Draft - Do Not Cite or Quote

-------
 1                  the authors noted that, in most cases, the BLM could predict acute metals toxicity with a
 2                  reasonable degree of accuracy.
              7.4.4.1  Freshwater Plants and Algae

 3                  In the 1977 Pb AQCD, the root system of plants was recognized as the major route of
 4                  uptake for Pb (U.S. EPA. 1977). Uptake and translocation studies of Pb in plants and
 5                  algae reviewed in the 1977 Pb AQCD and the 2006 Pb AQCD indicated that plants tend
 6                  to sequester larger amounts of Pb in their roots than in their shoots. Recent studies on
 7                  bioavailability of Pb to plants support the findings of the previous Pb AQCDs and
 8                  provide additional evidence for species-dependent differences in responses to Pb in water
 9                  and sediments.

10                  Most biouptake studies in aquatic plants and algae available since the 2006 Pb AQCD,
11                  were typically conducted at very high concentrations of Pb that are not representative of
12                  current levels of Pb typically encountered in freshwater. However, most of these
13                  exposures included a series of increasing concentrations of Pb and generally, Pb was
14                  accumulated in a dose-dependent manner. Studies in which high concentrations of Pb are
15                  used and an exposure-response relationship is observed may imply effects at lower
16                  concentrations but uncertainty remains to the extent to which effects would be observed
17                  at concentrations of Pb typically found in the environment. The role of modifying factors,
18                  such as the presence of other metals, on uptake rates as well as species differences in Pb
19                  uptake rates can  be determined from experimental Pb concentrations that are higher than
20                  measured Pb in the environment. Plants that are hyperaccumulators of Pb and other
21                  metals may be used for phytoremediation at highly contaminated sites and there is a large
22                  body of literature on uptake of very high concentrations of metals by different species.
23                  This chapter focuses on environmentally relevant concentrations of Pb and also those
24                  studies with doses or exposures in the range of one or two orders of magnitude above
25                  current or ambient conditions, as described  in the Preamble. In freshwater ecosystems in
26                  the U.S., the average  Pb  concentration in surface water is 0.5 (ig Pb/L (Table 7-2).
27                  however, total Pb in water has been measured as high as 2,000 (ig Pb/L where mining
28                  and smelting operations have affected streams (Table 3-11).Studies with freshwater algae
29                  available since the 2006  Pb AQCD, are primarily limited to nominal media exposures at
30                  high concentrations of Pb with metal quantified in tissues. For example, the microalgae
31                  Spirulina platensis was demonstrated to accumulate Pb from Zarrouk culture medium in
32                  a concentration-dependent manner with nominal initial concentrations of 5,000, 10,000,
33                  30,000, 50,000 and 100,000 jig Pb/L (Pb  in medium was measured every two days
34                  thereafter), following a 10-day incubation period (Arunakumara et al., 2008). Pb
35                  concentrations accumulated by algae appeared to decrease when culture time increased


      November 2012                            7-104                 Draft - Do Not Cite or Quote

-------
 1                  from 2 to 10 days. This may have occurred as a result of a gradual recovery of growth
 2                  and an addition of biomass that would have reduced the concentration of Pb in algal
 3                  tissue. An aquatic moss, Fontinalis antipyretica, accumulated up to an average of 622 mg
 4                  Pb/kg dry weight over a 7-day nominal exposure to 20,700 (ig Pb/L despite saturation of
 5                  intracellular Pb concentrations after 5 days of exposure (Rau et al., 2007). Interestingly,
 6                  experimentation with concurrent Cu and Pb exposure indicated that the presence of Cu
 7                  increased the uptake of Pb by the green algae Chlamydomonas reinhardtii (Chen et al.,
 8                  201 Oc). The authors noted that, in the case of Cu-Pb binary exposures, uptake rates of Pb
 9                  exhibited complex non-linear dynamics in other aquatic organisms as well.

10                  Additional uptake studies conducted since the 2006 Pb AQCD, include new information
11                  for freshwater macrophytes. When exposed to nominal water concentrations of up to
12                  20,700 (ig Pb/L, floating (non-rooted) coontail plants (Ceratophyllum demersum)
13                  accumulated an average Pb concentration of 1,748 mg Pb/kg after 7 days, although this
14                  was not significantly higher than levels accumulated in the first day of exposure  (Mishra
15                  et al., 2006b). Induction of the antioxidant system improved the tolerance of the aquatic
16                  plant Najas indica for bioaccumulated Pb, allowing for increased biomass and the
17                  potential to accumulate additional Pb mass. High Pb accumulation (3,554 mg Pb/kg dry
18                  weight tissue following a 7-day exposure to 20,720 (ig Pb/L) was considered to be a
19                  function of plant morphology; as a submerged, floating plant, N.  indica provides a large
20                  surface area for the absorption of Pb  (Singh etal.. 2010).

21                  Given that atmospherically-derived Pb is likely to become sequestered in sediments
22                  (Section 7.2). uptake by aquatic macrophytes is a significant route of Pb removal from
23                  sediments, and a potential route for Pb mobilization into the aquatic food web. The rooted
24                  aquatic macrophyte Eleocharis acicularis was determined to be a hyperaccumulator of
25                  Pb in an 11-month bioaccumulation experiment with mine tailings. When grown in
26                  sediments containing 1,930 mg Pb/kg, the  maximum concentration of Pb in E. acicularis
27                  was determined to be 1,120 mg Pb/kg dry weight. However, calculated BCFs for Pb were
28                  all less than  one, indicating that Pb uptake, although high, was less efficient than for other
29                  metals present (Ha et al.. 2009).

30                  Aquatic plants inhabiting a wetland containing an average sediment Pb concentration of
31                  99 mg Pb/kg exhibited variable Pb tissue concentrations, but these do not appear to be
32                  related to macrophyte type (e.g., submerged, floating, emergent, etc.). Consequently, the
33                  authors concluded that uptake of Pb by aquatic plants appears to be dependent on species,
34                  at the exclusion of habitat or type. For instance, among the submerged plant species,
35                  Ceratophyllum demersum accumulated the greatest amount of Pb (22 mg Pb/kg dry
36                  weight), while Potamogeton malainus tissue contained the least amount of Pb, 2.4 mg
37                  Pb/kg dry weight (Bi et al.. 2007). Tissues of the floating plants Azolla imbricata and
      November 2012                            7-105                 Draft - Do Not Cite or Quote

-------
 1                  Spirogyra communis were found to contain 12 and 20 mg Pb/kg dry weight, respectively,
 2                  while emergent macrophytes Scirpus triqueter and Alternantheraphiloxeroides
 3                  accumulated 1.4 and 10 mg Pb/kg dry weight. Fritioff and Greger (2006) determined that
 4                  anywhere from 24-59% of the total Pb taken up by Potamogeton natans aquatic plants
 5                  was sequestered in the cell wall  fraction, depending on plant tissue and environmental Pb
 6                  concentration. More importantly, no translocation of Pb was observed when plant tissues
 7                  (leaf, stem, root) were exposed to Pb solutions separately (Fritioff and Greger, 2006).

 8                  Dwivedi et al. (2008) reared nine different species of aquatic plants in a fly-ash
 9                  contaminated medium containing approximately 7 mg Pb/kg dry weight. Not only did
10                  species exhibit different Pb accumulation efficiencies but they also compartmentalized
11                  sequestered Pb differently. The submerged macrophyte Hydrilla verticillata accumulated
12                  the greatest amount of Pb (approximately 180 mg Pb/kg dry weight tissue), but Pb was
13                  sequestered solely in the shoot tissue. In contrast, other plant species accumulated
14                  between 15  and 100 mg Pb/kg dry weight (Ranunculus scloralus andMarsilea
15                  quadrifolia) with the majority compartmentalizing the metal in root tissue, except for
16                  C. demersum andM quadrifolia, which also utilized shoot tissue for Pb storage (Dwivedi
17                  et al.. 2008).

18                  Pb concentrations in the root, leaf, and stem tissues of three aquatic plant species were
19                  found to correlate most closely with the concentration of the exchangeable Pb fraction
20                  (e-g-, the fraction of Pb that is easily and freely leachable  from the sediment). Authors
21                  noted that seasonal variations can alter the amount of Pb present in the exchangeable
22                  fraction, and that Pb was more likely than Cd or Cu to remain tightly bound to sediments,
23                  and therefore the relationship between total sediment Pb and Pb in aquatic plant tissues
24                  was weaker (Ebrahimpour and Mushrifatu 2009).

25                  Lemna sp., a free floating macrophyte, incubated in a water extract of waste ash
26                  containing 19 (ig Pb/L accumulated  3.5 mg Pb/kg dry weight over 7 days of exposure.
27                  Slight toxic effects, including suppression of growth, were observed over this exposure
28                  period, but this may have been a result of exposures to multiple metals in the water
29                  extract, including Cr, Mn, Cu, and Zn (Horvat et al., 2007). Lemna sp. was also
30                  demonstrated to be effective in the biosorption of Pb from solution, even in the presence
31                  of sediments (1 g per 700 mL water). Over 7 days of exposure to 3,600 and
32                  7,000 (ig Pb/L, plant biomass was found to contain an average of 2,900 and 6,600 mg/kg
33                  (wet weight) Pb, respectively, versus 200 and 300 mg/kg  (dry weight) in sediment (Kurd
34                  and Sternberg. 2008).

35                  Young Typha latifolia, another rooted macrophyte, were grown in analytically verified
36                  concentrations of 5,000 and 7,500 (ig/L Pb-spiked sediment for 10 days to determine
37                  their value as metal accumulators. Within the exposure period, plants exposed to the


      November 2012                            7-106                  Draft - Do Not Cite or Quote

-------
 1                  lower concentration were able to remove 89% of Pb, while 84% of the Pb present in the
 2                  higher treatment was taken up by T. latifolia. Pb concentrations measured in root and leaf
 3                  tissue ranged from 1,365 to 4,867 mg Pb/kg and 272 to 927 mg Pb/kg, respectively, and
 4                  were higher at the greater Pb exposure (Alonso-Castro et al. 2009).

 5                  Uptake studies available for aquatic macrophytes since the 2006 Pb AQCD, include some
 6                  studies where Pb was measured in field collected plants growing in metal-contaminated
 7                  areas. Common reeds (Phragmites australis) grown in metal-impacted aquatic
 8                  environments in Sicily, Italy, preferentially accumulated Pb in root and rhizome tissues
 9                  (Bonanno and Lo Giudice. 2010). Pb concentrations in water and sediment averaged
10                  0.4 (ig Pb/L and 2.7 mg Pb/kg. These levels yielded root and rhizome concentrations of
11                  17 and 15 mg Pb/kg, respectively, whereas stem and leaf Pb concentrations were lower
12                  (9.9 and  13 mg Pb/kg). These tissue concentrations were significantly correlated to both
13                  water and sediment concentrations (Bonanno and Lo Giudice. 2010). Conversely, the
14                  semi-aquatic  plant Ammania baccifera, grown in mine tailings containing 35 to 78 mg
15                  Pb/kg, did not accumulate analytically detectable levels of Pb in either root or shoot
16                  tissues, despite the fact that other metals (Cu, Ni, Zn) were bioaccumulated (Das and
17                  Maiti. 2007). This would indicate that at low/moderate environmental Pb concentrations,
18                  some plant species may not bioaccumulate significant (or measurable) levels of Pb.

19                  The average concentration of Pb in the tissues of rooted aquatic macrophytes (Callitriche
20                  verna, P. natans, C. demersum, Polygonum amphibium, Veronica beccabunga) collected
21                  from two metals-polluted streams in Poland (average sediment concentration 38 to 58 mg
22                  Pb/kg) was less than 30 mg Pb/kg. Pb bioaccumulation in plants was significantly
23                  correlated with sediment Pb concentrations (Samecka-Cymerman and Kempers. 2007). A
24                  similar significant correlation was established between reed sweet grass root Pb
25                  concentration and sediment Pb concentrations (Skorbiowicz. 2006).

26                  Pb tissue concentrations of aquatic plants P. australis and Ludwigia prostrata collected
27                  from wetlands containing an average of 52 mg Pb/kg in surficial sediments were
28                  predominantly in root tissues, indicating poor translocation of Pb from roots. In the
29                  former, Pb decreased from an average of 37 mg Pb/kg in roots to 17, 14, and
30                  12 mg Pb/kg  in rhizome, stem and leaf tissues, respectively, while L. prostrata Pb tissue
31                  concentrations decreased from  77 mg Pb/kg in fibrous root to 7 and 43 mg Pb/kg in stem
32                  and leaf tissues (Yang et al.. 2008a). The authors proposed that this diminished transfer
33                  ability explained the  relatively  low BCFs for Pb uptake in these two species, when
34                  compared with those of other metals.

35                  Despite no significant seasonal effect on surface water Pb concentrations, shining
36                  pondweed (Potamogeton lucens), a rooted aquatic macrophyte grown in an urbanized
37                  metal-contaminated lake in Turkey, exhibited seasonal alterations in Pb tissue
      November 2012                            7-107                 Draft - Do Not Cite or Quote

-------
 1                  concentrations. Average water Pb concentrations were 28 (ig Pb/L in spring, 27 (ig Pb/L
 2                  in summer, and 30 (ig Pb/L in autumn. Over this same time period, root tissue Pb
 3                  concentrations significantly increased from 6 mg Pb/kg dry weight in spring, to 9 mg
 4                  Pb/kg dry weight in summer, and to 10 mg Pb/kg dry weight in autumn (Duman et al..
 5                  2006). No differences were detected in stem Pb concentrations between spring and
 6                  summer (approximately 4 mg Pb/kg dry weight), but stem Pb concentrations were found
 7                  to be significantly higher in autumn (6 mg Pb/kg dry weight). In the same system,
 8                  P. australis plants accumulated the most Pb during winter: 103, 23, and 21 mg Pb/kg dry
 9                  weight in root, rhizome, and shoot tissue, respectively, in sediments containing 13 mg
10                  Pb/kg dry weight. By contrast, Schoenoplectus lacustris accumulated maximum rhizome
11                  and stem Pb concentrations of 5.1 and 7.3 mg Pb/kg dry weight in winter, but  sequestered
12                  the greatest amount of Pb in root tissues during the spring (30 mg Pb/kg dry weight) at a
13                  comparable sediment concentration, 18 mg Pb/kg dry weight (Duman et al.. 2007). The
14                  authors suggest that this indicated that metal uptake was regulated differently between
15                  species.

16                  Tree species that inhabit semi-aquatic environments have also been shown to absorb Pb
17                  from Pb-contaminated sediments. Bald-cypress trees (Taxodium distichum) growing in
18                  sediments of a refinery-impacted bayou in Louisiana accumulated significantly greater
19                  amounts of Pb than did trees of the same species growing in bankside soil, despite the
20                  lower Pb concentrations of sediments. Bankside soils contained greater than 2,700 mg
21                  Pb/kg versus concentrations of 10 to 424 mg Pb/kg in sediments, yet Pb concentrations in
22                  trees averaged 4.5 and 7.8 mg Pb/kg tissue, respectively (Devall et al.. 2006). The authors
23                  theorized that Pb was more readily released from sediments and that soil dispersion to the
24                  swamp sediments provides additional, if periodic, loads of Pb into the system. Willow
25                  seedlings planted in Pb-contaminated  sediment were more effective at removing Pb from
26                  the media than a diffusive gradient in thin film technique predicted (Jakl et al.. 2009).
27                  The  authors proposed that the plant's active mobilization of nutrients from soil during
28                  growth also resulted in increased Pb uptake and sequestration.

29                  Given that sediments are a significant sink for Pb entering aquatic systems, it is not
30                  surprising that rooted macrophytes bioaccumulate significant quantities of the metal.
31                  Although there are some similarities to Pb accumulation observed in terrestrial plants
32                  (e-g-, preferential sequestration of the  metal in root tissue), Pb appears to be more
33                  bioavailable in sediment than it is in soil. This may be a result of differences in plant
34                  physiology between aquatic and terrestrial plants (e.g., more rapid growth or more
35                  efficient assimilation of nutrients and ions from a water-saturated medium). While rooted
36                  macrophytes are likely to be chronic accumulators of Pb sequestered in sediments, aerial
37                  deposition of Pb into aquatic systems may result in pulsed inputs of labile Pb that would
38                  be available for uptake by floating macrophytes and algae.
      November 2012                            7-108                 Draft - Do Not Cite or Quote

-------
              7.4.4.2  Freshwater Invertebrates

 1                  Uptake and subsequent bioaccumulation of Pb in freshwater invertebrates varies greatly
 2                  between species and across taxa as previously characterized in the 2006 Pb AQCD. This
 3                  section expands on the findings from the 1986 Pb AQCD and 2006 Pb AQCD on
 4                  bioaccumulation and sequestration of Pb in aquatic invertebrates. In the case of
 5                  invertebrates, Pb can be bioaccumulated from multiple sources, including the water
 6                  column, sediment, and dietary exposures, and factors such as proportion of bioavailable
 7                  Pb, lifestage, age, and metabolism can alter the accumulation rate. In this section, new
 8                  information on Pb uptake from freshwater and sediments by invertebrates will be
 9                  considered, followed by a discussion on dietary and water routes of exposure and factors
10                  that influence species-specific Pb tissue concentrations such as invertebrate habitat and
11                  functional feeding group.

12                  In a recent uptake study in freshwater mussels available since the 2006 Pb AQCD, the
13                  Eastern elliptic  mussel (Elliptic complanatd) was shown to accumulate Pb rapidly from
14                  water and then reach an equilibrium with exposure level and tissue concentration by two
15                  weeks following average daily exposures of 1, 4, 14, 57 or 245 (ig Pb/L as Pb nitrate
16                  (Mosher et al. 2012). Tissue concentrations of Pb increased at an exposure-dependent
17                  rate for the first 14  days and then did not change significantly for the remainder of the 28-
18                  day exposure although mussels continued to accumulate Pb. At the end of the exposure
19                  period, average Pb  in tissue ranged from 0.33  to 898 mg Pb/kg. The authors concluded
20                  that the mussels were likely eliminating Pb via pseudo feces and through storage of Pb in
21                  shell.

22                  The 2006 Pb AQCD (U.S. EPA. 2006b) summarized studies of uptake of Pb from
23                  sediment by aquatic invertebrates and noted that sediment pore water, rather than bulk
24                  sediment,  is the primary route of exposure. However, a recent study suggests that in the
25                  midge, Chironomus riparius, total metal concentrations in bulk sediment are better
26                  predictors of metal accumulation than dissolved metal concentrations in sediment pore
27                  water based on bioaccumulation studies using contaminated sediments from six different
28                  sites (Roulier et al.. 2008a). Vink (2009) studied six river systems  and found that, for a
29                  range of metals, uptake by benthic organisms  (the oligochaete, Limnodrilus (Family
30                  Tubificidae) and the midge, C. riparius) from the sediment pore water (as compared with
31                  surface water) was  observed only occasionally, and solely for Pb. The physiological
32                  mechanisms of Pb uptake are still unclear but it is suggested that uptake and elimination
33                  of Pb obey different mechanisms than for other heavy metals.

34                  The 2006 Pb AQCD recognized the potential importance of the dietary uptake pathway
35                  as a source of Pb exposure for invertebrates. Specifically, in a study with the freshwater
36                  amphipod Hyalella azteca, dietary exposure was found to contribute to the chronic

      November 2012                            7-109                 Draft - Do Not Cite or Quote

-------
 1                  toxicity of Pb, while acute toxicity was unaffected (Besser et al., 2004). Since the
 2                  2006 Pb AQCD, additional studies have considered the relative importance of water and
 3                  dietary uptake of Pb in aquatic invertebrates. A stable isotope technique was used to
 4                  simultaneously measure uptake of environmentally relevant concentrations of Pb
 5                  (10.4 (ig Pb/L) in the water column by the freshwater cladoceran D. magna directly from
 6                  water and through food, the green algae Pseudokirchneriella subcapitata. (Komjarova
 7                  and Blust 2009a). D. magna accumulated the metal from both sources, but the relative
 8                  proportion of uptake from each source changed over the exposure period. After the first
 9                  day of exposure, 12% of accumulated Pb was determined to have been absorbed from
10                  dietary (algal) sources, but this percentage decreased by day four of exposure to 4%. Pb
11                  absorbed from water exposure only resulted in Daphnia body burdens of approximately
12                  62.2 mg Pb/kg dry weight (300 (imol Pb/kg dry weight), and was similar to the amount
13                  absorbed by algae (Komjarova and Blust. 2009a). In a comparison of dietary and
14                  waterborne exposure as sources of Pb to aquatic invertebrates, no correlation between Pb
15                  uptake and dietary exposure was observed in the amphipod H. azteca fBorgmann et al..
16                  2007).

17                  Stable isotope analysis was to used measure uptake and elimination simultaneously in
18                  net-spinning caddisfly larvae (Hydropsyche sp.) exposed to aqueous Pb concentrations of
19                  0.2 (control) or 0.6 (ig Pb/L for 18 days (Evans et al., 2006). The measured uptake
20                  constant for Pb in this study was 7.8 g/dry weight per day, and the elimination rate
21                  constant of 0.15/day for Pb-exposed larvae was similar in both presence and absence of
22                  the metal in the water. Tissue concentrations ranged from approximately 15 to 35 mg
23                  Pb/kg. Hydropsychid Pb BCFs ranged from 41 to 65, and averaged 54, indicating a
24                  relatively high accumulation when compared to other metals tested (average BCF of 17
25                  for Cd, 7.7 for Cu, and 6.3 for Zn) (Evans et al.. 2006).

26                  Recent reports on Pb distribution in freshwater organisms generally support the findings
27                  of the 2006 Pb AQCD that Pb is primarily sequestered in the gills, hepatopancreas, and
28                  muscle. Uptake of Pb by the crayfish Cher ax destructor exposed to nominal
29                  concentration of 5,000 (ig Pb/L as Pb nitrate for 21 days resulted in accumulation at the
30                  highest concentration in gill,  followed by exoskeleton >mid-gut gland >muscle
31                  >hemolymph (Morris et al.. 2005). Body burden analysis following  96 hour nominal
32                  exposure to 50, 100 and 500 (ig Pb/L as Pb nitrate in the freshwater snail Biomphalaria
33                  glabrata indicated that bioaccumulation increased with increasing concentrations of Pb
34                  and the highest levels were detected in the digestive gland (Ansaldo et al.. 2006).

35                  When the relationship between invertebrate habitat (epibenthic and benthic) and
36                  environmental Pb bioaccumulation was investigated, De Jonge et al. (2010) determined
37                  that different environmental fractions of Pb were responsible for invertebrate uptake and
      November 2012                            7-110                 Draft - Do Not Cite or Quote

-------
 1                  exposure. Pb uptake by benthic invertebrate taxa was not significantly correlated to AVS
 2                  Pb levels, but rather to total sediment concentrations (De Jonge et al.. 2009). Conversely,
 3                  epibenthic invertebrate Pb body burdens were better correlated to AVS concentrations,
 4                  rather than total Pb sediment concentrations (De Jonge et al.. 2010). For instance, the
 5                  biologically available Pb (e.g., bound to metal-rich granules or metallothioneins)
 6                  accumulated by the oligochaete Tubifex tubifex was determined to correlate with
 7                  sediment SEM-AVS Pb concentrations (De Jonge et al., 2011). Similarly, Desrosiers et
 8                  al. (2008) reported that Pb accumulation by chironomid larvae from St. Lawrence river
 9                  sediments was significantly correlated to both total Pb and reactive  Pb sediment
10                  concentrations.

11                  Both inter- and intra-specific difference in Pb uptake and bioaccumulation may occur in
12                  macroinvertebrates of the same functional-feeding group. Cid et al.  (2010) reported
13                  significant differences in Pb bioaccumulation between field collected Ephoron virgo
14                  mayflies and Hydro psyche sp, caddisflies, with only the mayfly exhibiting increased Pb
15                  tissue concentrations when collected from Pb-contaminated sites; the caddisfly Pb tissue
16                  concentrations were similar between reference and Pb-contaminated areas. The authors
17                  also examined the lifestage specific accumulation of Pb for E. virgo mayflies, and
18                  although there was no statistical difference in Pb tissue concentrations between different
19                  lifestages, Pb bioaccumulation did change as  mayflies aged (Cid et  al., 2010).

20                  Reported BAF values for Pb in aquatic invertebrates from the 2006 Pb AQCD ranged
21                  from 499 to 3,670 [Table AX7-2.3.2 (U.S. EPA. 2006c)1. Since the 2006 Pb AQCD,
22                  additional BAF values have been established  for invertebrates in field studies which tend
23                  to be higher than BCF values calculated in laboratory exposures (Casas et al.. 2008;
24                  Gagnon and Fisher. 1997). A complicating factor in establishing BAF values is that
25                  laboratory studies usually assess uptake in water-only or sediment only exposures while
26                  field studies take into account dietary sources of Pb as well as waterborne Pb resulting in
27                  BAF values that are frequently 100-1,000 times larger than BCF values for the same
28                  metal and species (DeForest et al., 2007). The EPA Framework for  Metals Risk
29                  Assessment states that the latest scientific data on bioaccumulation  do not currently
30                  support the use of BCFs and BAFs when applied as generic threshold criteria for the
31                  hazard potential of metals (U.S. EPA. 2007c). See Section 7.3.3 for further discussion.

32                  As reviewed by Wang and Rainbow (2008) and supported by additional studies reviewed
33                  in the present document, there are considerable differences between species in the
34                  amount of Pb taken up from the environment and in the levels of Pb retained in the
35                  organism. The bioaccumulation and subsequent toxicity of Pb to aquatic organisms
36                  (Section 7.4.5) are closely linked to the environmental fate of the metal under variable
      November 2012                             7-111                  Draft - Do Not Cite or Quote

-------
 1                  environmental conditions (Sections 3.3 and 7.2) as they are highly dependent upon the
 2                  proportion of free metal ions in the water column.
              7.4.4.3  Freshwater Vertebrates

 3                  Uptake of Pb by vertebrates considered here includes data from fish species as well as a
 4                  limited amount of new information on amphibians and aquatic mammals. The
 5                  bioaccessibility and bioavailability of Pb is affected by abiotic and biotic modifying
 6                  factors considered in Sections 7.4.2 and 7.4.4. In fish, Pb is taken up from water via the
 7                  gills and from food via ingestion. Amphibians and aquatic mammals are exposed to
 8                  waterborne Pb primarily through dietary sources. In the 2006 Pb AQCD, dietary Pb was
 9                  recognized as a potentially significant source of exposure to all vertebrates since Pb
10                  adsorbed to food, particulate matter and sediment can be taken up by aquatic organisms.

11                  Since the 2006 Pb AQCD, tissue accumulation of Pb via gill and dietary uptake has been
12                  further characterized in freshwater fish and new techniques such as the use of stable
13                  isotopes  have been applied to further elucidate bioaccumulation of Pb. For example,
14                  patterns of uptake and subsequent excretion of Pb in fish as measured by isotopic ratios
15                  of Pb in each tissue can determine whether exposure was due to relatively long term
16                  sources (which favor accumulation in bone) or short term sources (which favors
17                  accumulation in liver) (Miller et al.. 2005). Recent information since the 2006 Pb AQCD,
18                  on uptake of Pb by fish from freshwater is reviewed below, followed by studies on
19                  dietary uptake as a route of Pb exposure. Next, tissue accumulation patterns in fish
20                  species are reported with special consideration of the anterior intestine as a newly
21                  identified target of Pb from dietary exposures. Finally, studies that report Pb tissue
22                  concentrations in amphibians, reptiles and freshwater mammals are considered.


                    Freshwater Fish

23                  Pb uptake in freshwater fish is accomplished largely via direct uptake of dissolved Pb
24                  from the water column through gill surfaces and by ingestion of Pb-contaminated diets.
25                  According to the data presented in the 2006 Pb AQCD (U.S. EPA. 2006b). accumulation
26                  rates of Pb are influenced by both environmental factors, such as water pH, DOC, and
27                  Ca2+ concentrations, and by species-dependent factors, such as metabolism, sequestration,
28                  and elimination capacities. The effects of these variables on Pb bioaccumulation in fish
29                  are largely identical to the effects observed for invertebrates (discussed above).

30                  Pb in fish is primarily found in bone, gill, blood, kidney and scales (Spry and Wiener.
31                  1991). Since the 2006 Pb AQCD, multiple studies on uptake of Pb from water by fathead
      November 2012                            7-112                  Draft - Do Not Cite or Quote

-------
 1                  minnow and subsequent tissue distribution have been conducted. Spokas et al. (2006)
 2                  showed that Pb accumulates to the highest concentration in gill when compared to other
 3                  tissues over a 24-day exposure. This pattern was also observed in larval fathead minnows
 4                  exposed to 26 (ig Pb/L for 10-30 days, where gill exhibited the highest Pb concentration
 5                  compared to carcass, intestine, muscle and liver (Grosell et al., 2006a). In the larval
 6                  minnows, Pb concentration in the intestine exhibited the highest initial accumulation of
 7                  all tissues on day 3 but then decreased for the remainder of the experiment while
 8                  concentrations in the other organs continued to increase. By day 30, gill tissue exhibited
 9                  the highest Pb concentration (approximately 120 mg Pb/kg), followed by whole fish and
10                  carcass (whole fish minus gill, liver, muscle and intestine) Pb concentrations
11                  (approximately 70 to 80 mg Pb/kg). However, in considering overall internal Pb body
12                  burden, nearly 80% was largely concentrated in the bone tissue, while gill contributed
13                  <5%.

14                  In another study with fathead minnow, chronic (300 day)  exposure to 120 (ig Pb/L
15                  resulted in accumulation of approximately 41 mg Pb/kg tissue, although this number was
16                  decreased from initial body burdens of greater than 104 mg Pb/kg at test initiation (Mager
17                  et al.. 2010). Tissue distribution at 300 days was consistent with Grosell et al. (2006a)
18                  with highest concentration in gill, followed by kidney, anterior intestine, and carcass.
19                  Addition of humic acid and carbonate both independently reduced uptake of Pb in these
20                  fish over the exposure time period. Interestingly, fathead minnow eggs collected daily
21                  during 21 day breeding assays that followed the chronic exposure described above
22                  accumulated similar levels of Pb from the test solutions regardless of Pb concentration or
23                  water chemistry (e.g., addition of humic acid and carbonate) (Mager et al., 2010). Direct
24                  acute exposure from water rather than parental transfer accounted for the majority of the
25                  Pb accumulation in eggs. Similarly, exposure offish to 32.5 (ig Pb/L in base water for
26                  150 days resulted in fathead minnow whole body concentrations of approximately 31 mg
27                  Pb/kg, with the most rapid accumulation rate occurring within the first 10 days of
28                  exposure, followed by an extended period of equilibrium  (Mager etal.. 2008). In this
29                  same study, fish were tested in two additional treatments: 36.7 (ig Pb/L in hard water
30                  (Ca2+ 500 (iM) or 38.7 (ig Pb/L in humic acid supplemented water (4 mg/L). While the
31                  addition of humic acid significantly reduced Pb bioaccumulation in minnows  (to
32                  approximately 10.4 mg Pb/kg on a whole body basis), Ca2+ sulfate did not alter uptake.
33                  Despite the fact that Ca2+-mediated Pb toxicity occurred in larval fathead minnow, there
34                  was no concurrent effect on whole body Pb accumulation.

35                  Uptake studies in other freshwater teleosts have generally followed the pattern of Pb
36                  uptake described above for fathead minnow. In the cichlid, Nile tilapia (Oreochromis
37                  niloticus), Pb accumulated significantly in gill (45.9 ±34.4 (ig/g dry weight at
38                  2,070 (ig Pb/L),  57.4 ±26.1 (ig/g dry weight at 4,100 (ig Pb/L) and liver (14.3 (ig/g dry
      November 2012                            7-113                  Draft - Do Not Cite or Quote

-------
 1                   weight at2,070 jig Pb/L) and 10.2 (ig/g dry weight at 4,100 jig Pb/L) during a 14-day
 2                   nominal exposure (as Pb nitrate) (Atli and Canli. 2008). In rainbow trout exposed to
 3                   100 (ig Pb/L (as Pb acetate) for 72 hours, the accumulation in tissues was gill >kidney
 4                   >liver and this same pattern was observed in all concentrations tested
 5                   (100-10,000 ng Pb/L) (Suicmez et al.. 2006). In contrast to uptake in teleosts, in
 6                   Pb-uptake studies with the Chondrostei fish Chinese Sturgeon (Acipenser sinensis),
 1                   muscle tissue accumulated higher levels  of Pb than gills (Hou et al., 2011).

 8                   Sloman et al. (2005) investigated the uptake of Pb in dominant-subordinate pairings of
 9                   rainbow trout exposed to 46 (ig/L or 325 \ig Pb/L (as Pb nitrate) for 48 hours. Significant
10                   Pb accumulation in gill, liver and kidney was only observed in the highest concentration.
11                   Pb accumulated preferentially in liver of subordinate trout when compared to dominant
12                   trout. Brown trout (Salmo truttd) exposed to aqueous Pb concentrations ranging from 15
13                   to 46 (ig Pb/L for 24 days accumulated 6 mg Pb/kg dry weight in gill tissue and Pb
14                   concentrations in liver tissue reached 14  mg Pb/kg dry weight. Interestingly,  Pb in gill
15                   tissue peaked on day 11 and decreased thereafter, while liver Pb concentrations increased
16                   steadily over the exposure period, which may indicate translocation of Pb in brown trout
17                   from gill to liver (Heier et al., 2009).

18                   Zebrafish (Danio rerio) Pb  uptake rates from media containing 5.2 (ig Pb/L was
19                   significantly increased  by neutral pH (versus a pH of 6 or 8) and by Ca2+ concentrations
20                   of 0.5 mM; uptake rate of Pb was increased from 10 L/kg-h to 35 L/kg-h by increasing pH
21                   from 6 to 7, and from 20 L/kg-h to 35 L/kg-h by increasing Ca2+ concentration from 0.1 to
22                   0.5 mM  (Komjarova and Blust 2009c). This study also demonstrated that zebrafish gill
23                   tissue is  the main uptake site for the metal, as Pb concentrations in these tissues were up
24                   to eight times as high as that in  other tissues.

25                   The Eurasian silver crucian carp (Carassius auratus) collected from a pond containing an
26                   average of 1,600 mg Pb/kg  in the sediments exhibited increased average Pb whole body
27                   burden of 36.5 mg Pb/kg dry weight (range 12 to 68 mg Pb/kg dry weight) (Khozhina
28                   and Sherriff. 2008). Pb was primarily sequestered in skin, gill, and bone tissues, but was
29                   also detected at elevated levels in muscle and liver tissues, as  well as in eggs. Two fish
30                   species (Labeo rohita and Ctenopharyngodon idella) collected from the Upper Lake of
31                   Bhopal,  India with average  Pb concentration 30 (ig Pb/L in the water column contained
32                   elevated Pb tissue concentrations (Malik et al.. 2010). However, while liver and kidney
33                   Pb concentrations were similar between the two species (1.5 and 1.1 mg Pb/kg tissue and
34                   1.3 and 1.0 mg Pb/kg tissue for C. idella and L. rohita, respectively), they accumulated
35                   significantly different amounts of Pb in gill and muscle tissues. C. idella accumulated
36                   more than twice the Pb in these tissues (1.6 and 1.3 mg Pb/kg) than did  L. rohita (0.5 and
37                   0.4 mg Pb/kg).
      November 2012                             7-114                  Draft - Do Not Cite or Quote

-------
 1                   The studies reviewed above generally support the conclusions of the 2006 Pb AQCD
 2                   (U.S. EPA. 2006b) that the gill is a major site of Pb uptake in fish and that there are
 3                   species-dependent differences in the rate and pattern of Pb accumulation. As indicated in
 4                   the 2006 Pb AQCD, exposure duration can be a factor in Pb uptake from water. In a
 5                   30-day exposure study, Nile tilapia fmgerlings had a three-fold increase in Pb uptake at
 6                   the gill on day 30 compared to Pb concentration in gill  at day 10 and 20 (Kamaruzzaman
 7                   et al., 2010). In addition to uptake at the gill, a time-dependent uptake of Pb into kidney
 8                   in rainbow trout exposed to 570 (ig Pb/L for 96 hours (Patel et al.. 2006) was observed.
 9                   Pb was accumulated preferentially in the posterior kidney compared to the anterior
10                   kidney. A similar pattern was observed by Alves and Wood (2006) in a dietary exposure.
11                   In catla (Catla catld) fmgerlings, the accumulation pattern of Pb was  kidney >liver >gill
12                   >brain >muscle in both 14 day and 60 day Pb exposures (Palaniappan et al., 2009). In
13                   multiple studies with fathead minnow at different exposure durations, tissue uptake
14                   patterns were similar at 30 days (Grosell et al., 2006a) and 300 days (Mager et al., 2010).
15                   In the larval minnows, Pb concentration in the intestine exhibited the  highest initial
16                   accumulation of all tissues on day 3 but then decreased for the remainder of the
17                   experiment while concentrations in the other organs continued to increase (Grosell et al..
18                   2006a). By day 30, gill tissue exhibited the highest Pb concentration followed by whole
19                   fish and carcass (whole fish minus gill, liver, muscle and intestine). The most rapid rate
20                   of Pb accumulation in this species occurs within the first 10 days of exposure (Mager et
21                   al.. 2008). African catfish (Clarias gariepinus) exposed to nominal Pb concentrations of
22                   50 to 1,000 (ig Pb/L (as Pb nitrate) for 4 weeks accumulated significant amounts of Pb in
23                   heart (520-600 mg Pb/kg), liver (150-242 mg Pb/kg), and brain (120-230 mg Pb/kg)
24                   tissues (Kudirat 2008). Doubling the exposure time to  8 weeks increased sequestration of
25                   Pb in these tissues as well as in skin (125-137.5 mg Pb/kg) and ovaries (30-60 mg Pb/kg).

26                   Since the 2006 Pb AQCD, several studies have focused on dietary uptake of Pb in
27                   teleosts. Metals have been shown to assimilate differently in tissues depending on the
28                   exposure  route (Rozon-Ramilo et al.. 2011;  Meyer etal.. 2005). Alves et al. (2006)
29                   administered a diet of three concentrations of Pb (7, 77 and 520 mg Pb/kg dry weight) to
30                   rainbow trout for 21 days. Doses were calculated to be  0.02 (ig Pb/day (control),
31                   3.7 (ig Pb/day (low concentration), 39.6 (ig Pb/day (intermediate concentration) and
32                   221.5 (ig  Pb/day (high concentration). Concentrations in the  study were selected to
33                   represent environmentally relevant concentrations in prey. After 21 days exposure to the
34                   highest concentration, Pb accumulation was greatest in the intestine, followed by carcass,
35                   kidney and liver leading the authors to hypothesize that the intestine is the primary site of
36                   exposure  in dietary uptake of Pb. All tissues, (gill, liver, kidney, intestine, carcass)
37                   sequestered Pb in a dose-dependent manner. The gills had the greatest concentration of
38                   Pb on day 7(8.0 mg Pb/kg tissue wet weight) and this accumulation decreased to
39                   2.2 mg Pb/kg tissue wet weight by the end of the experiment suggesting that the Pb was

      November 2012                             7-115                  Draft - Do Not  Cite or Quote

-------
 1                   excreted or redistributed (Alves et al., 2006). Furthermore, with increasing dietary
 2                   concentrations, the percentage of Pb retained in the fish decreased. Additionally, in this
 3                   study red blood cells were identified as a reservoir for dietary Pb. Plasma did not
 4                   accumulate significant Pb (0.012 mg Pb kg wet weight in the high dose), however, Pb
 5                   was elevated in blood cells (1.5 mg Pb kg wet weight in the high dose) (Alves et al..
 6                   2006).

 7                   Additional studies have supported the anterior intestine as a target for Pb in fish. Nile
 8                   tilapia exposed to dietary Pb for 60 days (105, 418, and 803 mg Pb/kg dry weight)
 9                   accumulated the greatest concentration of Pb in the intestine, followed by the stomach
10                   and then the liver (Dai et al.. 2009a). The amount of Pb in tissue increased with
11                   increasing dietary Pb concentration. In a 42 day chronic study of dietary uptake in
12                   rainbow trout, fish fed 45 or 480 mg Pb/kg, accumulated Pb preferentially in anterior
13                   intestine (Alves and Wood. 2006). Pb accumulation in the gut was followed by bone,
14                   kidney, liver, spleen, gill, carcass, brain and white muscle (Alves and Wood. 2006). Ojo
15                   and Wood (2007) investigated the bioavailability of ingested Pb within different
16                   compartments of the rainbow trout gut using an in vitro gut sac technique. Although a
17                   significant increase in Pb uptake was observed in the mid-intestines, this was determined
18                   to be much lower than Pb uptake rates via gill surfaces. However, given that intestinal
19                   uptake rate for Pb did not significantly differ from those derived for essential metals
20                   (e.g., Cu, Zn, and Ni), this uptake route is likely to be significant when aqueous Pb
21                   concentrations are low and absorption via gill surfaces is negligible (Ojo and Wood.
22                   2007).

23                   Following a chronic 63-day dietary exposure to Pb, male zebrafish had significantly
24                   increased Pb body burdens, but did not exhibit any significant impairment when
25                   compared with controls. Fish were fed diets consisting of field-collected Nereis
26                   diversicolor oligochaetes that contained 1.7 or  33 mg Pb/kg dry weight. This resulted in a
27                   daily Pb dose of either 0.1 or 0.4 mg Pb/kg (Boyle etal.. 2010). At the end of the
28                   exposure  period, tissue from male fish reared on the high-Pb diet contained
29                   approximately 0.6 mg Pb/kg wet weight, as compared with approximately 0.48 mg  Pb/kg
30                   wet weight in the low-Pb dietary exposure group. Pb level was elevated in female fish fed
31                   the high-Pb diet, but not significantly so.

32                   Ciardullo et al. (2008)  examined bioaccumulation of Pb in rainbow trout tissues
33                   following a 3-year chronic dietary exposure to the metal. Diet was determined to contain
34                   0.19 mg Pb/kg wet weight. Fish skin accumulated the greatest Pb concentrations (0.02 to
35                   0.05 mg Pb/kg wet weight), followed by kidney, gills, liver, and muscle. Pb accumulation
36                   in muscles (.005 mg Pb/kg) remained constant  over all sampled growth stages (Ciardullo
37                   et al., 2008).  The authors concluded that dietary Pb was poorly absorbed by rainbow
      November 2012                             7-116                 Draft - Do Not Cite or Quote

-------
 1                  trout. Comparison of dietary and water-borne exposures suggest that although
 2                  accumulation of Pb can occur from dietary sources, toxicity does not correlate with
 3                  dietary exposure,  but does correlate with gill accumulation from waterborne exposure
 4                  (Alves et al.. 2006). Comparison of uptake rates across the gut and gill have shown that
 5                  transporter pathways in the gill have a much higher affinity for Pb than do similar
 6                  pathways in the gut (Ojo and Wood. 2007).

 7                  Since the 2006 Pb AQCD, several field studies have considered Pb uptake and
 8                  bioaccumulation in fish as a tool for environmental assessment. Pb tissue concentrations
 9                  were elevated  in several species offish exposed in the field to Pb from historical mining
10                  waste, and  blood Pb concentrations were highly correlated with elevated tissue
11                  concentrations, suggesting that blood sampling may be a useful and potentially non-lethal
12                  monitoring technique (Brumbaugh et al.. 2005).

13                  This review of the recent literature indicates that the primary and most efficient mode of
14                  Pb absorption  for freshwater fish is assimilation of labile Pb via gill surfaces; recent
15                  research indicates that chronic dietary Pb exposure may result in some Pb
16                  bioaccumulation although it is not the predominant route of exposure.  Nevertheless, if
17                  benthic invertebrates comprise a large portion offish diets in chronically contaminated
18                  systems, assimilated Pb loads may be significant. This was demonstrated by Boyle et al.
19                  (2010).  who showed that laboratory diets consisting of less than one third field-collected
20                  Pb-contaminated invertebrates were  sufficient to raise fish tissue Pb levels. However,
21                  data from field sites suggest that fish accumulation of Pb from dietary sources is highly
22                  variable and may  be strongly dependent on the physiology of individual species and
23                  absorption  capacities.


                    Amphibians

24                  Since the 2006 Pb AQCD, there are a few recent field measurements and laboratory-
25                  based studies that consider uptake of Pb in amphibians. Whole body Pb measured in three
26                  species  of field-collected tadpoles in the Mobile-Tensaw River Delta in Alabama
27                  averaged 1.19  mg Pb/kg dry weight in Rana clamitans, 0.65 mg Pb/kg dry weight in
28                  Rana catesbeiana and 1.32 mg Pb/kg dry weight in Hyla cinerea fAlbrecht et al.. 2007 j.
29                  Blood-Pb levels in Ozark hellbender salamanders (Cryptobranchus alleganiensis
30                  bishopi), a  candidate species for the Endangered Species Act, ranged from 0.044 to
31                  0.055 mg/kg dry whole blood weight, in three rivers in Missouri (Huang et al., 2010). In
32                  the same study, Pb-blood levels were measured from Eastern hellbenders
33                  (Cryptobranchus  alleganiensis alleganiensis), a species of concern, collected from four
34                  rivers and ranged  from 0.075 to 0.088 mg  Pb/kg dry whole blood weight.
      November 2012                            7-117                  Draft - Do Not Cite or Quote

-------
 1                  In a chronic laboratory-based study with tadpoles of the Northern Leopard frog (Rana
 1                  pipiens), Pb tissue concentrations were evaluated following exposures to 3, 10, and
 3                  100 (ig Pb/L from embryo to metamorphosis. The tadpole tissue concentrations ranged
 4                  from 0.1 to 224.5 mg Pb/kg dry mass and were positively correlated to Pb concentrations
 5                  in the water (Chen et al., 2006b). Dose-dependent bioaccumulation of Pb was observed in
 6                  the livers of tadpoles of the African clawed frog (Xenopus laevis) exposed to nominal
 7                  concentrations ranging from 1.0 to 30,000 (ig Pb/L (3 to 115 mg Pb/kg wet weight) for
 8                  12 days (Mouchet et al.. 2007). Pb concentrations were measured in livers, bodies
 9                  without liver and whole bodies in Southern leopard frog (Rana sphenocephald) tadpoles
10                  exposed to Pb in sediment (45 to 7,580 mg Pb/kg dry weight) with corresponding pore
11                  water concentrations of 123 to 24,427 (ig Pb/L from embryonic stage to metamorphosis
12                  (Sparling et al.. 2006). There was 100% mortality at 3,940 mg Pb/kg and higher. In all
13                  body residues analyzed there was a significant positive correlation between Pb in
14                  sediment and Pb in sediment pore water. Concentrations of Pb in liver were similar to
15                  results with whole body and bodies without liver indicating that Pb is not preferentially
16                  sequestered in liver.


                    Reptiles

17                  Recent field surveys of Pb in water snakes since the 2006 Pb AQCD, indicate that Pb is
18                  bioaccumulated in several species. Water snakes spend time in terrestrial and aquatic
19                  habitats and could potentially be exposed to atmospherically deposited-Pb in both
20                  environments. Average Pb levels in whole body samples  of Eastern Ribbon Snakes
21                  (Thamnophis sauntus) collected from the Mobile-Tensaw River, a large watershed that
22                  drains more than 75% of Alabama were 0.35 ± 0.12 mg Pb/kg dry weight) (Albrecht et
23                  al.. 2007). Burger et al. (2007) measured Pb levels in blood, kidney, liver, muscle and
24                  skin from water snakes, (Nerodia sepedori) collected from an urban/suburban canal in
25                  New Jersey. Pb was highest in skin (0.467 mg Pb/kg wet weight) followed by kidney
26                  (0.343 mg Pb/kg wet weight) blood (0.108 mg Pb/kg wet weight), muscle (0.103 mg
27                  Pb/kg wet weight) and liver (0.063 mg Pb/kg wet weight). No interspecies differences
28                  were observed in blood Pb (range 0.04 to 0.1 mg Pb/kg) from field-collected banded
29                  water snakes (Nerodia fasciata), brown water snakes (N.  taxispilota) and cottonmouth
30                  (Agkistrodon piscivorus) from a reference area and an area contaminated by chemical and
31                  radiation releases from the 1950's to the 1980's at the Department of Energy's Savannah
32                  River site in South Carolina (Burger et al.. 2006). Cottonmouth and brown water snake
33                  from the exposed site had significantly higher levels of Pb in tail muscle when compared
34                  to the reference creek.
      November 2012                            7-118                 Draft - Do Not Cite or Quote

-------
                     Mammals

 1                   Pb bone levels in Eurasian otters (Lutra lutrd) measured in dead individuals collected in
 2                   southwest England fell by 73% between 1992 and 2004 (Chadwicket al.. 2011). Annual
 3                   mean bone Pb levels were 446 (ig Pb/kg in 1992 and 65 (ig Pb/kg in 2004. The 73%
 4                   decline of Pb in otter bones from 1992 to 2004 was found to coincide with legislative
 5                   controls on Pb emissions implemented in the  U.K. starting in 1986. A positive correlation
 6                   with stream  sediment Pb and bone Pb was also observed in this study. The strength of
 7                   this correlation decreased with increasing Ca2+ in streams.
              7.4.4.4  Food Web

 8                  In the 2006 Pb AQCD, trophic transfer of Pb through aquatic food chains was considered
 9                  to be negligible (U.S. EPA. 2006c). Concentrations of Pb in the tissues of aquatic
10                  organisms were found to be generally higher in algae and benthic organisms and lower in
11                  higher trophic-level consumers indicating that Pb was bioaccumulated but not
12                  biomagnified (U.S. EPA, 2006c; Eisler. 2000). Recent literature since the
13                  2006 Pb AQCD, provides evidence of the potential for Pb to be transferred in aquatic
14                  food webs. Other studies indicate Pb is decreased with increasing trophic level. This
15                  section incorporates recent literature on transfer of Pb through freshwater aquatic food
16                  chains including the application of stable isotope techniques to trace the accumulation
17                  and dilution of metals through producers and consumers.

18                  Pb was transferred through at least one trophic level in El Niagara reservoir,
19                  Aguascalientes, Mexico, a freshwater ecosystem that lacks fishes (Rubio-Franchini et al..
20                  2008). Pb was quantified in sediment (0.55 mg Pb/kg to 21 mg Pb/kg), water (5.8 to
21                  39 (ig Pb/L), and zooplankton samples of this freshwater system.  BAFs were calculated
22                  for predatory and grazing zooplanktonic species. The BAF of the rotifer A. brightwellii
23                  (BAF 49,300) was up to four times higher than the grazing cladocerans D. similis (BAF
24                  9,022) andM micrura (BAF 8,046). According to the authors, since M. micrura are prey
25                  for A. brightwellii this may explain the biomagnifications of Pb observed in the predatory
26                  rotifer and provides evidence that Pb biomagnifies at intermediate trophic levels.

27                  The relative contribution of water and food as source of trace metals including Pb was
28                  investigated in the larvae of the alderfly Sialis velata fCroisetiere et al. 2006). Its prey,
29                  the midge (C. riparius) was reared in the laboratory and then exposed to trace elements in
30                  a metal-contaminated lake for one week prior to being fed to S. velata. During the one-
31                  week exposure period of C. riparius to the contaminated water, five of six trace  elements,
32                  including Pb, reached steady state  within C. riparius. Alderfly larvae were held in the lab
33                  in uncontaminated lake water and feed one of the treated C. riparius per day for up to six


      November 2012                             7-119                 Draft - Do Not Cite or Quote

-------
 1                   days to measure Pb uptake via prey. A separate group of alderfly larvae were exposed
 2                   directly to the contaminated lake water for six days and fed uncontaminated C. riparius
 3                   while a third group was exposed to Pb via prey and water. Trace metal concentrations in
 4                   S. velata that consumed contaminated C. riparius increased significantly compared to
 5                   S. velata in water-only exposures. Food was concluded to be the primary source of Pb
 6                   (94%) to these organisms, not Pb in the water.

 7                   The trophic transfer of Pb from the sediment dwelling polychaete worm N. diversicolor
 8                   to the invertebrate polychaete predator Nereis virens provides additional evidence for
 9                   assimilation of Pb by a predator and the potential for further transport up the food chain
10                   (Rainbow et al.. 2006). N. virens significantly accumulated Pb from a diet of
11                   N. diversicolor and there was a significant inverse linear relationship between the trophic
12                   transfer coefficient and prey Pb concentration. In the same study, another predator, the
13                   decapod Palaemonetes varians, did not significantly accumulate Pb from N. diversicolor
14                   indicating that trophic transfer is dependent on species-specific differences in metal
15                   assimilation efficiencies and accumulation patterns.

16                   In a recent dietary metal study, field-collected invertebrates representing ecologically
17                   relevant sources of Pb were fed to zebrafish, to assess bioavailability of this metal via
18                   food. The polychaete worm N. diversicolor was collected from two sites; an estuary
19                   contaminated with Pb and a reference site with low metal concentrations (Boyle et al..
20                   2010). Male zebrafish fed Pb-enriched N. diversicolor had significant increases in whole-
21                   body Pb burden when compared to zebrafish fed prey from the  reference site, brine
22                   shrimp or flake food diets. There was a trend toward increased Pb levels in females under
23                   the same dietary regimen. In this study, deposit feeding invertebrates were shown to
24                   mobilize sediment-bound metals in the food chain since zebrafish were exposed only to
25                   biologically incorporated metal.

26                   The concentration of Pb in the tissues of various aquatic organisms was measured during
27                   the biomonitoring of mining-impacted stream systems in Missouri. Generally, Pb
28                   concentrations decreased with increasing trophic level: detritus contained 20 to 60 mg
29                   Pb/kg dry weight, while periphyton and algae contained 1 to 30 mg Pb/kg dry weight;
30                   invertebrates and fish collected from the same areas exhibited Pb tissue concentrations of
31                   0.1 to 8 mg Pb/kg dry weight (Besser et al.. 2007). In addition,  Pb concentrations in
32                   invertebrates (snails, crayfish, and other benthos) were negatively correlated with Pb
33                   concentrations in detritus, periphyton, and algae. Fish tissue concentrations, however,
34                   were consistently correlated only with detritus Pb concentrations (Besser et al.. 2007).

35                   Other studies have traced Pb  in freshwater aquatic food webs and have found no  evidence
36                   of biomagnification of Pb with increasing trophic level. Watanabe et al. (2008) observed
37                   decreasing Pb concentrations through a stream macroinvertebrate food web in Japan from


      November 2012                             7-120                   Draft - Do Not Cite or Quote

-------
 1                  producers to primary and secondary consumers. In a Brazilian freshwater coastal lagoon
 2                  food chain, Pb was significantly higher in invertebrates than in fishes (Pereira et al..
 3                  2010).
 4                  Introduction of exotic species into an aquatic food web may alter Pb concentrations at
 5                  higher tropic levels. In Lake Erie, the invasive round goby (Neogobius melanostomus)
 6                  and the introduced zebra mussel (Dreissena polymorphd) have created a new benthic
 7                  pathway for transfer of Pb and other metals (Southward Hogan et al.. 2007). The goby is
 8                  a predator of the benthic zebra mussel, while the endemic smallmouth bass (Micropterus
 9                  dolomieui) feed on goby. Since the introduction of goby into the lake, total Pb
10                  concentrations have decreased in bass. The authors attribute this decrease of Pb in bass to
11                  changes in food web structure, changes in prey contaminant burden or declines in
12                  sediment Pb concentrations.
      7.4.5  Biological Effects of Pb in Freshwater Systems

13                  This section focuses on the studies of biological effects of Pb on freshwater algae, plants,
14                  invertebrates, fish and other biota with an aquatic lifestage (e.g., amphibians) published
15                  since the 2006 Pb AQCD. Key studies  from the 1977 Pb AQCD, the 1986 Pb AQCD and
16                  the 2006 Pb AQCD on biological effects of Pb are summarized where appropriate.
17                  Waterborne Pb is highly toxic to aquatic organisms with bioavailability and subsequent
18                  toxicity varying depending upon the species and lifestage tested, duration of exposure,
19                  the form of Pb tested, and water quality characteristics (e.g., pH, alkalinity, DOC)
20                  (Sections 7.4.2 and 7.4.3).

21                  The 2006 Pb AQCD  (U.S. EPA. 2006c) noted that the physiological effects of Pb in
22                  aquatic organisms can occur at the biochemical, cellular, and tissue levels of biological
23                  organization and include inhibition of heme formation, alterations of blood chemistry,
24                  and decreases in enzyme levels. A review of the more recent literature corroborated these
25                  findings, and added information about  induction of oxidative stress by Pb, alterations in
26                  chlorophyll, and changes in production and storage of carbohydrates and proteins. Recent
27                  studies available since the 2006 Pb AQCD further consider effects of Pb on reproduction
28                  and development, growth and survival  of aquatic organisms. Alterations to these
29                  endpoints can lead to changes at the community and ecosystem levels of biological
30                  organization such as decreased abundance,  reduced taxa richness, and shifts in species
31                  composition (Section 7.1). Effects on reproduction, growth and survival are reported in
32                  additional species with some effects occurring in sensitive freshwater organisms at or
33                  near ambient levels of Pb (Table 7-2). Because this review is focused on effects of Pb,
34                  studies reviewed for this section include only those for which Pb was the only, or
      November 2012                            7-121                  Draft - Do Not Cite or Quote

-------
 1                  primary, metal to which the organism was exposed. Areas of research not addressed here
 2                  include literature related to exposure to Pb from ingestion of shot or pellets. Biological
 3                  effects of Pb on freshwater algae and plant species are considered below, followed by
 4                  information on effects on freshwater invertebrates and vertebrates. All reported values are
 5                  from exposures in which concentrations of Pb were analytically verified unless nominal
 6                  concentrations are stated.
              7.4.5.1  Freshwater Plants and Algae

 7                  The toxicity of Pb to algae and plants has been recognized in earlier agency reviews of
 8                  this metal. In the  1977 Pb AQCD, differences in sensitivity to Pb among different species
 9                  of algae were observed and concentrations of Pb within the algae varied among genera
10                  and within a genus (U.S. EPA. 1977). The 1986 Pb AQCD (U.S. EPA. 1986b) reported
11                  that some algal species (e.g., Scenedesmus sp.) were found to exhibit physiological
12                  changes when exposed to high Pb concentrations in situ. The observed changes included
13                  increased numbers of vacuoles, deformations in cell organelles, and increased autolytic
14                  activity. Effects of Pb on algae reported in the 2006 Pb AQCD included decreased
15                  growth, deformation and disintegration of algae cells, and blocking of the pathways that
16                  lead to pigment synthesis, thus affecting photosynthesis. Observations in additional algal
17                  species since the 2006 Pb AQCD, support these findings and indicate that Pb exposure is
18                  associated with oxidative stress. All of these effects were observed at concentrations of
19                  Pb that exceed those found currently in most surface waters (Table 7-2).

20                  Recent studies available since the 2006 Pb AQCD, report additional mechanistic
21                  information on Pb toxicity to freshwater macrophytes as well as further evidence for
22                  effects on oxidative stress and growth endpoints. However, many of these studies were
23                  conducted at nominal concentrations of Pb, complicating the comparisons to Pb
24                  quantified in surface waters. Furthermore, their relevance to conditions encountered in
25                  natural environments is difficult to establish since modifying factors of bioavailability,
26                  such as DOC, are often absent from controlled exposures.

27                  The effect of Pb exposure on the structure and function of plant photosystem II was
28                  studied in giant duckweed, S. polyrrhiza  fLing and Hong. 2009J.  The Pb concentration of
29                  extracted photosystem II particles was found to increase with increasing Pb
30                  concentration, and increased Pb concentration was shown to decrease emission peak
31                  intensity at 340 nm, amino acid excitation peaks at 230 nm, tyrosine residues, and
32                  absorption intensities. This results in decreased efficiency of visible light absorption by
33                  affected plants. The authors theorized that Pb2+ may replace either Mg2+ or Ca2+ in
      November 2012                             7-122                  Draft - Do Not Cite or Quote

-------
 1                  chlorophyll or the oxygen-evolving center, inhibiting photosystem II function through an
 2                  alteration of chlorophyll structure.

 3                  Pb exposure in microalgae species has been linked to several effects, including disruption
 4                  of thylakoid structure and inhibition of growth in both Scenedesmus quadricauda and
 5                  Anabaena flos-aquae fArunakumara and Zhang. 2008J. Arunakumara et al. (2008)
 6                  determined the effect of aqueous Pb on the algal species S. platensis using solutions of
 7                  Pb nitrate. Exposures at 3,440 (ig Pb/L stimulated 10-day algal growth, growth was
 8                  inhibited at higher concentrations of 6,830, 21,800, 32,800 and 44,500 (ig Pb/L by 5, 40,
 9                  49, and 78%, respectively. In addition to growth inhibition, algal chlorophyll a and b
10                  content were significantly diminished at the three highest Pb exposures (Arunakumara et
11                  al., 2008). Although no specific morphological abnormalities were linked to Pb exposure,
12                  filament breakage was observed in S. platensis at Pb concentrations >50,000 (ig Pb/L.
13                  Since the  2006 Pb AQCD, the production of reactive oxygen species following Pb
14                  exposure has been measured directly in cells of the freshwater algae Chlamydomonas
15                  reinhardtii at nominal concentrations of Pb as Pb nitrate (0.02 to 52 (ig Pb/L) with the
16                  greatest response at 3.15 times more stained cells compared to the control sample
17                  following an exposure of 2.5 hours (Szivak et al., 2009). Although this study provides
18                  direct evidence for a mechanism of Pb-toxicity at the sub-organism level of biological
19                  organization, the relevance of the exposure method to conditions encountered in natural
20                  environments is  unknown. The concentration data are not reliable in this case since Pb
21                  concentrations were not quantified and the lowest reported values are below the
22                  analytical detection limit for Pb.

23                  At the time of the 1977 Pb AQCD, there  was  limited information available on Pb effects
24                  on aquatic macrophytes. For plants in general, Pb was recognized to affect
25                  photosynthesis, mitosis, and growth, however, the majority of studies reporting Pb
26                  toxicity were not conducted with plants grown under field conditions (U.S.  EPA. 1977).
27                  The mechanism  for Pb inhibition of photosynthesis was further elucidated in the  1986 Pb
28                  AQCD. Additional evidence of Pb effects on  plant growth was also observed, however,
29                  the available studies were conducted under laboratory conditions at concentrations that
30                  exceeded  Pb levels in the environment except near smelters or roadsides (U.S. EPA.
31                  1986b). In the 1986 Pb AQCD, EC50 values for plant growth were available for several
32                  aquatic plants with the lowest EC50 of 1,100 (ig Pb/L in Azolla pinnata exposed to
33                  Pb nitrate for 4 days. Effects of Pb on metabolic processes in aquatic plants reviewed in
34                  the 2006 Pb AQCD (U.S. EPA. 2006b) included nitrate uptake, nitrogen fixation,
35                  ammonium uptake and carbon fixation at concentrations of 20,000 (ig Pb/L and higher.

36                  New information is available on Pb effects on oxidative stress endpoints such as changes
37                  in antioxidant enzymes, lipid peroxidation and reduced glutathione in aquatic plant,
      November 2012                            7-123                  Draft - Do Not Cite or Quote

-------
 1                   algae, and moss species exposed to Pb, however most evidence is from studies with high
 2                   concentrations where Pb was not quantified in the exposure media. An aquatic moss,
 3                   F. antipyretica, exhibited increased SOD and ascorbate levels following a 2-day exposure
 4                   to nominal Pb chloride solutions of concentrations of 20, 200, 2,070, 20,700 and
 5                   207,200 (ig Pb/L. When exposure duration was increased to 7 days, only SOD activity
 6                   remained significantly increased by Pb exposure (Dazy et al.. 2009). Bell-shaped
 7                   concentration-response curves were commonly observed for the induction of antioxidant
 8                   enzymes in F. antipyretica. The chlorophyll, carotenoid,  and protein contents of the
 9                   aquatic macrophyte Elodea canadensis were significantly reduced following Pb
10                   accumulation at nominal exposures of 1,000 10,000 and 100,000 (ig Pb/L (Dogan et al..
11                   2009). This, along with the induction of some antioxidant systems and the reduction of
12                   growth at the highest two exposures, indicated that exposure to the metal caused
13                   significant stress, and that toxicity increased with exposure. In addition, native
14                   Myriophyllum quitense exhibited elevated antioxidant enzyme  activity (glutathione-S-
15                   transferase, glutathione reductase, peroxidase) following  transplantation in
16                   anthropogenically polluted areas containing elevated Pb concentrations. These were
17                   correlated with sediment Pb concentrations in the range of 5 to 23 mg Pb/g dry weight
18                   (Nimptsch et al.. 2005).

19                   Since the 2006 Pb AQCD, toxicity and oxidative stress were also observed in coontail
20                   (C. demersum) rooted aquatic macrophytes following 7-day nominal exposures to
21                   aqueous Pb 200 to 20,700 (ig Pb/L ,with increasing effects observed with greater
22                   exposure concentrations and times. Chlorosis and leaf fragmentation were evident
23                   following a 7-day exposure to the highest concentration, while induction of antioxidant
24                   enzymes (glutathione, superoxide dismutase, peroxidases, and  catalase) was observed at
25                   lower exposure concentrations and times. However, as the duration and concentration of
26                   Pb exposure was increased, activities of these antioxidant enzymes decreased (Mishra et
27                   al.. 2006b).

28                   Sobrino et al. (2010) observed reductions in soluble starch stores and proteins with
29                   subsequent increases in  free sugars and amino acids in Lemna gibba plants exposed
30                   nominally to Pb (50,000 to 300,000 (ig Pb/L); total phenols also increased with
31                   increasing Pb exposure. Authors noted that this species exhibited similar responses under
32                   extreme temperatures, drought, and disease. According to Odjegba and Fasidi (2006).
33                   nominal exposure to 18,600 (ig Pb/L as Pb nitrate for 21 days was sufficient to induce a
34                   gradual reduction of both chlorophyll and protein content in the macrophyte Eichhornia
35                   crassipes. Decreased proteins were theorized to be related to inefficient protein formation
36                   following disruption of nitrogen metabolism after Pb exposure (Odjegba and Fasidi.
37                   2006). Foliar proline (which is thought to act as an antioxidant) concentrations were
      November 2012                             7-124                  Draft - Do Not Cite or Quote

-------
 1                  found to increase in a concentration-dependent manner as Pb concentrations increase
 2                  from 20,720 to 1,036,000 ng Pb/L.

 3                  Following 72-hour aqueous exposure to 8,495 (ig Pb/L as Pb nitrate, phytochelatin and
 4                  glutathione concentrations in the freshwater algae Scenedesmus vacuolatus were
 5                  significantly increased over that of non-exposed algal cultures (Le Faucheur et al.. 2006).
 6                  The 72-hour Pb exposure also significantly reduced S.  vacuolatus growth, and of all the
 7                  metals tested (Cu, Zn, Ni, Pb, Ag,  As, and Sb), Pb was determined to be the most toxic to
 8                  the algae species. In the algae Chlamydomonas reinhardtii, phytochelatin concentrations
 9                  were lower than intracellular Pb and not sufficient to bind to accumulated metal
10                  following 72-hour exposure (Scheidegger et al.. 2011).

11                  In addition to oxidative stress responses, there is new information since the
12                  2006 Pb AQCD on growth effects  observed at high concentrations of Pb summarized in
13                  Table 7-5. Growth effects at the species level can lead  to effects at the population-level of
14                  biological organization and higher (Section 7.1.1). Root elongation was significantly
15                  reduced in a number of wetland plant species (Beckmannia syzigachne, Juncus effusus,
16                  Oenanthe javanica, Cyperusflabelliformis, Cyperus malaccensis, and Neyraudia
17                  reynaudiana) following nominal Pb exposures of 20,000 (ig Pb/L as Pb nitrate for 21
18                  days (Deng et al., 2009). Further, while both Zn and Fe exposures exerted some selective
19                  pressure on plants, the authors did not observe the same with Pb, leading them to theorize
20                  that concentrations of bioavailable Pb were not present in high enough quantities to have
21                  such an effect. Lemna sp. aquatic plants were determined to effectively sequester aqueous
22                  Pb at nominal exposures of 5,000 and 10,000 (ig Pb/L  in a 7-day experiment, however,
23                  15,000 (ig Pb/L resulted in plant mortality (Hurd and Sternberg. 2008). In another study
24                  with duckweed, Paczkowska et al.  (2007) observed that nominal Pb exposures of 2,070 to
25                  20,700 (ig Pb/L for 9 days stimulated the growth of Lemna minor cultures, although there
26                  was concurrent evidence of chlorosis and induction of antioxidant enzymes. Additionally,
27                  Cd was found to be more toxic than Pb, although the authors determined that this resulted
28                  from poor uptake of Pb by L. minor (Paczkowska et al., 2007). Pb exposure (as
29                  Pb nitrate) caused oxidative damage, growth inhibition, and decreased biochemical
30                  parameters, including photosynthetic pigments, proteins, and monosaccharides, in Wolffia
31                  arrhiza plants. Fresh weight of plants was reduced following both 7- and 14-day
32                  exposures to Pb concentrations greater than 2,120 (ig Pb/L while chlorophyll a content
33                  was decreased at 210 (ig Pb/L and higher (Piotrowska  et al.. 2010).

34                  Effects of Pb on algae reported in the 2006 Pb AQCD  (U.S. EPA. 2006b) included
35                  decreased growth, deformation and disintegration of algae cells, and blocking of the
36                  pathways that lead to pigment synthesis, thus affecting photosynthesis. Observations in
37                  additional algal species since the 2006 Pb AQCD support these findings. Effects on
      November 2012                            7-125                  Draft - Do Not Cite or Quote

-------
 1                  plants supported by additional evidence in this review and evidence from previous
 2                  reviews include oxidative damage, decreased photosynthesis and reduced growth.
 3                  Elevated levels of antioxidant enzymes are commonly observed in aquatic plant, algae,
 4                  and moss species exposed to Pb. All of the observed effects on aquatic macrophytes and
 5                  algae occur at concentrations not typically encountered in surface waters of the U.S.
              7.4.5.2  Freshwater Invertebrates

 6                  Few studies on biological effects of Pb in freshwater invertebrates had been conducted at
 7                  the time of the 1977 Pb AQCD. One study reported an effect on reproduction in Daphnia
 8                  magna at 30 ng Pb/L (U.S. EPA. 1977). In the 1986 Pb AQCD (U.S. EPA. 1986b).
 9                  increased mortality was observed in the freshwater snail Lymnaea palustris as low as
10                  19 (ig Pb/L and reproductive impairment was reported as low as 27 (ig Pb/L for
11                  Daphnia sp. Population-level endpoints of Pb reviewed in the 2006 Pb AQCD included
12                  reproduction, growth, and survival. Pb was recognized to be more toxic in longer-term
13                  exposures than shorter-term exposures with chronic toxicity thresholds for reproduction
14                  in water fleas (D. magna) ranging as low as 30 (ig Pb/L. In aquatic invertebrates, Pb has
15                  also been shown to affect stress responses and osmoregulation (U.S. EPA. 2006c). Recent
16                  evidence that supports previous findings of Pb effects on reproduction and growth in
17                  invertebrates is reviewed here as well  as limited studies on behavioral effects associated
18                  with Pb exposure. Some of these effects are observed in the range of Pb values found in
19                  surveys of U.S. surface waters (median 0.50 (ig Pb/L, range 0.04 to 30 (ig Pb/L), in the
20                  U.S. based on a synthesis of NAWQA data reported in the previous 2006 Pb AQCD
21                  (U.S. EPA. 2006c) (Table 7-2). The studies are generally presented in this section from
22                  responses  at the sub-organismal level of biological organization to consideration of
23                  endpoints  relevant to ecological risk assessment (growth, reproduction, survival).

24                  Recent literature strengthens the evidence indicating that Pb affects enzymes and
25                  antioxidant activity in aquatic invertebrates. These alterations at the sub-organismal level
26                  may serve as biomarkers for effects at the organism level and higher. In invertebrate
27                  species that have hemoglobin, ALAD  activity can be measured as a biomarker for Pb
28                  exposure.  In the  freshwater gastropod  B. glabrata and the freshwater oligochaete
29                  Lumbriculus variegatus a significant negative correlation between whole body tissue
30                  ALAD enzyme activity and increasing Pb was observed following 48-hour exposure to
31                  varying nominal concentrations of the metal (Aisemberg et al.. 2005). The concentration
32                  at which 50% of enzyme inhibition was measured was much lower in B. glabrata (23 to
33                  29 (ig Pb/L) than in L. variegatus (703 (ig Pb/L). A significant negative correlation was
34                  also observed between ALAD activity and metal accumulation by the organisms.  Sodium
35                  and potassium ATPase (Na+/K+ATPase) activity in gills of Eastern elliptic mussels was


      November 2012                            7-126                  Draft - Do Not Cite or Quote

-------
 1                  significantly reduced following a 28-day exposure to 57 (ig Pb/L and 245 (ig Pb/L
 2                  (Mosher et al.. 2012). A significant reduction in Na+ and significant increase in Ca2+ in
 3                  hemolymph was only observed at the highest concentration.

 4                  Studies of stress responses to Pb in invertebrates, conducted since the 2006 Pb AQCD,
 5                  include induction of heat shock proteins and depletion of glycogen reserves. Although
 6                  these stress responses are correlated with Pb exposure, they are non-specific and may  be
 7                  altered with exposure to any number of environmental stressors. Induction of heat shock
 8                  proteins in zebra mussel exposed to an average concentration of 574 (ig Pb/L for 10
 9                  weeks exhibited a 12-fold higher induction rate as compared to control groups (Singer et
10                  al.. 2005). Energetic reserves in the freshwater snail B. glabrata in the form of glycogen
11                  levels were significantly decreased by 20%, 57% and 78% in gonads compared to control
12                  animals following 96-hour exposures to nominal concentrations of 50, 100 and
13                  500 (ig Pb/L, respectively (Ansaldo et al.. 2006). Decreases in glycogen levels were also
14                  observed in the pulmonary and digestive gland region at 50 and 100 (ig Pb/L treatment
15                  levels. Pb did not exacerbate the effects of sustained hypoxia in the crayfish (C.
16                  destructor) exposed to 5,000 (ig Pb/L for 14 days while being subjected to decreasing
17                  oxygen levels in water (Morris et al.. 2005). The crayfish appeared to cope with Pb by
18                  lowering metabolic rates in the presence of the metal.

19                  The effect of Pb on osmoregulatory response has been studied since the 2006 Pb AQCD.
20                  The combined effect of Pb and hyperosmotic stress on cell volume regulation was
21                  analyzed in vivo and in vitro in the freshwater red crab, Dilocarcinus pagei (Amado et
22                  al.. 2006). Crabs held in either freshwater or brackish water lost 10% of their body weight
23                  after one day when exposed to 2,700 (ig Pb2+/L as Pb nitrate.  This weight loss was
24                  transient and was not observed during days 2-10 of the exposure. In vitro, muscle from
25                  red crabs exposed to hyperosmotic saline solution had increased ninhydrin-positive
26                  substances and muscle weight decreased in isosmotic conditions upon exposure to Pb
27                  indicating that this metal affects tissue volume regulation in crabs although the exact
28                  mechanism is unknown.

29                  Behavioral responses of aquatic invertebrates to Pb reviewed in the 2006 Pb AQCD (U.S.
30                  EPA. 2006b) included avoidance. A limited number of recent studies have considered
31                  additional behavioral endpoints. Feeding rate of the blackworm L. variegatus was
32                  significantly suppressed by day 6 of a 10 day sublethal test in Pb-spiked sediments
33                  (Penttinen et al.. 2008)  as compared to feeding rates at the start of the experiment.
34                  However, this decrease of approximately 50% of the initial feeding rate was also
35                  observed in the controls; therefore it is likely caused by  some other factor other than Pb
36                  exposure. Aqueous soil leachates containing multiple metals, including Pb, had no effect
37                  on D. magna mobility. Authors noted that although some concentrations (13  to
      November 2012                            7-127                  Draft - Do Not Cite or Quote

-------
 1                  686 (ig Pb/L) exceeded Canadian Environmental Quality Guidelines, no significant
 2                  correlation could be established between Pb exposure and D. magna mobility; in fact, the
 3                  cladocerans were more sensitive to Fe and Al in the leachate than to Pb (Chapman et al.,
 4                  2010).
 5                  Alterations in reproductive and developmental endpoints at the species level can lead to
 6                  effects at the population-level of biological organization and higher (Section 7.1.1). For
 7                  example, reduced fecundity may lead to a decreased population size and developmental
 8                  defects can compromise the ability of an organism to escape predation. Recent evidence
 9                  of reproductive and developmental effects of Pb on freshwater invertebrates available
10                  since the 2006 Pb AQCD, include data from previously untested species as well as
11                  further characterization of reproductive effects in commonly tested organisms such as
12                  Daphnia sp (Table 7-5).  However, many of these studies are conducted at nominal Pb
13                  concentration complicating direct comparison to Pb quantified in freshwater
14                  environments. Sublethal concentrations of Pb negatively affected the total number of
15                  eggs, hatching success and embryonic survival of the freshwater snail B. glabrata
16                  exposed to nominal concentrations of 50, 100, or 500 (ig Pb/L as Pb nitrate (Ansaldo et
17                  al., 2009). Following exposure of adult snails for 96 hours, adults were removed and the
18                  eggs were left in the Pb solutions. The total number of eggs was significantly reduced at
19                  the highest concentration tested  (500 (ig Pb/L). Time to hatching was doubled and
20                  embryonic survival was  significantly  decreased at 50 and  100 (ig Pb/L, while no embryos
21                  survived in the highest concentration. Theegala et al.  (2007) observed that the rate of
22                  reproduction was significantly impaired in Daphnia pulex at >500  (ig Pb/L in 21-day
23                  exposures at nominal concentrations of Pb. In a 21-day reproductive test in D. magna the
24                  number of neonates born per female was significantly reduced at nominal concentrations
25                  of 25, 250, and 2,500 (ig Pb/L (Ha and Choi. 2009). C. dubia reproduction was also
26                  impacted by a seven-day exposure to  50 to  500 (ig Pb/L. Both DOC, and, to a lesser
27                  degree, alkalinity were observed to ameliorate the effects  of Pb on C. dubia reproduction.
28                  As DOC increased from 100 (imol C/L to 400 and 600 (imol C/L, the calculated mean
29                  EC50 values for C. dubia reproduction increased from approximately 25 \ig Pb/L to
30                  200 (ig Pb/L and greater than 500 jig  Pb/L, respectively (Mager et al.. 2011 a).
31                  Reproductive variables including average lifespan, rate of reproduction, generation time
32                  and rate of population increase were adversely affected in the rotifer Brachionus patulus
33                  under conditions of increasing turbidity and Pb concentration (Garcia-Garcia et al..
34                  2007).

35                  In larvae of the mosquito, Culex quinquefasciatus, exposed to 50 (ig Pb/L, 100 (ig Pb/L
36                  or 200 (ig Pb/L (as Pb nitrate), exposure was found to significantly reduce hatching rate
37                  and egg-production at all concentrations and larval emergence rate at 200 (ig Pb/L
38                  (Kitvatanachai et al.. 2005). Larval emergence rates of 78% (FO), 86% (Fl) and 86% (F2)
      November 2012                            7-128                  Draft - Do Not Cite or Quote

-------
 1                  were observed in the control group while emergence rates decreased in each generation
 2                  46% (FO), 26% (Fl) and 58% (F2) in mosquitoes reared in a concentration of
 3                  200 (ig Pb/L. The time to first emergence also increased slightly to 10 days in the
 4                  Pb-exposed group as compared to the control group where emergence was first observed
 5                  on day 9. In the F2 generation of parents exposed to 200 (ig Pb/L, the ratio of female to
 6                  male offspring was 3.6:1.0. No effects were observed on oviposition preference of adult
 7                  females, larval weight or larval deformation.

 8                  Impacts to growth can lead to effects at the population-level of biological organization
 9                  and higher (Section 7.1.1). As noted in the 2006 Pb AQCD (U.S. EPA. 2006b). Pb
10                  exposure negatively affects the growth of aquatic invertebrates. Some studies reviewed in
11                  the previous Pb AQCD suggested that juveniles do not discriminate between the uptake
12                  of essential and non-essential metals (Arai et al.. 2002). In recent literature (summarized
13                  in Table 7-5). the freshwater pulmonate snail Lymnaea stagnalis has been identified as a
14                  species that is extremely sensitive to Pb exposure. Growth of juveniles was inhibited at
15                  EC20 <4  (ig Pb/L. (Grosell and Brix. 2009; Grosell et al.. 2006b). In L. stagnalis exposed
16                  to 18.9 (ig/L Pb for 21 days, Ca2+ influx was significantly inhibited and model estimates
17                  indicated 83% reduction in growth of newly hatched snails after 30 days at this exposure
18                  concentration (Grosell and Brix. 2009). The authors speculate that the high Ca2+ demand
19                  of juvenile L. stagnalis for shell formation and interference of the Ca2+uptake pathway by
20                  Pb result in the sensitivity of this species.

21                  In a study of the combined effects of temperature (22 °C or 32 °C), nominal Pb
22                  concentration (50, 100 and 200 (ig Pb/L as Pb chloride) and presence of a competitor, the
23                  population growth rate of two freshwater rotifer species, Brachionus havanaensis and
24                  B. rubens, as measured by quantifying the number of live rotifers for 15 days, responded
25                  to presence of stressors (Montufar-Melendez et al.. 2007). At the lowest temperature,
26                  B. rubens suppressed population growth of B. havanaensis at 50 (ig Pb/L and higher and
27                  B. rubens population growth did not increase at any Pb concentration at 32 °C, a
28                  temperature more suited for B. havanaensis. In  situ toxicity testing with the woodland
29                  crayfish  (Orconectes hylas) indicated that crayfish survival and biomass were
30                  significantly lower in streams impacted by Pb mining and that concentrations of Pb and
31                  other metals in water, detritus, macroinvertebrates, fish and crayfish were  significantly
32                  higher at mining sites (Allert et al., 2009a).

33                  Although Pb  is known to cause mortality when  invertebrates are exposed at sufficiently
34                  high concentrations, species that are tolerant of Pb may not exhibit significant mortality
35                  even at high concentrations of Pb. Odonates are highly tolerant of Pb with no significant
36                  differences in survival of dragonfly larvae Pachydiplax longipennis and Erythemis
37                  simplicicollis exposed for 7 days to nominal concentrations of Pb as high as
      November 2012                            7-129                  Draft - Do Not Cite or Quote

-------
 1                  185,000 (ig Pb/L (Tollett et al. 2009). This apparent tolerance to Pb may be even more
 2                  pronounced in natural environments where the presence of multiple modifying factors
 3                  (e-g-, pH, alkalinity, hardness, DOC) influences Pb bioavailability. Other species are
 4                  more sensitive to Pb in the environment and these responses are reviewed in
 5                  Section 7.4.6.
              7.4.5.3  Freshwater Vertebrates

 6                  The 1977 Pb AQCD reported on Pb effects to domestic animals, wildlife and aquatic
 7                  vertebrates. The available Pb studies were from exposure to Pb via accidental poisoning
 8                  or ingestion of Pb shot (U.S. EPA. 1977). Studies on aquatic vertebrates reviewed in the
 9                  1986 Pb AQCD were limited to hematological, neurological and developmental
10                  responses in fish (U.S. EPA. 1986b). In the 2006 Pb AQCD, effects on freshwater
11                  vertebrates included recent data for fish specifically considering the effects of water
12                  quality parameters on toxicity, as well as limited information on sensitivity of turtles and
13                  aquatic stages of frogs to Pb (U.S. EPA. 2006c). Biological effects of Pb on freshwater
14                  fish that have been studied since the 2006 Pb AQCD are reviewed here, and limited
15                  recent evidence of Pb effects on amphibians are considered. This section presents recent
16                  information available on the mechanism of Pb as a neurotoxicant in fish and effects of
17                  this metal on blood chemistry. Additional mechanisms of Pb toxicity have been
18                  elucidated in the gill and the renal system offish since the 2006 Pb AQCD. Further
19                  supporting evidence of reproductive effects of Pb on fish is discussed along with limited
20                  new information on behavioral effects of Pb.


                    Freshwater Fish

21                  Evidence of toxicity of Pb and other metals to freshwater fish goes back to early
22                  observations whereby contamination of natural areas by Pb mining lead to extirpation of
23                  fish from streams (U.S. EPA. 1977). At the time of the  1977 Pb AQCD, documented
24                  effects of Pb on fish included anemia, mucous secretion, functional damage to inner
25                  organs, physical deformities and growth inhibition. Additionally, the role of temperature,
26                  pH, hardness and other water quality parameters on Pb toxicity was discussed in the 1977
27                  Pb AQCD. The 1986 Pb AQCD reported that hematological and neurological responses
28                  were the most commonly observed effects in fish  and the lowest exposure concentration
29                  causing either hematological or neurological effects was 8  (ig Pb/L. These findings were
30                  additionally supported in the 2006 Pb AQCD, where observed effects of Pb on  fish
31                  included inhibition of heme formation, alterations in brain receptors, effects on blood
32                  chemistry, and decreases in some enzyme activities (U.S. EPA. 2006c). Functional
      November 2012                            7-130                  Draft - Do Not Cite or Quote

-------
 1                   responses resulting from Pb exposure included increased production of mucus, changes in
 2                   growth patterns, and gill binding affinities. According to Eisler (2000) and reviewed in
 3                   the 2006 Pb AQCD,  the general symptoms of Pb toxicity in fish include production of
 4                   excess mucus, lordosis, anemia, darkening of the dorsal tail region, degeneration of the
 5                   caudal fin, destruction of spinal neurons, ALAD inhibition, growth inhibition, renal
 6                   pathology, reproductive effects, growth inhibition and mortality.

 7                   Evidence of Pb effects on fish available since the 2006 Pb AQCD generally supports the
 8                   findings in previous Pb reviews and further elucidates the mechanisms of Pb-associated
 9                   toxicity on some physiological responses. At the sub-organism level, new information on
10                   Pb effects on DNA, specific enzymes, ionoregulation and other biochemical responses is
11                   presented followed by a discussion of new information on population-level endpoints
12                   (i.e., growth reproduction summarized in Table 7-5).

13                   Since the 2006 Pb AQCD evidence of direct interaction of Pb with  fish DNA has become
14                   available as well as additional studies on the genotoxic effects of Pb exposure to fish.
15                   Hong et al. (2007a) observed  covalent binding of Pb with kidney DNA from silver
16                   crucian carp (Carassius auratus gibelio) though extended X-ray absorption fine structure
17                   spectroscopy. This study suggests that exposure to Pb results in effects to DNA but the
18                   exposure method (in vitro) makes it difficult to estimate the natural environmental
19                   conditions that would be equivalent to the experimental one. In the  freshwater fish
20                   Prochilodus lineatus, blood, liver, and gill cells were sampled from fish treated with
21                   nominal concentration of 5,000 (ig Pb/L as Pb nitrate for 6, 24 and  96-hours and then
22                   DNA damage was assessed by comet assay (Monteiro et al., 2011). DNA breaks were
23                   observed in all cell types after 96-hour exposure. The concentrations used in this study
24                   were high compared to Pb concentrations currently encountered in freshwater (Table
25                   7-2). however, it presents supporting evidence for a possible mechanism of Pb toxicity to
26                   fish.

27                   Upregulation of antioxidant enzymes in fish is a well-recognized response to Pb
28                   exposure. Since  the last review, additional studies demonstrating antioxidant activity as
29                   well as evidence for production of reactive oxygen species following Pb  exposure are
30                   available. Silver crucian carp  injected with nominal concentration of 10,  20 or 30 mg
31                   Pb/kg wet weight Pb chloride showed a significant increase in the rate of production of
32                   superoxide ion and hydrogen  peroxide in liver (Ling and Hong. 2010). In the same fish,
33                   activities of liver SOD, catalase, ascorbate peroxidase, and glutathione peroxidase were
34                   significantly inhibited. Both glutathione and ascorbic acid levels decreased and
35                   malondialdehyde content increased with increasing Pb dosage, suggesting that lipid
36                   peroxidation was occurring and the liver was depleting antioxidants. Although this
37                   exposure pathway is unlikely  to be relevant for air related deposition of Pb, it provides
      November 2012                             7-131                  Draft - Do Not Cite or Quote

-------
 1                   evidence for the mechanism of toxicity (production of reactive oxygen species) and the
 2                   responses of antioxidant enzymes observed in this study are supported by findings in
 3                   studies from fish from nominal water-only exposures. For example, in the freshwater fish
 4                   Nile tilapia, liver catalase, liver alkaline phosphatase, Na+/K+ATPase, and muscle
 5                   Ca2+ATPase activities were quantified in various tissues following a 14-day exposure to
 6                   nominal concentrations (1,000, 2,000 and 4,000 jig Pb/L )of Pb nitrate (Atli and Canli.
 7                   2007). Liver catalase activity significantly increased in the 1,000 and 4,000 (ig Pb/L
 8                   concentrations while liver alkaline phosphatase activity was significantly increased only
 9                   at the 4,000 (ig Pb/L concentration. No significant change in alkaline phosphatase
10                   activity was observed in intestine or serum. Ca2+ATPase activity was significantly
11                   decreased in muscle. Na+/K+ATPase was elevated in gill in the highest concentration of
12                   Pb while all concentrations resulted in significant decreases of this enzyme in intestine.
13                   Serum alanine aminotransferase  and aspartate aminotransferase activities were elevated
14                   in Nile tilapia exposed to 50 (ig Pb/L in 4 and 21 day aqueous exposures while elevations
15                   in alkaline phosphatase and lactate dehydrogenase were only observed at 21 days (Firat et
16                   al.. 2011). In another study with Nile tilapia, Pb had no effect on glutathione measured in
17                   liver, gill, intestine, muscle and blood and liver metallothionein levels following a 14-day
18                   exposure to 1,000, 2,000 and 4,000 jig Pb/L concentrations of Pb as Pb nitrate (Atli and
19                   Canli. 2008V

20                   Metabolic enzyme activity in teleosts has also been measured following dietary
21                   exposures. Alves and Wood (2006) in a 42 day chronic dietary Pb study with 45 and
22                   480 mg Pb/kg found that gill Na+/K+ATPase activity was not affected in rainbow trout
23                   while increased Na+/K+ATPase was observed in the anterior intestine. Metabolic
24                   activities measured in liver and kidney of Nile tilapia following 60 day dietary
25                   administration of 100, 400, and 800 mg  Pb/kg indicated that alanine transaminase,
26                   aspartate transaminase, and lactate dehydrogenase activities significantly decreased in
27                   kidney in a concentration-dependent manner (Dai et al., 2009b) and increased in liver
28                   with increasing concentration of dietary Pb. In a subsequent study using the same
29                   exposure paradigm, the digestive enzymes amylase, trypsin and lipase in tilapia were
30                   inhibited by dietary Pb in a concentration-dependent manner (Dai et al.. 2009a). Lesions
31                   were also evident in histological sections from livers of Pb-exposed fish from this study
32                   and included irregular hepatocytes, cell hypertrophy,  and vacuolation although no
33                   quantification of lesions by dose-group was presented.

34                   There is also evidence for Pb exposure leading to changes in hepatic CYP450 content
35                   although relevance of these in vitro and injection studies to air related exposures to Pb is
36                   unknown. Pb was shown to inhibit hepatic cytochrome P450 in vitro in carp (C. carpio),
37                   silver carp (Hypothalmichtys molitrix) and wels catfish (Silurus glanis) in a
38                   concentration-dependent manner from 0 to 4 (ig/mL (Pb2+) (Henczova et al., 2008). The
      November 2012                             7-132                  Draft - Do Not Cite or Quote

-------
 1                  concentrations of Pb that resulted in 50% inhibition of EROD and 7-ethoxycoumarin-o-
 2                  deethylase (ECOD) isoenzymes varied with the fish species. Silver carp was the least
 3                  sensitive to the inhibitory effects of Pb (EROD 1.21, ECOD 1.52 (ig Pb/mL) while carp
 4                  EROD activity was inhibited at 0.76 (ig Pb/mL. Interaction of Pb with cytochrome P450
 5                  was verified by spectral changes using Fourier Transform Infrared (FTIR) spectroscopy.
 6                  In the same study, CYP450 content was elevated and EROD isoenzyme activities were
 7                  decreased in vivo in silver carp for two days following an injection of 2 mg Pb/kg as
 8                  Pb acetate and returned to control values by 6 days. Liver damage to African catfish
 9                  exposed to nominal concentrations of Pb (50-1,000 (ig Pb/L) for 4 or 8 weeks included
10                  hepatic vacuolar degeneration followed by necrosis of hepatocytes (Adeyemo. 2008b).
11                  The severity of observed histopathological effects in the liver was proportional to the
12                  duration of exposure and concentration of Pb.

13                  In environmental assessments of metal-impacted habitats, ALAD is a recognized
14                  biomarker of Pb exposure (U.S. EPA. 2006c). For example, lower ALAD activity has
15                  been significantly correlated with elevated blood Pb concentrations in wild caught fish
16                  from Pb-Zn mining areas although there are differences in species sensitivity (Schmitt et
17                  al., 2007b; Schmitt et al., 2005). Suppression of ALAD activity in brown trout
18                  transplanted to a metal contaminated stream was  linked to Pb accumulation on gills and
19                  in liver in a 23-day exposure  (Heier et al., 2009). Alves Costa et al. (2007) observed
20                  inhibition of ALAD in hepatocytes of the neotropical traira (Hoplias malabaricus)
21                  following dietary dosing of 21 mg Pb/kg every 5  days for 70 days. Cytoskeletal and
22                  cytoplasmic disorganization were observed in histopathological examination of affected
23                  hepatocytes. In fathead minnow exposed to Pb in either control water (33 (ig Pb/L),
24                  CaSO4 (37 (ig Pb/L) or (39 (ig Pb/L) humic acid-supplemented water for 30 days and
25                  subsequently analyzed by quantitative PCR analysis there were no significant changes in
26                  ALAD mRNA gene response leading  the authors to speculate that water chemistry alone
27                  does not influence this gene response  (Mager et al., 2008). In the same study, glucose-6-
28                  phosphate dehydrogenase, glutathione-S-transferase and ferritin were upregulated, in
29                  microarray analysis, however, no changes in whole body ion concentrations were
30                  observed (Mager etal.. 2008).

31                  In fish,  changes in blood chemistry associated with Pb exposure were noted in the
32                  2006 Pb AQCD (U.S. EPA. 2006b), however, only limited recent studies consider effects
33                  on blood parameters. In a 70-day feeding study with traira exposed to dietary doses
34                  (21 mg  Pb/kg as Pb nitrate via prey [Astyanax sp.]) each five days (corresponding to
35                  daily nominal doses of approximately 4 mg Pb/kg), there were no significant changes to
36                  leukocytes or hemoglobin concentration and volume (Oliveira Ribeiro et al., 2006).
37                  Significant differences in area, elongation and roundness of erythrocytes were observed
38                  in the Pb-exposed individuals using light microscopy image analysis. Other studies
      November 2012                            7-133                 Draft - Do Not Cite or Quote

-------
 1                  available since the 2006 Pb AQCD have only shown effects on blood chemistry at high
 2                  aqueous concentrations of Pb that are not representative of Pb concentrations in U.S.
 3                  surface waters. For example, in the African catfish packed cell volume decreased with
 4                  increasing nominal concentration of Pb (25,000 to 200,000 (ig Pb/L as Pb nitrate) and
 5                  platelet counts increased in a 96-hour exposure (Adevemo. 2007). Red blood cell counts
 6                  also decreased in some of the treatments when compared to controls, although the
 7                  response was not dose-dependent and so may not have been caused by Pb exposure.

 8                  Disruption of ionoregulation is one of the major modes of action of Pb toxicity. The gill
 9                  has long been recognized as a target of Pb in teleosts. Acute Pb toxicity at the fish gill
10                  primarily involves disruption of Ca2+ homeostasis as previously characterized in the
11                  2006 Pb AQCD  (Rogers and Wood. 2004; Rogers and Wood. 2003). In addition to this
12                  mechanism, Pb was found to induce ionoregulatory toxicity at the gill of rainbow trout
13                  through a binding of Pb with Na+/K+ATPase and rapid inhibition of carbonic anhydrase
14                  activity thus enabling noncompetitive inhibition of Na+ and Cl" influx (Rogers et al..
15                  2005). Alves et al. (2006) administered a diet of three concentrations of Pb (7, 77 and
16                  520 mg Pb/kg dry weight) to rainbow trout for 21 days, and measured physiological
17                  parameters including Na+ and Ca2+ influx rate from water. Dietary Pb had no effect on
18                  brachial Na+ and Ca2+ rates except on day 8 where Na+ influx rates were significantly
19                  elevated. These studies suggest that Pb is intermediate between purely Ca2+ antagonists
20                  such as Zn2+ and Cd2+ and disrupters of Na+ and Cl" balance such as Ag+ and Cu2+. This
21                  finding has implications for BLM modeling since it suggests that both  Ca2+ and Na+ need
22                  to be considered as protective cations for Pb toxicity. Indeed, protection from Pb toxicity
23                  by both Na+ and Ca2+ has been documented in freshwater fish (Komjarova and Blust
24                  2009b).

25                  Additional experiments conducted since the 2006 Pb AQCD provide supporting evidence
26                  for underlying mechanisms of Pb toxicity. It was previously established that long-term
27                  exposures of Pb  can impact gill structure and function. Histopathological observations of
28                  gill tissue in the  catfish (C.  gariepinus) following an 8-week aqueous exposure to
29                  nominal concentrations of Pb nitrate revealed focal areas of epithelial hyperplasia and
30                  necrosis at the lower exposure concentrations (50 (ig Pb/L and 100 (ig  Pb/L) (Adeyemo.
31                  2008a). Hyperplasia of mucous cells and epithelial cells were apparent in the tissue from
32                  fish exposed the highest concentrations of Pb in the study (500 (ig Pb/L and
33                  1,000 (ig Pb/L).  In vitro incubation of gill tissue from fathead minnow with Pb
34                  concentrations of 2,500, 12,500 and 25,000 (ig Pb/L for 60 minutes decreased the ratio of
35                  reduced glutathione to oxidized glutathione, indicating that lipid peroxidation at the gill
36                  likely contributes to Pb toxicity at low water hardness (Spokas et al.. 2006). It is difficult
37                  to extrapolate these observations to natural environments  due to the methods used for
38                  exposure and the use of nominal exposure concentrations.
      November 2012                             7-134                  Draft - Do Not Cite or Quote

-------
 1                   In addition to recent evidence of Pb interruption of Na+ and Cl" at the gill (Rogers et al.,
 2                   2005). Pb can interfere with the ionoregulation of Na+ and Cl" and tubular reabsorption of
 3                   Ca2+, Mg2+, glucose, and water in the teleost kidney (Patel et al., 2006). Renal parameters
 4                   including urine flow rate, glomerular filtration rate, urine pH, and ammonia excretion
 5                   were monitored in a 96-hour exposure of rainbow trout to analytically verified
 6                   concentration of 1,200 (ig Pb/L as Pb nitrate. Rates of Na+ and Cl" excretion decreased by
 7                   30% by 48 hours while Mg excretion increased two-to-three fold by 96 hours. Urine flow
 8                   rate was not altered by Pb exposure, although urinary Pb excretion rate was significantly
 9                   increased. After 24 hours of Pb exposure, the urine excretion rate of Ca2+ increased
10                   significantly  by approximately 43% and remained elevated above the excretion rate in the
11                   control group for the duration of the exposure. Glomerular filtration rate significantly
12                   decreased only during the last 12 hours of the exposure. Ammonia excretion rate
13                   increased significantly at 48 hours as urine pH correspondingly decreased. At the end of
14                   the experiment glucose excretion was significantly greater in Pb-exposed fish. Although
15                   the exposures in this study approached the  96-hour LC50, nephrotoxic effects of Pb
16                   indicate the need to consider additional binding sites for this metal in the development of
17                   biotic ligand  modeling (Patel et al.. 2006).  Additional evidence for Pb effects on ion
18                   levels were observed in serum of Nile tilapia; Na+ and Cl" were decreased and K+ levels
19                   were elevated following a 21 day nominal exposure to 50 (ig Pb/L as Pb nitrate (Tirat et
20                   al..2011).

21                   Neurological responses offish to Pb exposure were reported in the 1986 Pb AQCD (U.S.
22                   EPA. 1986b). Additional evidence of the neurotoxic  effects of Pb on teleosts has become
23                   available since  the 2006 Pb AQCD. The mitogen-activated protein kinases (MAPK),
24                   extracellular  signal-regulated kinase (ERK)l/2 and p38MAPK were identified for the first
25                   time as possible molecular targets for Pb neurotoxicity in a teleost (Leal et al., 2006). The
26                   phosphorylation of ERK1/2 and p38MAPK by Pb was determined in vitro and in vivo in the
27                   catfish (Rhamdia queleri). R. quelen exposed to a nominal concentration of 1,000 (ig Pb/L
28                   (as Pb acetate)  for two days showed a significant increase in phosphorylation of ERK1/2
29                   and p38MAPK  in the nervous system. Incubation of cerebellar slices for 3 hours in 1,035
30                   and 2,070 (ig Pb/L as Pb acetate also  showed significant phosphorylation of MAPKs.  The
31                   observed effects of Pb on the MAPK family of signaling proteins have implications for
32                   control of brain development, apoptosis and stress response. In the neotropical fish traira,
33                   muscle cholinesterase was significantly inhibited after 14 dietary doses of 21 mg Pb/kg
34                   wet weight (Rabitto et al.. 2005). Histopathological observations of brains of African
35                   catfish exposed to  nominal  concentrations of 500 (ig Pb/L or 1,000 (ig Pb/L Pb  as
36                   Pb nitrate for 4 weeks included perivascular edema, focal areas of malacia, and diffuse
37                   areas of neuronal degeneration (Adevemo. 2008b). As in the observed effects of Pb on
38                   gill function  and ionoregulation, it is difficult to assess the significance of these findings
39                   to fish in natural environments due to the methods used for exposure.

      November 2012                             7-135                  Draft - Do Not Cite or Quote

-------
 1                  Evidence from the 2006 Pb AQCD (U.S. EPA. 2006b) and earlier Pb reviews indicate
 2                  that Pb can impair both cognitive and motor function in fish. Reduced locomotion and
 3                  foraging ability were observed in Chinese sturgeon juveniles exhibiting abnormal body
 4                  curvature following nominal exposure to either 800 or 1,600 (ig Pb/L for 112 days (Hem
 5                  et al., 2011). Actual concentrations of Pb were quantified at the end of the 112-day
 6                  exposure period (30 to 50% of test media was renewed daily):  129 (ig Pb/L  (200 (ig Pb/L,
 7                  nominal), 458 (ig Pb/L (800 (ig Pb/L nominal), and 1,276 (ig Pb/L (1,600 (ig Pb/L,
 8                  nominal). These chondrostean fish gradually recovered from deformities during a
 9                  depuration period and were able to swim and forage effectively 6 weeks after transfer
10                  into clean water

11                  Since the 2006 Pb AQCD, several studies integrating behavioral  and physiological
12                  measures of Pb toxicity have been conducted on fish. Some of these observations are
13                  reported to occur at concentrations of Pb reported in freshwater. Zebrafish embryos
14                  exposed nominally to low concentrations of Pb as Pb chloride (2.0 and 6.0 (ig Pb/L
15                  prepared from serial dilutions of a stock solution) until 24 hours post-fertilization and
16                  then subsequently tested as larvae or adult fish exhibited behavioral disruptions in
17                  response to mechanosensory and visual stimuli (Rice et al., 2011) Although Pb was not
18                  measured in the water, Pb uptake in the embryos was quantified during the first 24 hours
19                  post-fertilization (approximately 0.08 nM/100 embryos at 2.0 jig Pb/L and 0.32 nM
20                  Pb/100 embryos at 6.0 (ig Pb/L). Startle response time in larvae measured as maximum
21                  head turn velocity and escape time decreased in a concentration-dependent pattern
22                  following a directional, mechanical stimulus (tapping). The pattern of escape swimming
23                  was altered in larvae of Pb-exposed embryos compared to controls. In the adult fish
24                  hatched from Pb-exposed embryos (6.0 \ig Pb/L), visual response to a rotating black bar
25                  against a white background (ability to detect contrast) was significantly degraded. These
26                  findings provide evidence for behavioral effects of Pb at concentrations lower than
27                  previously reported in fish (U.S. EPA. 2006c). however, aqueous exposure
28                  concentrations were not analytically verified.

29                  Sloman et al. (2005) investigated the effect of Pb on hierarchical social interactions and
30                  the corresponding monoaminergic profiles in rainbow trout. Trout were allowed to
31                  establish dominant-subordinate relationships for 24 hours, and then were exposed to
32                  46 (ig Pb/L or 325 (ig Pb/L (Pb nitrate) for 48 hours to assess effects on behavior and
33                  brain monoamines. In non-exposed fish, subordinate individuals had higher
34                  concentrations of circulating plasma cortisol and telencephalic 5-hydroxyindoleacetic
35                  acid/5-hydroxytryptamine (serotonin) (5-HIAA/5-HT) ratios. In the high concentration of
36                  Pb, there was significant uptake of Pb into gill, kidney and  liver when compared with the
37                  control group and dominant fish appeared to have elevated  hypothalamic 5-HIAA/5HT
38                  ratios. Uptake of Pb into the liver was higher in subordinate fish when compared to the
      November 2012                            7-136                  Draft - Do Not Cite or Quote

-------
 1                  dominant fish. No significant differences were observed in cortisol levels or behavior
 2                  after metal exposure.

 3                  Mager et al. (2010) conducted prey capture assays with 10 day old fathead minnow
 4                  larvae born from adult fish exposed to 120 (ig Pb/L for 300 days, then subsequently
 5                  tested in a breeding assay for 21 days. The time interval between 1st and 5th ingestion of
 6                  10 prey items (Artemia nauplif) was used as a measure of behavior and motor function of
 7                  offspring of Pb-exposed fish. Larvae were offered 10 Artemia and the number ingested
 8                  within 5 minutes was scored. The number of larvae ingesting 5 Artemia decreased within
 9                  the time period in offspring of Pb-exposed fish as compared to the control group, leading
10                  the authors to suggest this behavior is indicative of motor/behavioral impairment. In
11                  another study with fathead minnows, swimming performance measured as critical aerobic
12                  swim speed was significantly impaired in minnows in 24-hour acute (139 (ig Pb/L) and
13                  chronic 33 to 57 day (143 (ig Pb/L) exposures, however, no significant difference in
14                  swim speed was observed in chronic exposures to 33 (ig Pb/L (Mager and Grosell. 2011).

15                  Alterations in reproductive and developmental endpoints at the species level can lead to
16                  effects at the population-level of biological organization and higher (Section 7.1.1). For
17                  example, reduced fecundity may lead to a decreased population size and developmental
18                  effects may decrease the ability of a fish to escape predators or reduce spawning
19                  mobility. Reproductive and developmental effects of Pb in fish have been reported for
20                  several decades. In the 1977 Pb AQCD,  second generation brook trout (Salvelinus
21                  fontinalis) exposed to 235 or 474 (ig  Pb/L were shown to develop severe spinal
22                  deformities (lordoscoliosis) (U.S. EPA. 1977). Pb concentration of 120 (ig Pb/L produced
23                  spinal curvature in rainbow trout (Oncorhynchus mykiss) and spinal curvatures were
24                  observed in developing eggs of killifish as reviewed in the 1986 Pb AQCD (U.S. EPA.
25                  1986b). Recent studies on reproductive effects of Pb in fish from oocyte formation to
26                  spawning are summarized in Table 7-5.

27                  Reproductive performance of zebrafish as measured by incidence of spawning, numbers
28                  of eggs per breeding pair or hatch rate of embryos was unaffected following a 63 day diet
29                  of field-collected Pb-contaminated polychaetes that were representative of a daily dose of
30                  0.3-0.48 mg Pb/kg-day (dry weight diet/wet weight fish) through food (Boyle et al..
31                  2010). Mager et al. (2010) conducted 21 day breeding exposures at the end of chronic
32                  300 day toxicity testing with fathead minnow. Non-exposed breeders were switched to
33                  water containing Pb and Pb-exposed breeders  were moved to control tanks and effects on
34                  egg hatchability and embryo Pb  accumulation were assessed. Fish in the high Pb
3 5                  concentration with HCO3  (113 (ig Pb/L) and DOC (112 (ig Pb/L) and the low Pb
36                  concentration with HCO3" (31 (ig Pb/L) reduced total reproductive output, while a
37                  significant increase in average egg mass was observed in the high Pb HCO3" and DOC
      November 2012                            7-137                 Draft - Do Not Cite or Quote

-------
 1                  treatments as compared to egg mass size in controls and in low HCO3" and DOC
 2                  treatments with Pb. No significant differences were present between treatments in egg
 3                  hatchability.

 4                  The effects of metals on embryonic stage offish development in C. carpio and other
 5                  species were reviewed in Jezierska et al. (2009) and included developmental
 6                  abnormalities during organogenesis as well as embryonic and larval malformations. The
 7                  authors concluded that the initial period of embryonic development, just after
 8                  fertilization, and the period of hatching are the times at which developing embryos are
 9                  most sensitive to metals. Additional nominal exposure studies provide supporting
10                  evidence for embryo malformations associated with Pb-exposure. A significant
11                  concentration-dependent increase in morphological malformations was observed in
12                  African catfish embryos exposed to nominal concentrations of 100 (ig Pb/L, 300 (ig Pb/L
13                  or 500 (ig Pb/L Pb nitrate from 6 hours post-fertilization to  168 hours post-fertilization
14                  (Osman et al.. 2007b). Hatching was delayed with increasing Pb concentration and hatch
15                  success of the embryos decreased from 75% in the controls to 40% in the group exposed
16                  to 500 (ig Pb/L. Chinese sturgeon exposed to nominal concentrations of 200 (ig Pb/L,
17                  800 (ig Pb/L or 1,600 (ig Pb/L for 112 days (96 hour post-fertilized eggs through juvenile
18                  stages) exhibited body curvatures in the two highest concentrations (Hou et al.. 2011).
19                  During a 42 day depuration period in clean water following exposure, the degree of
20                  curvature in affected individuals decreased with decreasing tissue concentrations of Pb.

21                  Reproductive and endocrine effects of Pb have also been reported at the cellular level in
22                  fish, including alterations in gonadal tissue and hormone secretions that are associated
23                  with Pb-exposure, however, recent studies that report these  effects are limited to
24                  experiments where only nominal concentrations of Pb were tested. Histopathological
25                  observations of ovarian tissue in the African catfish following an 8-week aqueous
26                  exposure to Pb nitrate indicated necrosis of ovarian follicles at the lowest concentration
27                  tested (50 (ig Pb/L) (Adevemo. 2008a). Severe degeneration of ovarian follicles was
28                  observed in the highest concentrations of 500 (ig Pb/L and 1,000 (ig Pb/L. Chaube et al.
29                  (2010) considered the effects of Pb on steroid levels through 12  and 24 hour in vitro
30                  exposures of post-vitellogenic ovaries from the catfish (Heteropneustes fossilis) to
31                  nominal concentrations of Pb as Pb nitrate (0, 10, 100, 1,000, 3,000 and 10,000 (ig Pb/L).
32                  Progesterone, 17-hydroxyprogesterone, 17, 20 beta-dihydroxyprogesterone,
33                  corticosterone, 21-deoxycortisol and deoxycorticosterone were inhibited in a dose-
34                  dependent manner. Pb was stimulatory on the steroids estradiol-17-(3, testosterone and
35                  cortisol at low concentrations, and inhibitory at higher concentrations. The authors
36                  propose that the disruption of steroid production and altered hormone secretion patterns
37                  observed at the lower concentrations of Pb in this study are  suggestive of the potential for
38                  impacts to fish reproduction (Chaube et al.. 2010).
      November 2012                             7-138                  Draft - Do Not Cite or Quote

-------
 1                  There is also evidence for alterations in steroid levels associated with Pb exposure in
 2                  other species offish although these studies were all conducted with nominal
 3                  concentrations of Pb and the actual exposure concentrations were not verified. Carp
 4                  (Cyprinus carpio) exposed for 35 days to nominal concentration of 410 (ig Pb/L
 5                  experienced altered plasma cortisol and prolactin levels. Plasma cortisol levels
 6                  significantly increased throughout the study period while plasma prolactin increased up
 7                  to day 14 and then declined and was not significantly different from controls by the end
 8                  of the experiment (Ramesh et al.. 2009). Cortisol levels were significantly decreased in
 9                  Nile tilapia exposed to 50 (ig Pb/L (nominal) for 4 days but were followed by a return to
10                  control levels at 21 days of exposure (Firat et al., 2011). In a comparative  study between
11                  in vitro and in vivo estrogenic activity of Pb, vitellogenin was reported to be significantly
12                  induced in juvenile goldfish  (Carassius auratus) following 96-hour exposure to nominal
13                  concentration of 0.2 and 0.02 (ig Pb/L when compared to control fish (Isidori et al..
14                  2010). In the same study, estrogenicity of Pb was detected in vitro using a proliferation
15                  assay with estrogen receptor-positive human MCF-7 cells. The estrogenic effects of Pb
16                  reported by the authors were observed at concentrations at or below that of Pb typically
17                  encountered in freshwaters, however, actual concentrations of Pb were not measured and
18                  the reported concentrations were at or below analytical detection limits for Pb. The
19                  observations of effects of Pb on vitellogenin are interesting; however, additional studies
20                  are warranted considering the difficulty in maintaining these low concentrations of Pb.
21                  The relevance of the  observed in vitro activity to air related exposure to Pb in natural
22                  environments is unknown.

23                  Reduction of growth in fish was noted as an effect of Pb exposure in the 2006 Pb AQCD.
24                  Recent studies available since the 2006 Pb  AQCD do not present consistent evidence of
25                  growth reduction in fish associated with Pb (Table 7-5).  In a series of exposures in which
26                  Ca2+, DOC and pH were varied to assess effects on Pb toxicity to fathead minnows,
27                  Grosell et al. (2006a) observed a significant increase in growth in some groups exposed
28                  to higher concentrations, however, the increase in body mass was noted to have occurred
29                  in tanks with high mortality earlier in the exposure (Grosell et al.. 2006a). Fathead
30                  minnows exposed to 33  (ig Pb/L to test swimming performance had significantly greater
31                  body length and body mass compared to control fish following a mean Pb exposure
32                  duration of 41 days (range 33 to 57 days) (Mager and Grosell 2011).  In 30 day chronic
33                  tests in which a range of pH  values (6.4, 7.5 and 8.3) were tested with low
34                  (25-32  (ig Pb/L), intermediate (82-156 (ig Pb/L) and high (297-453 (ig Pb/L)
35                  concentrations of Pb, Mager et al. (20 lib)  did not observe growth impairment in fathead
36                  minnows at environmentally relevant concentrations of Pb. However, two 60-day early
37                  lifestage tests with rainbow trout showed differences in LOEC for reduced growth
38                  (Mebane et al.. 2008). In the first test, a 69-day exposure, the LOECs for mortality and
      November 2012                             7-139                  Draft - Do Not Cite or Quote

-------
 1                  reduced growth were the same (54 (ig Pb/L). In the second test, a 62-day exposure of Pb
 2                  to rainbow trout, the LOEC for fish length was 18 (ig Pb/L with an EC2o of >87 (ig Pb/L.

 3                  No effects on growth were observed in recently conducted feeding studies with fish.
 4                  Growth and survival were not significantly affected in juvenile rainbow trout, fathead
 5                  minnow and channel catfish (Ictalurus punctatus) fed a live diet of L. variegatus
 6                  contaminated with Pb (846-1,000 (ig Pb/L-g dry mass for 30 days). (Erickson et al..
 7                  2010). No effects on growth rates were observed in rainbow trout administered a diet
 8                  containing three concentrations of Pb (7, 77 and 520 mg Pb/kg dry weight) for 21 days
 9                  (Alves et al.. 2006) or in Nile tilapia fed diets with nominal concentration of 100, 400, or
10                  800 mg/kg Pb dry weight for 60 days (Dai et al.. 2009b).

11                  In one recent field study, faster growth rates were associated with lower whole-body trace
12                  element concentrations in salmon (Salmo salar) across several streams in New
13                  Hampshire and Massachusetts, U.S.,  regardless of whether accumulation was from prey
14                  items or from water (Ward etal. 2010). In sites where conditions in the streams were
15                  conducive to rapid salmon growth, Pb concentrations were 86% lower than in streams
16                  where salmon were smaller.
                    Amphibians

17                  Amphibians move between terrestrial and aquatic habitats and can therefore be exposed
18                  to Pb both on land and in water. The studies reviewed here are all aquatic or sediment
19                  exposures. Biological effects of Pb on amphibians in terrestrial exposure scenarios are
20                  reviewed in  Sections 7.3.3.3 and 7.3.4.3. Amphibians lay their eggs in or around water
21                  making them susceptible to water-borne Pb during swimming, breeding and
22                  development. In the 2006 Pb AQCD amphibians were considered to be relatively tolerant
23                  to Pb. Observed responses to Pb exposure included decreased enzyme activity
24                  (e.g., ALAD reduction) and changes in behavior summarized in Table AX7-2.4.3 (U.S.
25                  EPA. 2006c). Since the 2006 Pb AQCD, studies conducted within two orders of
26                  magnitude of the range of published Pb concentrations for surface waters and sediments
27                  of the U.S. (Section 7.2.3) have indicated sublethal effects on tadpole endpoints including
28                  growth, deformity, and swimming ability. Genotoxic and enzymatic effects of Pb
29                  following chronic exposures have been assessed in laboratory bioassays, however, these
30                  studies were limited to nominal exposures.

31                  The genotoxic potential of Pb to larvae of the frog (X. laevis) was assessed by
32                  determining the number of micronucleated erythrocytes per thousand (MNE)  following a
33                  12-day exposure to nominal concentrations of Pb as Pb nitrate (Mouchet et al.. 2007).
34                  The lowest Pb concentrations withX laevis (10 and 100 (ig Pb/L) did not exhibit
      November 2012                            7-140                 Draft - Do Not Cite or Quote

-------
 1                  genotoxic effects while both 1,000 and 10,000 (ig Pb/L significantly increased MNE to
 2                  14 and 202, respectively compared to the control (6 MNE). In another chronic genotoxic
 3                  study, erythrocytic micronuclei and erythrocytic nuclear abnormalities were significantly
 4                  increased with increasing Pb concentrations (700 (ig Pb/L, 1,400 (ig Pb/L,
 5                  14,000 ng Pb/L, 70,000 ng Pb/L) during 45, 60, and 75-day exposures of tadpoles Bufo
 6                  raddei fZhang et al.. 2007bj. The authors noted that the erythrocytic micronuclei and
 7                  erythrocytic nuclear abnormalities frequencies generally decreased with increasing
 8                  exposure time and that this may be indicative of regulation of genotoxic factors by
 9                  tadpoles.

10                  Endpoints of oxidative damage were measured in testes of the black-spotted frog (Rana
11                  nigromaculata) treated with nominal concentrations of 100 (ig Pb/L, 200 (ig Pb/L,
12                  400 (ig Pb/L, 800 (ig Pb/L or 1,600 ng Pb/L Pb nitrate by epidermal absorption for 30
13                  days (Wang and Jia. 2009). All doses significantly increased MDA, a product of
14                  oxidative stress, and glutathione levels were elevated in all but the lowest treatment
15                  group. In the same study, damage to DNA assessed by DNA tail length showed effects at
16                  >200 (ig Pb/L and DNA tail movement showed effects at >400 (ig Pb/L. The authors
17                  concluded that the effects on endpoints of oxidative stress and DNA damage detected in
18                  testes indicated a possible reproductive effect of Pb to black-spotted frogs. The exposure
19                  method and use of nominal concentration in this study make it difficult to determine the
20                  relevance of this study to exposure scenarios under natural environmental conditions.

21                  Various sublethal endpoints (growth, deformity, swimming ability, metamorphosis) were
22                  evaluated in northern leopard frog (R. pipiens) tadpoles exposed to nominal
23                  concentrations of 3, 10, and 100 (ig Pb/L as Pb nitrate from embryonic stage to
24                  metamorphosis (Chen et al., 2006b). In this chronic study, the concentrations represent
25                  the range of Pb found in surface freshwaters across the U.S. The lowest concentration of
26                  3 (ig Pb/L approaches the EPA chronic criterion for Pb of 2.5 (ig Pb/L at a hardness of
27                  100 mg/L or 4.5 (ig Pb/L at a hardness of 170 mg/L (U.S. EPA. 2002b). No effects were
28                  observed in the lowest concentration. In the 100 (ig Pb/L treatment, tadpole growth rate
29                  was slower (Gosner stages 25-30), 92% of tadpoles had lateral spinal curvature
30                  (compared with 6% in the control) and maximum swimming speed was significantly
31                  slower than the other treatment groups. In this study, Pb concentrations in the  tissues of
32                  tadpoles were quantified and the authors reported that they were within the range of
33                  reported tissue concentrations from wild-caught populations.

34                  The effects of Pb-contaminated sediment on early growth and development were assessed
35                  in the southern leopard frog (Sparling et al., 2006). Tadpoles exposed to Pb in sediment
36                  (45, 75, 180, 540, 2,360, 3,940, 5,520, and 7,580 mg Pb/kg dry weight) with
37                  corresponding sediment pore water concentrations of 123, 227, 589, 1,833, 8,121,  13,579,
      November 2012                            7-141                  Draft - Do Not Cite or Quote

-------
 1                  19,038 and 24,427 (ig Pb/L from embryonic stage to metamorphosis exhibited sublethal
 2                  responses to Pb in sediment at levels below 3,940 mg Pb/kg. There was 100% mortality
 3                  in the 3,940, 5,520 and 7,580 mg Pb/kg exposures by day 5. The authors noted that the
 4                  most profound effects of Pb on the tadpoles were on skeletal development. At 75 mg
 5                  Pb/kg, subtle effects on skeletal formation such as clinomely and brachydactyly were
 6                  observed. Skeletal malformations increased in severity at 540 mg Pb/kg and included
 7                  clinodactyly, brachymely and spinal curvature and these effects persisted after
 8                  metamorphosis. At the highest concentration with surviving tadpoles (2,360 mg Pb/kg)
 9                  all individuals displayed severe skeletal malformations that impacted mobility. Other
10                  sublethal effects of Pb observed in this study were reduced rates of early growth of
11                  tadpoles at concentrations < 540 mg Pb/kg and increased time to metamorphosis in the
12                  2,360mg Pb/kg (8,121(ig Pb/L sediment pore water) treatment.


                    Birds

13                  As reviewed in Koivula and Eeva (2010) measurement of enzymes associated with
14                  oxidative stress in birds is a well-established biomarker of exposure to metals, however,
15                  little is known about the effects of this stress response in wild populations or at higher
16                  levels of ecological organization. Changes in ALAD activity and other oxidative stress
17                  biomarkers at low levels of Pb exposure were recently documented in mallards and coots
18                  (Fulica atra) from a lagoon in Spain impacted by Pb shot (Martinez-Haro et al., 2011).
19                  ALAD ratio in mallards decreased linearly with blood Pb levels between 6 and
20                  40 (ig Pb/dL, and at Pb levels of <20 (ig Pb/dL effects on several antioxidant enzymes
21                  were observed in coots. Although the primary route of exposure to the birds was via
22                  ingestion of Pb shot, effects were observed lower than 20 (ig Pb/dL, the background level
23                  frequently applied to Pb exposures in birds (Martinez-Haro et al., 2011; Brown et al.,
24                  2006).

25                  Consideration of toxicity of Pb to vertebrate embryos that develop surrounded by a
26                  protective egg shell has been  expanded since the 2006 Pb AQCD. Pb treatment of
27                  mallard duck (Anas platyrhynchos}, eggs by immersion in an analytically verified
28                  concentration of 100 (ig Pb/L for 30 minutes  on day 0 of development did not increase
29                  malformations or mortality of embryos (Kertesz and Fancsi. 2003). However, immersion
30                  of eggs in 2,900 (ig Pb/L under the same experimental conditions resulted in increased
31                  rate of mortality and significant malformations including hemorrhages of the body,
32                  stunted growth, and absence of yolk sac circulatory  system (Kertesz  et al. 2006). The
33                  second study was conducted to emulate environmental levels of Pb following a dam
34                  failure in Hungary.
      November 2012                            7-142                 Draft - Do Not Cite or Quote

-------
      7.4.6  Exposure and Response of Freshwater Species

 1                  Evidence regarding exposure-response relationships and potential thresholds for Pb
 2                  effects on aquatic populations can inform determination of standard levels that are
 3                  protective of aquatic ecosystems. The Annex of the 2006 Pb AQCD (U.S. EPA. 2006c)
 4                  summarized data on exposure-response functions for invertebrates (Table AX7-2.4.1) and
 5                  fish (Table AX7-2.4.2). The recent exposure-response studies in this section expand on
 6                  the findings from the 2006 Pb AQCD with information on newly-tested organisms
 7                  (including microalgae, invertebrate, amphibian and fish species). Overall, new data for
 8                  freshwater invertebrates generally support the previous finding of sensitivity of juvenile
 9                  lifestages and indicates some effects of Pb observed in some species at concentrations of
10                  Pb reported in freshwater environments. (Table 7-2). All reported values are from
11                  exposures in which concentrations of Pb were analytically verified unless nominal
12                  concentrations are stated.

13                  The aquatic macrophyte Lemna minor (duckweed) exhibited a EC50 for growth inhibition
14                  of 6,800 (ig Pb/L in a 4-day exposure and 5,500 (ig Pb/L for a 7-day exposure to a range
15                  of Pb concentrations from 100 to 9,970 (ig Pb/L (Dirilgen. 2011). Growth (measured as
16                  biomass) was slightly increased at 100 and 200 (ig Pb/L and then decreased in subsequent
17                  concentrations. In an assay using nominal concentrations of Pb the aquatic freshwater
18                  microalgae Scenedesmus obliquus was significantly more sensitive to Pb exposure than
19                  Chlorella vulgaris algae, although these authors stated that both appeared to be very
20                  tolerant of the heavy metal. Laboratory 48-hour standard toxicity tests were performed
21                  with both of these species and respective EC50 values of 4,040 and 24,500 (ig Pb/L for
22                  growth as measured by cell division rate were derived (Atici et al., 2008).

23                  Exposure-response data for freshwater invertebrates provide evidence  for effects of Pb at
24                  concentrations of Pb encountered in U.S. surface waters. In the 2006 Pb AQCD, effects
25                  of Pb-exposure in amphipods (H. aztecd) and water fleas (D. magnd) were reported at
26                  concentrations as low as 0.45 (ig Pb/L. Effective concentrations for aquatic invertebrates
27                  were found to range from 0.45 to 8,000 (ig Pb/L. Since the 2006 Pb  AQCD, recent
28                  studies have identified the freshwater snail L. stagnalis as a species that is extremely
29                  sensitive to Pb exposure (Grosell and Brix. 2009; Grosell et al., 2006b). Growth of
30                  juvenile L. stagnalis was inhibited below the lowest concentration tested resulting in an
31                  EC20 of <4 (ig Pb/L. In the same study, the NOEC was 12 jig Pb/L and the LOEC was
32                  16 (ig Pb/L. In contrast, freshwater juvenile ramshorn snails M.  cornuarietis were less
33                  sensitive to Pb with the same LOEC for hatching rate and LC50, calculated to be about
34                  10,000 (ig Pb/L based on nominal exposure data (Sawasdee and Kohler. 2010).

35                  Additional studies on Pb effects in aquatic invertebrates published since the
36                  2006 Pb AQCD provide further evidence for differences in sensitivity  of different

      November 2012                            7-143                  Draft - Do Not Cite or Quote

-------
 1                  lifestages of aquatic organisms to Pb. In the freshwater mussel, L. siliquoidea (fatmucket)
 2                  a Pb concentration response was observed in which newly transformed (5-day-old)
 3                  juveniles were the most sensitive lifestage in a 96-hour toxicity test when compared to
 4                  acute and chronic results with other lifestages (Wang et al.. 2010e). The 96-hour EC50
 5                  values for the 5-day-old L. siliquoidea in two separate toxicity tests were 142 and
 6                  298 (ig Pb/L (mean EC50 220 (ig Pb/L) in contrast to older juveniles (2 months old) with
 7                  an EC50 >426 (ig/L. The 24-hour median effect concentration for glochidia (larvae) of
 8                  L. siliquoidea in 48-hour acute toxicity tests was >299 (ig/L. A 28-day exposure chronic
 9                  value of 10 (ig Pb/L was obtained from 2-month-old L. siliquoidea juveniles, and was the
10                  lowest genus mean chronic value ever reported for Pb (Wang et al., 2010e). A 96-hour
11                  test on newly transformed juveniles was also conducted on Lampsilis rafmesqueana
12                  (Neosho mucket), a mussel that is a candidate for the endangered species list. The EC50
13                  for this species was 188 (ig Pb/L.

14                  Different lifestages of chironomids have been shown to have varying sensitivity to Pb
15                  exposure in several studies available since the 2006 Pb AQCD. The acute toxicity of Pb
16                  to first-instar C. riparius larvae was tested in soft water, with hardness of 8 mg/L as
17                  CaCO3 (Bechard et al.. 2008). The 24-hour LC50 of 610 ng Pb/L for first instar
18                  C. riparius larvae was much lower than previous values reported for later instars in
19                  harder water. In a chronic test with Chironomus tentans, (8 day-old larvae exposed to Pb
20                  until emergence [approximately 27 days]), the NOEC was 109, and the LOEC was
21                  497 (ig Pb/L (Grosell et al., 2006b). The EC2o for reduced growth and emergence of the
22                  midge Chironomus dilutus was 28 (ig Pb/L, observed in a 55-day exposure, while the
23                  same species had a 96-hour LC50 of 3,323  (ig Pb/L (Mebane et al., 2008). In fourth instars
24                  of the freshwater midge larvae Chironomus javanus the 24, 48, 72 and 96 hour LC50's
25                  were  20,490, 6,530, 1,690 and 720 jig Pb/L, respectively (Shuhaimi-Othman et al.,
26                  201 Ic). This was comparable to the 96-hour LC50 (400 (ig Pb/L) in the midge larvae
27                  Culicoides furens (Vedamanikam and Shazilli. 2008a). In the same study, the 96-hour
28                  LC50  for Chironomus plumosus ranged from 8,300 (ig Pb/L to 16,210 (ig Pb/L under
29                  different temperatures indicating the role of environmental factors in modulation of
30                  toxicity and differences in sensitivity to Pb even among related species.

31                  Cladocerans are commonly tested aquatic organisms, with data from three species:
32                  D. magna, D. pulex and Ceriodaphnia dubia, representing approximately 70% of
33                  available metal toxicological literature on this group (Wong et al.. 2009). Recent studies
34                  have been conducted with C. dubia and acute toxicity values for other cladocerans as
35                  well as sublethal endpoints for D. magna are available. In a series of 48 hour acute
36                  toxicity tests with C. dubia conducted in a variety of natural waters across North
37                  America, LC50 values ranged from 29 to 1,180 (ig Pb/L  and were correlated with DOC
38                  (Esbaugh et al., 2011). Median lethal concentrations forMoina micrura (LC50
      November 2012                            7-144                 Draft - Do Not Cite or Quote

-------
 1                  690 (ig Pb/L), Diaphanosoma birgei (LC50 3,160 (ig Pb/L), andAlona rectangular (LC50
 2                  7,060 (ig Pb/L) indicate differences in sensitivity to Pb in these freshwater cladocerans
 3                  from Mexico (Garcia-Garcia et al.. 2006). Several additional studies available since the
 4                  2006 Pb AQCD report exposure response values for Daphnia that are based an nominal
 5                  data: an acute study of Pb with D. pulex identified a 48-hour LC50 of 4,000 (ig/L for this
 6                  species (Theegala et al.. 2007) and the EC50 for swimming inhibition in neonate
 7                  D. magna exposed to Pb nitrate for 24 hours was 18,153 (ig Pb/L (Ha and Choi. 2009).

 8                  Rotifers are among the most sensitive aquatic genera to Pb with wide variation in LC50
 9                  values reported between species (Perez-Legaspi and Rico-Martinez. 2001). For example,
10                  in the rotifer genus Lecane, a 22-fold difference in LC50 values was observed in 48-hour
11                  exposure to Pb between L. hamata, L. luna andZ. quadridentata. (Perez-Legaspi and
12                  Rico-Martinez. 2001). L. luna was most sensitive to Pb toxicity with a 48-hour LC50 of
13                  140 (ig Pb/L. In a 48-hour toxicity test with the rotifer Brachionus calyciflorus, an NOEC
14                  (194 (ig Pb/L), an LOEC (284  (ig Pb/L), and an EC20 of 125 (ig Pb/L was established for
15                  this species (Grosell et al., 2006b). The freshwater rotifer Euchlanis dilatata 48 hour
16                  LC50 was 35 (ig  Pb/L using neonates hatched from asexual eggs (Arias-Almeida and
17                  Rico-Martinez. 2011). In this study the authors estimated the NOEC to be 0.1  jig Pb/L
18                  and the LOEC to be 0.5 (ig Pb/L. In contrast, for rotifer Brachionus patulus neonates,  the
19                  24-hour LC50 was 6,150 (ig Pb/L, however, this value was based on nominal exposures
20                  (Garcia-Garcia et al.. 2007).

21                  Exposure-response assays on other freshwater species have been conducted since the
22                  2006 Pb AQCD. The 24-hour LC50 for larvae of C.  quinquefasciatus mosquitoes was
23                  180 (ig Pb/L (Kitvatanachai et al.. 2005). A 48-hour LC50 of 5,200 (ig Pb/L was observed
24                  in water-only exposures of the blackworm Lumbriculus variegatus fPenttinen et  al.,
25                  2008J. In the mayfly Baetis tricaudatus, the 96-hour LC50 was 664 (ig Pb/L (Mebane et
26                  al.. 2008). An EC2o value of 66 (ig Pb/L was derived for B. tricaudatus by quantifying the
27                  reduction in the  number of molts over a 10-day exposure to Pb (Mebane et al.. 2008).  The
28                  number of molts was significantly less than the control (average of 14 molts over 10
29                  days) at concentrations of 160  (ig Pb/L and higher with the lowest number of molts
30                  (average of 5.3 molts over 10 days) observed in the highest concentration (546 \ig Pb/L).
31                  In the freshwater ostracod Stenocypris major, the 96-hour LC50 was 526 (ig Pb/L
32                  (Shuhaimi-Othman et al.. 201 Ib). In another freshwater crustacean, the prawn
33                  Macrobrachium lancesteri, the 96-hour LC50 was 35 (ig Pb/L in soft water (<75 mg/L as
34                  CaCO3) (Shuhaimi-Othman et al.. 201 la).

35                  In the studies reviewed for the 2006 Pb AQCD, freshwater fish demonstrated adverse
36                  effects at concentrations ranging  from 10 to >5,400 (ig Pb/L, generally depending on
37                  water quality parameters (e.g., pH, hardness, salinity) (U.S. EPA. 2006c). Pb tended to be
      November 2012                            7-145                  Draft - Do Not Cite or Quote

-------
 1                  more toxic in longer-term exposures and correlated to Pb-uptake in tissues. Table
 2                  AX7-2.4.2 of the 2006 Pb AQCD summarizes effects of Pb to fish. A series of studies
 3                  published since the 2006 Pb AQCD have been conducted and have further elucidated the
 4                  influence of water chemistry parameters on Pb uptake and toxicity in fathead minnow
 5                  resulting in additional dose-response data for this species. Grosell et al. (2006b)
 6                  conducted a series of 30-day exposures with larval fathead minnow in which varying
 7                  concentrations of Ca2+ (as CaSO4) and DOC were tested. The effects of reduced pH (6.7)
 8                  and increased pH (8.1) compared to a control pH of 7.4 on Pb toxicity were also assessed
 9                  in this study.  DOC, CaSO4 and pH influenced Pb toxicity considerably over the range of
10                  water parameters tested. The 30-day LC50 for low hardness (19 mg CaSO4/L) in basic test
11                  water was 39 (ig dissolved Pb/L and the highest LC50 value (obtained from the protection
12                  from increased concentrations of DOC and CaSO4) was 1,903 (ig dissolved Pb/L (Grosell
13                  et al.. 2006a). This range in LC50 values for larval fathead minnows for differing pH and
14                  concentrations of DOC and hardness clearly demonstrates the importance of the
15                  chemistry of the exposure medium to Pb toxicity.

16                  Mager et al. (2010) conducted 300-day chronic toxicity tests at 35 and 120 (ig Pb/L with
17                  fathead minnow under conditions of varied DOC and alkalinity to assess the effects of
18                  these water quality parameters on fish growth and Pb-uptake. In additional tests with
19                  fathead minnow, Mager et al. (20lib) conducted both 96-hour acute and 30-day chronic
20                  tests to further characterize Ca2+, DOC, pH, and alkalinity values on Pb toxicity.
21                  Increased  Ca2+, DOC and NaHCO3 concentration afforded protection to minnows in acute
22                  studies. The role of pH in Pb toxicity is complex and likely involves Pb speciation and
23                  competitive interaction of fT with Pb2+ (Mager et al.. 201 Ib). In a series of 96-hour acute
24                  toxicity tests with fathead minnow conducted in a variety of natural waters across North
25                  America, LC50 values ranged from 41 to 3,598 (ig Pb/L and no Pb toxicity occurred in
26                  three highly alkaline waters (Esbaugh et al.. 2011).

27                  In the 2006 Pb AQCD, fish lifestage was recognized as an important variable in
28                  determining the sensitivity of these organisms to Pb. Recent data available since the
29                  2006 Pb AQCD (U.S. EPA. 2006c) support the findings of increased sensitivity of
30                  juvenile fish to Pb when compared to adults. Acute (96-hour) and chronic (60-day) early-
31                  lifestage test exposures were conducted with rainbow trout to develop acute-chronic
32                  ratios (ACR's) for this species (Mebane et al., 2008). Two early-lifestage chronic tests
33                  were conducted, the first with an exposure range of 12-384 (ig Pb/L (69 days) at 20 mg
34                  CaCO3/L water hardness and the second with an exposure range of 8 to 124 (ig Pb/L
35                  (62 days) and a water hardness of 29 mg CaCO3/L. In the 69-day test, the following
36                  chronic values were observed for survival: NOEC=24 (ig Pb/L, maximum acceptable
37                  toxicant concentration (MATC)=36 (ig Pb/L, EC10=26 (ig Pb/L, EC20=34 (ig Pb/L,  and
38                  LC50=55 (ig Pb/L. Results from the 62-day test,  with fish length as the  endpoint, were
      November 2012                            7-146                 Draft - Do Not Cite or Quote

-------
 1                  NOEC=8 ng Pb/L, MATC=12 ng Pb/L, EC10=7(ig Pb/L, EC20=102 ng Pb/L and
 2                  LC50=120 (ig Pb/L. In acute tests with rainbow trout run concurrently with the chronic
 3                  tests, 96-hour LC50 values were 120 and 150 (ig Pb/L, respectively. Data from this study
 4                  resulted in ACR's for trout lower than previously reported. The low ACR values were
 5                  due to the acute tests which produced LC50 values that were 10 to 25 times lower than
 6                  earlier studies with trout (Mebane et al.. 2008). The authors speculated that the lower
 7                  LC50 values were due to the age of the fish used in the study (two to four week  old fry)
 8                  and that testing with larger and older fish may not be protective of more sensitive
 9                  lifestages.

10                  There have been only a few recent exposure-response studies in amphibians since the
11                  2006 Pb AQCD. Southern leopard frog tadpoles exposed to Pb in sediment (45 to
12                  7,580 mg Pb/kg dry weight) with corresponding sediment pore water concentrations from
13                  123 to 24,427 (ig Pb/L from embryonic stage to metamorphosis exhibited concentration-
14                  dependent effects on survival (Sparling et al.. 2006). The LC50 value for Pb in sediment
15                  was 3,738 mg Pb/kg, which corresponds to  12,539 (ig Pb/L in sediment pore water. In the
16                  same study, concentration-dependent effects on skeletal development were observed. The
17                  40 day-ECso for deformed spinal columns in the tadpoles was  1,958 mg Pb/kg
18                  (corresponding to 6,734 (ig Pb/L sediment pore water) and the 60 day-EC50 was 579 mg
19                  Pb/kg (corresponding to 1,968 (ig Pb/L sediment pore water) (Sparling et al., 2006).
      7.4.7  Freshwater Community and Ecosystem Effects

20                  As discussed in the 1986 Pb AQCD (U.S. EPA. 1986b) and the 2006 Pb AQCD (U.S.
21                  EPA. 2006b). exposure to Pb is likely to have impacts in aquatic environments via effects
22                  at several levels of ecological organization (organisms, populations, communities, or
23                  ecosystems). These effects resulting from toxicity of Pb would be evidenced by changes
24                  in species composition and richness, in ecosystem function, and in energy flow. The
25                  2006 Pb AQCD concluded that, in general, there was insufficient information available
26                  for single materials in controlled studies to permit evaluation of specific impacts on
27                  higher levels of organization (beyond the organism). Furthermore, Pb rarely occurs as a
28                  sole contaminant in natural systems making the effects of Pb difficult to ascertain. New
29                  information on effects of Pb at the population, community, and ecosystem level is
30                  reviewed below.

31                  In laboratory studies reviewed in the 2006 Pb AQCD and in more recent studies, Pb
32                  exposure has been demonstrated to alter predator-prey interactions, as well as feeding and
33                  avoidance behaviors. In aquatic ecosystems there are field studies reviewed in the 1977
34                  Pb AQCD (U.S. EPA. 1977). the  1986 Pb AQCD (U.S. EPA. 1986b), the
      November 2012                            7-147                 Draft - Do Not Cite or Quote

-------
 1                   2006 Pb AQCD (U.S. EPA. 2006b) and more recent studies that report reductions of
 2                   species abundance, richness or diversity particularly in benthic macroinvertebrate
 3                   communities coexisting with other metals where the sources of Pb were from mining or
 4                   urban effluents. Additionally, field studies have linked Pb contamination to reduced
 5                   primary productivity and respiration, and to altered energy flow and nutrient cycling.
 6                   However, because of the complexity inherent in defining such effects, there are relatively
 7                   few available population, community, or ecosystem level studies that conclusively relate
 8                   Pb exposure to aquatic ecosystem effects. In addition, most of the available work is
 9                   related to point-source Pb contamination, with very few studies considering the effects of
10                   diffuse Pb pollution. Both plant species and habitat type were determined to be factors
11                   affecting the  rate of Pb accumulation from contaminated sediments. While the rooted
12                   aquatic plant E. canadensis was observed to accumulate the highest concentrations of Pb,
13                   the authors concluded that submerged macrophytes (versus emergent plants) as a group
14                   were the most likely to accumulate Pb and other heavy metals (Kurilenko and
15                   Osmolovskaya. 2006). This would suggest that certain types of aquatic plants, such as
16                   rooted and submerged species, may be more susceptible to aerially-deposited Pb
17                   contamination, resulting in  shifts in plant community composition as a result of Pb
18                   pollution.

19                   Alteration of macrophyte community composition was demonstrated in the presence of
20                   elevated surface water Pb concentrations at three lake sites impacted by mining effluents
21                   (Mishra et al., 2008). A total of 11 species of macrophytes were collected. Two sites
22                   located 500 meters and  1,500 meters downstream from the mine discharge point (study
23                   sites 2 and 3) exhibited  similar dissolved Pb concentrations (78 to 92 (ig Pb/L, depending
24                   on season) and contained six and eight unique macrophyte species, respectively. The site
25                   nearest the discharge point  of the mine effluent (study site 1) had the highest Pb
26                   concentrations (103 to  118  (ig Pb/L) and the lowest number of resident macrophyte
27                   species; these included E. crassipes, L. minor, Azolla pinnata and S. polyrrhiza. Based on
28                   analysis of plant tissue Pb concentrations, the authors theorized that certain species may
29                   be more able to develop Pb tolerant eco-types that can  survive at higher Pb
30                   concentrations (Mishra et al.. 2008). In field studies available for certain freshwater
31                   habitats, exposure to Pb has been shown to result in significant alterations of invertebrate
32                   communities. Macroinvertebrate community structure in mine-influenced streams was
33                   determined to be significantly correlated to Pb sediment pore water concentrations.
34                   Multiple invertebrate community indices, including Ephemeroptera, Plecoptera,
35                   Trichoptera (EPT) taxa richness, Missouri biotic index, and Shannon-Wiener diversity
36                   index, were integrated into  a macroinvertebrate biotic condition score (Poulton et al..
37                   2010). These scores were determined to be significantly lower at sample sites
38                   downstream from mining sites where Pb pore water and bulk sediment concentrations
39                   were elevated. Sediment Pb, Cd, and Zn levels were inversely correlated to mussel taxa

      November 2012                             7-148                  Draft - Do Not Cite or Quote

-------
 1                  richness in the Spring River basin encompassing sites in Kansas, Missouri and Oklahoma
 2                  overlapping a former Pb and Zn mining and processing area (Angelo et al.. 2007). In sites
 3                  upstream of the mining area, 21 to 25 species of mussels were present whereas in sites
 4                  downstream, only 6 to 8 species were observed.

 5                  Rhea et al. (2006) examined the effects of multiple heavy metals in the Boulder River,
 6                  Montana, watershed biofilm on resident macroinvertebrate assemblages and community
 7                  structure,  and determined that, among all the metals, biofilm Pb concentrations exerted
 8                  the greatest influence on the macroinvertebrate community indices. Pb biofilm
 9                  concentrations were significantly correlated with reduced EPT taxa richness, reduced
10                  EPT abundance, and an increase in Diptera species abundance. Interestingly, Pb
11                  concentrations in invertebrate tissues were correlated to an increase in Hydropsychidae
12                  caddisfly abundance, but this may have resulted from the intrinsically high variability in
13                  tissue Pb concentrations. The authors concluded that Pb-containing biofilm represented a
14                  significant dietary exposure for impacted macroinvertebrate species, thus altering
15                  invertebrate community metrics (Rhea et al.. 2006).

16                  Kominkova and Nabelkova (2005) examined ecological risks associated with metal
17                  contamination (including Pb) in small urban streams. Although surface water Pb
18                  concentrations in monitored streams were determined to be very low, concentrations of
19                  the metal in sediment were high enough to pose a risk to the benthic community (e.g., 34
20                  to 101 mg Pb/kg). These risks were observed to be linked to benthic invertebrate
21                  functional feeding group, with collector-gatherer species exhibiting larger body burdens
22                  of heavy metals than other groups (Kominkova and Nabelkova.  2005). In contrast,
23                  benthic predators and collector-filterers accumulated significantly lower metals
24                  concentrations. Consequently,  it is  likely that sediment-bound Pb contamination would
25                  differentially affect members of the benthic invertebrate community, potentially altering
26                  ecosystems dynamics.

27                  Invertebrate functional feeding group may also affect invertebrate Pb body burdens in
28                  those systems where Pb bioconcentration occurs. The predaceous zooplanktonic rotifer,
29                  A. brightwellii collected from a Pb-impacted freshwater reservoir in Mexico, contained
30                  384 ng Pb/mg and exhibited a water-to-tissue BCF of 49,344.  The authors theorized that
31                  Pb biomagnification may have been observed in this case because the cladoceran
32                  M. micrura is both a known Pb accumulator and a favorite prey item of the rotifer
33                  (Rubio-Franchini et al.. 2008). They showed thatM micrura had twice the Pb body
34                  burden of D. similis, another grazing cladoceran species present in the reservoir. These
35                  two species exhibited average Pb tissue concentrations  of 57 and 98 ng Pb/mg,
36                  respectively, with respective water column BCFs of 9,022 and 8,046. Conversely, an
37                  examination of the simultaneous uptake of dissolved Pb by the algae P. subcapitata and
      November 2012                             7-149                  Draft - Do Not Cite or Quote

-------
 1                   the cladoceran D. magna suggests that the dietary exposure route for the water column
 2                   filter-feeder is minor. Although Pb accumulated in the algal food source, uptake directly
 3                   from the water column was determined to be the primary route of exposure for D. magna
 4                   flComiarova and Blust. 2009c).

 5                   For many invertebrate species, sediment Pb concentrations may be the most important
 6                   driver in determining Pb uptake. For instance, while Hg and Cd body burdens in lentic
 7                   invertebrates were affected by lake ecological processes (e.g., eutrophication), a similar
 8                   effect was not observed for Pb concentrations in crayfish tissue, despite a high variability
 9                   between sites (Larsson et al.. 2007). Although this may be a result of differing
10                   bioaccumulation tendencies, the authors suggested that other factors, including the
11                   potential for sediment exposures, may be responsible  for Pb uptake in lentic
12                   invertebrates.

13                   A field survey of fishes in the Viburnum Trend Pb-Zn mining district in southeast
14                   Missouri available since the 2006 Pb AQCD, found that species richness and species
15                   density of riffle-swelling benthic fishes were negatively correlated with metal
16                   concentrations in pore water and in fish in mining impacted streams (Allert et al., 2009b).
17                   Density of Ozark sculpin (Coitus hypselurus) and banded sculpin (Coitus carolinae) were
18                   positively correlated with distance from mining sources.

19                   In addition to the ecological effects discussed above, there is  additional evidence that Pb
20                   exposure could alter bacterial infection (and potentially disease transmission) in certain
21                   fish species. Following 96-hour exposures to 4,000 (ig Pb/L, bacterial density in Channa
22                   punctatus fish was observed to be significantly altered when compared to non-exposed
23                   fish. Bacteria population densities in fish spleen, gills, liver, kidneys and muscle tissues
24                   were higher following Pb exposure, with bacterial abundance in the gills too numerous to
25                   quantify (Pathak and Gopal. 2009). In addition, bacteria inhabiting Pb-exposed fish were
26                   more likely to exhibit antibacterial resistance than colonies isolated from non-exposed
27                   fish. Although the mechanism remains unknown, this study suggests that Pb exposure
28                   may increase the likelihood of infection in fish, potentially affecting fish abundance and
29                   recruitment.

30                   In summary, despite the fact that alterations of macrophyte communities may be highly
31                   visible effects of increased sediment Pb concentrations, several recently published papers
32                   propose that ecological impacts on invertebrate communities  are also significant, and can
33                   occur at environmental Pb concentrations lower than those required to impact plant
34                   communities. High sediment Pb concentrations were linked to shifts in amphipod
35                   communities inhabiting plant structures, and potentially to alterations in ecosystem
36                   nutrient processing through selective pressures on certain invertebrate functional feeding
37                   groups (e.g., greater bioaccumulation and toxic effects in collector-gatherers versus


      November 2012                             7-150                  Draft - Do Not Cite or Quote

-------
 1                  predators or filter-feeders). Increased sediment pore water Pb concentrations were
 2                  demonstrated to likely be of greater importance to invertebrate communities, as well.
 3                  Interestingly, recent research also suggests that Pb exposure can alter bacterial
 4                  infestations in fish, increasing both microbial density and resilience, and potentially
 5                  increasing the likelihood of serious disease outbreak.
      7.4.8  Critical Loads in Freshwater Systems

 6                  The general concept and definition of critical loads is introduced in Section 7.1.3 of this
 7                  chapter [also see Section 7.4 of the 2006 Pb AQCD (U.S. EPA. 2006c)1. Critical load
 8                  values are linked to critical limits of Pb for endpoints/receptors of interest in the
 9                  ecosystems, such as blood Pb. Some important critical limits for Pb in aquatic ecosystems
10                  are discussed in this section along with information on aquatic critical loads for Pb.

11                  Unit World Models (UWM) have been used to calculate critical loads for metals in
12                  aquatic ecosystems. These models couple an ecotoxicity model, the BLM, to a
13                  speciation/complexation model, the Windermere Humic Adsorption Model (WHAM),
14                  then to the multi-species fate model, TRANsport-SPECiation (TRANSPEC).  Gandhi et
15                  al.  apply the UWM to estimate speciation/complexation, fate and critical loads using
16                  lakes of three different trophic status. A high percentage of colloidal-bound Pb was found
17                  in the eutrophic and mesotrophic lakes (75-80%) versus the oligotrophic lakes (2%),
18                  owing the high affinity of Pb to DOM. Pb concentrations were lowest for mesotrophic
19                  and highest for oligotrophic systems.  Critical loads were not calculated for Pb; however,
20                  for the other metals tested the critical load was lowest in the oligotrophic and highest in
21                  the eutrophic systems.

22                  A critical load of 39.0 g Pb/m2-yr (0.19 mol Pb/ m2-yr) was calculated for a generalized
23                  lake in the Sudbury area of the Canadian Shield using TICKET-UWM based on acute
24                  toxicity data for D.  magna. (Farley et al.. 2011). The model was set up to calculate
25                  critical loads of metals by specifying  free metal ion activity or the critical biotic ligand
26                  concentration. This critical load for Pb was much higher than for Cu, Ni and Zn and the
27                  authors attribute this difference to the strong binding of Pb to particulate organic matter
28                  and the sequestration  of PbCO3 in sediment.

29                  As  stated previously in Section 7.3.7. in the short term, metal emissions generally have
30                  greater effects on biota in freshwater systems than in terrestrial systems because metals
31                  are more readily immobilized in soils than in sediment. However, over the longer term,
32                  terrestrial systems may be more affected particularly by those metals with a long soil
33                  residence time, such as Pb. Thus, for a particular locale, either the terrestrial or the
34                  aquatic ecosystem at that site may have the lower critical load. Given the heterogeneity of

      November 2012                            7-151                  Draft - Do Not Cite or Quote

-------
 1                  ecosystems affected by Pb, and the differences in expectations for ecosystem services
 2                  attached to different land uses, it is expected that there will be a range of critical load
 3                  values for Pb for soils and waters within the U.S. Refer to Section 7.4.6 of the
 4                  2006 Pb AQCD for additional discussion of critical loads of Pb in aquatic systems.
      7.4.9  Characterization of Sensitivity and Vulnerability in Freshwater Systems

 5                  Data from the literature indicate that exposure to Pb may affect survival, reproduction,
 6                  growth, metabolism, and development in a wide range of freshwater aquatic species.
 7                  Often, species differences in metabolism, sequestration, and elimination rates control
 8                  relative sensitivity and vulnerability of exposed organisms. Diet and lifestage at the time
 9                  of exposure also contribute significantly to the determination of sensitive and vulnerable
10                  populations and communities. Further, environmental conditions in addition to those
11                  discussed as affecting bioavailability (Sections 7.4.3 and 7.4.4) may also alter Pb toxicity.
12                  The 2006 Pb AQCD (U.S. EPA. 2006b) reviewed the effects of genetics, age, and body
13                  size on Pb toxicity. While genetics appears to be a significant determinant of Pb
14                  sensitivity, effects of age and body size are complicated by environmental factors that
15                  alter metabolic  rates of aquatic organisms. A review of the more recent literature
16                  corroborated these findings, and identified seasonally-affected physiological changes and
17                  lifestage as other important determinants of differential sensitivity to Pb.
              7.4.9.1  Seasonally-Affected Physiological Changes

18                  A study by Duman et al. (2006) identified species and seasonal effects of Pb uptake in
19                  aquatic plants. P. australis accumulated higher root Pb concentrations than S. lacustris.
20                  Additionally, the P. australis Pb accumulation factor was significantly higher during the
21                  winter versus other seasons, while the Pb accumulation factor for S. lacustris was greatest
22                  in spring and autumn. The Pb accumulation factor for a third species, P. lucens, was
23                  greatest in autumn (Duman et al.. 2006). Most significantly, these changes in
24                  bioaccumulation were not linked with biomass increases, indicating that species-
25                  dependent seasonal physiological changes may control Pb uptake in aquatic macrophytes
26                  (Duman et al.. 2007). Significant interspecies differences in Pb uptake were observed for
27                  plants representing the same genus (Sargassum), indicating that uptake of Pb by aquatic
28                  plants also may be governed by highly species-dependent factors (Jothinayagi and
29                  Anbazhagan. 2009).

30                  Heier et al. (2009) established the  speciation of Pb in water draining from a shooting
31                  range in Norway and looked at the time dependent accumulation in brown trout. They
      November 2012                            7-152                  Draft - Do Not Cite or Quote

-------
 1                  found that high molecular weight (>10 kDa) cationic Pb species correlated with high flow
 2                  episodes and accumulation of Pb on gills and in the liver. Thus, high flow episodes can
 3                  remobilize metals from a catchment and induce stress to aquatic organisms.
              7.4.9.2  Increased Nutrient Uptake

 4                  Singh et al. (2010) proposed that metal-resistant plants have the capacity to not only up-
 5                  regulate antioxidant synthesis, but also have the ability to increase nutrient consumption
 6                  and uptake to support metal sequestration and detoxification via production of
 7                  antioxidants (Singh et al.. 2010). Therefore, it is likely that such plant species would be
 8                  significantly less susceptible to Pb exposure than those species without those abilities.
              7.4.9.3  Temperature and pH

 9                  Water temperature also appears to affect the toxicity of Pb to aquatic organisms, with
10                  higher temperatures leading to greater responses. Pb toxicity to crayfish increased 7 to
11                  10% when the water temperature was increased by 4 °C, and by 14% when the
12                  temperature increased by 7 °C. The authors determined that the increased toxicity was a
13                  result of the negative impact of Pb on crayfish respiration, which was exacerbated by the
14                  lower dissolved oxygen concentrations at higher water temperatures (Khan et al.. 2006).
15                  In a study of the combined effects of temperature and Pb concentration on two freshwater
16                  rotifer species, Brachionus havanaensis and B.  rubens, population growth was measured
17                  in three nominal concentrations of Pb as Pb chloride (50, 100 and 200 (ig Pb/L) for 15
18                  days at either 22°C or 32°C (Montufar-Melendez et al.. 2007). At 22°C, population
19                  growth of B. havanaensis was suppressed by B. rubens regardless of Pb treatment. At the
20                  higher temperature, there was no population increase of B. rubens at any Pb
21                  concentration. In the controls, population growth rates of B. havanaensis, but not
22                  B. rubens, increased with an increase in temperature. These studies highlight the role of
23                  temperature in Pb toxicity in organisms adapted to low temperatures.

24                  The sequestration ability of L. minor macrophytes was similarly impacted by increased
25                  surface water temperature; plants absorbed a maximum Pb concentration of 8.6 mg /g at
26                  30 °C, while uptake at 15 °C was only 0.3 mg/g (Uvsal and Taner. 2009). Decreased pH
27                  was also demonstrated to increase the uptake of environmental Pb in aquatic plants
28                  (Wang et al.. 2010b: Uysal and Taner. 2009). Lower pH was shown to result in increased
29                  sensitivity to Pb in juvenile fathead minnows in 30-day exposure to Pb of varying
30                  concentrations (Grosell et al.. 2006a). Additionally, Birceanu et al. (2008) determined
31                  that fish (specifically rainbow trout) were more susceptible to Pb toxicity in acidic, soft
      November 2012                            7-153                 Draft - Do Not Cite or Quote

-------
 1                   waters characteristic of sensitive regions in Canada and Scandinavia. Hence, fish species
 2                   endemic to such systems may be more at risk from Pb contamination than fish species in
 3                   other habitats.
              7.4.9.4 Lifestage

 4                   It is clear that certain stages of a life cycle are more vulnerable to Pb. A comparison of
 5                   C. riparius Pb LC50 values derived from toxicity tests with different instars indicates a
 6                   significant effect of lifestage on Pb sensitivity for aquatic invertebrates. Bechard et al.
 7                   (2008) calculated a first instar C. riparius 24-hour LC50 value of 613 (ig Pb/L, and
 8                   contrasted this value with the 24-hour and 48-hour LC50 values derived using later instar
 9                   larvae—350,000 and 200,000 (ig Pb/L, respectively. This disparity would suggest that
10                   seasonal co-occurrence of aquatic Pb contamination and sensitive early instars could have
11                   significant population-level impacts (Bechard et al.. 2008). Similarly, Wang et al. (2010e)
12                   demonstrated that the newly transformed juvenile mussels, L. siliquoidea and
13                   L. rafmesqueana, at  5 days old were more sensitive to Pb exposure than were glochidia
14                   or two to six month- old juveniles, suggesting that Pb exposure at particularly sensitive
15                   lifestages could have a significant influence on population viability (Wang et al., 2010e).

16                   Evidence for differences in susceptibility to Pb at distinct lifestages is also available for
17                   freshwater fish. In chronic (60-day) early-lifestage test exposures conducted with
18                   rainbow trout to develop ACR's for this species the study resulted in ACR's for rainbow
19                   trout lower than previously reported due to the acute tests which produced LC50 values
20                   that were 10 to 25 times lower than earlier studies  with trout. (Mebane et al., 2008). The
21                   authors speculated that the lower LC50 values were due to the age of the fish used in the
22                   study (two to four week old fry) and that testing with larger and older fish may not be
23                   protective of more sensitive lifestages. Post-hatching stages of the African catfish were
24                   more sensitive than the embryonic stage to Pb-exposure and the authors attributed this
25                   apparent protective effect to the presence of a hardened chorion in embryos (Osman et
26                   al.. 2007a).
              7.4.9.5 Species Sensitivity

27                   Species-specific Ca2+ requirements have been shown to affect the vulnerability of aquatic
28                   organisms to Pb. The snail, L. stagnalis, exhibits an unusually high Ca2+ demand due to
29                   CaCO3 formation required for shell production and growth, and exposure to Pb prevents
30                   the uptake of needed Ca2+, leading to toxicity. Consequently, aquatic species that require
31                   high assimilation rates of environmental Ca2+ for homeostasis are likely to be more


      November 2012                             7-154                  Draft - Do Not Cite or Quote

-------
 1                   sensitive to Pb contamination (Grosell and Brix. 2009). Grosell and colleagues also noted
 2                   that reduced snail growth following chronic Pb exposure was likely a result of reduced
 3                   Ca2+ uptake (Grosell et al.. 2006b).

 4                   There is some indication that molting may comprise an additional sequestration and
 5                   excretion pathway for aquatic animals exposed to Pb (Soto-Jimenez et al.. 201 la:
 6                   Mohapatra et al.. 2009; Tollett et al.. 2009; Bergev and Weis. 2007). Libellulidae
 7                   dragonfly nymphs (Tollett et al.. 2009) have been shown to preferentially sequester Pb in
 8                   exoskeleton tissue. Consequently, aquatic arthropod species and those species that shed
 9                   their exoskeleton more frequently may be able to tolerate higher environmental Pb
10                   concentrations than non-arthropods or slow-growing molting species, as this pathway
11                   allows them to effectively lower Pb body burdens.

12                   In contrast, the effect of Pb exposure on fish bacterial loads demonstrated by Pathak and
13                   Gopal (2009) suggest that infected fish populations may be more at risk to the toxic
14                   effects of Pb than healthier species. Aqueous Pb was demonstrated to both increase
15                   bacteria density in several fish organs and to improve the likelihood of antibacterial
16                   resistance (Pathak and Gopal. 2009).

17                   Tolerance to prolonged Pb exposure may develop in aquatic invertebrates and fish. Multi-
18                   generational exposure Pb appears to confer some degree of metal tolerance in
19                   invertebrates such as C. plumosus larvae; consequently, previous population Pb
20                   exposures may decrease species' susceptibility to Pb contamination fVedamanikam and
21                   Shazilli, 2008b). However, the authors noted that metal tolerant larvae were significantly
22                   smaller than larvae reared under clean conditions, and that transference of Pb-tolerant
23                   C. plumosus larvae to clean systems resulted in a subsequent loss of tolerance. Evidence
24                   of acclimation to elevated Pb in fathead minnow was suggested in the variations in
25                   ionoregulatory parameters that were measured on day 10 and 30 in fish exposed to
26                   115 (ig Pb/L for 30 days. At the end of the experiment,  whole body Ca2+ was elevated
27                   while Na+ and K+ recovered  from elevated levels at 30 days (Grosell et al., 2006a).

28                   A series of species sensitivity distributions constructed  by Brix et al. (2005) in freshwater
29                   systems indicated that sensitivity to Pb was greatest in crustacean species, followed by
30                   coldwater fish, and warmwater fish and aquatic insects, which exhibited a similar
31                   sensitivity. Further, analysis  of both acute and chronic mesocosm data sets indicated that
32                   Pb-contaminated systems exhibited diminished species  diversity and taxa richness
33                   following both types of exposure (Brix et al.. 2005). Wong  et al. (2009) constructed Pb
34                   species sensitivity distributions for both cladoceran and copepod freshwater species. A
35                   comparison of the two curves indicated that cladoceran species, as a group, were more
36                   sensitive to the toxic effects  of Pb than were copepods,  with respective hazardous
37                   concentration values for 5%  of the species (HC5) values of 35 and 77 (ig Pb/L. This


      November 2012                             7-155                 Draft - Do Not Cite or Quote

-------
 1                  difference in sensitivities would indicate that cladoceran species are more likely to be
 2                  impacted at lower environmental Pb concentrations than copepods, potentially altering
 3                  community structures or ecosystem functions (Wong et al., 2009).
              7.4.9.6  Ecosystem Vulnerability

 4                  Relative vulnerability of different freshwater ecosystems to effects of Pb can be inferred
 5                  from the information discussed above on species sensitivity and the influence of water
 6                  quality variables on the bioavailability and toxicity of Pb. It is, however, difficult to
 7                  categorically state that certain freshwater plant, invertebrate or vertebrate communities
 8                  are more vulnerable to Pb than others, since toxicity is dependent on many variables and
 9                  data from field studies are complicated by co-occurrence of other metals and alterations
10                  of pH, such as in mining areas. Aquatic ecosystems with low pH and low DOM are likely
11                  to be the most sensitive to the effects of atmospherically-deposited Pb. Examples of such
12                  systems are acidic, soft waters such as sensitive regions in Canada and Scandinavia
13                  (Birceanu et al.. 2008). In the U.S., aquatic systems that may be more sensitive to effects
14                  of Pb include habitats that are acidified due to atmospheric deposition of pollutants,
15                  runoff from mining activities or lakes and streams with naturally occurring organic acids.
16                  Hence, fish and invertebrate species endemic to such systems may be more at risk from
17                  Pb contamination than corresponding species in other habitats.
      7.4.10 Ecosystem Services Associated with Freshwater Systems

18                  Pb deposited on the surface of, or taken up by organisms has the potential to alter the
19                  services provided by freshwater biota to humans although the directionality of impacts is
20                  not always clear. For example, aquatic macrophytes provide a service by sequestering Pb.
21                  At the same time, the uptake of Pb by plants may result in toxicological effects associated
22                  with Pb exposure and decreased capacity of wetland species to remove contaminants. At
23                  this time, few publications address Pb impacts on ecosystem services associated with
24                  freshwater systems and most studies focus on wetlands rather than lakes and streams. Pb
25                  can affect the ecological effects in each of the four main categories of ecosystem services
26                  (Section 7.1.2) as defined by Hassan et al. (2005). These effects are sorted into ecosystem
27                  services categories and summarized here:

28                     • Supporting: food for higher trophic levels, biodiversity
29                     • Provisioning: clean drinking water, contamination of food by heavy metals,
30                       decline in health offish and other aquatic species
      November 2012                            7-156                 Draft - Do Not Cite or Quote

-------
 1                     • Regulating: water quality
 2                     • Cultural: ecosystem and cultural heritage values related to ecosystem integrity
 3                       and biodiversity, wildlife and bird watching, fishing
 4                  Freshwater wetlands are sinks for atmospheric Pb as well as Pb from terrestrial runoff
 5                  (Landre etal.. 2010; Watmough and Dillon. 2007). Several studies have addressed the
 6                  response of natural wetlands to Pb (Odum. 2000; Gambrell. 1994). Recent reviews of
 7                  pollution control (Mander and Mitsch. 2009) or removal of metals (Marchand et al.,
 8                  2010) by constructed wetlands and phytoremediation of metals by wetland plants (Rai.
 9                  2008) indicate that these systems can remove Pb from the aquatic environment and are
10                  important for water quality, sediment stabilization, nutrient cycling and shelter for
11                  aquatic biota. The use of plants as a tool for immobilization of Pb and other metals from
12                  the environment is not limited to wetland species. Recent advances in the
13                  phytoremediation  of metals are reviewed in Dickinson et al. (2009). The impact of Pb on
14                  ecological services provided by specific components of aquatic systems has been
15                  considered in a limited number of studies. For example, Theegala et al. (2007) discuss the
16                  high uptake rate of Pb by D. pulex as the basis for a possible Daphnia-based remediation
17                  for aquatic systems.

18
      7.4.11      Synthesis of New Evidence for Pb Effects in Freshwater Ecosystems

19                  This synthesis of the effects of Pb on freshwater ecosystems covers information from the
20                  publication of the 2006 Pb AQCD to present. It is followed in Section 7.4.12 by
21                  determinations of causality that take into account evidence dating back to the 1977 Pb
22                  AQCD.

23                  Evidence assessed in the present document supports the findings of the previous Pb
24                  AQCDs that waterborne Pb is highly toxic to freshwater organisms, with toxicity varying
25                  with species and lifestage, duration of exposure, form of Pb, and water quality
26                  characteristics. The studies that are available for freshwater plants, invertebrates and
27                  vertebrates include studies where Pb concentration was analytically verified and those
28                  that reported nominal concentrations (Table 7-5). Many of the studies that report nominal
29                  concentrations in media are uptake studies that subsequently quantify Pb in tissues,
30                  however, measurement of Pb in water or sediment at the beginning of an exposure is
31                  desirable when comparing laboratory studies to concentrations of Pb in freshwater
32                  systems. As reported in Section 7.2.3 and Table 7-2. the median and range of Pb
33                  concentrations in surface waters  (median 0.50 ug Pb/L, range 0.04 to 30 ug Pb/L) and
      November 2012                            7-157                  Draft - Do Not Cite or Quote

-------
 1                   sediments (median 28 mg Pb/kg dry weight, range 0.5 to 12,000 mg Pb/kg dry weight) in
 2                   the U.S. based on a synthesis of NAWQA data was reported in the previous
 3                   2006 Pb AQCD (U.S. EPA. 2006c).

 4                   Recent studies provide further evidence for the role of modifying factors such as pH,
 5                   DOC and hardness on the effects of Pb on plants, invertebrates and vertebrates. The same
 6                   Pb concentration added to water or sediment produces far greater effects under some
 7                   conditions, than others. Many studies reviewed in the ISA included concentrations that
 8                   were higher than Pb found near contaminated areas. However, when multiple
 9                   concentrations were used, effects gradually increased with increasing Pb exposure.
10                   Effects at lower concentrations can be implied from many studies since an exposure-
11                   response relationship to Pb was observed, although uncertainty remains in relating these
12                   findings to reported concentrations of Pb in freshwater. Many studies only report an LC50
13                   value when an LOEC or LCio would be more relevant for consideration of effects on
14                   organisms since an effect occurring at the LC50 would most likely not maintain a stable
15                   population. Most available studies only report acute toxicity and are conducted at higher
16                   concentrations of Pb than found in  sampling from U.S. surface waters (Table 7-2).
17                   however, exposure to Pb in freshwater systems is most likely characterized as a chronic
18                   low dose exposure.


                     Plants

19                   Most recent studies on effects of Pb in freshwater algal species reviewed in Section
20                   7.4.5.1 were conducted with nominal media exposures and effect concentrations greatly
21                   exceeded Pb reported in surface water. In studies where Pb was quantified, effect
22                   concentrations for growth (EC50) for aquatic macrophytes were much higher than
23                   currently reported ambient Pb, however, some sublethal endpoints such as effects on
24                   chlorophyll were observed at lower concentrations. For example, chlorophyll a content
25                   was significantly inhibited at 210 (ig Pb/L and higher in W.  arrhiza (Piotrowska et al..
26                   2010). An increase in biomass was reported in L. minor exposed to 100 or 200 (ig Pb/L
27                   with inhibition observed at higher concentrations (Dirilgen. 2011). There were also
28                   numerous studies conducted at nominal Pb concentration that report effects on enzyme
29                   activities and protein content in freshwater aquatic plant species. Exposure-response
30                   relationships in which increasing concentrations  of Pb lead to increasing effects were
31                   consistently observed for freshwater aquatic plants.

32                   Recent studies on bioavailability of Pb in aquatic plants and algae support the findings of
33                   previous Pb AQCDs that plants tend to sequester larger amounts of Pb  in their roots than
34                   in their shoots and provide additional evidence for species differences in
35                   compartmentalization of sequestered Pb and responses to Pb in water and sediments.
      November 2012                             7-158                  Draft - Do Not Cite or Quote

-------
 1                   Given that atmospherically-derived Pb is likely to become sequestered in sediments,
 2                   uptake by aquatic plants is a significant route of Pb removal from sediments, and a
 3                   potential route for Pb mobilization into the aquatic food web. Although there are some
 4                   similarities to Pb accumulation observed in terrestrial plants (e.g., preferential
 5                   sequestration of the metal in root tissue), Pb appears to be more bioavailable in sediment
 6                   than it is in soil.


                     Invertebrates

 7                   The largest body of evidence for effects of Pb at or near concentrations of this metal
 8                   found in surveys of surface waters of the U.S. is for invertebrates and recent studies
 9                   reviewed in Sections 7.4.5.2 and 7.4.6 further support this observation. Exposure-
10                   response relationships in which increasing concentrations of Pb lead to increasing effects
11                   were  consistently observed for freshwater invertebrates. Among the most sensitive
12                   species, growth of juvenile freshwater snails L. stagnalis was inhibited at an EC20 of
13                   <4 (ig Pb/L. (Grosell and Brix. 2009: Grosell et al. 2006b). A chronic value of
14                   10 (ig Pb/L obtained in 28-day exposures of 2-month-old fatmucket mussel,
15                   L. siliquoidea juveniles was the lowest genus mean chronic value ever reported for Pb
16                   (Wang etal.. 2010e). The 96-hour EC50 values for 5-day-old juveniles in two separate
17                   toxicity tests with this species were 142 and 298 (ig Pb/L (mean EC50 220 (ig Pb/L).

18                   Recent studies (Sections 7.4.5.2 and 7.4.6) have further elucidated the role of water
19                   quality on Pb toxicity. In freshwater invertebrates some effects were observed at
20                   concentrations occasionally encountered in U.S. surface waters (Table 7-2). In a 7-day
21                   exposure of the cladoceran C. dubia to 50 to 500 (ig Pb/L, increased DOC leads to an
22                   increase in mean EC50 for reproduction ranging from approximately 25 (ig Pb/L to
23                   >500 (ig Pb/L (Mager et al.. 201 la). The 48-hour LC50 values for the cladoceran C.  dubia
24                   tested in eight natural waters across the U.S. varied from 29 to 1,180 (ig Pb/L and were
25                   correlated with DOC (Esbaugh etal.. 2011).

26                   Additional new evidence reviewed in Sections 7.4.5.2 and 7.4.6 for effects near the upper
27                   range of concentrations of Pb available from surveys of U.S. surface waters include
28                   studies with rotifer, midge and mayfly species. The freshwater rotifer E. dilatata 48 hour
29                   LC50  was 35 (ig Pb/L using neonates hatched from asexual eggs (Arias-Almeida and
30                   Rico-Martinez. 2011). An EC2o for reduced growth and emergence of the midge
31                   C. dilutus was reported to be 28 (ig Pb/L, observed in a 55-day exposure, while the same
32                   species had a 96-hour LC50 of 3,323 (ig Pb/L (Mebane  et al.. 2008) The ECio for molting
33                   in the mayfly B. tricaudatus was 37  (ig Pb/L (Mebane et al., 2008). All of these effect
34                   concentrations provide additional  evidence for Pb effects on freshwater invertebrates.
      November 2012                             7-159                  Draft - Do Not Cite or Quote

-------
                    Vertebrates

 1                  For freshwater fish (Sections 7.4.5.3 and 7.4.6). most recent studies available since the
 2                  2006 Pb AQCD, were conducted with fathead minnow, P. promelas, or rainbow trout,
 3                  O. mykiss. In a series of 96-hour acute toxicity tests with fathead minnow conducted in a
 4                  variety of natural waters across North America, LC50 values ranged from 41 to
 5                  3,598 (ig Pb/L (Esbaugh et al.. 2011). Reproductive effects associated with water quality
 6                  parameters were also noted with this species (Mager et al.. 2010). In trout, no effects of
 7                  Pb were observed in dietary studies. In chronic aqueous exposures with trout the
 8                  following endpoints were reported: NOEC=24 (ig Pb/L, ECi0=26 (ig Pb/L,
 9                  EC2o=34 (ig Pb/L, and LC50=55 (ig Pb/L. In a separate test with the same species an
10                  NOEC=8 ng Pb/L, EC10=7(ig Pb/L, EC20=102 ng Pb/L and LC50=120 ng Pb/L were
11                  reported. In acute tests with rainbow trout run concurrently with the chronic tests,
12                  96-hour LC50 values were 120 and 150 \ig Pb/L, respectively (Mebane et al.. 2008).
13                  These reported effects provide additional evidence for toxicity of Pb to fish and chronic
14                  NOEC and ECio values reported for trout, a sensitive species, are within the upper range
15                  of Pb currently reported in U.S. surface waters (Table 7-2).

16                  In Section 7.4.5.3. a study with the frog R. pipiens exposed nominally to Pb, tissue
17                  concentrations were quantified at the end of the study and found to be in the  range of Pb
18                  tissue concentrations in wild-caught tadpoles. Growth rate was significantly slower in the
19                  100  (ig Pb/L nominal concentration and more than 90% of tadpoles developed lateral
20                  spinal curvature. Time to metamorphosis was also delayed at this treatment level.


                    Food  Web

21                  In the 2006 Pb AQCD, trophic transfer of Pb through aquatic food chains was considered
22                  to be negligible.  Concentrations of Pb in the tissues of aquatic organisms were generally
23                  higher in algae and benthic organisms than in higher trophic-level consumers indicating
24                  that  Pb was bioaccumulated but not biomagnified (U.S. EPA. 2006c: Eisler. 2000). Some
25                  studies published since the 2006 Pb AQCD, (see Section 7.4.4.4) support the potential for
26                  Pb to be transferred in aquatic food webs, while other studies indicate that Pb
27                  concentration decreases with increasing trophic level (biodilution).


                    Community and Ecosystem  Effects

28                  New evidence of effects of Pb at the community and ecosystem levels of biological
29                  organization reviewed in Section 7.4.7 include shift in community composition in
30                  macrophytes. Effects on reproduction, growth or survival (summarized in Table 7-5) may
31                  lead to effects  at the population-level of biological organization and higher. Additional
      November 2012                            7-160                  Draft - Do Not Cite or Quote

-------
 1                  evidence for community and ecosystem level effects of Pb have been observed primarily
 2                  in microcosm studies or field studies with other metals present.
      7.4.12 Causal  Determinations for Pb in Freshwater Systems

 3                  In the following sections, organism-level effects on reproduction and development,
 4                  growth and survival are considered first since these endpoints can lead to effects at the
 5                  population level or above and are important in ecological risk assessment.
 6                  Neurobehavioral effects are considered next followed by sub-organismal responses
 7                  (hematological effects, physiological stress) for which Pb has been shown to have an
 8                  impact in multiple species and across taxa, including humans. Causal determinations for
 9                  terrestrial, freshwater and saltwater ecological effects are summarized in Table 7-3.
              7.4.12.1     Reproductive and Developmental Effects-Freshwater Biota

10                  Evaluation of the findings in previous Pb AQCDs and recent literature on Pb effects in
11                  aquatic fauna indicates that exposure to Pb is associated with reproductive effects at or
12                  near ambient concentrations of this metal (Table 7-2) in some freshwater species.
13                  Impaired fecundity at the organismal level can result in a decline in abundance and/or
14                  extirpation of populations, decreased taxa richness, and decreased relative or absolute
15                  abundance at the community level (Suter etal. 2005; U.S. EPA. 2003a). Various
16                  endpoints have been measured in freshwater organisms to assess the effect of Pb on
17                  fecundity, development and hormone homeostasis. However, there are typically only
18                  limited studies available from different taxa. Recent evidence available since the
19                  2006 Pb AQCD for effects of Pb on reproductive endpoints in freshwater invertebrates
20                  and vertebrates is summarized in Table 7-5.

21                  There are no studies reviewed in the ISA or previous Pb AQCDs on development and
22                  reproductive effects of Pb in freshwater aquatic algae or macrophytes.

23                  Experimental data from freshwater invertebrates provide evidence for increasing
24                  reproductive effects associated with increasing exposure to Pb. The exposure-response
25                  relationship is used in judging causality (Table I Preamble). Reproductive effects of Pb in
26                  freshwater aquatic invertebrates are well-characterized in previous Pb AQCDs, the draft
27                  Ambient Water Quality Criteria for Pb (U.S. EPA. 2008b) and in the current ISA and
28                  have been observed at or near current ambient concentrations (median 0.5 (ig Pb/L, range
29                  0.04 to 30 (ig Pb/L) (U.S. EPA. 2006c) in some species in laboratory exposures.  In the
30                  1986 Pb AQCD reproductive effects were reported to begin at 19 (ig Pb/L for the
31                  freshwater snail Lymnaeapalustris and 27 (ig Pb/L for Daphnia sp. (U.S. EPA. 1986b).

      November 2012                            7-161                 Draft - Do Not Cite or Quote

-------
 1                  In the 2006 Pb AQCD (U.S. EPA. 2006b) the number of neonates per surviving adult was
 2                  significantly decreased in the amphipod H. azteca during chronic 42-day exposures to Pb
 3                  (Besser et al., 2005). In the group exposed to Pb in water-only exposures, the LOEC for
 4                  reproductive effects was 16 (ig Pb/L while in amphipods receiving both water-borne and
 5                  dietary Pb the LOEC for reproduction was 3.5 (ig Pb/L.

 6                  New evidence in freshwater invertebrates (Table 7-5 and Section 7.4.5.2) show
 7                  consistency of the observed association between reproductive endpoints and Pb exposure.
 8                  In the freshwater rotifer B. calyciflorus, reproductive output was measured as total
 9                  number of individuals and intrinsic growth rate. The EC2o for number of rotifers was
10                  125 (ig Pb/L and the 48 hour EC2o for intrinsic rate of population increase was
11                  307 (ig Pb/L with an NOEC of 194 (ig Pb/L (Grosell et al.. 2006b). In a 7-day exposure
12                  of the cladoceran C. dubia to 50 to 500 (ig Pb/L, increased DOC leads to an increase in
13                  mean EC50 for reproduction ranging from approximately 25 (ig Pb/L to >500 (ig Pb/L
14                  (Mager et al.. 201 la). Additional reproductive impairment endpoints for freshwater
15                  cladocerans are reported in Table 6 of the draft Ambient Aquatic Life Water Quality
16                  Criteria for Pb (U.S. EPA. 2008b). It is not clear how these laboratory-derived values for
17                  freshwater invertebrates compare to Pb exposures in natural systems due to the role of
18                  modifying factors (i.e., pH, hardness, and DOC) which affect Pb speciation and
19                  bioavailability, however, results under controlled conditions have consistently shown
20                  reproductive effects of Pb in sensitive taxa (amphipods, cladocerans) at concentrations at
21                  or near Pb quantified in freshwater environments.

22                  In freshwater aquatic vertebrates there is evidence for reproductive and developmental
23                  effects of Pb. Pb exposure in frogs has been demonstrated to delay metamorphosis,
24                  decrease larval size and produce skeletal malformations. For example, in northern
25                  leopard frog R. pipiens exposed to nominal concentrations of 100 \ig Pb/L from
26                  embryonic stage to metamorphosis, maximum swimming speed was significantly slower
27                  than other treatment groups and 92% of tadpoles exposed to 100 (ig Pb/L had lateral
28                  spinal curvature (compared with 6% in the control)  (Chen et al., 2006b). Pb tissue
29                  concentrations were quantified in frogs following exposure and fell within the range of
30                  tissue concentrations in wild-caught tadpoles.

31                  The weight of evidence for reproductive and developmental effects in freshwater
32                  vertebrates is from studies with fish. Pb AQCDs have reported developmental effects in
33                  fish, specifically spinal deformities in brook trout (Salvelinus fontinalis) exposed to
34                  119 ng Pb/L for three generations (U.S. EPA. 1977). and in rainbow trout as low as
35                  120 ng Pb/L (U.S.  EPA. 1986b). Reproductive behaviors of fathead minnows including
36                  reduced time spent in nest preparation by males, increased interspawn periods and
37                  reduced oviposition by females was observed following a 4-week exposure to
      November 2012                            7-162                 Draft - Do Not Cite or Quote

-------
 1                  500 ng Pb/L fWeber. 1993). In the 2006 Pb AQCD (U.S. EPA. 2006b). decreased
 2                  spermatocyte development in rainbow trout was reported at 10 (ig Pb/L and, in fathead
 3                  minnow testicular damage occurred at 500 (ig Pb/L. In a recent study, reproductive
 4                  effects in fathead minnows were influenced by water chemistry parameters (alkalinity
 5                  and DOC) in breeding exposures following 300 day chronic toxicity testing with Pb
 6                  (Maaeretal.. 2010). Specifically, in fish treated in both 35 and 120 jig Pb/L with HCO3"
 7                  and with 120 (ig Pb/L with DOC, total reproductive output was decreased and average
 8                  egg mass production increased as compared to egg mass size in controls and in low
 9                  HCO3" and DOC treatments with Pb. No significant differences were present between
10                  treatments in egg hatchability. In a feeding study, Reproductive performance was
11                  unaffected in zebrafish exposed to Pb-via consumption of contaminated prey (Boyle et
12                  al., 2010). In fish, there is evidence for alteration of steroid profiles and additional
13                  reproductive parameters although most of the available studies were conducted using
14                  nominal concentrations of Pb.

15                  Observations of Pb toxicity to reproductive and developmental endpoints in freshwater
16                  fauna are further supported by evidence in terrestrial invertebrates and vertebrates
17                  (Section 7.3.12.1). marine invertebrates (Section 7.4.21.1) and from laboratory animals
18                  (Section 5.8). Pb appears to affect multiple endpoints associated with reproduction and
19                  development in aquatic invertebrates and vertebrates. A few sensitive invertebrate taxa
20                  (amphipods, cladocerans) have been identified where effects are observed in laboratory
21                  studies at concentrations of Pb that occur in the environment. Overall, there is a dearth of
22                  information on reproductive effects of Pb in natural environments,  however, the weight
23                  of evidence is sufficient to conclude that there is a causal relationship between Pb
24                  exposures and developmental and reproductive effects in freshwater invertebrates and
25                  vertebrates. In freshwater plants, the evidence is inadequate to conclude that there is a
26                  causal relationship between Pb exposures and plant developmental and reproductive
27                  effects.
              7.4.12.2     Growth Effects-Freshwater Biota

28                  Alterations in the growth of an organism can impact population, community and
29                  ecosystem level variables. Growth is a commonly measured endpoint in aquatic plants,
30                  however, reported effects typically occur at concentrations that exceed Pb quantified in
31                  freshwater habitats. Growth effects of Pb on plants include visible growth responses and
32                  reduction of photosynthetic rate, inhibition of respiration, cell elongation, root
33                  development or premature senescence (U.S. EPA. 1986b). In the 2006 Pb AQCD (U.S.
34                  EPA. 2006b). both freshwater algae and plants had EC50 values for growth in the range of
35                  1,000 to >100,000  (ig Pb/L with minimal inhibition of growth observed as low as


      November 2012                            7-163                 Draft - Do Not Cite or Quote

-------
 1                  100 (ig Pb/L (U.S. EPA. 2006c). The most sensitive aquatic macrophyte reported in the
 2                  2006 Pb AQCD was A. pinnata with an EC50 for relative growth rate of 1,100 (ig Pb/L
 3                  following a 4-day exposure to Pb (Gauretal. 1994). An LOEC of 25 jig Pb/L for
 4                  reduced chlorophyll in Coontail (Ceratophyllum demersum), and 50 (ig Pb/L in Cattail
 5                  (71. latifolid) following 12-day exposure to Pb (as Pb acetate) were the lowest reported
 6                  concentrations of growth-related effects in freshwater plants in the draft Ambient Aquatic
 7                  Life Water Quality Criteria for Pb (U.S. EPA. 2008b) and were near the upper range of
 8                  Pb values reported from sampling of U.S. surface waters (Table 7-2). Additional growth
 9                  studies in freshwater algae and plants summarized in Table 6 of the draft Ambient
10                  Aquatic Life Water Quality Criteria for Pb and Table 7-5 of the present document report
11                  growth effects in laboratory studies at concentrations that exceed measured levels of Pb
12                  in the aquatic environment (U.S. EPA. 2008b).

13                  Most of the evidence for growth effects of Pb in freshwater biota is for invertebrates.
14                  Some of these studies  report inhibition of growth in sensitive species occurring at or near
15                  the current upper range of Pb in surface waters (median 0.50 (ig Pb/L, range 0.04 to
16                  30 (ig Pb/L) (U.S. EPA. 2006c). Growth effects of Pb on aquatic  invertebrates are
17                  reviewed in the draft Ambient Aquatic Life Water Quality Criteria for Pb (U.S. EPA.
18                  2008b) and the 2006 Pb AQCD. The lowest reported LOEC for growth in the
19                  2006 Pb AQCD (16 (ig Pb/L) was in amphipods (H. azteca) in a 42-day chronic exposure
20                  (Besser et al. 2005).

21                  Recent studies provide additional evidence for effects on growth  of freshwater aquatic
22                  invertebrates at < 10 (ig Pb/L. Growth effects observed in invertebrates underscores the
23                  importance of lifestage to overall Pb sensitivity. In general, juvenile organisms are more
24                  sensitive than adults. Growth of juvenile freshwater snails L. stagnalis was inhibited
25                  below the lowest concentration tested resulting in an EC2o <4 (ig  Pb/L (Grosell and Brix.
26                  2009; Grosell et al.. 2006b). In the same study, the NOEC was 12 jig Pb/L and the LOEC
27                  was 16 (ig Pb/L. The authors indicated that the observed effect level for Pb was very
28                  close to the current U.S. EPA water quality criteria for Pb (3.3  (ig Pb/L normalized to test
29                  water hardness) (Grosell and Brix. 2009). In the freshwater mussel, fatmucket  (L.
30                  siliquoided) juveniles were the most sensitive lifestage (Wang et  al., 2010e). In this
31                  study, growth of juvenile mussels at the end of a 28-day exposure in 17 (ig Pb/L was
32                  significantly reduced from growth in the controls. A chronic value of 10 (ig Pb/L in 2-
33                  month-old fatmucket juveniles was the lowest genus mean chronic value  ever reported
34                  for Pb. The ECi0 and EC20 for reduced growth and emergence of the midge C. dilutus in a
35                  55-day exposure were  28 (ig Pb/L and 55 (ig Pb/L, respectively, while the same species
36                  had a 96-hour LC50 of 3,323 (ig Pb/L (Mebane et al.. 2008) The EC10 and EC20 for
37                  molting in the mayfly B. tricaudatus were 37 (ig Pb/L and 66 (ig  Pb/L, respectively
38                  (Mebane et al.. 2008).  In natural freshwater systems the effects of Pb are  influenced by
      November 2012                            7-164                 Draft - Do Not Cite or Quote

-------
 1                  additional factors (i.e., pH, hardness, and DOC) which may modulate the toxicity of Pb
 2                  observed under laboratory conditions.

 3                  Evidence for growth effects of Pb in freshwater aquatic vertebrates is limited to a few
 4                  studies in amphibians and fish. In the 2006 Pb AQCD growth effects of Pb were reported
 5                  in frogs at concentrations typically higher than currently found in the environment. A
 6                  recent study supports findings of growth effects in frogs and suggests that these effects
 7                  may be occurring at lower concentrations: the growth rate of tadpoles of the northern
 8                  leopard frog exposed nominally to 100 (ig Pb/L from embryo to metamorphosis was
 9                  slower than the growth rate of the controls (Chen et al.. 2006b). In this study, Pb
10                  concentrations in the tissues of tadpoles were quantified and the authors reported that
11                  they were within the range of reported tissue concentrations reported in wild-caught
12                  populations.

13                  Reports of Pb-associated growth effects in freshwater fish are inconsistent (Mager. 2012).
14                  In a review cited in the 2006 Pb AQCD, general symptoms of Pb toxicity in fish included
15                  growth inhibition (Eisler. 2000) however, other studies with Pb have shown no effects on
16                  growth (Mager. 2012). In the studies reviewed for the current ISA no growth effects were
17                  observed in fish exposed to Pb via dietary intake. Recent aqueous exposure studies with
18                  fathead minnows showed significant increases  in body length and body mass following
19                  chronic low Pb exposure, however, the authors noted that some effects were observed in
20                  tanks with high mortality early in the exposure (Mager and Grosell. 2011; Grosell et al.,
21                  2006a). Other studies with fathead minnows have shown growth reductions with Pb
22                  exposure, however, concentrations of observed effects typically exceeded the 96-hour
23                  LC50 value (Mager. 2012: Mager etal.. 2010: Grosell et al.. 2006a). Two 60-day early
24                  lifestage tests with rainbow trout showed differences in LOEC for reduced growth
25                  (Mebane et al.. 2008). In the first test, a 69-day exposure, the LOECs for mortality and
26                  reduced growth were the same (54 (ig Pb/L). In the second test, a 62-day exposure of Pb
27                  to rainbow trout, the LOEC for fish length was 18 (ig Pb/L with an EC2o  of >87 (ig Pb/L.

28                  Evidence of effects of Pb exposure on growth in terrestrial plants (Section 7.3.12.2) is
29                  highly coherent with evidence from freshwater plants.  Although there is a lack of
30                  evidence in freshwater plants for growth effects at concentrations of Pb typically
31                  encountered in U.S. surface waters, several studies suggest that minimal  growth
32                  inhibition can occur within one to two orders of magnitude of the reported range for
33                  freshwater. Due to the concentration-response relationship observed between Pb exposure
34                  and freshwater plants, growth is  likely impacted at lower, more ecologically relevant
35                  ECio or LOEC values, than the typically reported EC50 values which may not be  suitable
36                  for a maintaining a sustainable population.
      November 2012                            7-165                 Draft - Do Not Cite or Quote

-------
 1                  There is a large body of evidence to support growth effects of Pb on aquatic plants at
 2                  concentrations that greatly exceed those typically found in U.S. surface waters. Less
 3                  evidence is available at current concentrations of Pb measured in U.S. surface waters and
 4                  within one to two orders of magnitude above the range of these measured values. The
 5                  available evidence is, however, sufficient to conclude that a causal relationship is likely
 6                  to exist between Pb exposures and growth effects in freshwater plants.  The evidence is
 7                  sufficient to conclude that there is a causal relationship between Pb exposures and growth
 8                  effects in aquatic invertebrates. Available evidence is inadequate to conclude that there is
 9                  a causal relationship between Pb exposures and growth effects in aquatic vertebrates.
              7.4.12.3     Survival-Freshwater Biota

10                  The relationship between Pb exposure and survival has been well demonstrated in
11                  freshwater species as presented in Section 7.4.5 and Table 7-5 of the present document
12                  and in the previous Pb AQCDs. Pb exposure can either result in direct lethality or
13                  produce sublethal effects that diminish survival probabilities. Survival is a biologically
14                  important response that can have a direct impact on population size. However, the
15                  concentration typically reported at which there is 50% mortality of test organisms (LC50)
16                  is a poor measure for consideration of effects at ecologically-relevant concentrations.
17                  LC50 is a measure for acute toxicity whereas Pb effects on ecosystem receptors are likely
18                  characterized as a chronic, cumulative exposure rather than acute exposure. Furthermore,
19                  a scenario in which 50% of a population does not survive is likely not a sustainable
20                  population. From the LC50 data on Pb in this review and previous Pb AQCDs, a wide
21                  range of sensitivity to Pb is evident across taxa and within genera. However, the LC50 is
22                  usually much higher than current environmental levels of Pb in the U.S, even though
23                  physiological dysfunction that adversely impacts the fitness of an organism often occurs
24                  at concentrations well below lethal ones. When available, LCio, NOEC or LOEC are
25                  therefore reported.

26                  There are no studies reported in the previous Pb AQCDs or the current ISA for aquatic
27                  plants that indicate phytotoxicity at current concentrations of Pb in freshwater
28                  environments.

29                  There are considerable  data available on toxicity of Pb to aquatic invertebrates as
30                  reviewed in the previous Pb AQCDs and Ambient Water Quality Criteria for Lead (U.S.
31                  EPA. 1985) (U.S. EPA. 2008b). Table AX7-2.4.1 from the 2006 Pb AQCD summarizes
32                  LC50 data and other endpoints for freshwater and marine invertebrates (U.S. EPA.
33                  2006c). Recent studies  available since the 2006 Pb AQCD and draft Aquatic Life Water
34                  Quality Criteria for Pb that report mortality data are summarized in Table 7-5. Freshwater
      November 2012                             7-166                  Draft - Do Not Cite or Quote

-------
 1                  invertebrates are generally more sensitive to Pb exposure than other taxa, with survival
 2                  impacted in a few species at or near concentrations that are encountered in aquatic
 3                  environments (Table 7-2). These impacted species may include candidate or endangered
 4                  species. For example, the freshwater mussel L. rafinesqueana (Neosho mucket), is a
 5                  candidate for the endangered species list. The EC50 for foot movement (a measure of
 6                  viability) for newly transformed juveniles of this species was 188 (ig Pb/L. (Wang et al..
 7                  2010e).

 8                  Most of the evidence for Pb effects on survival in freshwater invertebrates is from
 9                  sensitive species of gastropods, amphipods, cladocerans and rotifers (Sections 7.4.5.2 and
10                  7.4.6). In some of these organisms, increased mortality is observed in the upper range of
11                  Pb concentration values found in surveys of U.S. surface waters (median 0.50 (ig Pb/L,
12                  range 0.04 to 30 (ig Pb/L) (U.S. EPA. 2006c). although the toxicity of Pb is highly
13                  dependent upon water quality variables such as DOC, hardness and pH. In the 1986 Pb
14                  AQCD, increased mortality was reported in the freshwater gastropod Lymnaeapalutris at
15                  Pb concentration as low as 19 (ig Pb/L effectively  reducing total biomass production
16                  (Borgmann et al.. 1978). Toxicity testing with amphipods under various water conditions
17                  indicate these organisms are sensitive to Pb at <10 (ig Pb/L (U.S. EPA. 2006c) and the
18                  present document). A 7 day LC50 of 1 (ig Pb/L was observed in soft water with the
19                  amphipod H. azteca ^Borgmann et al.. 2005,). In this same species, the 96-hour LC50 for
20                  Pb at pH of 5 was 10 (ig Pb/L (Mackie. 1989). In 42-day chronic exposures ofH. azteca
21                  exposed to  Pb via water and diet, the LC50 was 16  (ig Pb/L (Besser et al.. 2005). At
22                  higher pH and water hardness, amphipods are less  sensitive to Pb (U.S. EPA. 2006c). In a
23                  series of 48 hour acute toxicity tests with the cladoceran C. dubia conducted in a variety
24                  of natural waters across North America, LC50 values ranged  from 29 to 1,180 (ig Pb/L
25                  (NOEC range 18 to <985 (ig Pb/L) and were most significantly influenced by DOC and
26                  water ionic strength (Esbaugh et al.. 2011).  In the 2006 Pb AQCD the range of 48 hour
27                  LC50 values for C. dubia were reported from 280 to >2,700 (ig Pb/L when tested at
28                  varying pH levels (U.S. EPA. 2006c). In the rotifer genus Lecane,  a 22-fold difference in
29                  LC50 values was observed in 48-hour exposure to Pb between L. hamata, L. luna and
30                  L. quadridentata. (Perez-Legaspi and Rico-Martinez. 2001). L. luna was most sensitive to
31                  Pb toxicity with a 48-hour LC50 of 140  (ig Pb/L. In neonate rotifers, E. dilatata the 48-
32                  hour LC50 was 35 (ig Pb/L (Arias-Almeida and Rico-Martinez. 2011). A wide range of
33                  LC50 values were reported for chironomid species  (Table 7-5). however, the available
34                  evidence suggests these freshwater invertebrates are less sensitive to Pb than amphipods,
35                  cladocerans and rotifers. Other freshwater invertebrates such as odonates may be tolerant
36                  of Pb concentrations that greatly exceed concentrations of Pb reported in environmental
37                  media. Some invertebrates are able to detoxify Pb  such as through sequestration of Pb in
38                  the exoskeleton which is subsequently molted.
      November 2012                            7-167                 Draft - Do Not Cite or Quote

-------
 1                  There is considerable historic information on Pb toxicity to freshwater fish. Early
 2                  observations from mining areas where Pb and other metals were present indicated local
 3                  extinction offish from streams (U.S. EPA. 1977). The lowest LC50 for fish reported in the
 4                  1977 Pb AQCD was 1,000 (ig Pb/L in soft water for rainbow trout O. mykiss (reclassified
 5                  from Salmo gairdneri} following 96-hour exposure to Pb (U.S. EPA. 1977). Additional
 6                  LC50 values for freshwater fish are summarized in the 1985 Ambient Water Criteria for
 7                  Pb (U.S. EPA. 1985) and the draft Ambient Aquatic Life Water Quality Criteria for Pb
 8                  (U.S. EPA. 2008b). An LC50 of 236 jig Pb/L adjusted to a total hardness of 50 mg/L
 9                  CaCO3 was reported for O. mykiss in the draft Ambient Aquatic Life Water Quality
10                  Criteria for Pb.

11                  More recently reviewed studies using fish have considered the role of water quality
12                  variables and bioavailability on Pb toxicity. Higher toxicity tends to  occur in acidic
13                  waters where more free-Pb ion is available for uptake. The interactive effects of Pb
14                  concentration and water quality variables on toxicity may result in equivalent toxicity for
15                  a broad range of Pb concentrations. In a series of 96-hour acute toxicity tests with
16                  juvenile fathead minnow conducted in a variety of natural waters across North America,
17                  LC50 values ranged from 41 to 3,598 (ig Pb/L and no Pb toxicity occurred in three highly
18                  alkaline waters (Esbaugh et al.. 2011). In the 2006 Pb AQCD, the  96-hr LC50 values in
19                  fathead minnow ranged from 810->5,400 (ig Pb/L in varying pH and hardness (U.S.
20                  EPA. 2006c).

21                  Decreased survival is also a function of age of the fish. Thirty day LC50 values for larval
22                  fathead minnows ranged from 39 to 1,903 (ig Pb/L in varying concentrations of DOC,
23                  CaSO4 and pH (Grosell et al.. 2006b). In a recent study of rainbow trout fry at 2 to 4
24                  weeks post swim-up, the 96-hour LC50 was 120 (ig Pb/L at a hardness of 29  mg/L as
25                  CaCO3, a value much lower than in testing with older fish (Mebane et al.. 2008). In the
26                  same study, two chronic (>60 day) tests were conducted with rainbow trout and the
27                  NOECs for survival were 24 and 87 (ig Pb/L and the LOECs were 54 and 125 (ig Pb/L,
28                  respectively. In contrast to aqueous exposures, 30 day dietary studies with rainbow trout
29                  fathead minnow, and channel catfish fed a live diet of L. variegatus contaminated with Pb
30                  showed no statistically significant effects on survival  (Erickson et al.. 2010).

31                  Freshwater fish are less sensitive to Pb than freshwater invertebrates, however,  recent
32                  studies have highlighted the importance of considering pH, hardness and additional
33                  modifying factors in assessing toxicity since effects can vary over several orders of
34                  magnitude. Fish mortalities occur above the concentrations of Pb encountered in U.S.
35                  surface waters although, in some cases, the observed effects may be just above the upper
36                  measured  range of Pb in some aquatic environments (Table 7-2). Furthermore, although
37                  LC50 values are the most commonly reported, effects are occurring at lower
      November 2012                            7-168                 Draft - Do Not Cite or Quote

-------
 1                  concentrations. A more relevant indication of exposure impacts would be an LCi0 or
 2                  LOEC, however, these values are not always provided. The evidence is sufficient to
 3                  conclude that there is a causal relationship between Pb exposures and survival in
 4                  freshwater invertebrates and vertebrates. The evidence is inadequate to conclude that
 5                  there is a causal relationship between Pb exposures and survival in freshwater plants.
              7.4.12.4     Neurobehavioral Effects-Freshwater Biota

 6                  Evidence from laboratory studies and limited data from field studies reviewed in this
 7                  chapter, in the draft Ambient Aquatic Criteria document for Pb which updates the 1985
 8                  Ambient Water Quality Criteria for Pb (U.S. EPA. 1985). and in previous Pb AQCDs
 9                  have shown effects of Pb on neurological endpoints in aquatic animal taxa. These include
10                  changes in behaviors that may decrease the overall fitness of the organism such as
11                  avoidance responses, decreased ability of an organism to capture prey or escape
12                  predators, and alterations in feeding behaviors. Evidence of alteration in behaviors at the
13                  level of the organism is a potential endpoint for effects at population or community levels
14                  of biological organization (U.S. EPA. 2003a).

15                  In the 1977 Pb AQCD behavioral impairment of a conditioned response (avoidance of a
16                  mild electric shock) in goldfish was observed as low as 70 (ig Pb/L (Weir and Hine.
17                  1970). In the 2006 Pb AQCD several studies were reviewed in which Pb was shown to
18                  affect predator-prey interactions, including alteration in prey size choice and delayed prey
19                  selection in juvenile fathead minnows following 2-week pre-exposure to 500 (ig Pb/L
20                  (Weber. 1996). In limited studies available on worms, snails, tadpoles, hatchling turtles
21                  and fish there is evidence that Pb may affect the ability to escape or avoid predation. For
22                  example, in the tubificid worm T. tubifex the 96 hour EC50 for immobilization was
23                  42 (ig Pb/L (Khangarot.  1991). Some organisms exhibit behavioral avoidance while
24                  others do not seem to detect the presence of Pb (U.S. EPA.  2006c). Additional behavioral
25                  endpoints  reported in the Draft Ambient Aquatic Life Quality Criteria for Pb include an
26                  EC50 of 140 (ig Pb/L for feeding inhibition in the freshwater cladoceran C. dubia and
27                  deceased learning acquisition in bullfrogs at 500 (ig Pb/L (31.51 (ig Pb/L adjusted to a
28                  total hardness of 50 mg/L CaCO3). All of these effects occur at concentrations that
29                  exceed Pb concentration values found in surveys of U.S. surface waters although within
30                  the range of Pb detected near some mining-disturbed areas (Table 3-11).

31                  Recent information since the 2006 Pb AQCD provides evidence for Pb  impacts on
32                  behaviors  that may affect feeding and predator avoidance in freshwater environments at
33                  concentrations near the range of Pb detected in U.S.  surface waters (Table 7-2 and
34                  Section 7.4.5.3). Prey capture ability was decreased in 10 day old fathead minnow larvae
      November 2012                            7-169                 Draft - Do Not Cite or Quote

-------
 1                  born from adult fish exposed to 120 (ig Pb/L for 300 days, then subsequently tested in a
 2                  21-day breeding assay (Mager et al.. 2010). Another study in fish reported effects at
 3                  low (ig Pb/L concentration, however, the findings are not considered as strong evidence
 4                  for causality since exposure concentrations in water were not analytically verified.
 5                  Specifically, zebrafish embryos exposed nominally to concentrations of Pb (2.0 to
 6                  6.0 (ig Pb/L) until 24 hours post-fertilization and then subsequently tested as larvae
 7                  exhibited decreased startle response time and altered pattern of escape swimming (Rice et
 8                  al.. 2011). In adult fish raised from the exposed embryos (6.0 (ig Pb/L), the ability to
 9                  detect visual contrast was degraded. Although this study was conducted with nominal
10                  concentration of Pb in media, uptake of Pb by embryos was quantified and more  Pb was
11                  measured in tissues of embryos exposed to the higher concentration of Pb when
12                  compared to the lower exposure concentration. Additional studies are needed in fish to
13                  support these initial findings of effects on ecologically relevant behavioral impairments.

14                  Findings in  laboratory animals support the limited evidence for neurobehavioral effects
15                  of Pb in freshwater invertebrates and vertebrates. In animal toxicological studies  Pb
16                  induced changes in learning and memory (Section 5.3.2.3). as well as attention and motor
17                  skills (Section 5.3.3.1). New behaviors induced by exposure to Pb reviewed in Chapter_5
18                  that are relevant to effects of Pb observed in freshwater systems include effects on visual
19                  and auditory sensory systems and changes in structure and function of neurons and
20                  supporting cells in the brain. Mechanisms that include the displacement of physiological
21                  cations, oxidative stress and changes in neurotransmitters and receptors are also
22                  reviewed. Central nervous system effects in fish recognized in previous Pb AQCDs
23                  include effects on spinal neurons and brain receptors. New evidence from this review
24                  identifies the MAPKs ERK1/2 and pSS^15 as possible molecular targets for Pb
25                  neurotoxicity in catfish (Leal et al.. 2006). Evidence in terrestrial ecosystems
26                  (Section 7.3.12.4) is not as extensive, but it is highly coherent with findings in aquatic
27                  ecosystems. Overall, the evidence from available studies on neurobehavioral effects  of Pb
28                  in aquatic systems is limited, but sufficient to conclude that a causal relationship  is likely
29                  to exist between Pb exposures and neurobehavioral effects in aquatic invertebrates and
30                  vertebrates.
              7.4.12.5     Hematological Effects-Freshwater Biota

31                  Hematological responses are commonly reported effects of Pb exposure in aquatic
32                  invertebrates and vertebrates. Anemia was recognized as a symptom of chronic Pb
33                  poisoning in fish in the 1977 Pb AQCD and has been subsequently reported in various
34                  fish species using common hematological endpoints (e.g., red blood cell counts,
35                  hematocrit, Hb concentrations) (Mager. 2012). In the 1986 Pb AQCD, hematological
      November 2012                            7-170                  Draft - Do Not Cite or Quote

-------
 1                  effects of Pb exposure on fish included decrease in red blood cells and inhibition of
 2                  ALAD (U.S. EPA. 1986b). Inhibition of ALAD activity under various test conditions is
 3                  reported in Table 6 of the draft Ambient Aquatic Life Water Quality Criteria for Pb for
 4                  freshwater fish species (O. mykiss (Rainbow Trout), S. fontinalis (Brook Trout),
 5                  C. auratus (Goldfish) and Lepomis gibbosus (Pumpkinseed)) (U.S. EPA. 2008b). In these
 6                  studies, Rainbow Trout was the most sensitive with inhibition of ALAD activity reported
 7                  in multiple studies within the upper range of Pb in surface waters of the U.S. (median
 8                  0.5 ng Pb/L, range 0.04 to 30 jig Pb/L) (U.S. EPA. 2006c).

 9                  Laboratory studies with freshwater invertebrates have also indicated considerable species
10                  differences in ALAD activity in response to Pb. For example, the concentration at which
11                  50% ALAD inhibition was measured in the freshwater gastropod B. glabrata (23 to
12                  29 (ig Pb/L) was much lower than in the freshwater oligochaete L. variegatus
13                  (703 (ig Pb/L) based on nominal  exposure data (Aisemberg  et al.. 2005).

14                  Findings in laboratory studies are additionally supported by evidence from field-collected
15                  organisms providing coherence to the observations of Pb effects on ALAD activity. In
16                  environmental assessments of metal-impacted habitats, ALAD is a recognized biomarker
17                  of Pb exposure (U.S. EPA. 2006c). ALAD activity is negatively correlated with total Pb
18                  concentration in freshwater bivalves, and lower ALAD activity has been correlated with
19                  elevated blood Pb levels in field-collected fish as well. Further evidence from the
20                  2006 Pb AQCD and this review of Pb effects on ALAD enzymatic activity, including
21                  effects in bacteria, amphibians and additional field and laboratory studies on freshwater
22                  fish, confirms that the decreased  activity in this enzyme is an indicator for Pb exposure
23                  across a wide range of taxa and that a common mode of action is likely for invertebrates
24                  and  vertebrates. The  finding that the hematological system is a target for Pb  in natural
25                  systems is also supported by some evidence of Pb-induced alterations of serum profiles
26                  and  changes in white blood cell counts in fish (U.S. EPA. 2006c) and amphibians. This
27                  evidence is strongly coherent with evidence from terrestrial vertebrates (Section
28                  7.3.12.5). It is also coherent with observations from human  epidemiologic and animal
29                  toxicology studies (Section 5.7) where there is consistent evidence that exposure to Pb
30                  induces adverse effects on hematological endpoints, including altered heme  synthesis
31                  mediated through decreased ALAD and ferrochelatase activities, decreased red blood cell
32                  survival and function, and increased red blood cell oxidative stress. The overall weight of
33                  epidemiologic and toxicological evidence for humans was sufficient to conclude that a
34                  causal relationship exists between exposure to Pb and hematological effects  (Section 5.7).

35                  Based on observations in freshwater organisms and additionally supported by findings in
36                  terrestrial systems, saltwater invertebrates (Section 7.4.21.5). and by toxicological and
37                  epidemiologic evidence on human health effects, evidence is sufficient to conclude that
      November 2012                             7-171                  Draft - Do Not Cite or Quote

-------
 1                  there is a causal relationship between Pb exposures and hematological effects in
 2                  freshwater vertebrates. Evidence is sufficient to conclude that a causal relationship is
 3                  likely to exist between Pb exposures and hematological effects in freshwater
 4                  invertebrates.
              7.4.12.6     Physiological Stress-Freshwater Biota

 5                  Building on the body of evidence presented in the 2006 Pb AQCD (U.S. EPA. 2006c)
 6                  recent studies provide consistent and coherent evidence of upregulation of antioxidant
 7                  enzymes and increased lipid peroxidation associated with Pb exposure within one or two
 8                  orders of magnitude above current or ambient conditions in many species of freshwater
 9                  plants, invertebrates and vertebrates. A few studies provide  evidence of effects at
10                  concentrations of Pb encountered in some sediments of the U.S. (Table 7-2). In aquatic
11                  plants, increases of antioxidant enzymes with Pb exposure occur in algae, mosses, and
12                  floating and rooted aquatic macrophytes. Most available evidence for antioxidant
13                  responses in aquatic plants are from laboratory studies lasting from 2 to 7 days and at
14                  concentrations higher than typically found in the environment. However, data from
15                  transplantation experiments from non-polluted to polluted sites indicate that elevated
16                  enzyme activities are associated with Pb levels measured in sediments. For example, the
17                  freshwater macrophyte Myriophyllum quitense exhibited elevated antioxidant enzyme
18                  activity (glutathione-S-transferase, glutathione reductase, peroxidase) following
19                  transplantation in anthropogenically polluted areas containing elevated Pb concentrations.
20                  These were correlated with sediment Pb concentrations in the range of 5 to 23 mg Pb/g
21                  dry weight (Nimptsch et al. 2005).There is evidence for antioxidant activity in response
22                  to Pb exposure in freshwater invertebrates (i.e., bivalves). Markers of oxidative damage
23                  are also observed in fish, amphibians and mammals in laboratory studies. Across all
24                  organisms, there are differences in the  induction of antioxidant enzymes that appear to be
25                  species-dependent.

26                  Additional stress responses to Pb in a few aquatic invertebrates have been reported since
27                  the 2006 Pb AQCD, and included elevated heat shock proteins, osmotic stress, lowered
28                  metabolism and decreased glycogen levels associated with Pb exposure. Although these
29                  stress responses are correlated with Pb exposure, they are non-specific and may be altered
30                  with exposure to any number of environmental stressors. Heat shock protein induction
31                  has been observed in zebra mussels exposed to 500 (ig Pb/L for 10 weeks (Singer et al..
32                  2005). Crayfish exposed for 14 days to 500 (ig Pb/L exhibited a Pb-induced
33                  hypometabolism under conditions of environmental hypoxia in the presence of the metal
34                  (Morris et al., 2005). Glycogen levels in the freshwater snail B. glabrata were
      November 2012                             7-172                  Draft - Do Not Cite or Quote

-------
 1                   significantly decreased following 96-hour exposures at 50 (ig/L and higher (Ansaldo et
 2                   al.. 2006V

 3                   Evidence for stress responses observed in freshwater plants, invertebrates and vertebrates
 4                   is coherent with findings in terrestrial species (Section 7.3.12.6) and saltwater
 5                   invertebrates (Section 7.4.21.6). It is also coherent with evidence from human and
 6                   experimental animal studies of oxidative stress following impairment of normal metal ion
 7                   functions (Section 5.2.4). Upregulation of antioxidant enzymes and increased lipid
 8                   peroxidation are considered to be reliable biomarkers of stress, and provide evidence that
 9                   Pb exposure induces a stress response in those organisms which may increase
10                   susceptibility to other stressors and reduce individual fitness. Evidence is sufficient to
11                   conclude that a causal relationship is likely to exist between Pb exposures and
12                   physiological stress in freshwater aquatic plants, invertebrates and vertebrates.
              7.4.12.7     Community and Ecosystem Level Effects-Freshwater Biota

13                   Most direct evidence of community- and ecosystem-level effects in freshwater systems is
14                   from heavily contaminated sites where Pb concentrations are higher than typically
15                   observed environmental concentrations for this metal. Impacts of Pb on aquatic habitats
16                   that receive runoff from contaminated areas have been studied for several decades. For
17                   aquatic systems, the literature focuses on evaluating ecological stress from Pb originating
18                   from urban and mining effluents rather than atmospheric deposition. Ecosystem-level
19                   field studies are complicated by the confounding of Pb exposure with other factors such
20                   as the presence of trace metals and acidic deposition and by the  variability inherent in
21                   natural systems.  In natural systems, Pb is often found co-existing with other stressors, and
22                   observed effects may be due to cumulative toxicity.

23                   In laboratory studies and simulated ecosystems, where it is possible to isolate the effect
24                   of Pb, this  metal has been shown to alter competitive behavior of species, predator-prey
25                   interactions and contaminant avoidance. These dynamics may change species abundance
26                   and community structure at higher levels of ecological organization. Uptake of Pb into
27                   aquatic and terrestrial organisms and  subsequent effects on mortality, growth,
28                   developmental and reproductive endpoints at the organism level are expected to have
29                   ecosystem-level consequences, and thus provide consistency and plausibility for causality
30                   in ecosystem-level effects.

31                   In aquatic ecosystems, field studies reviewed in the 2006 Pb AQCD (summarized in
32                   Table AX7-2.5.2) and more recent studies report reductions of species abundance,
33                   richness or diversity. This is particularly the case for benthic macroinvertebrate
34                   communities where sources of Pb were mining or urban effluents, and Pb coexisted with

      November 2012                             7-173                  Draft - Do Not Cite or Quote

-------
 1                  other metals. The results often indicate a correlation between the presence of one or more
 2                  metals and the negative effects observed. For example, in the 2006 Pb AQCD, the Coeur
 3                  d'Alene River watershed in Idaho, U.S. was used as a case study for Pb effects at the
 4                  population and community level. Significant negative correlations were observed
 5                  between Pb in water column and total taxa richness and EPT taxa richness in the river. In
 6                  a simulated aquatic microcosm a reduction in abundance and richness of protozoan
 7                  species was observed with increasing Pb concentration from 50 to 1,000 (ig Pb/L
 8                  (Fernandez-Leborans and Antonio-Garcia. 1988).

 9                  Since the 2006 Pb AQCD, there is further evidence for effects of Pb in sediment-
10                  associated communities. Sediment-bound Pb contamination appears to differentially
11                  affect members of the benthic invertebrate community, potentially altering ecosystems
12                  dynamics in small urban streams (Kominkova and Nabelkova. 2005). Although surface
13                  water Pb concentrations in monitored streams were determined to be very low,
14                  concentrations of the metal in sediment were high enough to pose a risk to the benthic
15                  community (e.g., 34 to  101 mg Pb/kg). These risks were observed to vary with benthic
16                  invertebrate  functional feeding group, with collector-gatherer species exhibiting larger
17                  body burdens of heavy metals than benthic predators and collector-filterers.

18                  In a recent study conducted since the 2006 Pb AQCD, changes to aquatic plant
19                  community composition have been observed in the presence of elevated surface water Pb
20                  concentrations at three lake sites impacted by mining effluents. The site with highest Pb
21                  concentration (103-118 (ig Pb/L) had lowest number of aquatic plant species when
22                  compared to sites with lower Pb concentrations (78-92 (ig Pb/L) (Mishraet al.. 2008).
23                  Certain types of plants such as rooted and submerged aquatic plants may be more
24                  susceptible to aerially deposited Pb resulting in shifts in Pb community composition.
25                  High Pb sediment concentrations are linked to shifts in amphipod communities inhabiting
26                  plant structures.

27                  Avoidance response to Pb exposure varies widely in different species and this could
28                  affect community composition and structure and species abundance. For example, frogs
29                  and toads lack avoidance response while snails and fish avoid higher concentrations of Pb
30                  (U.S. EPA. 2006c).

31                  In the Annex to the 2006 Pb AQCD, the Coeur d'Alene River basin in  Idaho was
32                  presented as a case study for a watershed impacted by Pb and other metals. A significant
33                  negative correlation was observed between Pb in water column (0.5 to  30 (ig Pb/L) and
34                  total taxa richness, EPT taxa richness, and the number of metal-sensitive mayfly species
35                  (Maret et al.. 2003). Additional lines of evidence including mine density, metal
36                  concentrations, and bioaccumulation in caddisfly tissue were included. Since the
37                  2006 Pb AQCD, additional research at this site and model development has resulted in


      November 2012                            7-174                 Draft - Do Not Cite or Quote

-------
 1                  further characterization of the effects of Pb on waterfowl and other aquatic organisms in
 2                  this heavily contaminated ecosystem. Mean Pb concentrations in Coeur d'Alene sediment
 3                  range from 14 to 5,009 mg Pb/kg dry weight (Spears et al., 2007). Modeling of sediment
 4                  and Pb levels in waterfowl predict a sediment Pb effects range of 147-944 mg Pb/kg dry
 5                  weight and a mortality effects level of 1,652 mg/kg dry weight (Spears et al. 2007). In a
 6                  6-week feeding study with mallard (Anas platyrhynchos) ducklings, ingestion of
 7                  Pb-contaminated sediments from the Coeur d' Alene basin was shown to result in
 8                  decreased brain growth and altered brain chemistry (Douglas-Stroebel et al. 2004). These
 9                  findings support previous observations of altered behavior and hematological,
10                  hepatotoxic, and histopathological endpoints in waterfowl from Lake Coeur d'Alene that
11                  ingest Pb contaminated sediments and vegetation during feeding.

12                  In addition to the evidence from microcosm and field studies presented above, effects on
13                  reproduction (Section 7.4.12.1). growth (Section 7.4.12.2) and survival (Section 7.4.12.3)
14                  have been clearly demonstrated in freshwater species. These endpoints can have effects at
15                  the population-level and community-level of biological organization which may lead to
16                  ecosystem-level impacts. Although the evidence is strong for effects of Pb on growth,
17                  reproduction and survival in certain species in experimental settings at or near the range
18                  of Pb concentrations reported in surveys of U.S. freshwater systems, considerable
19                  uncertainties exist in generalizing effects observed under small-scale, particular
20                  conditions up to predicted effects at the ecosystem level of biological organization. In
21                  many cases it is difficult to characterize the nature and magnitude of effects and to
22                  quantify relationships between ambient concentrations of Pb and ecosystem response due
23                  to presence of multiple stressors, variability in field conditions and to differences in Pb
24                  bioavailability at that level of organization. Bioavailability of Pb is influenced by pH,
25                  alkalinity, total suspended solids, and DOC among other factors and can vary greatly in
26                  natural environments. Nevertheless, evidence of ecosystem effects in aquatic systems is
27                  coherent with similar evidence in terrestrial systems, and based on the cumulative
28                  evidence from laboratory studies and field observations, a causal relationship is likely to
29                  exist between Pb exposures and the alteration of species richness, species composition
30                  and biodiversity in freshwater aquatic ecosystems.
      7.4.13 Introduction to Bioavailability and Biological Effects of Pb in Saltwater
             Ecosystems

31                  Saltwater ecosystems include salt marsh, estuaries, embayments, beaches, and other
32                  coastal areas; and encompass a range of salinities from just above that of freshwater to
33                  that of seawater. These ecosystems may receive Pb contributions from direct atmospheric
34                  deposition and/or via runoff from terrestrial systems. A range of 0.01 to 27 (ig Pb/L
      November 2012                            7-175                  Draft - Do Not Cite or Quote

-------
 1                  including coastal areas, estuaries and open ocean was reported by Sadiq (1992) with the
 2                  higher values associated with sites involving human activity, however, these values are
 3                  not specific to the U.S. (Table 7-2). In an earlier publication, levels of Pb in the North
 4                  Atlantic and North Pacific surface waters ranged from 0.005 to 0.4 (ig Pb/L but the range
 5                  of values in coastal waters and estuaries were approximately equal to the range of Pb in
 6                  freshwater (Leland and Kuwabara. 1985). Additional information on Pb levels in water is
 7                  available in Sections 7.2.3 and 3.6. The 2006 Pb AQCD provided an overview of
 8                  regulatory considerations  for water and sediments in addition to consideration of
 9                  biological effects and major environmental factors that modify the response of marine
10                  organisms to Pb exposure. Regulatory guidelines for Pb in saltwater have not changed
11                  since the 2006 Pb AQCD  and  are summarized below. This section is followed by new
12                  information on bioavailability and biological effects of Pb in saltwater since the
13                  2006 Pb AQCD.

14                  The most recent ambient water quality criteria for Pb in saltwater were released in 1985
15                  (U.S. EPA. 1985) by the EPA Office of Water which employed  empirical regressions
16                  between observed toxicity and water hardness to develop hardness-dependent equations
17                  for acute and chronic criteria. These criteria are published pursuant to Section 304(a) of
18                  the Clean Water Act and provide guidance to states and tribes to use in adopting water
19                  quality standards for the protection of aquatic life and human health in surface water. The
20                  ambient water quality criteria for Pb are currently expressed as a criteria maximum
21                  concentration (CMC) for acute toxicity and criterion continuous concentration (CCC) for
22                  chronic toxicity (U.S. EPA. 2009b). In saltwater, the CMC is 210 (ig Pb/L and the CCC
23                  is 8.1 (ig Pb/L. The 2006 Pb AQCD summarized two approaches for establishing
24                  sediment criteria for Pb based on either bulk sediment or equilibrium partitioning as
25                  reviewed in the present document in Section 7.4.2.

26                  In the following sections,  recent information available since the  2006 Pb AQCD on Pb in
27                  marine and estuarine ecosystems will be presented. Throughout the sections, brief
28                  summaries of conclusions from the 1977 Pb AQCD (U.S. EPA.  1977). the  1986 Pb
29                  AQCD (U.S. EPA. 1986b) and the 2006 Pb AQCD (U.S. EPA. 2006b) are included
30                  where appropriate. Recent research on the bioavailability and uptake of Pb  into saltwater
31                  organisms including plants, invertebrates and vertebrates is presented in Section 7.4.14.
32                  Toxicity of Pb to marine flora and fauna including growth, reproductive and
33                  developmental effects (Section 7.4.15) are followed with data on exposure and response
34                  of saltwater organisms (Section 7.4.16). Responses at the community and ecosystem
35                  levels of biological organization are reviewed in Section 7.4.17 followed by
36                  characterization of sensitivity  and vulnerability of saltwater ecosystem components
37                  (Section 7.4.18) and a discussion of ecosystem services (Section 7.4.19). The saltwater
38                  sections conclude with a synthesis of the new data for Pb effects on saltwater plants,
      November 2012                            7-176                 Draft - Do Not Cite or Quote

-------
 1                  invertebrates and vertebrates (Section 7.4.20) and causal determinations based on
 2                  evidence from previous Pb AQCDs and recent studies (Section 7.4.21).
      7.4.14 Bioavailability of Pb in Saltwater Systems

 3                  Bioavailability was defined in the 2006 Pb AQCD as "the proportion of a toxin that
 4                  passes a physiological membrane (the plasma membrane in plants or the gut wall in
 5                  animals) and reaches a target receptor (cytosol or blood)". In 2007, EPA took cases of
 6                  bioactive adsorption into consideration and revised the definition of bioavailability as
 7                  "the extent to which bioaccessible metals absorb onto, or into, and across biological
 8                  membranes of organisms, expressed as a fraction of the total amount of metal the
 9                  organism is proximately exposed to (at the sorption surface) during a given time and
10                  under defined conditions" (U.S. EPA. 2007c)

11                  Factors affecting bioavailability of Pb to marine organisms are the same as those in
12                  freshwater systems (Sections 7.4.2 and 7.4.4). However, although routes of exposure and
13                  physiological mechanisms for storage and excretion influence uptake of metals by all
14                  organisms, they may be different in marine organisms, particularly for ion transport
15                  mechanisms (Niyogi and Wood. 2004). Marine environments are characterized by higher
16                  levels of ions, such as Na+, Ca2+, and Mg2+, which compete for potential binding sites on
17                  biotic ligands such as gills, thereby generally reducing the effective toxicity of metal ions
18                  as compared to freshwater environments. However, because the concentrations of these
19                  ions are relatively constant, bioavailability may be more predictable in marine systems
20                  that are little influenced by freshwater than in freshwater systems, varying mostly with
21                  amount and type of dissolved organic matter. In estuaries and embayments, changing
22                  salinities and proximity to anthropogenic loading of pollutants add to the complexity of
23                  predicting Pb speciation in these dynamic systems. BLMs (Figure 7-3) now being
24                  developed for marine organisms are functionally similar to those applied to freshwater
25                  organisms (Section 7.4.4).

26                          Although in freshwater systems the presence of humic acid is considered to
27                  reduce the bioavailable fraction of metals in freshwater, there is evidence that
28                  DOC/DOM does not have the same effect on free Pb ion concentration in marine systems
29                  (see Section 7.4.2.4 for detailed discussion).  For the sea urchin P.  lividus, the presence of
30                  humic acid increased both the uptake and toxicity of Pb possibly by enhancing uptake of
31                  Pb via membrane Ca2+ channels (Sanchez-Marin et al., 2010b). This also was observed in
32                  the marine diatom Thalassiosira weissflogii,  where humic acids absorbed to cell surfaces
33                  increased metal uptake (Sanchez-Marin et al., 2010b). Formation of a ternary complex
34                  that is better absorbed by biological membranes was another proposed mechanism that
      November 2012                            7-177                 Draft - Do Not Cite or Quote

-------
 1                  could describe the increased bioavailability to marine invertebrates of Pb bound to humic
 2                  acid (Sanchez-Marin et al.. 2007).

 3                  Sanchez-Marin et al. (2011) subsequently have shown that different components of DOM
 4                  have different effects on Pb bioavailability in marine systems. Their initial research using
 5                  Aldrich humic acid found that increasing humic acid concentrations increased Pb uptake
 6                  by mussel gills and increased toxicity to sea urchin larvae in marine environments
 7                  (Sanchez-Marin et al.. 2007). In contrast, a subsequent investigation found that fulvic
 8                  acid reduced Pb bioavailability in marine water (Sanchez-Marin et al.. 2011). The
 9                  contradictory effects of different components of DOM on marine bioavailability likely
10                  reflect their distinct physico-chemical characteristics. More hydrophobic than fulvic acid,
11                  humic acid may adsorb directly with cell membranes and enhance Pb uptake through
12                  some (still unidentified) mechanism (Sanchez-Marin et al.. 2011). Pb AVS-measurements
13                  were also determined to accurately predict uptake by mussels (Mytilus sp.) in the
14                  presence of 2.5 to 20 mg/L fulvic acid (Sanchez-Marin et al.. 2011). However, the effects
15                  of DOM on Pb bioavailability to mussels were underpredicted by AVS Pb concentration
16                  measurements, potentially as a result of adsorption of DOM-Pb complexes.

17                  Based on the above, BLMs (see Section 7.4.4 and Figure 7-3) used to predict
18                  bioavailability of Pb to aquatic organisms (Pi Toro et al.. 2005). may require
19                  modifications for application to marine organisms. Of particular importance is the finding
20                  that in marine aquatic systems, surface water DOM was found to increase (rather than
21                  decrease) uptake of Pb by fish gill structures, potentially through the alteration of
22                  membrane Ca2+ channel permeability. Veltman et al. (2010) proposed integrating BLM
23                  and bioaccumulation models in order to more accurately predict metal uptake by fish and
24                  invertebrates, and calculated metal absorption efficiencies for marine fish species from
25                  both types of models. They noted that affinity constants for Ca2+, Cd, Cu, Na, and Zn
26                  were highly similar across different aquatic species, including fish and invertebrates
27                  (Veltman et al.. 2010). These findings suggest that the BLM can be integrated with
28                  bioaccumulation kinetics to account for both environmental chemical speciation and
29                  biological and physiological factors in both marine and freshwater systems.
              7.4.14.1      Saltwater Plants and Algae

30                  In the 1977 Pb AQCD, the cordgrass Spartina alterniflora was found to reduce by a small
31                  amount the quantity of Pb in sediments (U.S. EPA. 1977). Limited data on marine algal
32                  species  reviewed in the 1986 Pb AQCD and 2006 Pb AQCD provided additional
33                  evidence for Pb uptake. Recent data available since the 2006 Pb AQCD includes Pb
34                  bioaccumulation studies conducted with five species of marine algae, (Tetraselmis chuii,
      November 2012                            7-178                 Draft - Do Not Cite or Quote

-------
 1                  Rhodomonas satinet, Chaetoceros sp., Isochrysis galbana and Nannochloropsis
 1                  gaditana). In this study it was demonstrated that bioaccumulation rates varied with
 3                  species following 72-hour exposure to Pb. /. galbana accumulated the lowest
 4                  concentrations of Pb (0.01 and 0.6 pg Pb/cell at water concentrations of 51 and
 5                  6,348 (ig Pb/L), while Chaetoceros sp. was observed to be the most efficient Pb
 6                  bioaccumulator, adsorbing 0.04 and 54 pg Pb/cell at 51 and 6,348 (ig Pb/L (Debelius et
 7                  al.. 2009).

 8                  Recent uptake studies of Pb in plants associated with marine environments are also
 9                  available. The roots of two salt marsh species, Sarcocornia fruticosa and Spartina
10                  maritima, significantly accumulated Pb, to maximum concentrations of 2,870 mg Pb/kg
11                  and 1,755 mg Pb/kg, respectively (Caetano et al., 2007). Roots had similar isotopic
12                  signature to those of sediments in vegetated zones indicating that Pb uptake by plants
13                  reflects the input in sediments. BCFs for Pb in root tissue from mangrove tree species
14                  range between 0.09 and  2.9, depending on the species and the habitat, with an average
15                  BCF of 0.84. The average BCF for mangrove species leaf tissue was considerably less
16                  (0.11), as these species are poor translocators of Pb (MacFarlane et al.. 2007).
              7.4.14.2     Saltwater Invertebrates

17                  Uptake and subsequent bioaccumulation of Pb in marine invertebrates varies greatly
18                  between species and across taxa as previously characterized in the 2006 Pb AQCD. This
19                  section expands on the findings from the 2006 Pb AQCD on bioaccumulation and
20                  sequestration of Pb in saltwater invertebrates. In the case of invertebrates, Pb can be
21                  bioaccumulated from multiple sources, including the water column, sediment, and dietary
22                  exposures, and factors such as proportion of bioavailable Pb, lifestage, age, and
23                  metabolism can alter the accumulation rate. In this section, new information on Pb uptake
24                  and subsequent tissue  and subcellular distribution will be considered, followed by a
25                  discussion on dietary and water routes of exposure and strategies for detoxification of Pb
26                  in marine invertebrates.

27                  In marine invertebrates, sites for Pb accumulation include the gill and digestive
28                  gland/hepatopancreas. The gills were the main sites  of Pb accumulation in pearl oyster,
29                  Pinctada fucata followed by mantle, in 72-hour exposures to 103.5 (ig Pb/L (Jing et al..
30                  2007). Following a 10-day exposure to 2,500 (ig Pb/L as Pb nitrate, accumulation of Pb
31                  was higher in gill than digestive gland ofMytilus edulis: after a 10 day depuration, Pb
32                  content was decreased in the gills and digestive gland of these mussels (Einsporn et al..
33                  2009). In blue crabs, Callinectes sapidus, collected from a contaminated and a clean
      November 2012                            7-179                  Draft - Do Not Cite or Quote

-------
 1                   estuary in New Jersey, U.S., the hepatopancreas was found to be the primary organ for Pb
 2                   uptake (Reichmuth et al.. 2010).

 3                   There is more information now on the cellular and subcellular distribution of Pb in
 4                   invertebrates than there was at the time of writing the 2006 Pb AQCD. Specifically,
 5                   localization of Pb at the ultrastructural level has been assessed in the marine mussel (M
 6                   edulis) through an antibody-based detection method (Einsporn et al.. 2009; Einsporn and
 7                   Koehler. 2008). Dissolved Pb was detected mainly within specific lysosomal structures in
 8                   gill epithelial cells and digestive gland cells and was also localized in nuclei and
 9                   mitochondria. Transport of Pb is thought to be via lysosomal granules associated with
10                   hemocytes (Einsporn et al.. 2009). In the digestive gland of the variegated scallop
11                   (Chlamys varia), Pb was also mainly bound to organelles, (66% of the total metal burden)
12                   (Bustamante and Miramand. 2005). In the digestive gland of the cephalopod Sephia
13                   officinalis, (cuttlefish) most of the Pb was found in the organelles (62%) (Bustamante et
14                   al.. 2006). In contrast, only 7% of Pb in the digestive gland of the octopus (Octopus
15                   vulgaris) was associated with the fraction containing nuclei, mitochondria, lysosome and
16                   microsomes: the majority of Pb in this species was found in cytosolic proteins (Raimundo
17                   et al.. 2008).

18                   Metian et al. (2009) investigated the uptake and bioaccumulation of 210Pb in variegated
19                   scallop and king scallop to determine the major accumulation route (seawater or food)
20                   and then assess subsequent tissue distribution. Dietary Pb from phytoplankton in the diet
21                   was poorly assimilated (<20%) while more than 70% of Pb in seawater was retained in
22                   the tissues. In seawater, 210Pb was accumulated more rapidly in variegated scallop than
23                   king scallop and soft tissue distribution patterns differed between the species. Variegated
24                   scallop accumulated Pb preferentially in the digestive gland (50%) while in king scallop,
25                   Pb was equally distributed in the digestive gland, kidneys, gills, gonad, mantle, intestine,
26                   and adductor muscle with each tissue representing 12-30% of 210Pb body load. An
27                   additional test with Pb-spiked sediment in king scallop showed low bioaccumulation
28                   efficiency of Pb from spiked sediment.

29                   Recently, several studies have attempted to establish biodynamic exposure assessments
30                   for various contaminants. In an in situ metal kinetics field study with the mussel
31                   M. galloprovincialis, simultaneous measurements of metal concentrations in water and
32                   suspended particles with mussel biometrics and physiological indices were conducted to
33                   establish uptake and excretion rates in the natural environment (Casas et al., 2008). The
34                   mean logarithmic ratio of metal concentration in mussels (ng/kg of wet-flesh weight) to
35                   metal concentration in water (ng/L) was found to be 4.3  inM galloprovincialis, based on
36                   the rate constants of uptake and efflux in a series of transplantation experiments between
      November 2012                             7-180                  Draft - Do Not Cite or Quote

-------
 1                  contaminated and clean environments. Equilibrium concentrations of Pb in mussels
 2                  leveled out at approximately 30 days with a concentration of 6.7 mg Pb/kg.

 3                  The protective barrier against Pb toxicity formed by the egg structure in some
 4                  invertebrates was recognized in the 2006 Pb AQCD. Consideration of toxicity of Pb to
 5                  embryos that develop surrounded by a protective egg shell has been expanded since the
 6                  2006 Pb AQCD. In a study with cuttlefish (S. officinalis) eggs, radioisotopes were used to
 7                  assess the permeability of the egg to Pb at low exposure concentrations (210Pb activity
 8                  concentration corresponding to 512 (ig/L Pb) (Lacoue-Labarthe et al., 2009). Retention
 9                  and diffusion properties of the cuttlefish egg change throughout the development of the
10                  embryo and since the eggs are fixed on substrata in shallow coastal waters they may be
11                  subject to both acute and chronic Pb exposures. In the radiotracer experiments, 210Pb was
12                  never detected in the internal compartments of the egg during the embryonic
13                  development stage, while concentrations in the eggshell increased throughout the 48-day
14                  exposure. These results are consistent with a study of cuttlefish eggs collected from
15                  natural environments in which Pb was only detected in the eggshell. These studies
16                  indicate that the cuttlefish egg provides a protective barrier from Pb toxicity (Miramand
17                  et al.. 2006).

18                  Aquatic invertebrate strategies for detoxifying Pb were reviewed in the 2006 Pb AQCD
19                  and include sequestration of Pb  in lysosomal-vacuolar systems, excretion of Pb by some
20                  organisms, and deposition of Pb to molted exoskeleton. Molting of the exoskeleton can
21                  result in depuration of Pb from the body (see Knowlton et al. (1983) and Anderson et al.
22                  (1997). as cited in the 2006 Pb AQCD). New research has provided further evidence of
23                  depuration of Pb via molting in  invertebrates. Mohapatra et al. (2009) observed that Pb
24                  concentrations in body tissues were lower in the newly molted mud crabs (Scylla serrata)
25                  than in the pre-molt, hard-shelled crabs. However, the carapace of hard shelled crabs had
26                  lower concentrations of Pb than the exuvium of the soft shell crabs, leading the authors to
27                  speculate that some of the metal might be partially excreted during the molting process,
28                  rather than entirely through shedding of the previous exoskeleton. Bergey and Weis
29                  (2007) showed that differences in the proportion of Pb stored in exoskeleton and soft
30                  tissues changed during intermolt and immediate postmolt in two populations of fiddler
31                  crabs (Ucapugnax) collected from New Jersey. One population from a relatively clean
32                  estuary eliminated an average of 56% of Pb total body burden during molting while
33                  individuals from a site contaminated by metals eliminated an average of 76% of total Pb
34                  body burden via this route. Pb distribution within the body of crabs from the clean site
35                  shifted from exoskeleton to soft tissues prior to molting. The authors observed the
36                  opposite pattern of Pb distribution in fiddlers from the contaminated site where larger
37                  amounts of Pb were depurated in the exoskeleton. The exact dynamics of Pb depuration
38                  through molting in crabs are  thus still not completely characterized.
      November 2012                            7-181                  Draft - Do Not Cite or Quote

-------
              7.4.14.3     Saltwater Vertebrates

                    Saltwater Fish

 1                  In comparison to freshwater fish, fewer studies have been conducted on Pb uptake in
 2                  marine fish. Since marine fish drink seawater to maintain osmotic homeostasis, Pb can be
 3                  taken up via both gills and intestine (Wang and Rainbow. 2008). Pb was significantly
 4                  accumulated in gill, liver, plasma, kidney, rectal gland, intestine, skin, and muscle of the
 5                  elasmobranch spotted dogfish (Scyliorhinus canicula) exposed to 2,072 (ig Pb/L for one
 6                  week (De Boeck et al., 2010). In contrast to Pb distribution patterns in freshwater
 7                  teleosts, high Pb concentrations were present in this species in the skin and rectal gland.
 8                  Egg cases of the spotted dogfish exposed to 210Pb in seawater for 21 days, accumulated
 9                  radiolabeled Pb rapidly and the metal was subsequently detected in embryos indicating
10                  the permeability of shark eggs to  Pb in coastal environments (Jeffree et al., 2008). A
11                  study of Pb bioaccumulation in five marine fish species (Chloroscombrus chrysurus,
12                  Sardinella aurita, Ilisha africana, Galeoides decadactylus, Caranx latus) found that
13                  C. chrysurus was an especially strong bioaccumulator, yielding Pb concentrations of 6 to
14                  10 mg Pb/kg (Gnandi et al., 2006). However, C. chrysurus metal content was not
15                  correlated to the Pb concentrations along the mine tailings gradient from which they were
16                  collected (8.5 and 9.0 (ig Pb/L for minimum  and maximum tissue concentrations,
17                  respectively). This lack of correlation was also observed for fish species that were
18                  considered to be weaker Pb bioaccumulators, indicating that unidentified sources of Pb
19                  (e.g., in sediments or in dietary sources) may be contributing to Pb uptake by marine fish.

20                  In grunt fish H. scudderi, exposed to Pb via dietary uptake through a simulated marine
21                  food chain, mean total Pb body burden increased from 0.55 to 3.32 mg Pb/kg in a 42-day
22                  feeding study (Soto-Jimenez et al.. 201 Ib). Pb was accumulated to the highest relative
23                  concentration in liver with less than 3% of total Pb accumulated in gills.  Most of the Pb
24                  based on total body mass was accumulated in skeleton, skin, scales and muscle.

25                  The 2006 Pb AQCD considered detoxification mechanisms in fish including mucus
26                  production and Pb removal by shedding of scales in which Pb is chelated with keratin.
27                  Since the 2006 Pb AQCD review, additional Pb detoxification mechanisms in marine fish
28                  have been further elucidated. Mummichog (Fundulus heteroclitus) populations in metal-
29                  polluted salt marshes in New York exhibited different patterns of intracellular
30                  partitioning of Pb although body burden between sites was not significantly different
31                  (Goto and Wallace. 2010). Mummichogs at more polluted sites stored a higher amount of
32                  Pb in metal rich granules as compared to other detoxifying cellular components such as
33                  heat-stable proteins, heat-denaturable proteins and organelles.
      November 2012                            7-182                  Draft - Do Not Cite or Quote

-------
                    Marine Mammals
 1                  Studies that consider uptake of Pb in aquatic mammals are limited. Kannan et al. (2006)
 2                  compared trace element concentrations in livers of free-ranging sea otters (Enhydra lutris
 3                  nereis) found dead along the California coast. They detected Pb in all individuals
 4                  sampled (N=80) in a range of 0.019 to 1.06 mg Pb/kg. The otters were classified by cause
 5                  of death (infectious causes, non-infectious causes, emaciated condition) and trace element
 6                  patterns of tissue distribution were compared. Livers from emaciated otters had
 7                  significantly elevated levels of Pb compared to non-diseased individuals.
              7.4.14.4     Marine Food Web

 8                  As discussed in Section 7.4.4.4 trophic transfer of Pb through aquatic food chains was
 9                  considered to be negligible in the 2006 Pb AQCD (U.S. EPA. 2006c). Measured
10                  concentrations of Pb in the tissues of aquatic organisms were found to be generally higher
11                  in algae and benthic organisms and lower in higher trophic-level consumers, indicating
12                  that Pb was bioaccumulated but not biomagnified (U.S. EPA. 2006c; Eisler. 2000).
13                  Recent literature since the 2006 Pb AQCD, provides evidence of the potential for Pb to
14                  be transferred in marine food webs while other studies indicate Pb is decreased with
15                  increasing trophic level. This section incorporates recent literature on transfer of Pb
16                  through marine food chains.

17                  In a dietary study using environmentally realistic concentrations of Pb in prey through
18                  four levels of a simplified marine food chain, biological responses including decreased
19                  growth and survival and changes in behavior were observed at different trophic levels.
20                  However, the concentration of Pb did not increase along the trophic gradient (Soto-
21                  Jimenez et al.. 201 Ib; Soto-Jimenez et al.. 201 la). The base of the simulated food chain
22                  was the microalgae Tetraselmis suecica (phytoplankton) grown in 20 (ig Pb/L.
23                  Pb-exposed cultures of T. suecica had significantly less cell divisions per day  (growth),
24                  biomass and total cell concentrations than control microalgae at 72 hours of exposure.
25                  The microalgal cultures were then fed to Artemia franciscana (crustacean, brine shrimp)
26                  which were then fed to Litopenaeus vannamei  (crustacean, whiteleg shrimp) and finally
27                  to Haemulon scudderi (fish, grunt). Effects on behavior, growth and survival were
28                  observed in shrimp and in grunt fish occupying the intermediate and top levels of the
29                  simulated marine food chain. The authors speculate that the species used in the simulated
30                  food chain were able to regulate and eliminate Pb (Soto-Jimenez et al.. 201 Ib).

31                  Partial evidence for biomagnification was observed in  a subtropical lagoon in  Mexico
32                  with increases of Pb concentration occurring in 14 of the 31 (45.2%) of trophic
33                  interactions considered (Ruelas-Inzunza and Paez-Osuna. 2008). The highest rate of


      November 2012                            7-183                  Draft - Do Not Cite or Quote

-------
 1                  transference of Pb as measured in muscle tissue occurred between the prey species
 2                  whiteleg shrimp (Litopenaeus vannamei) and mullet (Mugil cephalus) to pelican
 3                  (Pelecanus occidentalis).

 4                  Other studies have traced Pb in marine food webs and have found no evidence of
 5                  biomagnification of Pb with increasing trophic level. In the southeastern Gulf of
 6                  California, Mexico,  Pb was not positively transferred (biomagnification factor <1)
 7                  through primary producers (seston, detritus) and 14 consumer species in a lagoon food
 8                  web (Jara-Marini et  al., 2009). In a planktonic food web in Bahia Blanca estuary,
 9                  Argentina, Pb levels in macrozooplankton and mesozooplankton exhibited temporal
10                  fluctuations, however no biomagnification was observed between mesozooplankton and
11                  macrozooplankton (Fernandez Severini et al., 2011). It is important to note, however, that
12                  even in the absence  of biomagnification, aquatic organisms can bioaccumulate relatively
13                  large amounts of metals and become a significant source of dietary metal to their
14                  predators (Fairbrother et al.. 2007: Reinfelder et al.. 1998).
      7.4.15 Biological Effects of Pb in Saltwater Systems

15                  This section focuses on the studies of biological effects of Pb on marine and estuarine
16                  algae, plants, invertebrates, fish and mammals published since the 2006 Pb AQCD. Key
17                  studies from the 1977 Pb AQCD, the 1986 Pb AQCD and the 2006 Pb AQCD on
18                  biological effects of Pb are summarized where appropriate. Biological effects of Pb on
19                  saltwater algae and plant species are considered below, followed by information on
20                  effects on marine invertebrates and vertebrates. Alterations to reproduction, growth and
21                  survival of saltwater organisms can lead to changes at the community and ecosystem
22                  levels of biological organization such as decreased abundance, reduced taxa richness, and
23                  shifts in species composition (Section 7.1). New evidence for Pb effects on reproduction,
24                  growth and survival in saltwater plants, invertebrates and vertebrates is summarized in
25                  Table 7-6. In general, Pb toxicity to saltwater organisms is less well characterized than
26                  toxicity of Pb in freshwater ecosystems due to the fewer number of available studies on
27                  marine species. Because this review is focused on effects of Pb, studies reviewed for this
28                  section include only those for which Pb was the only, or primary, metal to which the
29                  organism was exposed. All reported values are from exposures in which concentrations
30                  of Pb were analytically verified unless nominal concentrations are stated.
      November 2012                            7-184                 Draft - Do Not Cite or Quote

-------
              7.4.15.1      Saltwater Algae and Plants

 1                  New evidence on toxicity of Pb to marine algae indicates that species exhibit varying
 2                  sensitivities to Pb in saltwater. The lowest 72-hour EC50 for growth inhibition reported for
 3                  marine algae was 105 (ig Pb/L in Chaetoceros sp (Debelius et al.. 2009). The microalgae
 4                  T. suecica, had statistically significant decreased biomass yield, growth rate and cell
 5                  count following 72 hours nominal exposure to 20 (ig Pb/L (Soto-Jimenez et al.. 201 Ib).
 6                  Pb tested at nominal concentrations up to -2,000  (ig Pb/L over a 14-day period did not
 7                  affect photosynthetic activity in seven species of marine macroalgae (Ascophyllum
 8                  nodosum, Fucus vesiculosus,  Ulva intestinalis, Cladophora rupestris, Chondrus crispus,
 9                  Palmaria palmate, Polysiphonia lanosa) as measured by pulse amplitude modulation
10                  chlorophyll fluorescence yield although Pb was readily accumulated by these species
11                  (Baumann et al.. 2009). In a recent review of the production of phytochelatins and
12                  glutathione by marine phytoplankton in response  to metal stress, Kawakami et al. (2006)
13                  included several studies in which Pb exposure was shown to induce glutathione and
14                  phytochelatin at high concentrations in a few species.
              7.4.15.2      Saltwater Invertebrates

15                  No studies with marine invertebrates were reviewed in the 1977 Pb AQCD or the 1986
16                  Pb AQCD. Effects of Pb on marine invertebrates reported in the 2006 Pb AQCD included
17                  impacts on embryo development in bivalves with an EC50 of 221 (ig Pb/L for
18                  embryogenesis, gender differences in sensitivity to Pb in copepods and increasing
19                  toxicity with decreasing salinity in mysids. Survival, growth and reproduction are
20                  affected by Pb in marine organisms. Pb has also been shown to affect stress responses,
21                  antioxidant activity and osmoregulation.

22                  Recent literature strengthens the evidence indicating that  Pb affects enzymes and
23                  antioxidant activity in marine invertebrates. Most of these studies only report nominal
24                  concentrations of Pb. Activity of enzymes associated with the immune defense system in
25                  the mantle of pearl oyster were measured at 0, 24, 48 and 72 hour nominal exposure to
26                  100 (ig Pb/L (Jing et al.. 2007). Activity of AcPase, a lysosomal marker enzyme, was
27                  detected at 24 hours and subsequently decreased. Phenoloxidase activity was depressed
28                  compared with controls and remained significantly lower than control after 72 hours of
29                  exposure to Pb. Increased SOD activity was observed in the mantle but decreased with
30                  time, although always remaining higher than in the control animals (Jing et al.. 2007).
31                  Activity of Se-dependent glutathione peroxidase did not change with Pb exposure. SOD,
32                  catalase, and glutathione peroxidase were significantly reduced at environmentally
33                  relevant concentrations of Pb (2 (ig Pb/L as measured in Bohai Bay, China) in the
      November 2012                            7-185                  Draft - Do Not Cite or Quote

-------
 1                   digestive gland of the bivalve Chlamys farreri fZhang et al., 20101x). In contrast,
 2                   Einsporn et al. (2009) observed no change in catalase activity in the digestive gland and
 3                   gill of blue mussel M edulis following nominal exposure to 2,500 (ig Pb/L as Pb nitrate
 4                   for 10 days and again following a 10 day depuration period. However, in this same
 5                   species, glutathione-S-transferase activity was elevated in the gills after Pb exposure and
 6                   remained active during depuration while no changes to glutathione-S-transferase activity
 7                   were observed in the digestive gland. In black mussel (M galloprovincialis) exposed 10
 8                   days to sublethal nominal concentrations of Pb, fluctuations in SOD activity were
 9                   observed over the length of the exposure and MDA levels were increased in mantle and
10                   gill (Vlahogianni and Valavanidis. 2007). Catalase activity was decreased in the mantle
11                   of these mussels but fluctuated in their gills, as compared with the control group. In the
12                   bivalve C. farreri exposed to Pb, there was induction of lipid peroxidation measured as
13                   MDA of 24% and a 37% reduction in 7-ethoxyresorufm-o-deethylase (EROD) activity
14                   when compared to controls (Zhang et al., 2010b). In red fingered marsh crab,
15                   Parasesarma erythrodactyla, collected from sites along an estuarine lake in New South
16                   Wales, Australia, elevated glutathione peroxidase activity was correlated with individuals
17                   with higher metal body burdens (MacFarlane et al.. 2006).

18                   ALAD is a recognized biomarker of exposure across a wide range of taxa including
19                   bacteria (Korean et al., 2007). invertebrates and vertebrates. Since the 2006 Pb AQCD,
20                   there are additional studies measuring changes in ALAD activity in field-collected
21                   bivalves and crustaceans from saltwater habitats. In the bivalve Chamelea gallina
22                   collected from the coast of Spain, ALAD inhibition was greater with higher
23                   concentrations of Pb measured in whole tissue (Kalman et al., 2008). In another study
24                   conducted in Spain, ALAD activity was negatively correlated with total Pb concentration
25                   in seven marine bivalves (C. gallina, Mactra corallina, Donax trunculus, Cerastoderma
26                   edule, M. galloprovincialis, Scrobicularia plana and Crassostrea angulata). However,
27                   the authors of this study indicated the need to consider variability of responses between
28                   species when using ALAD as a biomarker for Pb (Company et al.. 2011). Pb content
29                   varied significantly among species and was related to habitat (sediment versus substrate)
30                   and feeding behavior.

31                   Behavioral responses of aquatic invertebrates to Pb reviewed in the 2006 Pb AQCD
32                   included avoidance. A limited number of recent studies have considered additional
33                   behavioral endpoints in marine organisms. Valve closing speed was used as a measure of
34                   physiological alterations due to Pb exposure in the Catarina scallop (Sobrino-Figueroa
35                   and Caceres-Martinez. 2009). The average valve closing time increased from under one
36                   second in the control group to 3 to 12 seconds in juvenile scallops exposed to analytically
37                   verified concentrations of Pb as Pb nitrate (40 (ig/L to 400 (ig/L) for 20 days. Damage to
      November 2012                             7-186                  Draft - Do Not Cite or Quote

-------
 1                   sensory cilia of the mantle was observed following microscopic examination of
 2                   Pb-exposed individuals.

 3                   Since the 2006 Pb AQCD, limited studies on marine invertebrates have indicated effects
 4                   of Pb on reproduction. In a long term (approximately 60 days) sediment
 5                   multigenerational bioassay with the estuarine-sediment dwelling amphipod Elasmopus
 6                   laevis, onset to reproduction was significantly delayed at 118 mg Pb/kg compared to
 7                   controls. In the higher concentrations, start of offspring production was delayed further; 4
 8                   days in 234 mg Pb/kg and 8 days in 424 mg Pb/kg (Ringenary et al., 2007). Fecundity
 9                   and time of first offspring production was also reduced with increasing Pb concentration
10                   in sediment above 118 mg Pb/kg. The authors indicate that this concentration is below
11                   the current marine sediment regulatory guideline for Pb (218 mg Pb/kg sediment)
12                   (NOAA. 1999) and that reproductive endpoints are more sensitive than survival in this
13                   species. Exposure of gametes to Pb prior to fertilization resulted in a decrease of the
14                   fertilization rates of the marine polychaete Hydroides elegans (Gopalakrishnan et al..
15                   2008). In sperm pretreated in 97 (ig Pb/L filtered seawater for 20 minutes, fertilization
16                   rate decreased by approximately 70% compared to controls.  In a separate experiment,
17                   eggs were pretreated with Pb prior to addition of an untreated sperm suspension. The
18                   fertilization rate of eggs pretreated in 48 (ig Pb/L filtered seawater decreased to 20% of
19                   the control. In another test with H. elegans in which gametes were not pre-treated, but
20                   instead added directly to varying concentrations of Pb for fertilization, there appears to be
21                   a protective effect following fertilization due to the formation of the fertilization
22                   membrane during the first cell division that may prevent Pb from entering the oocytes
23                   (Gopalakrishnan et al., 2007).

24                   As noted in the 2006 Pb AQCD and supported by recent studies, Pb exposure negatively
25                   affects the growth of marine invertebrates. Wang et al., (2009d_) observed growth of
26                   embryos of the Asian Clam (Meretrix meretrix) was significantly reduced by Pb with an
27                   EC50 of 197 (ig/L. In juvenile Catarina scallop, Argopecten ventricosus, exposed to  Pb for
28                   30 days, the EC50 for growth was 4,210 (ig Pb/L (Sobrino-Figueroa et al., 2007). Rate of
29                   growth of the deposit feeding Capitella sp. polychaetes decreased significantly from the
30                   controls in 3  and 6-day exposures, however, the observed changes did not exhibit a  clear
31                   dose response with increasing Pb concentration (Horng et al.. 2009).

32                   Although Pb is known to cause mortality when invertebrates are exposed to sufficiently
33                   high concentrations, some species may not exhibit significant mortality even at high
34                   concentrations. In a 10-day Pb-spiked sediment exposure (1,000 mg Pb/kg and
35                   15 (ig Pb/L dissolved Pb in pore water), 100% of individuals of the Australian estuarine
36                   bivalve Tellina deltoidalis survived (King et  al.. 2010). In the deposit feeding Capitella
37                   sp., polychaetes, exposure to varying concentrations of Pb associated with spiked
      November 2012                             7-187                  Draft - Do Not Cite or Quote

-------
 1                  sediment up to 870 mg Pb/kg had no effect on survival (Horng et al., 2009). No
 2                  differences in adult survival were observed in 28 and 60 day sediment exposures to a
 3                  range of Pb concentrations from 58 mg Pb/kg to 424 mg Pb/kg in the amphipod E. laevis
 4                  (Ringenary et al. 2007). Other species are more sensitive to Pb and these responses are
 5                  reviewed in Section 7.4.16.
              7.4.15.3     Saltwater Vertebrates

                    Saltwater Fish

 6                  There is a dearth of information in previous Pb AQCDs on Pb effects in saltwater fish.
 7                  Recent data available since the 2006 Pb AQCD include a study with a marine
 8                  elasmobranch. De Boeck et al. (2010) exposed the spotted dogfish (S. canicula) to
 9                  2,072 (ig Pb/L for one week and measured metallothionein induction in gill and liver
10                  tissue, and the electrolytes Na, K, Ca2+ and Cl, in plasma. No effects were observed in
11                  Pb-exposed fish for any of the physiological variables measured in this  study, although
12                  Pb was detected in all organs (De Boeck et al.. 2010).

13                  Since the 2006 Pb AQCD, several studies integrating behavioral and physiological
14                  measures of Pb toxicity have been conducted on marine fish. The  ornate wrasse
15                  (Thalassoma pavo) was exposed nominally to sublethal (400 (ig Pb/L) or a maximum
16                  acceptable toxicant concentration (1,600 (ig Pb/L) dissolved in seawater for one week to
17                  assess the effects of Pb on feeding and motor activities (Giusi et al.. 2008). In the
18                  sublethal concentration group, hyperactivity was elevated 36% over controls. In the high
19                  concentration, a 70% increase in hyperactivity was observed and hyperventilation
20                  occurred in 56% of behavioral observations. Elevated expression of heat shock protein
21                  70/90 orthologs was detected in the hypothalamus and mesencephalic areas of the brains
22                  of Pb-treated fish and neuronal damage was observed in the  posterior hypothalamic area
23                  and optic tectum. No changes in feeding activity were noted between non-treated and
24                  treated fish.

25                  Additional behavioral studies in fish consider effects of dietary Pb. The grunt fish
26                  H. scudderi, occupying the top level of a simulated marine food chain, exhibited lethargy
27                  and decreased food intake during the last week of a 42-day feeding study (Soto-Jimenez
28                  et al., 20 lib). The  fish were fed white shrimp exposed to Pb via brine shrimp that were in
29                  turn fed microalgae cultured at a nominal concentration of 20 (ig Pb/L. Pb was quantified
30                  in shrimp and fish. The authors noted a few of the fish exposed to  Pb via dietary transfer
31                  through the food chain were observed surfacing and speculated that this behavior was air
32                  breathing as a response to stress.
      November 2012                            7-188                 Draft - Do Not Cite or Quote

-------
 1                  Evidence for reproductive effects of Pb in saltwater fish is limited to a field study in
 2                  which decreased oocyte diameter and density in the toadfish (Tetractenos glaber) were
 3                  associated with elevated levels of Pb in the gonad offish collected from contaminated
 4                  estuaries in Sydney, Australia (Alquezar et al.. 2006). The authors  state this is suggestive
 5                  of a reduction in egg size which ultimately may lead to a decline in female reproductive
 6                  output.


                    Mammals

 7                  Although Pb continues to be detected in tissues of marine mammals in U.S. coastal
 8                  waters (Bryan et al., 2007; Stavros et al., 2007; Kannan et al.. 2006) few studies exist that
 9                  consider biological effects associated with Pb exposure. Pb effects on immune variables,
10                  including cell viability, apoptosis, lymphocyte proliferation, and phagocytosis were tested
11                  in vitro on phagocytes and lymphocytes isolated from the peripheral blood of bottlenose
12                  dolphin (Tursiops truncates} (Camara Pellisso et al., 2008). No effects on viability of
13                  immune cells, apoptosis, or phagocytosis were observed in 72-hour exposure to nominal
14                  concentrations of 1,000, 10,000, 20,000 and 50,000 (ig Pb/L. Proliferative response of
15                  bottlenose dolphin leukocytes was significantly reduced at 50,000  (ig Pb/L, albeit by only
16                  10% in comparison to the control. This in vitro exposure with nominal concentrations of
17                  Pb is likely not relevant for assessing effects of atmospherically-deposited Pb on marine
18                  mammals, however, no additional studies  were available for review on the effects on Pb
19                  on these organisms.
      7.4.16 Exposure and Response of Saltwater Species

20                  Evidence regarding exposure-response relationships and potential thresholds for Pb
21                  effects on saltwater populations can inform determination of standard levels that are
22                  protective of marine ecosystems. The Annex of the 2006 Pb AQCD (U.S. EPA. 2006c)
23                  summarized data on exposure-response functions for invertebrates (Table AX7-2.4.1)
24                  (Table AX7-2.4.2). The recent exposure-response studies reviewed in this section expand
25                  on earlier findings with information on microalgal and invertebrate species. Studies
26                  specific to growth, reproduction and survival endpoints are summarized in Table  7-6. All
27                  reported values are from exposures in which concentrations of Pb were analytically
28                  verified unless nominal concentrations are stated.

29                  A series  of 72-hour Pb toxicity tests were conducted with five marine microalgae species
30                  (T. chuii, R. salina, Chaetoceros sp., /. galbana and N. gaditana) to determine the relative
31                  Pb sensitivities as measured by growth inhibition. The respective 72-hour EC50 values
32                  derived were 2,640, 900,  105, 1,340, and 740 (ig Pb/L (Debelius et al.. 2009). The


      November 2012                            7-189                  Draft - Do Not Cite or Quote

-------
 1                  authors noted that species cellular size, sorption capacity, or taxonomy did not explain
 2                  differences in sensitivity to Pb, leaving the mechanism of response still open to question.

 3                  In the deposit feeding polychaete, Capitella sp. an LOAEL of 85 mg Pb/kg sediment was
 4                  established in 3 day and 6 day growth experiments (Horng et al., 2009). Other studies of
 5                  marine invertebrates published since the 2006 Pb AQCD (U.S. EPA. 2006c) have
 6                  indicated differences in sensitivity of different lifestages of aquatic organisms to Pb. In a
 7                  series of seawater and sediment exposures using adult and juvenile amphipods Melita
 8                  plumulosa, juveniles were more sensitive to Pb than adults (King et al., 2006). In the
 9                  seawater-only exposures, the 96-hour LC50 was 1,520 (ig Pb/L for juveniles and
10                  3,000 (ig Pb/L for adults. In comparison,10 day juvenile sediment test results were LC50
11                  1,980, NOEC 580 and LOEC 1,020 mg Pb/kg dry weight compared to the LC50, NOEC,
12                  and LOEC value for the adults exposed in sediment (3,560 mg Pb/kg dry weight). A 24-
13                  hour LC50 of 4,500 (ig Pb/L for adult black mussel (M galloprovincialis) suggests that, in
14                  general, juvenile bivalves are more sensitive to Pb exposure than adults although this
15                  value was based on nominal exposure data (Vlahogianni and Valavanidis. 2007).

16                  Since the 2006 Pb AQCD , Pb toxicity to larval stages of marine species has been
17                  assessed at sublethal and lethal concentrations. The effective concentrations at which Pb
18                  resulted in 50% of abnormal embryogenesis of the Asian clam (M meretrix) was
19                  297 (ig Pb/L. The 96-hour LC50 for larvae of the same species was 353 (ig Pb/L (Wang et
20                  al., 2009d). In comparison, juvenile Catarina scallop  (A. ventricosus) had a LC50 of
21                  830 (ig Pb/L in a 96-hour exposure (Sobrino-Figueroa et al.. 2007). In the marine
22                  polychaete H. elegans, EC50 values of gametes, embryos, larvae (blastula to trochophore
23                  and larval settlement), and adults, exhibited dose-responses to Pb that reflected the
24                  differential sensitivity of various lifestages of this organism (Gopalakrishnan et al..
25                  2008). The EC50 values for sperm and egg toxicity were 380 and 690 (ig Pb/L
26                  respectively. Larval settlement measured as the metal concentration causing 50%
27                  reduction in attachment was most sensitive to Pb with an EC50 of 100 (ig Pb/L, while the
28                  EC50 for abnormal development of embryos was 1,130 (ig Pb/L. The LC50 values for
29                  adult worms in 24-hour and 96-hour tests were 25,017 and 946 (ig Pb/L, respectively.
30                  Manzo et al. (2010) established a LOEC of 500 (ig Pb/L and a maximum effect at
31                  3,000 (ig Pb/L in an embryotoxicity assay with sea urchin P. lividus exposed to nominal
32                  concentrations of Pb. The EC50 for developmental defects in this species was
33                  1,250 (ig Pb/L with a NOEL of 250 (ig Pb/L. In a study using  nominal concentrations of
34                  Pb, morphological deformities were observed in 50% of veliger larvae of blacklip
35                  abalone (Haliotis rubra)  at 4,100 (ig Pb/L following a 48-hour exposure, suggesting this
36                  species is not as sensitive to Pb as other marine invertebrate larvae (Gorski  and
37                  Nugegoda. 2006).
      November 2012                            7-190                 Draft - Do Not Cite or Quote

-------
      7.4.17 Community and Ecosystem Effects in Saltwater Systems

 1                  As discussed in the 1986 Pb AQCD and the 2006 Pb AQCD (U.S. EPA. 2006c).
 2                  exposure to Pb is likely to have impacts in aquatic environments via effects at several
 3                  levels of ecological organization (organisms, populations, communities, or ecosystems).
 4                  But fewer studies explicitly consider community and ecosystem-level effects in marine
 5                  and brackish waters than in freshwater. Reduced species abundance and biodiversity of
 6                  protozoan and meiofauna communities were observed in laboratory microcosm studies
 7                  with marine  water and marine sediments reviewed in the 2006 Pb AQCD as summarized
 8                  in Table AX7-2.5.2 (U.S. EPA. 2006c). In a laboratory study with larval mummichogs
 9                  reviewed in the 2006 Pb AQCD, feeding and predator avoidance behaviors were altered
10                  in this marine fish species following a 4-week exposure to Pb. Observations from field
11                  studies reviewed in the 2006 Pb AQCD included findings of a negative correlation
12                  between Pb and species richness and diversity indices of macroinvertebrates associated
13                  with estuary sediments and changes in species distribution and abundance in fish,
14                  crustaceans and macroinvertebrates correlated with Pb levels in marine sediments. The
15                  2006 Pb AQCD concluded that, in general, information from controlled studies for single
16                  pollutants was insufficient to  permit evaluation of specific impacts on higher levels of
17                  organization (beyond the organism). In studies from natural saltwater ecosystems, Pb
18                  rarely occurs as a sole  contaminant making its effects difficult to ascertain. New
19                  information on effects  of Pb at the population, community and ecosystem level in coastal
20                  ecosystems is reviewed below.

21                  The faunal composition of seagrass beds in a Spanish coastal saltwater lagoon was found
22                  to be impacted by Pb in sediment, plants, and biofilm (Marin-Guirao et al., 2005).
23                  Sediment Pb concentrations ranged from approximately 100 to 5,000 mg Pb/kg and
24                  corresponding biofilm  concentrations were 500 to 1,600 mg Pb/kg, with leaf
25                  concentrations up to 300  mg Pb/kg. Although multiple community indices (abundance,
26                  Shannon-Wiener diversity, Simpson dominance index) did not vary from site to site,
27                  multivariate  analysis and similarity analysis indicated significant differences in
28                  macroinvertebrate communities between sites with different sediment, biofilm, and leaf
29                  Pb concentrations. Differences were largely attributable to three amphipod species
30                  (Microdeutopus sp., Siphonoecetes sabatieri,  Gammarus sp.). This indicates that,
31                  although seagrass abundance  and biomass were unaffected by Pb exposure, organisms
32                  inhabiting these plants still may be adversely impacted.

33                  Caetano et al. (2007) investigated the mobility of Pb in salt marshes using total content
34                  and stable isotope signature. They found that roots had similar isotopic signature to
35                  sediments in vegetated zones  indicating that Pb uptake by plants reflects the input in
36                  sediments. At one site, there was a high anthropogenic Pb content while at the other
      November 2012                            7-191                  Draft - Do Not Cite or Quote

-------
 1                  natural mineralogical sources dominated. The roots of S. fruticosa and S. maritima
 1                  significantly accumulated Pb, having maximum concentrations of 2,870 mg Pb/kg and
 3                  1,755 mg Pb/kg, respectively, indicating that below-ground biomass played an important
 4                  role in the biogeochemical cycling of Pb.

 5                  Exposure to three levels of sediment Pb contamination (322, 1,225, and 1,465 mg Pb/kg
 6                  dry weight) had variable effects on different species within a marine nematode
 7                  community (Mahmoudi et al.. 2007). Abundance, taxa richness, and species dominance
 8                  indices were altered at all Pb exposures when compared with unexposed communities.
 9                  Further, while the species Oncholaimellus mediterraneus dominated control communities
10                  (14% of total abundance), communities exposed to low and medium Pb concentrations
11                  were dominated by Oncholaimus campylocercoides (36%) andMarylynnia stekhoveni
12                  (32%), and O. campylocercoides  (42%) and Chromadorina metulata (14%), respectively.
13                  Communities exposed to the highest Pb sediment concentrations were dominated by
14                  Spirinia gerlachi (41%) and Hypodontolaimus colesi (29%). Given this, the authors
15                  concluded that exposure to Pb significantly reduced nematode diversity and resulted in
16                  profound restructuring of the community structure.

17                  In another laboratory microcosm  experiment with nematodes, nematode diversity and
18                  community structure was altered  with a mean number of 8 genera present in microcosms
19                  contaminated with Pb compared to the control with 20 genera. The spiked sediments used
20                  in the study were collected from the Swartkop River estuary, South Africa. Pb (3 to
21                  6,710 mg Pb/kg sediment dry weight) was tested alone and in combination with Cu, Fe,
22                  and Zn (Gyedu-Ababio and Baird. 2006). The synergistic effect of the four metals on
23                  nematode community structure was greater than the individual metals and the effects of
24                  Pb could not be distinguished from Cu, Fe and Zn.
      November 2012                            7-192                 Draft - Do Not Cite or Quote

-------
      7.4.18 Characterization of Sensitivity and Vulnerability in Saltwater Species

 1                  Species differences in metabolism, sequestration, and elimination rates have been shown
 2                  to control relative sensitivity and vulnerability of exposed organisms and effects on
 3                  survival, reproduction, growth, metabolism, and development. Diet and lifestage at the
 4                  time of exposure also contribute significantly to the determination of sensitive and
 5                  vulnerable populations and communities. Further, environmental conditions in addition to
 6                  those discussed as affecting bioavailability may also alter Pb toxicity. The
 7                  2006 Pb AQCD (U.S. EPA. 2006c) reviewed the effects of genetics, age, and body size
 8                  on Pb toxicity. While genetics appears to be a significant determinant of Pb sensitivity,
 9                  effects of age and body size are complicated by environmental factors that alter metabolic
10                  rates of saltwater organisms. A review of the more recent literature corroborated these
11                  findings, and identified seasonal physiological changes and lifestage as other important
12                  determinants of differential  sensitivity to Pb.
              7.4.18.1      Seasonally Affected Physiological Changes

13                  Couture et al. (2010) investigated seasonal and decadal variations in Pb sources to
14                  mussels (M. edulis) from the French Atlantic shoreline. Pb concentrations in the mussels
15                  were 5-66 times higher than the natural background value for the north Atlantic. The
16                  206Pb/207Pb signature indicated that the bioaccumulated Pb was anthropogenic in origin.
17                  The signature was not, however, the same as that emitted in western Europe, as a result of
18                  leaded gasoline combustion, although that was a major emission source to the atmosphere
19                  during a large part of the study period (1985-2005). Instead, it was most similar to that of
20                  Pb released into the environment from wastewater treatment plants, municipal waste
21                  incinerators and industries such as metal refineries and smelters. Thus continental runoff
22                  rather than atmospheric deposition was identified as the main source of Pb to the French
23                  coastal area. The strong seasonal variations in 206Pb/208Pb were used to conclude that
24                  re suspension of Pb triggered by high river runoff events was a key factor affecting
25                  bioaccumulation of Pb inM edulis.

26                  In another monitoring study, Pearce and Mann (2006) investigated variations in
27                  concentrations of trace metals in the U.K. including Pb in the shells of pod razor shell
28                  (Ensis siliqud). Pb concentration varied from 3.06-36.2 mg Pb/kg and showed a regional
29                  relationship to known sources, e.g., former metal mining  areas such as Cardigan Bay,
30                  Anglesey, and industrial activity in Liverpool  Bay. Seasonal variations were also found
31                  for  Pb in both Cardigan Bay and Liverpool Bay, relating to increased winter fluxes of Pb
32                  (and other metals) into the marine environment. In contrast, levels of Pb and other metals
      November 2012                            7-193                 Draft - Do Not Cite or Quote

-------
 1                  were highest in summer and lowest in winter in oysters Crassostrea corteziensis collected
 2                  from Sonora, Mexico (Garcia-Rico et al.. 2010).

 3                  Carvalho et al. (2011) quantified 210Pb inM galloprovincialis sampled at coastal
 4                  locations in Portugal and noted that the apparent seasonal fluctuation in radionuclide
 5                  concentrations in mussel soft tissues was mostly attributable to changes in physiological
 6                  condition (i.e., fat content, gonadal development) and not to radionuclide body burden
 7                  fluctuation. The authors caution that since concentrations of contaminants are dependent
 8                  upon tissue composition, corrections for mussel physiological condition are need to
 9                  compare results from different seasons and different locations.
              7.4.18.2     Lifestage

10                  Lifestages of the marine polychaete H. elegans including embryogenesis, sexual
11                  maturation, and offspring development were shown to be differentially affected by Pb
12                  exposure. Pb water concentrations of 91 (ig Pb/L and greater significantly affected
13                  fertilization and embryonic development, but the greatest effects were exhibited by 24-
14                  hour-old larvae (Gopalakrishnan et al.. 2007). The authors  suggested that timing of Pb
15                  exposure may have different impacts on marine polychaete populations, if life cycles are
16                  offset (Gopalakrishnan et al., 2007). Further, given that the adult lifestage is sedentary,
17                  reduction of the mobile early lifestage as a result of Pb exposures may disproportionally
18                  affect sessile polychaetes. For instance, larval settlement was significantly reduced at Pb
19                  exposures of 48 (ig Pb/L and greater (Gopalakrishnan et al.. 2008).
              7.4.18.3     Species Sensitivity

20                  Both inter- and intra-specific differences in Pb uptake and bioaccumulation may occur in
21                  macroinvertebrates of the same functional feeding group. Data from 20 years of
22                  monitoring of contaminant levels in filter-feeding mussels of the Mytilus species and
23                  Crassostrea virginica oysters in coastal areas of the U.S. through the National Oceanic
24                  and Atmospheric Administration (NOAA) Mussel Watch program indicate that Pb is on
25                  average three times higher in mussels than in oysters (Kimbrough et al., 2008). Limpet
26                  (Patella sp.) from the Lebanese Coast had Pb BAF values ranging from 2,500 to 6,000
27                  and in the same field study Pb BAF values for a mussel (Brachidontes variabilis) ranged
28                  from 7,500-8,000 (Nakhle et al.. 2006).

29                  There is some indication that molting may comprise an additional sequestration and
30                  excretion pathway for aquatic animals exposed to Pb (Soto-Jimenez et al.. 201 la;
31                  Mohapatra et al.. 2009: Tollett et al.. 2009: Bergev and Weis. 2007). Crab species

      November 2012                            7-194                  Draft - Do Not Cite or Quote

-------
 1                   U. pugnax (Bergey and Weis. 2007) and Scylla serrata fMohapatra et al., 2009A and
 2                   white shrimp L. vannamei (Soto-Jimenez et al.. 201 la) have been shown to sequester Pb
 3                   preferentially in exoskeleton tissue, where it is later shed along with other tissue.
 4                   Consequently, aquatic arthropod species and those species that shed their exoskeleton
 5                   more frequently may be able to tolerate higher environmental Pb concentrations than
 6                   non-arthropods or slow-growing molting species, as this pathway allows them to
 7                   effectively lower Pb body burdens.

 8                   Some tolerant species offish (e.g., mummichog) have the ability to sequester
 9                   accumulated Pb in metal-rich granules or heat-stable proteins (Goto and Wallace. 2010).
10                   Fish with such abilities are more likely to thrive in Pb-contaminated environments than
11                   other species.
      7.4.19 Ecosystem Services Associated with Saltwater Systems

12                  Pb deposited on the surface of (or taken up by) organisms has the potential to alter the
13                  services provided by saltwater biota to humans although the directionality of impacts is
14                  not always clear. For example, oysters and mussels provide a service by sequestering Pb.
15                  At the same time, the uptake of Pb by these bivalves may result in toxicological effects
16                  associated with Pb exposure and decreased value of shellfish as a commodity. At this
17                  time, a few publications address Pb impacts on ecosystem services associated with
18                  saltwater ecosystems. Pb can affect the ecological effects in each of the four main
19                  categories of ecosystem services (Section 7.1.2) as defined by Hassan et al. (2005). These
20                  effects are sorted into ecosystem services categories and summarized here:

21                     • Supporting: food for higher trophic levels, biodiversity
22                     • Provisioning: contamination of food by heavy metals, decline in health offish
23                       and other aquatic species
24                     • Regulating: water quality
25                     • Cultural: ecosystem and cultural heritage values related to ecosystem integrity
26                       and biodiversity, wildlife and bird watching, fishing
27                  A few recent studies explicitly consider the impact of Pb and other heavy metals on
28                  ecosystem services provided by salt marsh (Gedan et al., 2009)  and estuaries (Smith et
29                  al.. 2009b). These systems are natural sinks for metals and other contaminants. Pb can be
30                  toxic to salt marsh plant species and decaying plant detritus may result in resuspension of
31                  Pb into the aquatic food chain (Gedan et al.. 2009). Salt marsh and estuaries provide
32                  habitat and breeding areas for both terrestrial and marine wildlife and are locations for
      November 2012                            7-195                 Draft - Do Not Cite or Quote

-------
 1                  bird watching. Using a modeling approach designed to assess the degree of risk of Pb and
 2                  Hg to wading birds in estuarine habitats in the U.K., the authors found a high probability
 3                  that Pb poses an ecologically relevant risk to dunlin, Calidris alpina fSmith et al., 2QQ9b).
 4                  However, the authors noted that a major source of uncertainty in this study was the
 5                  NOAEL values for Pb.

 6                  The impact of Pb on ecological services provided by specific components of aquatic
 7                  systems has been considered in a limited number of studies. Recent research has
 8                  suggested that dietary Pb (i.e., Pb adsorbed to sediment, particulate matter, and food) may
 9                  contribute to exposure and toxicity in primary and secondary order consumers (including
10                  humans). Aquatic fauna can take up and bioaccumulate metals. If the bioaccumulating
11                  species is a food source, the uptake of metals may make it toxic or more dangerous for
12                  people or other wildlife to consume. For example, oysters and mussels bioaccumulate Pb
13                  from anthropogenic sources, including atmospheric deposition, and are a food source that
14                  is widely consumed by humans and wildlife (Couture et al.. 2010). Their capacity to
15                  bioaccumulate Pb makes them good bioindicators of environmental contamination and
16                  they have been used as monitors of coastal pollutants by the NOAA Mussel Watch
17                  program since 1986. Although bioaccumulation may render aquatic fauna toxic to
18                  consumers, bioaccumulation is a way to sequester the metals and remove them from
19                  waters and soils. Sequestration for this purpose is itself an ecosystem service and has
20                  been quantified. For example, the total ecological services value of a constructed
21                  intertidal oyster (Crassostrea sp.) reef in improving water quality and sequestering metals
22                  including Pb was calculated in the Yangtze River estuary to be about $500,000 per year
23                  (Quan et al.. 2009).
      7.4.20 Synthesis of New Evidence for Pb Effects in Saltwater Systems

24                  This synthesis of the effects of Pb on saltwater ecosystems covers information from the
25                  publication of the 2006 Pb AQCD (U.S. EPA. 2006c) to present. It is followed in
26                  Section 7.4.21 by determinations of causality that take into account evidence dating back
27                  to the 1977 Pb AQCD. In general, evidence for toxicity to saltwater organisms is less
28                  well characterized than toxicity of Pb in freshwater ecosystems due to the fewer number
29                  of available studies on marine species. The studies that are available for marine plants,
30                  invertebrates and vertebrates include studies where Pb concentration was analytically
31                  verified and those that reported nominal concentrations (Table 7-6). Many of the studies
32                  that report nominal concentrations in media are uptake studies that subsequently quantify
33                  Pb in tissues; however, measurement of Pb in water or sediment at the beginning of an
34                  exposure is desirable when comparing laboratory studies to concentrations of Pb in
35                  marine systems. In Section 7.2.3 and Table 7-2. a range of 0.01 to 27 (ig Pb/L was


      November 2012                            7-196                  Draft - Do Not Cite or Quote

-------
 1                   reported for saltwater, including estuaries and open ocean, with the higher values
 2                   associated with sites involving human activity (Sadiq. 1992).

 3                   Most studies on marine organisms reviewed in the present document included
 4                   concentrations that were higher than Pb encountered in seawater. However, when
 5                   multiple concentrations were used, effects generally increased with increasing Pb
 6                   exposure. Effects at lower concentrations can be implied from many reported studies
 7                   since an exposure response relationship to Pb was observed. In marine and estuarine
 8                   systems, exposure to Pb from air is most likely characterized as a chronic low dose
 9                   exposure, however, most studies only report an acute LC50 value when an LOEC or LCio
10                   would be more appropriate measurement for consideration of effects on organisms since
11                   an effect occurring at the LC50 value would most likely not maintain a stable population.


                     Plants

12                   Only a few studies were available since the 2006 Pb AQCD, that consider effects of Pb
13                   on marine algae (Section 7.4.15.1). A 72-hour EC50 for growth inhibition was reported in
14                   the marine algae Chaetoceros sp. at 105 (ig Pb/L (Debelius et al.. 2009). A study with the
15                   green alga T. suecica reports a statistically significant decease in growth rate, total dry
16                   biomass and final cell concentration between control cultures and algae cultured in
17                   20 (ig Pb/L (Soto-Jimenez et al., 201 Ib). Both of these studies suggest growth effects at
18                   or near the highest recorded values of Pb in seawater (27 (ig Pb/L), however, effects are
19                   likely to occur at  lower concentrations since only EC50 values are reported.


                     Invertebrates

20                   In saltwater invertebrates (Section 7.4.15.2 and 7.4.16) there are studies that consider
21                   Pb-effects on supporting endpoints (stress responses, hematological effects and
22                   neurobehavior) as well as studies that assess Pb impacts to reproduction, growth, and
23                   survival; endpoints that have the potential to  alter population, community and
24                   ecosystem—levels of biological organization. Many studies, especially those that
25                   consider enzymatic responses to Pb exposure, were conducted with nominal Pb
26                   concentrations. Two of these studies; Jing et al. (2007) and Zhang et al., (201 Ob) consider
27                   Pb nominal exposures at  100 (ig Pb/L or lower and reported significant decreases in
28                   antioxidant enzyme activity. The Zhang et al. (201 Ob) study observed effects on
29                   enzymatic activity at a nominal exposure of 2 (ig Pb/L. Although these effects are near
30                   reported Pb concentrations in seawater they were not analytically verified.

31                   Other studies that report sub-organismal responses in saltwater organisms have quantified
32                   Pb exposure. Field studies with bivalves collected off the coast of Spain correlated
      November 2012                             7-197                  Draft - Do Not Cite or Quote

-------
 1                  ALAD activity with measured levels of Pb in tissue (Company et al.. 2011; Kalman et al..
 2                  2008). An increase in valve closing time with increasing Pb exposure in the range of 40
 3                  to 400 (ig Pb/L was observed in the scallop, A.ventricosus (Sobrino-Figueroa and
 4                  Caceres-Martinez. 2009). Although the concentrations in this study exceed reported
 5                  levels of Pb in seawater, the lower range is near 27 (ig Pb/L reported by Sadiq (1992).

 6                  Evidence for effects on reproduction, growth and survival in marine invertebrates (Table
 7                  7-6) are primarily from studies in which Pb in the exposure media was quantified. In the
 8                  amphipod, E. laevis, onset to reproduction was significantly delayed at 118 mg/Pb kg
 9                  sediment; a concentration that the authors indicate is below the current marine sediment
10                  regulatory guideline for Pb (218 mg Pb/kg sediment) (Ringenary et al.. 2007; NOAA.
11                  1999). In the same study, no effects of Pb on adult survival in 28 and 60 day sediment
12                  exposures were observed. In another study with amphipods, juvenile M. plumosa were
13                  more sensitive than adults in 10-day sediment exposures with an NOEC of 580 mg Pb/kg
14                  dry weight compared to an NOEC of 3,560 mg Pb/kg dry weight for adults (King et al..
15                  2006). Effects of Pb on gametes of the marine polycheate H. elegans were observed at
16                  48 (ig Pb/L (Gopalakrishnan et al.. 2008). a concentration near the upper range of Pb in
17                  seawater reported by Sadiq (1992). Specifically, fertilization rate of eggs pretreated with
18                  48 (ig Pb/L decreased to 20% of control. Life stages ofH. elegans varied in their
19                  sensitivity to Pb with the most sensitive period being larval settlement with an EC50 of
20                  100  (ig Pb/L.

21                  There are only a few recent studies that considered effects of Pb on growth of marine
22                  invertebrates (Sections  7.4.15.2 and 7.4.16).  In the polychaete Capitella sp. growth was
23                  decreased significantly  from controls, however, there was not a clear dose-response
24                  relationship between increasing Pb concentrations and observed effects (Horng et al..
25                  2009). The authors reported a LOAEL of 85 mg Pb/kg in the  sediment exposure. In the
26                  Asian Clam M. meretrix, an EC50 of 197 (ig Pb/L was reported for growth (Wang et al..
27                  2009d). Other marine invertebrate growth effects were observed at much higher Pb
28                  concentrations (Table 7-6).

29                  Survival was a less sensitive endpoint in marine invertebrates than reproduction or
30                  growth with no effects reported at concentrations typically observed in seawater (Table
31                  7-6). In the amphipodM plumulosa an NOEC of 400 (ig Pb/L for juveniles and an
32                  NOEC of 850 (ig Pb/L was reported for adults in 96-hour seawater only exposures (King
33                  et al.. 2006). In 10 day sediment tests with the same species, juveniles were also more
34                  sensitive than adults. Other concentrations at which survival effects were reported in
35                  marine invertebrates also greatly exceeded concentrations of Pb typically found in
36                  seawater.
      November 2012                            7-198                  Draft - Do Not Cite or Quote

-------
                    Vertebrates

 1                  There is not sufficient new evidence for saltwater vertebrates especially for reproductive,
 2                  growth and survival endpoints that may have relevance to the population-level of
 3                  biological organization and higher.


                    Food Web

 4                  Some studies published since the 2006 Pb AQCD (see Section 7.4.14.4) support the
 5                  potential for  Pb to be transferred in saltwater food webs, while other studies have found
 6                  no evidence for biomagnification.


                    Ecosystem Level Effects

 7                  Evidence for effects at higher levels of biological organization in saltwater habitats is
 8                  primarily supported by observations in a small number of microcosm and field studies
 9                  where shifts in community structure are the most commonly observed effects of Pb
10                  (Section 7.4.17). Effects on reproduction, growth or survival (summarized in Table 7-6)
11                  may lead to effects at the population-level of biological organization and higher.
      7.4.21 Causal Determinations for Pb in Saltwater Systems

12                  In the following sections, organism-level effects on reproduction and development,
13                  growth and survival are considered first since these endpoints can lead to effects at the
14                  population level or above and are important in ecological risk assessment.
15                  Neurobehavioral effects are considered next followed by sub-organismal responses
16                  (hematological effects, physiological stress) for which Pb has been shown to have an
17                  impact in multiple species and across taxa, including humans. Causal determinations for
18                  terrestrial, freshwater and saltwater ecological effects are summarized in Table 7-3.
              7.4.21.1     Reproductive and Developmental Effects-Saltwater Biota

19                  Reproductive effects of Pb have been reported in a few marine organisms and the
20                  majority of the available studies are with invertebrate species. In a study reviewed in the
21                  2006 Pb AQCD (U.S. EPA. 2006c). embryo development in two commercial bivalves
22                  Ruditapes decussatus and M. galloprovincialis was inhibited by Pb (Beiras and
23                  Albentosa. 2003). In R. decussatus an EC50 range of 156 to 312 (ig Pb/L and LOEC of
24                  156 (ig Pb/L were observed for inhibition of embryonic development while in
      November 2012                            7-199                 Draft - Do Not Cite or Quote

-------
 1                  M. galloprovincialis the EC50 was 221 (ig Pb/L and the LOEC was 50 (ig Pb/L. Larvae of
 2                  the mussel M. edulis were sensitive to Pb exposure with an EC50 of 476 (ig Pb/L for
 3                  abnormal development of embryos following 48-hour exposure to Pb during
 4                  embryogenesis (Martin et al.. 1981). The LOEC for embryogenesis in the marine bivalve
 5                  M. galloprovincialis was 50 (ig Pb/L with an EC50 for embryogenesis of 221 (ig Pb/L
 6                  (Beiras and Albentosa. 2003).

 7                  Recent evidence for reproductive effects of Pb on marine invertebrates is summarized in
 8                  Table 7-6. In the marine polychaete H. elegans an EC50 of 261 (ig Pb/L was observed for
 9                  unhatched or abnormal larvae following 20 hour incubation with Pb (Gopalakrishnan et
10                  al.. 2008). The EC50 for the metal concentration causing 5% reduction in larval
11                  attachment was 100 (ig Pb/L. The EC50 values for sperm and egg toxicity were 380 and
12                  692 (ig Pb/L, respectively. The EC50 for embryogenesis in the clamM meretrix was
13                  297 (ig Pb/L  (Wang et al., 2009d). In a multigenerational bioassay with the marine
14                  amphipod E.  laevis, statistically significant delays in onset of reproduction (4 to 8 days),
15                  sexual maturation and first offspring were observed at concentrations of 188 mg Pb/kg
16                  sediment and higher (Ringenary et al.. 2007). The authors indicate that this concentration
17                  is below the current sediment regulatory guideline for Pb (218 mg Pb/kg sediment)
18                  (TSfOAA. 1999) and that reproductive effects are a more sensitive endpoint than lethality.
19                  Although LC50 values are typically reported for Pb effects on reproductive endpoints in
20                  saltwater invertebrates, a concentration dependent relationship between reproductive
21                  impairment and increasing concentration of Pb is reported in most studies. This exposure-
22                  response relationship implies that effects on reproduction are occurring at concentrations
23                  lower than the LC50 value.

24                  Reproductive effects are only characterized in a few species and endpoints for marine
25                  systems. The weight of the current evidence for reproductive effects is limited to
26                  laboratory-based studies with saltwater invertebrates in which observed effects occur at
27                  Pb concentrations that are higher than Pb concentrations encountered in the marine
28                  environment. Evidence for reproductive effects of Pb on marine plant species is limited to
29                  one study on the red alga (Champia parvuld) reviewed in the draft Ambient Aquatic Life
30                  Water Quality Criteria for Pb (U.S. EPA. 2008b). In one study from a saltwater fish,
31                  field-collected smooth toadfish (T. glaber) from metal contaminated estuaries in Sydney,
32                  Australia had elevated Pb levels in gonad and decreased oocyte diameter and density.
33                  Evidence is, therefore, inadequate to conclude that there is a causal relationship for
34                  reproductive  effects in saltwater plants, and vertebrates. The available studies on marine
35                  invertebrates are suggestive that there is a causal relationship between Pb exposure and
36                  reproductive  effects.
      November 2012                            7-200                  Draft - Do Not Cite or Quote

-------
              7.4.21.2     Growth Effects-Saltwater Biota

 1                  There are few studies that measure growth effects of Pb on marine organisms; available
 2                  information is limited to marine flora and invertebrates. Growth studies in saltwater plant
 3                  species are summarized in Table 4 and Table 6 of the draft Ambient Aquatic Life Water
 4                  Quality Criteria for Pb (U.S. EPA. 2008b) and Table 7-6 of the present document.

 5                  Diatoms are among the most sensitive algae; however, growth effects are typically
 6                  observed at concentrations of Pb higher than the range of values available from saltwater
 7                  locations [0.01 to 27 (ig Pb/L,  (Sadiq.  1992)1. In studies available since the draft Ambient
 8                  Aquatic Life Water Quality Criteria for Pb, the lowest 72-hour EC50 for growth inhibition
 9                  reported in marine diatoms was 105 (ig Pb/L in Chaetoceros sp (Debelius et al., 2009)
10                  and the growth of the green alga T. suecica exposed nominally to 20 (ig Pb/L was 40%
11                  lower than control cultures (Soto-Jimenez et al., 201 Ib). The microalgae was the base of
12                  a simulated marine food chain including primary, secondary and tertiary level consumers
13                  and effects on survival were observed  at the higher trophic levels that originated from Pb
14                  exposure via consumption of the primary  producer. The majority of growth effects
15                  reported in saltwater algae exceed concentrations of Pb in  seawater by several orders  of
16                  magnitude. Effects of Pb on growth in two species of brown algae, Fucus vesiculosus and
17                  Fucus serratus are summarized in Table 6 of the draft Ambient Aquatic Life Water
18                  Quality Criteria for Pb (U.S. EPA. 2008b). Concentrations where growth impairment was
19                  observed in these species greatly exceed available values for Pb measured in seawater.

20                  In saltwater invertebrates, evidence for growth effects is limited to a few species at
21                  concentrations that exceed Pb  concentrations reported in seawater. Growth inhibition in
22                  the bivalve Macoma balthica (EC50=453.4 (ig Pb/L) is reported in Table 6 of the draft
23                  Ambient Aquatic Life Water Quality Criteria for Pb (U.S.  EPA. 2008b). Recent studies
24                  include Wang et al., (2009d_) in which  observed growth of embryos of the Asian Clam
25                  (M meretrix) was significantly reduced by Pb with an EC50 of 197 (ig Pb/L. In juvenile
26                  Catarina scallop, A. ventricosus, exposed to Pb for 30 days, the EC50 for growth was
27                  4,210 (ig Pb/L (Sobrino-Figueroa et al., 2007). Rate of growth of the deposit feeding
28                  polychaete Capitella sp. exposed to Pb-spiked sediments from polluted estuaries
29                  decreased significantly from the control; however, changes were inconsistent with
30                  increasing concentration of Pb (Horng et al.. 2009). Evidence  is therefore inadequate  to
31                  conclude that there is a causal  relationship between Pb exposure and growth effects in
32                  saltwater plants,  invertebrates  and vertebrates.
      November 2012                            7-201                 Draft - Do Not Cite or Quote

-------
              7.4.21.3     Survival-Saltwater Biota

 1                  There are no studies reported in the previous Pb AQCDs or the current ISA for aquatic
 2                  plants that indicate phytotoxicity at or near current concentrations of Pb in saltwater [0.01
 3                  to 27 ng Pb/L, (Sadiq. 1992)1.

 4                  Mortality data for saltwater invertebrate species are summarized in the draft Ambient
 5                  Aquatic Life Water Quality Criteria for Pb (U.S. EPA. 2008b) and reported LC50 values
 6                  greatly exceed Pb concentrations encountered in seawater. Recent studies available since
 7                  the 2006 Pb AQCD, and draft Aquatic Life Water Quality Criteria for Pb that report
 8                  mortality data are summarized in Table 7-6. In general, marine fauna are less sensitive to
 9                  this metal than freshwater fauna and the highest toxicity is observed in juveniles. A 144-
10                  hour LC50 of 680 (ig Pb/L was reported for juvenile scallop A. ventricosus (Sobrino-
11                  Figueroa et al., 2007) and a 96-hour LC50 of 353 (ig Pb/L for embryos of the clam
12                  M. meretrix (Wang et al.. 2009d). In the amphipod M. plumulosa, juveniles were more
13                  sensitive to Pb than adults in 96 hour  seawater-only exposures and 10 day sediment
14                  exposures (King et al.. 2006). The 96-hour LC50 was  1,520 jig Pb/L and the NOEC was
15                  400 (ig Pb/L for juveniles in comparison to adults (96-hour LC50 =3,000 (ig Pb/L;
16                  NOEC=1,680 (ig Pb/L). In the 10-day sediment exposures, the NOEC for juveniles was
17                  580 mg Pb/kg dry weight compared to an adult NOEC of 3,560 mg Pb/kg dry weight. In
18                  10-day exposures to Pb nitrate spiked sediment, all individuals of the bivalve
19                  T. deltoidalis survived at 1,000 mg/Pb kg with 15 (ig  Pb/L dissolved in pore water (King
20                  et al., 2010). No effects on survival were observed in either the amphipod E. laevis
21                  exposed 60 days to Pb-spiked sediment up to 424 mg Pb/kg (Ringenary et al.. 2007). or
22                  in the polychaete Capitella sp. exposed to sediment for 3 or 6 days up to 871 mg Pb/kg
23                  (Horng et al.. 2009).

24                  Effects of Pb on survival have  been demonstrated though a simulated marine food chain
25                  in which the primary producer, the microalgae T. suecica, was exposed nominally to
26                  20 (ig Pb/L and subsequently fed to brine shrimp A. franciscana, (mean Pb content 12 to
27                  15 mg Pb/kg) which were consumed by white-leg shrimp L. vannamei, itself consumed
28                  by grunt fish H. scudderi representing the top of the marine food chain (Soto-Jimenez et
29                  al., 201 Ib). Survival of brine shrimp was 25 to 35% lower than the control and both
30                  white shrimp and grunt fish had significantly higher mortalities than controls.

31                  Data on Pb toxicity to eight species of marine fishes are summarized in Table 1 of the
32                  draft Ambient Aquatic Life Water Quality Criteria for Pb (U.S. EPA. 2008b). All of the
33                  LC50 values for these fish (range  1,500 to 315,000 (ig Pb/L) greatly exceed
34                  concentrations of Pb reported in seawater. Additionally, in the 2006 Pb AQCD (U.S.
35                  EPA. 2006c) the acute toxicity of Pb to plaice (Pleuronectes platessa) was reported to
36                  range from 50 (ig Pb/L to 300,000 (ig Pb/L depending on the form of Pb (Eisler. 2000).

      November 2012                           7-202                  Draft - Do Not Cite or Quote

-------
 1                  The existing evidence on toxicity of Pb to marine vertebrates is limited to laboratory-
 2                  based studies conducted under different salinities and exposure conditions. Considerable
 3                  uncertainties exist in applying laboratory observations to actual conditions in the field
 4                  where other modulating factors can affect Pb bioavailability and toxicity.

 5                  Although evidence exists for increased mortality of marine fish at very high
 6                  concentrations of Pb, the focus of the causal determinations are on studies where effects
 7                  were observed within one to two orders of magnitude of Pb measured in the environment
 8                  (Preamble Table II).  Evidence is therefore inadequate to conclude that there is a causal
 9                  relationship between Pb and survival in saltwater plants, invertebrates and vertebrates.
              7.4.21.4     Neurobehavioral Effects-Saltwater Biota

10                  In marine organisms evidence for neurobehavioral effects of Pb is limited to a few studies
11                  on bivalves and fish. In a study reviewed in the 2006 Pb AQCD (U.S. EPA. 2006c). prey
12                  capture rate and predator avoidance was affected in mummichogs starting at 300 (ig Pb/L
13                  (Weis and Weis. 1998). Recent studies support previous findings of decreased ability to
14                  escape predation associated with Pb exposure. In juvenile Catarina scallops exposed to
15                  Pb (40 (ig/L to 400 (ig/L) for 20 days, the average valve closing time increased from
16                  under one second in the control group to  3 to 12 seconds in juvenile scallops A decrease
17                  in valve closing speed in these bivalves may impact escape swimming behaviors
18                  important for predator avoidance (Sobrino-Figueroa and Caceres-Martinez. 2009).
19                  Behavioral effects in grunt fish H. scudderi, occupying the top level of a simulated
20                  marine food chain included lethargy and  decreased food intake in a 42-day feeding study
21                  (Soto-Jimenez et al.. 201 Ib). These fish were fed white shrimp exposed to Pb via brine
22                  shrimp that were initially fed microalgae cultured at a nominal concentration of
23                  20 (ig Pb/L. In the same study, surfacing, reduction of motility, and erratic swimming
24                  were observed in the white shrimp after 30 days of exposure to Pb via diet. The ornate
25                  wrasse, T. pavo, was exposed nominally to sublethal (400 (ig Pb/L) or a maximum
26                  acceptable toxicant concentration (1,600  (ig Pb/L) dissolved in seawater for one week to
27                  assess the effects of Pb on feeding and motor activities (Giusi et al.. 2008). In the
28                  sublethal concentration group, hyperactivity was elevated 36% over controls. In the high
29                  concentration, a 70% increase in hyperactivity was observed and hyperventilation
30                  occurred in 56% of behavioral observations, however, no changes in feeding activity
31                  were noted between non-treated and treated fish.

32                  Most of the evidence for neurobehavioral changes in marine organisms is observed with
33                  concentrations of Pb that exceed the range of Pb values available for saltwater of 0.01 to
34                  27 (ig Pb/L (Sadiq. 1992)1. with the exception of the food chain study discussed above in
      November 2012                            7-203                  Draft - Do Not Cite or Quote

-------
 1                  which behavioral effects were observed in shrimp and their fish predators following
 2                  ingestion of microalgae cultured in nominal concentration of 20 (ig Pb/L and then
 3                  quantified in prey (Soto-Jimenez et al.. 201 Ib). Marine species are typically
 4                  underrepresented in toxicity testing of behavioral endpoints with metals. There are
 5                  considerable uncertainties in applying observations from laboratory-based studies to field
 6                  scenarios including the role of environmental factors such as salinity and DOM on Pb
 7                  bioavailability. Evidence is inadequate to conclude that there is a causal relationship
 8                  between Pb exposures and neurobehavioral endpoints in saltwater invertebrates and
 9                  vertebrates.
              7.4.21.5     Hematological Effects-Saltwater Biota

10                  Evidence for hematological effects of Pb on saltwater organisms is limited primarily to
11                  field monitoring studies on bivalves. Several recent field studies using a multi-biomarker
12                  approach to study the sources and impacts of Pb in marine environments have measured
13                  ALAD activity in bivalve species and found positive correlations between increased
14                  tissue Pb levels and ALAD inhibition (e.g., Company et al. (2011). Kalman et al (2008)).
15                  Generally, these studies have noted that Pb content varies significantly among species
16                  and is related to habitat and feeding behavior. There is precedent, especially in Europe,
17                  for the inclusion of ALAD as a biomarker of exposure to Pb in marine invertebrates. The
18                  mechanism of ALAD inhibition in response to Pb exposure is likely mediated through a
19                  common pathway in both marine and freshwater invertebrates (Section 7.4.12.5) as well
20                  as in terrestrial species  (Section 7.3.12.5) and humans (Section 5.7). Evidence is
21                  therefore, suggestive of a causal relationship between Pb exposure and hematological
22                  effects in saltwater invertebrates. Evidence is inadequate to conclude that there is a causal
23                  relationship between hematological effects and saltwater vertebrates.
              7.4.21.6     Physiological Stress-Saltwater Biota

24                  Most studies on physiological stress responses in marine invertebrates are laboratory-
25                  based exposures where effects are observed at Pb concentrations that exceed those known
26                  to occur in seawater [0.01 to 27 (ig Pb/L (Sadiq. 1992). Table 7-21. However, some
27                  recent evidence for invertebrate antioxidant response in bivalves and crustaceans
28                  indicates effects may occur at Pb concentrations that are detected in the marine
29                  environment. For example,  SOD, catalase, and glutathione peroxidase activities were
30                  significantly reduced in the digestive gland of the marine bivalve C. farreri at 2 (ig Pb/L
31                  (as measured in Bohai Bay, China) (Zhang et al.. 2010b). In red fingered marsh crabs,
32                  P. erythrodactyla collected from an estuarine lake in Australia, elevated glutathione

      November 2012                            7-204                  Draft - Do Not Cite or Quote

-------
 1                  peroxidase activity was correlated with individuals with higher metal body burdens
 2                  (MacFarlane et al.. 2006).

 3                  Additional evidence from environmental monitoring studies that compared biomarker
 4                  responses between reference and contaminated sites indicated a correlation between the
 5                  amount of Pb with changes in antioxidant enzyme activity [e.g., (Serafim etal. 2011;
 6                  Cravo et al.. 2009)1. Marine bivalves are the organisms typically sampled for these
 7                  biomonitoring studies since both metals and enzymatic activities can be readily measured
 8                  in these invertebrates. Although these studies show clear evidence of alterations in
 9                  antioxidant stress markers in response to marine pollution, these effects cannot be
10                  attributed solely to Pb in the environment due to the presence of other metals and
11                  contaminants. Evidence for stress responses in marine organisms is typically limited to
12                  invertebrates, however, elevated expression of heat shock protein orthologs were reported
13                  for the first time in the hypothalamic and mesencephalic brain  regions of Pb-treated fish
14                  (Giusi et al.. 2008).

15                  Evidence for physiological stress responses in saltwater invertebrates are supported by
16                  evidence in freshwater species (Section 7.4.12.6) and terrestrial species (Section 7.3.12.6)
17                  as well as in humans and experimental animal studies of oxidative stress following
18                  impairment of normal metal ion functions (Section 5.2.4). Stress responses may increase
19                  susceptibility to other stressors and reduce individual fitness. Evidence is suggestive of a
20                  causal relationship between Pb exposures and physiological stress in saltwater
21                  invertebrates. The evidence is inadequate to conclude that there is a causal relationship
22                  between Pb exposure and physiological stress in saltwater plants and vertebrates.
              7.4.21.7     Community and Ecosystem Level Effects-Saltwater Biota

23                  No studies on community and ecosystem level effects of Pb in marine systems were
24                  reviewed in the 1977 Pb AQCD (U.S. EPA. 1977). or the 1986 Pb AQCD (U.S. EPA.
25                  1986a). Observations from field studies reviewed in the 2006 Pb AQCD (U.S. EPA.
26                  2006c) included findings of a negative correlation between Pb and species richness and
27                  diversity indices of macroinvertebrates associated with estuary sediments (summarized in
28                  Table AX7-2.5.2 of the 2006 Pb AQCD). Additional findings in marine environments
29                  included changes in species distribution and abundance in fish, crustaceans and
30                  macroinvertebrates correlated with Pb levels in marine sediments.

31                  New evidence for community and ecosystem level effects of Pb in saltwater ecosystems
32                  includes laboratory microcosm studies as well as observations from field-collected
33                  sediments, biofilm and plants in which changes in community structure were observed. In
34                  a recent study, significant differences in macroinvertebrate communities associated with

      November 2012                            7-205                 Draft - Do Not Cite or Quote

-------
 1                  seagrass beds were reported between sites with different sediment, biofilm, and leaf Pb
 2                  concentrations (Marin-Guirao et al.. 2005). Sediment Pb concentrations ranged from
 3                  approximately 100 to 5,000 mg Pb/kg and corresponding biofilm concentrations were
 4                  500 to 1,600 mg Pb/kg, with leaf concentrations up to 300 mg Pb/kg. In a laboratory
 5                  microcosm experiment conducted with estuarine sediments from South Africa, total
 6                  meiofauna density decreased (range 3 to 5 taxa) after 32 days in Pb-treated (1,886 to
 7                  6,710 (ig/Pb g sediment dry weight) sediments compared to 9 taxa in the control (3 (ig/Pb
 8                  g sediment dry weight) (Gvedu-Ababio and Baird. 2006). In a microcosm experiment,
 9                  exposure to three levels of sediment Pb contamination (322, 1,225, and 1,465 mg Pb/kg
10                  dry weight) significantly reduced marine nematode diversity and resulted in profound
11                  restructuring of the community structure (Mahmoudi et al.. 2007).

12                  There is not sufficient information at this time to characterize and to quantify
13                  relationships between ambient concentrations of Pb and response in saltwater
14                  communities and ecosystems. Fewer studies are available for saltwater organisms when
15                  compared to freshwater systems. There are likely differences in uptake and
16                  bioaccumulation in marine species due to physiological characteristics for adaptation in
17                  salt water. Additional uncertainties in evaluating the effects of Pb in marine environments
18                  include the presence of multiple stressors, inherent natural variability, and differences in
19                  Pb bioavailability across saltwater ecosystems. Evidence is inadequate to establish if
20                  there is a causal relationship between Pb exposures and the alteration of species richness,
21                  species composition and biodiversity in saltwater ecosystems.
      7.5        Causal Determinations for Ecological Effects of Pb

22                  This section summarizes the key conclusions regarding causality for welfare effects of
23                  Pb. Causal determinations for reproductive, growth, survival, neurobehavioral,
24                  hematological and physiological stress endpoints are presented separately for terrestrial,
25                  freshwater and saltwater organisms (Sections 7.3.12, 7.4.12. and 7.4.21). In Section 2.7.3.
26                  causal determinations for the same endpoints are further integrated across terrestrial,
27                  freshwater and saltwater taxa. Evidence considered in establishing causality was drawn
28                  from findings presented in the 1977 (U.S. EPA. 1977). 1986 (U.S. EPA. 1986a) and
29                  2006 Pb AQCDs (U.S. EPA. 2006c). integrated with an exhaustive review of more recent
30                  evidence. The causal statements for terrestrial, freshwater and saltwater effects are
31                  divided into two categories: (1) endpoints that are commonly used in ecological risk
32                  assessment (reproduction, growth and survival) because they clearly can lead to
33                  population-level (e.g., abundance, production, extirpation), community-level (taxa
34                  richness, relative abundance) and ecosystem-level effects (Anklev etal.. 2010; Suter et
35                  al.. 2005). and (2) organism and sub-organism responses such as neurobehavioral effects,

      November 2012                            7-206                 Draft - Do Not Cite or Quote

-------
 1                   hematological effects and physiological stress. There are many different effects at the
 2                   molecular and cellular levels, and chronic toxicity of Pb in ecosystems is thus likely
 3                   attained through multiple modes of action. Furthermore, the effects of Pb on ecosystems
 4                   necessarily begin with some initial effects at the molecular level of specific organisms
 5                   within the ecosystem (U.S. EPA. 1986b).

 6                   Experimental settings for studies used in making causal determinations for the ecological
 7                   effects of Pb include controlled exposures in the laboratory, microcosm experiments and
 8                   field observations. Controlled exposure studies in laboratory or small-to medium-scale
 9                   field settings provide the most direct evidence for causality, but their scope of inference
10                   may be limited. In contrast, microcosms and field studies where exposure is not
11                   controlled include potentially confounding factors (e.g., other metals) or factors known to
12                   interact with exposure (e.g., pH), thus increasing the uncertainty in associating effects
13                   with exposure to Pb specifically. A large majority of the available studies of Pb
14                   exposures are laboratory toxicity tests on single species, in which an organism is exposed
15                   to a known concentration of Pb and the effect on a specific endpoint is evaluated. These
16                   studies provide evidence for a temporal sequence between Pb exposure and an effect, an
17                   aspect important in judging causality (Table I Preamble). As detailed in the Framework
18                   for Causality (see Preamble), coherence between different types of studies also provides
19                   strong support to a determination of causality. Evidence from laboratory studies
20                   conducted under controlled conditions provides the largest amount of information used in
21                   the causal determinations summarized in Table 7-3. but their coherence with microcosm
22                   and field-based studies plays an important role in those determinations. Biological
23                   gradients (Table I Preamble) are often found in studies of the effects of Pb, and add
24                   support to causality where present. For some ecological endpoints, support for causal
25                   determinations is additionally supported by toxicological findings reviewed in the
26                   chapters of the ISA that evaluate evidence for human health effects associated with Pb
27                   exposure, particularly when a common mode of action is documented.

28                   The amount of Pb  in ecosystems is a result of a number of inputs and it is not  currently
29                   possible to determine the contribution of atmospherically-derived Pb from total Pb in
30                   terrestrial, freshwater or saltwater systems. The causal determinations are, therefore, not
31                   specific to Pb from atmospheric deposition since atmospherically-derived Pb may
32                   ultimately be present in water, sediments, soils and biota (Section 7.2 and Figure 7-1).
33                   The causal determinations encompass findings of studies at concentrations of Pb reported
34                   from environmental media (Table 7-2). and up to one to two orders of magnitude above
35                   the range of these values (Preamble Table II). Studies at the upper range of Pb
36                   concentrations are generally conducted at and near heavily exposed sites such as mining
37                   and metal industries-disturbed areas. Studies at those higher concentrations were used
38                   only when they were part of a range of concentrations that also included more typical
      November 2012                             7-207                 Draft - Do Not Cite or Quote

-------
 1                  values, or when they informed understanding of modes of action and illustrated the wide
 2                  range of sensitivity to Pb across taxa.

 3                  The exposure values at which Pb elicits a specific effect in terrestrial and aquatic systems
 4                  are difficult to establish, due the influence of other environmental variables on Pb
 5                  bioavailability and toxicity and to substantial differences among biological species in
 6                  their sensitivity to Pb. In the 1977 Pb AQCD (U.S. EPA. 1977). no correlation could be
 7                  established between toxic effects in invertebrates, fish, birds or small mammals and
 8                  environmental concentrations of Pb. At the time of the 1986 Pb AQCD additional data
 9                  were available on toxicity but there was still little information on the exposure values that
10                  can cause toxic effects in small mammals or birds (U.S.  EPA. 1986b). In the
11                  2006 Pb AQCD (U.S. EPA, 2006c) several studies on effects of Pb exposure on natural
12                  ecosystem structure and function advanced the characterization of Pb levels in the
13                  environment that occur near contaminated sites (i.e., smelters, mining,  industry).
14                  According to the 2006 Pb AQCD, natural terrestrial ecosystems near significant Pb
15                  sources exhibited a number of ecosystem-level effects, including decreased species
16                  diversity, changes in floral and faunal community composition, and decreasing vigor of
17                  terrestrial vegetation. These findings were summarized in Table AX7-2.5.2 of the Annex
18                  to the 2006 Pb AQCD (U.S. EPA. 2006c). The 2006 Pb  AQCD concluded that, in
19                  general, there was insufficient information available for single materials in controlled
20                  studies to permit evaluation on higher levels of biological organization (beyond the
21                  organism). Furthermore, Pb rarely occurs as a sole contaminant in natural systems
22                  making the effects of Pb difficult to ascertain. Recent information available since the
23                  2006 Pb AQCD, includes additional field studies in both terrestrial and aquatic
24                  ecosystems, but the connection between air concentration and ecosystem exposure
25                  continues to be poorly characterized for Pb and the contribution of atmospheric Pb to
26                  specific sites is not clear.
      November 2012                             7-208                  Draft - Do Not Cite or Quote

-------
Table 7-3    Summary of Pb causal determinations for plants, invertebrates and
              vertebrates.
    Level
Effect
Terrestrial3    Freshwater3  Saltwater3
Community
and
Ecosystem
1 Population-Level Endpoints
Sub-organismal Organism-Level Responses
Responses
Community and Ecosystem Effects
Reproductive and Developmental Effects-Plants
Reproductive and Developmental Effects-
Invertebrates
Reproductive and Developmental Effects-
Vertebrates
Growth-Plants
Growth-Invertebrates
Growth-Vertebrates
Survival-Plants
Survival- Invertebrates
Survival- Vertebrates
Neurobehavioral Effects-Invertebrates
Neurobehavioral Effects- Vertebrates
Hematological Effects-Invertebrates
Hematological Effects-Vertebrates
Physiological Stress-Plants
Physiological Stress-Invertebrates
Physiological Stress-Vertebrates
Likely Causal
Inadequate
Causal
Causal
Causal
Likely Causal
Inadequate
Inadequate
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Causal
Causal
Likely Causal
Causal
Inadequate
Inadequate
Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Causal
Likely Causal
Likely Causal
Likely Causal
Inadequate
Inadequate
Suggestive
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Inadequate
Suggestive
Inadequate
Inadequate
Suggestive
Inadequate
aBased on the weight of evidence for causal determination in Table II of the ISA Preamble. Ecological causal
determinations are based on doses or exposures generally within one to two orders of magnitude of the range of Pb
currently measured in the environment (Table 7-2).
November 2012
                          7-209
      Draft - Do Not Cite or Quote

-------
7.6       Supplemental Material
Table 7-4



Species
Terrestrial plants, invertebrates and vertebrates; growth, reproduction and survival.

Exposure Exposure
Concentration Concentration
(Nominal) (Measured)


Exposure
Method


Modifying
factors
Reference3
(Published since
Effect the 2006 Pb
Effects on Endpoint Concentration AQCD)
Plants
Buckwheat
(Fagopyrum
esculentum)














Canola
(Brassica
napus)




Contaminated
soil:
HCI extractable:
6,643 mg Pb/kg
Acetate
extractable:
832 mg Pb/kg
Water leachate:
0.679 mg Pb/kg

Control soil:
HCI extractable:
5 mg Pb/kg
Acetate
extractable: ND
Water leachate:
ND
0; 22; 45; and 67
mg Pb/kg





Plants were
grown for 8
weeks in
contaminated soil
collected from a
shooting range,
and control soil.










Plants of four
cultivars were
grown for 40
days in soil
amended with
Pb chloride.

Contaminated soil
Sand: 62.3%
Silt: 36.7%
Clay: 1 .0%
pH:6.0
CECM3.0
Control soil
Sand: 87.7%
Silt: 12.3%
Clay: ND
pH: 6.3
CEC:7.6












Growth: Tamura et al.
No effect on growth (2005)
Survival:
No effect on survival













Growth: Ashraf et al. (201 1 )
Shoot and root dry weight
decreased with increasing
Pb
Zn, Cu, Fe, Mn content
decreased with increasing
Pb.
                                                                    N, P, K, and Ca^+content
                                                                    decreased to a lesser
                                                                    degree.
November 2012
7-210
Draft - Do Not Cite or Quote

-------
Species
Chinese
cabbage
(Brassica
pekinensis)
Corn (Zea
mays)
Grass pea
(Lathyrus
sativus)
Exposure Exposure
Concentration Concentration Exposure Modifying
(Nominal) (Measured) Method factors
46; 874; 1 ,703 Plants were
mg Pb/kg dry soil grown for 12
days in soil
amended with
Pb acetate.
0; 0.007; 0.7; 7 Seeds were
mg Pb/L germinated on
paper soaked in
P-sulfate.
Plants were
grown for 21
days in washed
sand with
Pb-sulfate
amended nutrient
solution.
16; 31; 63; 125; Plants were
188mgPb/L grown in soil
amended with
Pb nitrate.
Effect
Effects on Endpoint Concentration
Growth:
Shoot biomass decreased
with increasing Pb (91% and
84% of lowest exposure).
Reproduction:
Germination%, germination
index, plant decreased with
increasing Pb.
Growth:
Shoot length, plant dry
weight, water use efficiency
decreased with increasing
Pb.
Reproduction:
Germination decreased with
increasing Pb (control 100%,
highest exposure 30%).
Chromosomal abnormalities
increased with increasing Pb
(control 0%, highest
exposure 72%).
Growth:
Shoot length decreased with
increasing Pb (highest
exposure was 50% of
control).
Reference3
(Published since
the 2006 Pb
AQCD)
Xiong et al. (2006)
Ahmad et al.
(201 1 )
Kumar and Tripathi
(2008)
November 2012
7-211
Draft - Do Not Cite or Quote

-------



Species
Lettuce
(Lactuca
sativa)








Mustard
(Brassica
juncea)



Exposure Exposure
Concentration Concentration
(Nominal) (Measured)
2,000 mg Pb/kg,
but soil was
mixed with 50%
V/V vermiculite
following
amendment with
Pb nitrate.
Tissue Pb:
3.22-233
n-l« DK/U«
mg ru/kg




0;31;62; 124;
186; 249; 311
mg Pb/L




Exposure
Method
Plants were
grown for 40
days in 21 soils
with varying
native CEC, OC,
pH, and
amorphous Fe
and Al oxides,
which were then
all amended with
Pb nitrate and
mixed with 50%
V/V vermiculite.


Plants were
grown for 60
days in field soil
amended with
Pb acetate.


Modifying
factors
After amendment:
pH: 3.8-7.8
CEC:
3.01-32.04cmolc/kg
OC:
5 - 30 g/kg
Fe/AI oxides:
0.009-0.1 95 mol/kg












Effect
Effects on Endpoint Concentration
Reproduction:
Germination 50 - 92%
Germination was greater in
amended soils.
Growth:
2.5-88.5% of control
In the presence of the same
amount of Pb(NO3)2, OC was
the main determinant of
effects, although CEC had a
strong influence, but
mediated by its effect on pH
and Fe/AI oxides
Growth:
Root and shoot length
decreased with increasing
Pb, and the decrease was
greater with time.
Reference3
(Published since
the 2006 Pb
AQCD)
Dayton et al.
(2006)









John et al. (2009)




                                                                                        After 60 days, roots were two
                                                                                        times longer in controls than
                                                                                        in the highest Pb exposure
                                                                                        shoot length was 75%
                                                                                        greater.
Radish 0; 21; 105
(Raphanus mg Pb/L
sativus)



Plants were
grown for 35
days in sand with
a full nutrient
solution
amended with
Pb nitrate.
Growth:
Leaf area, root volume,
shoot and root dry weight
decreased with
increasing Pb.
(total dry weight at
91 mn Ph/l was 3D% smaller
Gopal et al. (2008)





                                                                                        than control, 52% smaller at
                                                                                        105mg Pb/L).
November 2012
7-212
Draft - Do Not Cite or Quote

-------

Species
Wheat
(Triticum
aestivum)
Lettuce
(Lactuca
sativa)





Exposure Exposure
Concentration Concentration
(Nominal) (Measured)
69-9,714
mg Pb/kg





Exposure Modifying
Method factors
Wheat plants pH: 4.25-7.26
were grown for 6 oc. 6 2.47 Q%
weeks in
undisturbed core
samples from
four locations in
each two
Pb-contaminated
sites
Lettuce seeds
were germinated
in the leachate.
Reference3
(Published since
Effect the 2006 Pb
Effects on Endpoint Concentration AQCD)
Growth: Chapman et al.
No effects were found on (201 0)
germination or growth of
either species.




Invertebrates
Cabbage        0.87 mg Pb/L in
aphid           watering solution
(Brevicoryne     used for plants
brassicae)
   Aphids were
   reared for several
   generations on
   radish and
   cabbage plants
   grown in soil
   amended with
   Pb nitrate.
               Reproduction:
               In aphids fed
               Pb-contaminated plants,
               development time was
               longer, and relative fecundity
               and rate of population
               increase were lower than in
               control aphids.
               Survival:
               Mortality was higher in
               exposed aphids, both adults
               and offspring.
November 2012
7-213
Draft - Do Not Cite or Quote

-------



Species
Collembolan
(Folsomia
Candida)





















Collembolan
(Folsomia
Candida)



Exposure Exposure
Concentration Concentration
(Nominal) (Measured)
Approximate
range was 12 mg
Pb/kg soil to
1 5,000 mg Pb/kg
soil;
Pore water
approximately
0.002 Pb/L to
1 ,000 mg Pb/L
(Concentrations
were measured,
but not reported).
















0; 100; 200; 400;
800; 1,600; 3,200
mg Pb/kg dry soil




Exposure
Method
Springtails were
reared for 28
days in soil
collected at
seven locations
along each of
three transects
with increasing
Pb concentration
s within each
transect
(21 locations).
Lowest
concentration
soils from each of
the three
transects were
then amended
with Pb nitrate to
match the
gradient, and one
set of the
amended
samples were
then leached, for
a total of
57 concentration
sofPb.
Springtails were
reared for 1 0
days in field soil
amended with
Pb chloride.


Modifying
factors
pH was constant in
transects, but
decreased with
increasing addition
of Pb(NO3)2 in both
amended and
amended-and-
leached soils.
pH decreased by 3
units in the highest
addition, regardless
of subsequent
leaching.























Effects on Endpoint
Reproduction:
Reproduction decreased by
up to 50% in transect soils
Amended soils
Pb concentrations 2,207 mg
Pb/kg or lower never had a
significant effect on
reproduction.

















Reproduction:
Hatching success decreased
with increasing Pb.




Effect
Concentration
Transect A 28 day
EC50 in mg Pb/kg dry
weight:
native: >5,690
amended: 2,570
amended and
leached: 2,060
Transect B 28 day
EC50 in mg Pb/kg dry
weight:
native: > 14, 400
amended: 3,210
amended and
leached: 2,580
Transect C 28 day
EC50 in mg Pb/kg dry
weight:
native: >5,460
amended: 2,160
amended and
leached: 2,320







EC50 (hatching):
2,361
mg Pb/kg dry soil


Reference3
(Published since
the 2006 Pb
AQCD)
Lock et al. (2006)























Xu et al. (2009b)




November 2012
7-214
Draft - Do Not Cite or Quote

-------

Species
Collembolan
(Sinella
curviseta)



Collembolan
(Paronychiuru
s kimi)





Exposure Exposure
Concentration Concentration
(Nominal) (Measured)
0; 100; 200; 400;
800; 1,600; 3,200
mg Pb/kg dry soil



Toxicity run: 100;
500; 1,000; 2,000
mg Pb/kg
Reproduction run:
0; 250; 500;
1,000; 2,000;
3,000 mg/kg Pb





Exposure Modifying
Method factors
Springtails were
reared for 28
days in field soil
amended with
Pb chloride.



Springtails were
reared for 28
days on artificial
soil amended
with Pb chloride
in two separate
runs.





Effects on Endpoint
There was a small effect of
Pb on survival and growth,
and a stronger effect on
reproduction.



Survival:
Survival decreased with
increasing Pb.
Reproduction:
Offspring production and
instantaneous rate of
increase values decreased
with increasing Pb.





Effect
Concentration
Survival:
1 ,838 mg Pb/kg
Reproduction:
642 mg Pb/kg
EC50:
3,21 2 mg/kg Pb
Body Size:
4,094 mg Pb/kg
Survival LC50:
7 day:
1 ,322 mg Pb/kg
28 day:
1 ,299 mg Pb/kg
EC50
28 day: 428 mg Pb
/kg
NOEC
reproduction: EC50
28 day: 428 mg
Pb/kg
NOEC: 250 mg
Pb/kg
LOEC: 500 mg
Pb/kg
Reference3
(Published since
the 2006 Pb
AQCD)
Xu et al. (2009a)



Son et al. (2007)




November 2012
7-215
Draft - Do Not Cite or Quote

-------

Exposure
Concentration
Species (Nominal)
Collembolans 10; 50; 100; 500;
(Sinella coeca, 1 ,000 mg Pb/kg
Folsomia
Candida)
















Earthworm
(Eisenia
andrei)











Exposure
Concentration
(Measured)




















2,000 mg Pb/kg in
soil
Internal
concentration of
Pb in earthworms
varied among the
amended soils,
between 29 and
782 mg Pb/kg dry
weight.








Exposure
Method
Springtails were
reared for 42 or
45 days in
artificial soil
amended with
Pb nitrate.














Earthworms were
reared for 28
days in 21 soils
with varying
native CEC ,OC,
„[! nn/J
pn, ana
amorphous
Fe and Al oxides,
which were then
all amended with
Pb nitrate.








Modifying
factors




















After amendment
pH:
3.8-7.8
CEC:
3.01 - 32.04
cmolc/kg
OC:
5 - 30 g/kg
Fe/AI oxides:
0.009 -0.1 95 mol/kg









Effects on Endpoint
S. coeca:
Survival:
Mortality significantly
increased with increasing
concentration in adult
population.
Reproduction:
Juvenile production not
significantly compromised at
10-500 mg Pb/kg, reduced at
1 ,000 mg Pb/kg


F. Candida:
Survival:
Increase in mortality with
increasing concentration
Reproduction:
Juvenile production not
significantly reduced
between 10-500 mg Pb/kg,
significant effect at 1 ,000 mg
Pb/kg
Survival:
Mortality ranged between 0
and 100%.
In the presence of potentially
lethal amounts of Pb, the
main determinant of mortality
was pH, with little or no
effect from OC, CEC, or
Fe/AI oxides.
Reproduction:
Reproduction relative to
controls ranged between 0
and 167%.
Effects of Pb on reproduction
are dependent principally on
Fe/AI oxides, with some
influence of CEC


Effect
Concentration
LC50:
S. coeca: Could not
be determined
F. Candida: Could
not be determined

EC50 reproduction:
S. coeca:
490 mg Pb/kg Pb on
dry soil
F. Candida:
Could not be
calculated with
accuracy; Ranged
from 500-1 ,000
mg Pb/kg



















Reference3
(Published since
the 2006 Pb
AQCD)
Mentaetal. (2006)



















Bradham et al.
(2006)











November 2012
7-216
Draft - Do Not Cite or Quote

-------



Species
Earthworm
(Eisenia fetida)







Earthworm
(Eisenia fetida)





Earthworm
(Eisenia fetida)
















Exposure
Concentration
(Nominal)









0; 300; 711;
1 ,687; 2,249
mg Pb/kg




Soil!:
0; 355; 593; 989;
1 ,650 mg Pb/kg
Soil 2:
59; 297; 593;
2,965 mg Pb/kg

Soil 3:
386; 771 ; 1 ,929;
3,857 mg Pb/kg








Exposure
Concentration Exposure
(Measured) Method
18-9,311 Earthworms were
mg Pb/kg reared for 1 4
days in OECD-
standard toxicity
testing soil with
7 concentrations
of Pb, either one
or 10 earthworms
per container.



Mean: 79% of Earthworms were
nominal reared for 14
days in soil
amended with
five levels of
Pb nitrate and
artificially aged.
Earthworms were
reared for 28
days in three
soils amended
with five levels of
Pb nitrate without
aging, after which
they were
removed from the
containers.
Containers were
then kept in the
same conditions
for another 28
days, after which
cocoons were
extracted.


Modifying
factors
pH decreased with
increasing Pb nitrate
addition
NH3 concentration
increased with
Pb concentration
and time





pH 6.72 prior to
amendment
OC 0.7%

CEC 11 meq/100g


PH
6.72; 5.48; 6.75
(prior to
amendment)
OC
0.7; 1.2; 5.2%

CEC
11;8;27meq/100g








Reference3
(Published since
Effect the 2006 Pb
Effects on Endpoint Concentration AQCD)
Survival: LC50 (multiple- Currie et al. (2005)
Mortality increased from 0 to occupancy):
1 00% with increasing Pb, 2 662 mg Pb/kg at
with 1 00% reached at 4,500 7'd a9nd 9
mg Pb/kg after 7 days and
1 4 days. Number of worms 2>589 m9 Pb/k9 at
per container had no effect 14 days or
on mortality. 2,827 mg Pb/kg at
Growth. both 7 and 1 4 days
Worm weight decreased with
increasing Pb, and faster in
multiple-worm containers.
Survival: Jones etal.
Mortality was only observed (2009b)
at the highest exposure.




Reproduction: Jones etal.
Soil!: (200*)
Juvenile and cocoon count
decreased from 19 and 45,
respectively, to near 0 with
increasing Pb.

Soil 2:
Cocoon count decreased to
40% of control at highest Pb.
Soil 3:
Cocoon count was 0 at all
concentrations.




November 2012
7-217
Draft - Do Not Cite or Quote

-------



Species
Earthworm
(Pheretima
guillelmi)













Earthworms
(Eisenia
andrei,
Lumbricus
ru bell us,
Aporrectodea
caliginosa)










Exposure Exposure
Concentration Concentration
(Nominal) (Measured)
Toxicity run:
0, 1,000; 1,400;
2,000; 2,800;
3,800; 5,400;
7,500 mg Pb/kg
dry weight (form
of Pb not
reported).

Sublethal toxicity
run:
1,000; 1,400;
1,800; 2,500
mg Pb/kg dry
weight (form of Pb
not reported).
0; 1,000; 3,000;
4,000; 5,000;
7,500; 10,000
mg Pb/kg













Exposure
Method
Earthworms were
reared for 1 4
days in OECD-
standard soil
amended with Pb
(form of Pb
unreported).
There were two
runs with
different
concentrations.





Earthworms were
reared for 28
days in sterilized
Kettering Loam
watered with
Pb nitrate
solution.











Modifying
factors
















Temperature:
20 °C (E. andrei);
15°C(L rubellus
and A. caliginosa)
pH
Day 7 : 4.57-5.83,
Day 28: 4.71-5.83,
increasing with
decreasing Pb










Effects on Endpoint
Survival:
Mortality increased with
increasing Pb.(0% in control,
100% at 7,500 mg Pb/kg
after 14 days).
Growth:
Weight decreased with
increasing Pb(2.5 g in
control, 1 .4 g at 5.400 mg
Pb/kg after 14 days).
Reproduction:
Sperm abnormalities
increased with increasing Pb
(10% of control, 21% at
2,500 mg Pb/kg after 14
days).
Growth:
Weight decreased with
increasing concentration and
time, severity of weight
decrease varied with
species.
Survival:
Mortality increased with
increasing concentration and
time, and varied with species

100% mortality for all
species at higher
concentrations after 28 days.


Reference3
(Published since
Effect the 2006 Pb
Concentration AQCD)
Zheng and Li
(2009)














LC50: Langdon et al.
E. andrei: (2005)
5,824 mg Pb/kg
L. rubellus:
2,867 mg Pb/kg
A. caliginosa:
2,747mg Pb/kg

EC50:
E. andrei:
2,841 mg Pb/kg

L. rubellus:
1 ,303 mg Pb/kg
A. caliginosa:
1 ,208 mg Pb/kg
November 2012
7-218
Draft - Do Not Cite or Quote

-------

Exposure
Concentration
Species (Nominal)
Nematode
(Caenorhabditi
s elegans)



Nematode 5; 10; 16; 21
(Caenorhabditi mg Pb/L
s elegans)


Nematode 0.5; 16; 41 mg
(Caenorhabditi Pb/L
s elegans)





Exposure
Concentration Exposure Modifying
(Measured) Method factors
0.5; 10; 21 Nematodesat
mg Pb/L various
developmental
stages were
exposed to
Pb(NO3)2 for four
hours.
Late larval
nematodes (L4)
were exposed for
one or three
days.


Nematodes were
placed for 48
hours in growing
medium with
4 concentrations
ofPb.
Nematodes were
placed for three
days in growth
medium
amended with
Pb nitrate.




Effect
Effects on Endpoint Concentration
Reproduction:
Brood size decreased with
increasing Pb , but the
decrease was smaller with
increasing developmental
age.
Generation time increased
with increasing Pb, and the
increase was smaller with
increasing developmental
age.
These effects were greater in
late larval nematodes when
exposure duration increased
from four hours to one and
three days.
Survival:
No effect


Growth:
Life span, body size
decreased with increasing
Pb.
Reproduction:
Generation time and brood
size increased with
increasing Pb.
All effects were present and
of comparable magnitude in
progeny of exposed
nematodes.
Reference3
(Published since
the 2006 Pb
AQCD)
Guo et al. (2009)



Vigneshkumar et
al. (In Press)


Wang and Peng
(2007)




November 2012
7-219
Draft - Do Not Cite or Quote

-------

Species
Snail
(Achatina
achatina)



Exposure
Concentration
(Nominal)





Exposure
Concentration
(Measured)
1.33;
70.98;
134.61;
339.40;
674. 86;
1,009.22;
1 ,344.39 mg
Pb/kg

Exposure Modifying
Method factors
Snails were
reared for 1 2
weeks on a diet
amended with
Pb chloride.


Effect
Effects on Endpoint Concentration
Survival:
no effect
Growth:
Small decrease in feeding at
highest exposure, small
decrease in weight gain with
Reference3
(Published since
the 2006 Pb
AQCD)
Ebenso and
Ologhobo (2009a)


                                                                                       increasing Pb (over 12
                                                                                       weeks, snails in the highest
                                                                                       exposure gained 12% less
                                                                                       weight than in the lowest
                                                                                       exposure).
Snail
(Achatina
achatina)








Snail (Theba 0; 50; 100; 500;
pisana) 1 ,000; 5,000;
10,000; 15,000
mg Pb/Kg




Snails
(Cantareus
aspersus,
Helix aspersa)





0.56;
20.37;
200.42;
1,200.30
mg Pb/Kg














Total Soil Pb:
1 740-2060 mg
Pb/kg
CaCI2extractable;
4-80 mg Pb/kg
Dissolved
(estimated):
0.007-0.09 mg
Pb/L
Snails from
laboratory source
were reared for
12 weeks in
bottomless
enclosures at
four locations
within the
grounds of an
abandoned
battery factory.
Snails were
reared for 5
weeks on
Pb-amended
diet.



Snails were
reared for 7 - 9
weeks in field soil
amended with
varying amounts
of Pb-sulfate,
clay, peat, and
CaCO3.

pH: 4.42 -6.29,
decreasing with
increasing Pb
OC: 1.39-3.45%,
decreasing with
increasing Pb

CEC: 3.32 -5.37
cmol/kg, increasing
with increasing Pb.









Clay content
11 - 1 6%
Organic matter
1.2-1 0%

pH
AC ~7 AC,
4.D - / .4y


Growth:
Feeding, weight gain and
shell thickness all decreased
with increasing Pb (13; 17;
and 19% lower in highest
exposure than in lowest).





Growth:
Feeding and weight gain
decreased with increasing
Pb and time (snails in 0
added Pb gained 45% more
weight than in highest Pb).
Survival:
No effect
Growth:
No effect
Survival:
No effect





Ebenso and
Ologhobo (2009b)









EI-Gendy et al.
(201 1 )






Pauget et al.
(201 1 )







November 2012
7-220
Draft - Do Not Cite or Quote

-------

Exposure
Concentration
Species (Nominal)

Exposure
Concentration
(Measured)


Exposure Modifying
Method factors
Reference3
(Published since
Effect the 2006 Pb
Effects on Endpoint Concentration AQCD)
Vertebrates
Japanese
quail (Coturnix
coturnix
japonica)








Pied
flycatchers
(Ficedula
hypoleuca)














Drinking water
Pb:
0; 5; 50 mg Pb/L

Tissue Pb:
0;1.1; 10.7
mg/kg wet weight





Blood Pb in
nestlings at
mining site
while active:
41 mg Pb/100kg
wet weight;

after closing:
29 mg Pb/100kg
wet weight.
Blood Pb in
nestlings at
reference site
while active:
2 mg Pb/1 00 kg
wet weight;
after closing:
0.4 mg Pb/1 00 kg
wet weight.
Quails were
given
Pb-amended
water for 7
weeks.







Data were
collected in wild
flycatchers near
a Pb mine and at
a reference site
for three years
while the mine
was active, and
for three years
five years after
mine closing.







Growth: Main and Smits
Feed intake and growth rate (201 1 )
not affected.
Survival:
Morbidity/mortality was lower
in highest exposure than in
control.
Incidence of pericarditis,
airsacculitis, perihepatitis,
and arthritis was lower in
highest exposure than in
control.
Reproduction: Berglund et al.
Clutch size and breeding (201 0)
success were lower at the
mine site, but did not change
after closure of the mine
(clutch size
5.6 reference,
4.9 mining site;
breeding success
80% reference site,
76% mining site).
Nestling mortality was higher
at the mine site , and
increased after closure
(5% reference site,
1 1 % mining site while active;
1 1 % reference site,
26% mining site after
closure).
November 2012
7-221
Draft - Do Not Cite or Quote

-------

Species
Pig (Sus
domestica)




Exposure Exposure
Concentration Concentration
(Nominal) (Measured)
Feed Pb
control:
unreported
exposed:
1 0 mg Pb/kg
Blood Pb
control:
1 .44 |jg/dL
exposed:
2.08 |jg/dL

Exposure Modifying
Method factors
Pigs were reared
for 120 days with
Pb-sulfate-
amended feed.




Effect
Effects on Endpoint Concentration
Growth:
Significant decrease in body
weight, average day gain,
average day feed intake, and
feed efficiency.
Increase in feed conversion
ratio.
Reproduction:
No effect on ovary and
uterus weight
Reference3
(Published since
the 2006 Pb
AQCD)
Yu et al. (2005)



References included are those which were published since the 2006 Pb AQCD.
November 2012
7-222
Draft - Do Not Cite or Quote

-------
Table 7-5 Freshwater plants, invertebrates and vertebrates; growth, reproduction and survival.
Species
Concentration
Exposure Method
Modifying
factors
Effects on Endpoint
Effect Concentration
Reference3
(Published
since the
2006 Pb
AQCD)
Algae/Plants
Blue-green algae
(Spirulina
(Arthrospira)
platensis)
Microalgae
(Scenedesmus
obliquus)
Microalgae
(Chlorella vulgaris)
Duckweed
(Lemna minor)
Duckweed
(Lemna minor)
5,000; 10,000;
30,000; 50,000;
and
1 00,000 ug Pb/L
(nominal
concentration at
day 0 and then Pb
in media was
measured every
two days
thereafter).
5,000to
300,000 ug Pb/L
(nominal)

100; 200; 490;
900; 2,000; 5,020;
7,990; and
9,970 ug Pb/L
(measured)
2,070; 10,360;
20,700; and
1 03,600 ug Pb/L
(nominal)
10-day exposure to
Pb nitrate in Zarrouk
liquid medium.
48 and 96-hour acute
toxicity test with
Pb nitrate in BG11
medium.
4 day or 7-day
exposures to
Pb chloride in static
test conditions with
Jacob culture medium
under continuous
illumination.
9-day exposures to
Pb nitrate in a growth
chamber on Knopp's
medium under a 14-
hour photoperiod.
Temperature:
25 ± 1 °C
pH:
7.0
Light dark cycle of
14:10 hours.
Plants were
incubated at
normal room
temperature
(not provided)
Temperature:
25 ± 2 °C
PH
6.0

Growth:
10-day algal growth (measured
turbidimetrically at 560 nm) was
stimulated by 3.7% in the lowest
concentration, growth was inhibited
at higher concentrations of 30,000;
50,000; and 100,000 ug Pb/L by 40;
49; and 78%, respectively.
Chlorophyll a and b content were
significantly diminished at the three
highest exposures.
Growth:
In growth studies (measured as cell
division rate) S. obliquus was
significantly more sensitive to Pb
exposure than C. vulgaris
Growth:
Growth (measured as biomass) of
the duckweed was promoted up to
103% at 100 ug Pb/L and
200 ug Pb/L. However, growth was
inhibited monotonically at all other
test levels with increasing
concentrations. Overall, the relative
growth rate was reduced to 37-38%
at the highest concentration.
Growth: At lower Pb doses, growth
was slightly stimulated. Fresh weight
was lower by 65% at the highest
dose. Pb-induced chlorosis occurred
and the enzymes of the antioxidative
system were modified due to Pb
exposure in all concentrations.
LC50:
75,340 ug/L
48 hour EC50:
4,040 ug Pb/L
S. obliquus
48 hour EC50:
24,500 ug/L
C. vulgaris
4 day EC50:
6,800 ug Pb/L
7 day EC50:
5,500 ug/L

Arunakumara
et al. (2008)
Atici et al.
(2008)

Dirilgen (2011)
Paczkowska
et al. (2007)
November 2012
7-223
Draft - Do Not Cite or Quote

-------



Species
Duckweed
(Woffia arrhiza)






Waterweed
(Elodea canadensis)




Wetland plants
(Beckmannia
syzigachne,
Alternanthera
philoxeroides,
Juncus effusus,
Oenanthe
javanica,
Cyperus
flabelliformis,
Cvoerus
\SJ/ftl,l UVJ
malaccensis,
Polypogon fugax,
Leersia hexandra,
Panicum
paludosum,
Neyraudia
reynaudiana)



Concentration
210; 2,120;
20,720; and
207,200 ug Pb/L
(analytically
verified)




1,000; 10,000;
and 100,000 ug/L
(nominal)




20,000 ug/L
(nominal)
Metals were
analyzed in plant
samples.
















Exposure Method
14-day exposure to Pb
as Pb nitrate in sterile
1/50 dilution of
Hutner's medium, day
to night cycle of 16:8
hours.



Plants were exposed 5
days to Pb as
Pb acetate in a 10%
nutrient solution and
then assayed for
pigment content, total
ascorbic acid, and
protein content.
Field-collected tillers or
seedlings (from various
locations in China) for
each species were
used in 21 -day
experiments to
determine Pb tolerance
as inferred from
measuring the
elongation of the
longest root in a
hydroponic system in a
Pb nitrate solution.








Modifying
factors
Temperature:
25 ± 0.5 °C
pH'
r
7.0





Temperature:
25-27°C
pH
6.5-6.7

10% Hoagland &
Arnon nutrient
solution
















Reference3
(Published
since the
2006 Pb
Effects on Endpoint Effect Concentration AQCD)
Growth: Piotrowska et
Biomass decreased proportionally al. (2010)
with increasing Pb concentration;
Chlorophyll a content was
significantly inhibited at 210 ug Pb/L
and greater; carotenoid,
monosaccharide and protein content
significantly decreased in higher
concentrations.
Growth: Dogan et al.
The chlorophyll, carotenoid, and (2009)
protein contents of £. canadensis
were significantly reduced following
Pb accumulation.


Growth: Deng et al.
Root elongation was significantly (2009)
reduced in a number of wetland
species (6. syzigachne, J. effusus,
O. javanica, C. flabelliformis,
C. malaccensis, and
N. reynaudiana). Metal tolerance was
related to root anatomy and spatial
pattern of radical oxygen loss.










November 2012
7-224
Draft - Do Not Cite or Quote

-------


Species


Concentration


Exposure Method

Modifying
factors


Effects on Endpoint


Effect Concentration
Reference3
(Published
since the
2006 Pb
AQCD)
Invertebrates
Rotifer
(Brachionus
calyciflorus)



Rotifer
(Brachionus patulus)











Rotifer
(Euchlanis dilatata)




67; 194; 284; 390;
and 700 |jg Pb/L
(measured)



1,250; 2,500;
4,000; 5,000; and
8.000 ug Pb/L
(nominal) for
acute toxicity
tests.
Chronic
exposures used
nominal
concentration of
60 and
600 ug Pb/L with
varying turbidity
levels.
0.1; 0.5; 50; 100;
250; 1 ,000;
2,500 ug Pb/L
(analytically
verified, actual
concentrations not
reported)


Cysts of rotifers were
obtained from Florida
Aqua farms in Dade
City, Florida, U.S.
Tests with Pb nitrate
were performed in total
darkness for 48 hours.

24-hour exposures to
Pb chloride in the
presence and absence
of sediments using
rotifers originally
isolated from the
Chimaliapan wetland,
Toluca, Mexico.
Three week chronic
toxicity tests were also
conducted.



48-hour acute toxicity
tests with rotifer
neonates exposed to
Pb nitrate in synthetic
moderately hard water.
Adult rotifers were
collected in a reservoir
in Aguascalientes,
Mexico.
Temperature:
25 ± 1 °C
pH:
8.19


Temperature:
20 °C











Temperature:
25 ± 2 °C
pH:
7.5

Hardness:
80-100mg/L
/>_/>PI
oauUs

Reproduction:
The total number of rotifers and the
intrinsic rate of population increase
exhibited concentration-dependent
responses at the end of the 48 hour
incubation period.

Reproduction:
In chronic tests, net reproductive rate
and rate of population increase
decreased under conditions of
increasing turbidity and Pb
concentration.
Survival:
24-hour LC50 reported for this
species. In chronic tests, average life
span and life expectancy at birth
decreased under conditions of
increasing turbidity and Pb
concentration.

Survival:
Based on 48-hour LC50 E. dilatata is
among the most sensitive rotifer
species to Pb. E. dilatata may be a
more suitable test organism for
ecotoxicology in Mexico, where this
study was conducted, instead of
D. magna, a species that is not been
found in Mexico reservoirs.
EC2o for number of
rotifers: 125 ug Pb/L
48-hour EC2o for intrinsic
rate of population
increase: 307 ug Pb/L
NOEC:194ugPb/L
LOEC: 284 ug Pb/L
24-hour LC50:
6, 150 ug/L











48-hour
NOEC: 0.1 ug/L
LOEC: 0.5 ug/L
LC 50: 35 ug/L
(estimated from
analytically verified
concentrations)

Grosell et al.
(2QQ6b)




Garcia-Garcia
etal. (2007)











Arias-Almeida
and Rico-
Martinez (2011)




November 2012
7-225
Draft - Do Not Cite or Quote

-------


Species
Cladoceran
(Ceriodaphnia dubia)









Cladoceran
(Ceriodaphnia dubia)








Cladoceran
(Diaphanosoma
birgei)

Cladoceran
(Moina micrura)


Cladoceran
(Nona rectangular)



Concentration
Measured but not
reported.
Predicted
concentration of
major Pb
chemical species
in the natural
water bioassays is
provided in Table
4 of Esbaugh et
al. (2011).



A range of 5 to 6
Pb concentrations
(measured but not
reported) were
prepared with
varying pH,
hardness and
alkalinity.





2,000to
5,500 ug Pb/L
(analytically
verified)
1 ,000 to
8,000 ug Pb/L
(analytically
verified)
3,000to
1 0,000 ug Pb/L
(analytically
verified)


Exposure Method
Acute toxicity of Pb to
C. dubia was assessed
in 48-hour exposures
to two lab generated
reference waters and
eight natural waters
from across North
America selected to
include a range of
water quality
parameters. Waters
were spiked with
varying concentrations
of Pbas Pb nitrate.
Chronic 7-day static
renewal 3 times per
week in 2:1
dechlorinated, aerated
tap waterdeionized
water to determine the
effects of hardness (as
CaSO4 and MgSO4),
alkalinity, pH, and
DOM on Pb toxicity.



24-hour exposure to
Pb chloride in
moderately hard water.

24-hour exposure to
Pb chloride in
moderately hard water

24-hour exposure to
Pb chloride in
moderately hard water


Modifying
factors
Temperature: 26°C
Water chemistry of
the field-collected
waters are reported
in Table 1 of
Esbaugh et al.
(201 1 ) including pH
(range 5.5 to 8.5),
Ca2+ (range 24 to
1,934uM), DOC
(range 36 to 1 ,244
uM) and hardness
(range 4 to 298
mg/L)
Temperature:
25 °C
pH:
6.4-8.2
Hardness:
22-524 mg/L






Temperature:
23 °C
pH:
7.0-7.5
Temperature:
23 °C
pH:
7.0-7.5
Temperature:
23 °C
pH:
7.0-7.5


Effects on Endpoint
Survival:
LC50 values ranged from 29 to
1,180 ug Pb/L. Sensitivity to Pb
varied greatly with water chemistry.
DOC was correlated with protection
from acute toxicity.







Survival:
DOM and alkalinity have a protective
effect against chronic toxicity of Pb.
CaSO4 and MgSO4 do not have a
protective influence of water
hardness; Pb toxicity increased at
elevated Ca2+ and Mg2+. Low pH
increases the toxicity of Pb.
Reproduction:
Increased DOC leads to an increase
in mean EC50 for reproduction
ranging from approximately
25 ug Pb/L to >500 ug Pb/L.
Survival:
LC50 reported


Survival:
M. micrura was more sensitive to Pb
than D. birgei and A. rectangular.

Survival:
A. rectangular was more resistant to
Pb than the other species tested



Effect Concentration
48 -hour LC50 range:
29 to 1,1 80 ug Pb/L
NOEC range: 18 to
<985 ug Pb/L.
LOEC range: 52 to
1 ,039 ug Pb/L







Control base water EC20:
45 ug Pb/L
5.0 mM CaSO4 EC20:
22 ug Pb/L
32 mg/L DOM EC20:
523 ug Pb/L
2.5 mM NaHCO3 EC20:
73 ug Pb/L
Additional EC20 and all
EC50 values were reported
in the study.

24 hour LC50
3,160ug Pb/L


24 hour LC50
690 ug Pb/L


24 hour LC50
7,000 ug Pb/L

Reference3
(Published
since the
2006 Pb
AQCD)
Esbaugh et al.
(2011)









Mageret al.
(2011 a)








Garcia-Garcia
et al. (2006)


Garcia-Garcia
et al. (2006)


Garcia-Garcia
et al. (2006)

November 2012
7-226
Draft - Do Not Cite or Quote

-------


Species
Cladoceran
(Daphnia magna)





Cladoceran
(Daphnia pulex)






Ostracod
(Stenocypris major)












Concentration
Acute test:
Concentrations
not provided
Chronic test:
25 |jg Pb/L
250 |jg Pb/L
2,500 |jg Pb/L
(nominal)
Acute test:
250; 500; 1 ,000;
2,000;
5,000 |jg Pb/L
(nominal)
Chronic test:
250; 500;
1 ,000 |jg Pb/L
(nominal).
475; 1,160; 3,410;
4, 829;
8,972 |jg Pb/L
(measured)












Exposure Method
24 hour acute toxicity
test and 21 day toxicity
test with Pb nitrate,
static renewal every
two days.



48-hour acute toxicity
test and two 21 -day
exposures to Pb nitrate
in dechlorinated tap
water.




96-hour static renewal
with Pb nitrate in
dechlorinated tap
water. Ostracods were
collected from a filter
system of a fish pond
in Bangi, Selangor,
Malaysia.







Modifying
factors
Temperature:
20 ± 1 °C













Temperature:
28-30 °C
pH:
6.5 ±0.01

Conductivity:
244.3 ± 0.6 uS/cm
DO: 6.3 ± 0.06
mg/L
Total hardness:
15.6 mg/L as
CaCO3
Light dark cycle of
12:12 hours.


Effects on Endpoint
Reproduction:
Significant concentration-dependent
decrease in number of neonates per
female; Significant long-term effects
on reproduction. Negative correlation
between hemoglobin gene
expression and reproduction
outcomes.
Reproduction:
Reproduction rates (cumulative
neonates) significantly decreased at
1 ,000 ug Pb/L in the first chronic
toxicity test and at 500 ug Pb/L in a
second test.
Survival:
LC50 values reported

Survival:
LC50 values reported










Reference3
(Published
since the
2006 Pb
Effect Concentration AQCD)
24-hour EC50 (immobility): Ha and Choi
1 8,1 53 ug Pb/L (2009)





48-hour LC50: Theegala et al.
4,000 ug Pb/L (2QQ7J






24-hour LC50: Shuhaimi-
6 583 ua Pb/L Othman et al.
,ou ,„ (201 1b)
48-hour LC50:
2,886 ug Pb/L
72-hour LC50:

1,491 ug Pb/L
96-hour LC50:
526 ug Pb/L




November 2012
7-227
Draft - Do Not Cite or Quote

-------



Species
Midge
(Chironomus dilutus,
formerly C. tentans)





















Midge
(Chironomus
riparius)








Concentration Exposure Method
29; 57; 75; 1 1 5; 96-hour static renewal
128; 152 ug Pb/L test
(measured) 20 day midge life cycle
test in Pb chloride
spiked water, flow
through, and
emergence at 55 days.
The C. dilutus culture
was initially started with
egg cases from Aquatic
Biosystems, Fort
Collins, CO, USA.













Logarithmic range 24-hour acute toxicity
from 0 to tests with first-instar
25,000 ug Pb/L larvae exposed to
(nominal) Pb nitrate in synthetic
soft water. C. riparius
culture was from egg
masses from
Environment Canada.


Modifying
factors
Average ± SD
(range):
Temperature:
22.2 ± 1.0 °C
(1 9.7-24.4 °C)
pH:
7.26 ±0.21
(6.9 - 7.7)
Hardness:
32 ± 3.2 mg/L
as CaCO3

Alkalinity:
31 ± 3.0 mg/L
as CaCO3

Conductivity:
76 ± 4.9 us
DO:
7.8 ± 0.8 mg/L





Temperature:
20 °C
Water hardness:
8 mg/Lof CaCO3,







Effects on Endpoint
Growth:
Growth and emergence decreased
as concentration increased.
Reproduction:
No effect

Survival:
No effect


















Survival:
Concentration-dependent decrease
in survival with increasing Pb.
Of the five metals tested in the study
(Cd, Cu, Pb, Ni, Zn), Pb was most
toxic to first instar C. riparius.





Effect Concentration
96-hour LC50:
3,323 ug Pb/L
Survival
NOEC: 152 ug Pb/L
LOEC: >152 ug Pb/L
MATC: >1 52 ug Pb/L
Weight
NOEC: 57 ug Pb/L
LOEC: 75 ug Pb/L
MATC: 65 ug Pb/L
EC10(95%):15ugPb/L
EC20(95%):28ugPb/L

Fecundity
NOEC: 152 ug Pb/L
LOEC: >1 52 ug Pb/L
MATC: >1 52 ug Pb/L
EC10(95%):>152ugPb/L
EC20(95%): >152ug Pb/L
Emergence
NOEC: 1 15 ug Pb/L
LOEC: 128ugPb/L
MATC: 121 ug Pb/L
EC10: 28 ug Pb/L
EC20: 55 ug Pb/L
24-hour LC 50:
613 ug Pb/L






Reference3
(Published
since the
2006 Pb
AQCD)
Mebane et al.
(2008)






















Bechard et al.
(2008)






November 2012
7-228
Draft - Do Not Cite or Quote

-------


Species
Midge
(Chironomus
javanus)








Midge
(Culicoides furens)










Concentration
430; 580,; 1,330;
2,460;
7,670 |jg Pb/L
(measured)








100; 290; 510;
800;
2,800 ug Pb/L
(measured)










Exposure Method
4-day exposure with
fourth instar larvae to
Pb nitrate in aerated,
filtered, dechlorinated
tap water with static
aerated renewal at
2 days. Larvae were
collected from a filter
system of a fish pond
in Bangi, Selangor,
Malaysia.


Series of 96-hour
exposures to
Pb chloride in
dechlorinated water
under several
temperature ranges.









Modifying
factors
Temperature:
28-30 °C
pH:
6.51 ± 0.01
Conductivity:
244.3 ± 0.6 uS/cm
DO:
6.25 ± 0.06 mg/L
Total hardness
(Mg2+ and Ca2+):
15.63 ±2.74 mg/L
as CaCO3
1st experiment:
Temperature:
25-28 °C
2nd experiment:
Temperature:
20-26 °C

3rd experiment:
Temperature:
10, 15,20,23,25,
28, 30, 35,
40 ± 0.5 °C




Effects on Endpoint
Survival:
LC50 values reported for this species.









Survival:
Higher and lower temperatures
brought about increased toxicities.
LC50 values generally increased in
10-25°C and decreased in 28-40°C.
40°C temperature produced 100%
mortality.








Effect Concentration
24 hour LC50:
20,490 ug Pb/L
48 hour LC50:
6,530 ug Pb/L
72-hour LC50:
1 ,690 ug Pb/L
96 hour LC50:
790 ug Pb/L




96 hour LC50 values:
28-25 °C*: 400 ug Pb/L
26-20 °C*: 300 ug Pb/L
25 °C: 400 ug Pb/L
35 °C-25 °C*: 500 ug Pb/L
23 °C: 700 ug Pb/L
20 °C: 400 ug Pb/L
15 °C: 400 ug Pb/L
10 °C: 357 ug Pb/L

'temperature decreased
over the duration of the
experiment
Reference3
(Published
since the
2006 Pb
AQCD)
Shuhaimi-
Othman et al.
(2011c)








Vedamanikam
and Shazilla.
(2008a)








November 2012
7-229
Draft - Do Not Cite or Quote

-------


Species
Midge
(Chironomus
plumosus)










Oligochaete worm
(Lumbriculus
variegatus)

Mayfly
(Baetis tricaudatus)












Concentration
3,000; 5,400;
8,200; 30,000;
54,000 |jg Pb/L
(measured)










1,300; 3,200;
8,000; 20,000;
50,000 ug Pb/L
(nominal)
69; 103; 160; 222;
350; 546 ug Pb/L
(measured)












Exposure Method
Series of 96-hour
exposures to
Pb chloride in
dechlorinated water
under several
temperature ranges.









24 and 48-hour
exposures to Pb nitrate
spiked water from
Lake Vesijarvi, Finland
96-hour static renewal
test and 10 day chronic
study with aerated
Pb chloride spiked
water, static renewal
every 48 hours.
Mayflies were collected
from the South Fork
Coeurd'Alene River,
Idaho.







Modifying
factors
1st experiment:
Temperature:
25-28°C
2nd experiment:
20-26°C
3rd experiment:
10, 15, 20, 23, 25,
28, 30, 35,
40 ± 0.5 °C






Temperature:
20 °C


Mean ± SD
Temperature:
9.3 ± 0.67 °C
nU-l-
pH.
6.64 ±0.18
Hardness:
20.7 ± 0.58 mg/L
as CaCOS
Alkalinity:
19.8± 1.04 mg/L
as CaCOS
Conductivity:
47.7 ± 1 .72 us
DO:
10.1 ±0.45 mg/L


Effects on Endpoint
Survival:
Higher and lower temperatures
brought about increased toxicities
40°C temperature produced 100%
mortality
LC50 values generally increased in
10-25°C and decreased in 28-40°C








Survival:
48-hour LC50 reported


Growth:
Consistent dose-dependent
reductions in mayfly growth (as
number of molts); growth decreased
with increased Pb exposure.
Survival:
96-hour EC50 reported for this
species. Reduced molting endpoint
more sensitive than mortality
endpoint.





Reference3
(Published
since the
2006 Pb
Effect Concentration AQCD)
96 hour LC50 values: Vedamanikam
28-25°C*' 16 200 ua Pb/L and Snazilla
' ' (2008a)
26-20°C*: 8,300 ug Pb/L 	
25 °C: 9,500 ug Pb /L
35 °C: 700 ug Pb/L
30 °C: 700 ug Pb/L
28 °C: 900 ug Pb/L
25 °C: 900 ug Pb/L
23 °C: 700 ug Pb/L
20 °C: 600 ug Pb/L
15 °C: 600 ug Pb/L
10 °C: 500 ug Pb/L
'temperature decreased
over the duration of the
experiment
48-hour LC50: Penttinen et al.
5,200 ug Pb/L (2008)


96-hour EC50 Mebane et al.
664 ug Pb/L (2008)
Survival:
NOEC: 222 ug Pb/L
LOEC: 350 ug Pb/L
MATC: 279 ug Pb/L
EC1() (95%): 169ugPb/L
EC20(95%):23ugPb/L
Molting :
NOEC: 103ugPb/L
LOEC: 160ugPb/L
MATC: 130ug Pb/L
EC10(95%):37ugPb/L
EC20(95%):66 ug Pb/L


November 2012
7-230
Draft - Do Not Cite or Quote

-------


Species
Mosquito
(Culex
quin quefascia tus)





Neosho mucket
(Lampsilis
rafinesqueana)









Concentration
Acute test:
100; 150; 200;
250 ug Pb/L
(analytically
verified)
50; 150; and
200 ug Pb/L
for reproductive
studies
Concentrations
were measured
and used to
calculate EC50
values, reported
in supplemental
data.








Exposure Method
24-hour acute toxicity
test and several tests
to assess reproductive
endpoints. All tests
were conducted with
Pb nitrate in distilled
water


24 and 48-hour
exposure with 5 day
old juveniles obtained
from adults collected
from Spring River, KS,
U.S.







Modifying
factors
Temperature:
25 ± 2 °C
pH:
7





Temperature:
20 ± 1 °C
pH:
7.2-7.6
DOC:
>7.0 mg/L
Hardness:
40-48 mg/L
as CaCO3
Alkalinity:
30-35 mg/L
as CaCO3


Effects on Endpoint
Reproduction:
Hatching rate significantly decreased,
lower emergence rates, larval
development from L1 to adults took
longer.
Survival:
24 hour LC50 reported


Survival:
Neosho mucket is a candidate
species for U.S. federal endangered
and threatened status. Toxicity
testing with newly transformed
juveniles indicated that this species is
sensitive to Pb exposure.






Reference3
(Published
since the
2006 Pb
Effect Concentration AQCD)
Survival: Kitvatanachai et
24 hour LC50: aL ^^>
1 80 ug Pb/L





24 hour EC50: Wang et al.
5 day old juveniles (201 Oe)
>507 ug Pb/L
48 hour EC50:
5 day old juveniles:
1 88 ug Pb/L






November 2012
7-231
Draft - Do Not Cite or Quote

-------
Species
Fatmucket mussel
(Lampsilis
siliquoidea)












Concentration Exposure Method
For 28-day 24 and 48-hour
exposure: exposure with
0.04; 2.9; 6.1; 17; glochidia, 96-hour
36; 83 ug Pb/L exposure with 5 day
(measured) old, and 2 or 6 month
old juveniles and
28-day exposure with 2
or 4 month old mussels
in reconstituted soft
water. Tests were
conducted with
glochidia and juveniles
obtained from adults
collected from the
Silver Fork of Perche
Creek, MO, U.S.





Modifying
factors
Temperature:
20 ± 1 °C
pH'
r
7.2-7.6
DOC:
>7.0 mg/L
Hardness:
40-48 mg/L
as CaCO3
Alkalinity:
30-35 mg/L
as CaCO3








Effects on Endpoint
Growth:
Growth of juvenile mussels in the
17 ug Pb/L concentration was
statistically significantly reduced
compared to growth in the controls at
the end of 28 days. Growth was not
assessed in the higher
concentrations due to mortality.

Survival:
The 24-hour EC50 values for glochidia
and 96-hour EC50 values for 2 and 6
month old juveniles were much
higher than 96 hour LC50 value for 5
day old newly transformed juveniles.
Genus mean chronic value was the
lowest value ever reported for Pb.
Survival was based on foot
movement within a 5-minute
observation period.
Effect Concentration
24 and 48 hour EC50:
glochidia
>400 ug Pb/L (test 1 )
>299 ug Pb/L (test 2)
48 hour EC50:
5 day old juveniles
465 ug Pb/L (test 1)
392 ug Pb/L (test 2)
96 hour EC50:
5 day old juveniles
142 ug Pb/L (test 1)
298 ug Pb/L (test 2)
24 and 48 hour EC50:
2 month old juveniles
>426 ug Pb/L
4 day EC50: >83 ug Pb/L

10 day EC50: >83 ug Pb/L
21 day EC50: 29 ug Pb/L
28 day EC50: 20 ug Pb/L
Reference3
(Published
since the
2006 Pb
AQCD)
Wang et al.
(201 Oe)












                                                                                                                   28-day NOEC
                                                                                                                   Juvenile fatmucket:
                                                                                                                   6.1 ug/L
                                                                                                                   28-day LOEC
                                                                                                                   Juvenile fatmucket:
                                                                                                                   17 ug/L
                                                                                                                   Genus mean chronic
                                                                                                                   value
                                                                                                                   10 ug Pb/L
Snail
(Lymnaea stagnalis)
4; 12; 16; 42; 113;
and 245 ug Pb/L
(measured)
30-day exposure with
newly hatched snails
(<24-hour old) in
artificial fresh water
with Pb nitrate.
Temperature:        Growth:
23 ± 1 °C           Newly hatched snails exhibited
                   greatly reduced growth in response
                   to Pb exposure
                   Survival:
                   No Pb-induced mortality was
                   observed.
EC20 <4 ug Pb/L
NOEC: 12 ug Pb/L
LOEC: 16 ug Pb/L
Grosell et al.
(2006b)
November 2012
                            7-232
                                    Draft - Do Not Cite or Quote

-------


Species
Snail
(Lymnaea stagnalis)







Snail
(Marisa cornuarietis)






Snail
(Biomphalaria
glabrata)













Concentration
1st Experiment:
<0.5 (control),
2.7 and
18.9 |jg Pb/L
(measured)
2nd Experiment:
1.3 and
7.5 ug Pb/L
(measured)


5,000;1 0,000; and
15,000 ug Pb/L
(nominal)






50; 100; and
500 ug Pb/L
(nominal)













Exposure Method
Pb exposures were
performed with juvenile
snails (~ 1 g) for 21
days and then 1 4 days
in dechlorinated tap
water under flow-
through conditions.




5-day, 6-day, and
10-day exposure to
Pb chloride in
deionized or double-
distilled water. Snail
strain used for egg
production was from
the Zoological Institute
in Frankfurt, Germany.


96-hour acute
laboratory bioassays













Modifying
factors
Dechlorinated City
of Miami tap water
([Na+] -1.1 mmol/L
[Ca2+] -0.31
mmol/L
[Cf] -1.03 mmol/L
[HCO3~] -0.68
mmol/L,
[DOC]~200 umol/L
pH~7.7 at room
temperature
Temperature:
24 ± 1 °C
pH:
-7.5
Conductivity:
-800 uS/cm




Dechlorinated
continuously
aerated tap water:
Temperature:
22 °C
pH:
7.1 ±0.2

Total hardness:
65 ± 3 mg CaCO3/L
Alkalinity:
29 ± 2 mg CaCO3/L
Conductivity"
230 ± 17 uS


Effects on Endpoint Effect Concentration
Growth:
In juveniles exposed to 1 8.9 ug/L Pb
for 21 days, Ca2+ influx was
significantly inhibited and model
estimates indicated 83% reduction in
growth of newly hatched snails after
30 days at this exposure
concentration
Survival:
No Pb-induced mortality was
observed
Reproduction: LOEC: 10,000 ug Pb/L
Significant delay in hatching at
1 0,000 ug Pb/L
Growth:
Significantly delayed development
(reduced visible tentacles, eye
formation) at 15,000 ug Pb/L.
No effect on fresh weight.
Survival:
Significantly increased mortality at
15,000 ug Pb/L
Reproduction:
Significant decrease in number of
eggs laid at 500 ug Pb/L.
Survival:
Embryonic survival was 12% of the
number of eggs laid by the control
group at 100 ug Pb/L. Time to
hatching increased 3 fold from the
control. No embryos survived the
highest concentration.




Reference3
(Published
since the
2006 Pb
AQCD)
Grosell and Brix
(2009)







Sawasdee and
Kohler (201 0)






Ansaldo et al.
(2009)












November 2012
7-233
Draft - Do Not Cite or Quote

-------
Species
Prawn
(Macrobrachium
lancesteri)
Crayfish
(Orconectes hylas)
Concentration
22; 31; 48; 126;
170|jgPb/L
(measured)
Reference sites:
0.03 ug Pb/L
Mining sites:
0.12 to
1 .59 ug Pb/L
Downstream sites
0.03 to
0.04 ug Pb/L
Exposure Method
4-day exposures in
Pb chloride in aerated,
filtered, dechlorinated
tap water, static
aerated, renewal at
2 days. Prawns were
purchased from
aquarium shops in
Bani, Selangor,
Malaysia.
in situ 28-day exposure
with juvenile crayfish in
streams impacted by
Pb mining and
reference sites in
Missouri, USA.
Modifying
factors
Temperature:
28-30 °C
pH:
6.51 ±0.01
Conductivity:
244.3 ± 0.6 uS/cm
DO:
6.25 ± 0.06 mg/L
Total hardness
(Mg2+ and Ca2+):
15.63 ±2.74 mg/L
as CaCO3
Water quality
parameters were
measured at each
site
Temperature
23 to 26 °C
pH:
7.9 to 8.1
Conductivity:
282 to 858
DO:
6.3 to 8.4 mg/L)
Alkalinity 141 to
182 mg/L as
CaCO3
Turbidity: 0.4 to 0.6
NTU
Sulfate:
0.3 to 304 mg/L
Other metals were
present
downstream of
mining sites
Effects on Endpoint
Survival:
LC50 increased with decrease in
mean exposure concentration
Survival:
Crayfish survival and biomass were
significantly lower in streams
impacted by Pb mining Metal
concentrations were negatively
correlated with caged crayfish
survival.
Effect Concentration
24 hour LC50
85.9 ug Pb/L
48 hour LC50:
58.5 ug Pb/L
72 hour LC50:
45.5 ug Pb/L
96 hour LC50
35.0 ug Pb/L

Reference3
(Published
since the
2006 Pb
AQCD)
Shuhaimi-
Othman et al.
(2011 a)
Allertetal.
(2009a)
November 2012
7-234
Draft - Do Not Cite or Quote

-------



Species



Concentration



Exposure Method


Modifying
factors
Reference3
(Published
since the
2006 Pb
Effects on Endpoint Effect Concentration AQCD)
Vertebrates
Fathead minnow
(Pimephales
promelas)
























Measured:
Mean ± SEM
Tap low Pb:
28 ± 1.1 |jg Pb/L
Tap high Pb:
105 ±4.8 |jg Pb/L
HCOs'lowPb:
31 ± 1 .2 |jg Pb/L
HC03"highPb:
113 ±4.6 |jg Pb/L
Humic low Pb:
30 ± 1 .4 |jg Pb/L
Humic high Pb:
112 ±4.5 ug Pb/L

















4-day, 10-day, 30-day,
150-day, and 300-day
exposures in Pb nitrate
spiked dechlorinated
tap water with static
renewal to study the
effects of DOC and
alkalinity on Pb toxicity.
Breeding assays (21
days) were also
performed.




















Temperature:
22 ±1 °C
Tap H2O:
Hardness:
91 mg/L
pH:
8.1
+500 |JM NaHC03:
DOC:
257 uM
Hardness:
93 mg/L
pH:
8.3
NaHCO3:

Hardness:
93 mg/L
pH:
8.3
+4 mg/L HA:

Hardness:
(-JQ n-i/N/l
yo mg/L
pH:
8.0
Humic:
Hardness:
93 mg/L
pH:
8.0
Growth: Mager et al.
No statistically significant growth (29-1Q)
differences observed at any age due
to water chemistry alone; DOC
addition strongly protected against
Pb accumulation; increased alkalinity
reduced whole body Pb burdens;.
Growth inhibited at 4 days, but
recovered by 30 days in high Pb
concentration.
Reproduction:
HCO3- reduced 21 day total
reproductive output (reduced clutch
size and number of clutches
produced); addition of HCO3" alone
actually increased reproductive
output; significantly higher fecundity
in HCO3" treatment; Egg attachment
low in both tap water and HCO3"
treatments; HA promoted
attachment; No statistically significant
differences in egg hatchability. HCO3"
and humic acid treatments increased
average egg mass; no effects on
hatchability in the HCO3"and humic
acid treatment;






November 2012
7-235
Draft - Do Not Cite or Quote

-------
Species
Fathead minnow
(Pimephales
promelas)
























Concentration Exposure Method
<0.1 - 3,605 30 day flow through
ug Pb/L exposure to Pb nitrate
(measured) to determine to the
Pb was quantified effects of Ca*' '- numic
in fish tissues in a acid and PH (6-3 and
separate set of 8-3) on pb
experiments. accumulation and
toxicity in juvenile
fathead minnows.






















Modifying
factors
Exposure media
were made up from
a base-water
consisting of 2:1
deionized water:
dechlorinated tap
water.
Temperature: 23
°C and had various
levels of Ca2+,
humics, and pH
values:
0.5; 1 ; 2 mM Ca2+
2; 4; 8; 16mg
humic

6.3; 8.3 pH














Effects on Endpoint
Growth:
No growth inhibition was observed in
any treatment. An increase in growth
was observed in groups exposed to
higher Pb concentrations where there
were high initial mortalities.
Survival:
For most treatments, mortalities
occurred during the first 5 to 7 days
of exposure. The lowest tolerance
was observed at low pH (6.8).
Addition of DOC or CaSO4
decreased Pb toxicity.


















Effect Concentration
30 day LC50, EC20,
and LOEC values*:
LC 50 in ug Pb/L:
0.5 mM Ca2+: 91
1.0 mM Ca2+: 104
2 mg humic: 255
4 mg humic: 443
8 mg humic: 832
16 mg humic: 1903
pH 6.3: 4.5
pH 8.3: 13
EC 20 in ug Pb/L:
0.5 mM Ca2+: 47
1.0 mM Ca2+251
2 mg humic: 189
4 mg humic: 319
8 mg humic: 736
16 mg humic: 1729
pH 6.3: 2.1
pH 8.3: 8.7
LOEC in ug Pb/L:
0.5 mM Ca?+: 40
1.0mMCa2+:107
2 mg humic: 199
4 mg humic: 475
8 mg humic: 919
16 mg humic: 1751
pH 6.3: 6.2
pH8.3:15
*4 day and 1 0 day LC50,
LC2o, and LOEC values also
reported in the paper.
Reference3
(Published
since the
2006 Pb
AQCD)
Grosell et al.
(2006a)
























November 2012
7-236
Draft - Do Not Cite or Quote

-------


Species
Fathead minnow
(Pimephales
promelas)












Fathead minnow
(Pimephales
promelas)









Concentration
Measured but not
reported. Figure 1
of Esbaugh et al.
(2011) plots the
relationship
between
dissolved and
nominal Pb
concentrations in
three waters with
low Pb solubility.
Predicted
concentration of
major Pb
chemical species
in the natural
water bioassays is
provided in
Table 4 of
Esbaugh et al.
(2011)
0.2 ±0.1 ugPb/L
(control)
33 ± 4 ug/L
(chronic low)
143 ± 14 ug/L
(chronic high)
(measured)






Exposure Method
Acute toxicity of Pb to
juvenile P. promelas
(<24 hours old) was
assessed in 96-hour
static renewal
exposures to two lab
generated reference
waters and seven
natural waters from
across North America
selected to include a
range of water quality
parameters. Waters
were spiked with
varying concentrations
of Pbas Pb nitrate.





33 to 57-day exposures
in dechlorinated tap
water and deionized
water to Pb nitrate to
assess swimming
performance.







Modifying
factors
Temperature: 26
°C
Water chemistry of
the field-collected
waters are reported
in Table 1 of
Esbaugh et al.
(201 1 ) including pH
(range 5.5 to 8.5),
Ca2+ (range 24 to
1 ,934 uM), DOC
(range 36 to 1 ,244
uM) and hardness
(range 4 to 298
mg/L)





Temperature:
21 ± 1 °C
pH:
7.50 ±0 .03
Total CO2:
543 ± 69 umol/L
DOC:
108± 4umol
carbon/L
Hardness:
26 ± 3 mg/L


Effects on Endpoint Effect Concentration
Survival: 96-hour LC50 range:
LC50 values ranged from 41 to 41 to 3.598 U9 pb/L
3,598 ug Pb/L. NOEC: range
DOC had the strongest protective 1 4 to 2,271 ug Pb/L.
effect. LOEC: range
The lowest LC50 occurred in the pH 42 to 5>477 U9 pb/L
5.5 water.
No Pb toxicity was observed in three
alkaline natural waters.










Growth:
Fish from the chronic low exposure
were significantly larger than control
fish (increased mass and body
length).






Reference3
(Published
since the
2006 Pb
AQCD)
Esbaugh et al.
(2011)












Magerand
Grosell (201 1 )







November 2012
7-237
Draft - Do Not Cite or Quote

-------



Species
Fathead minnow
(Pimephales
promelas)








Zebra fish
(Dan/o rerio)
















Concentration
Fish were feed
L. variegatus
exposed via water
to 628 |jg Pb/L.








Mean daily dietary
dose of 0.417
(0.3-0.48)
or 0.1
(0.07-0.14)
mg Pb/kg/day
(measured)















Exposure Method
Juvenile fish fed a live
diet of the oligochaete
L. variegatus for
30 days contaminated
with Pb.








63 day dietary
exposure with
Pb-enriched
polychaete, Nereis
diversicolor.
Adult zebrafish were
fed a daily dose of 1 %
flake food (dry wet
diet/wet weight fish),
1% brine shrimp, and
1 % N. diversicolor
collected from either
Gannel Estuary,
Cornwall, U.K., an
estuary with legacy Pb
contamination, or
Blackwater Estuary,
Essex, U.K., (reference
site)


Modifying
factors
Water from Lake
Superior that was
subsequently
filtered and UV-
treated.
Temperature:
25 °C
pH:
7.5-8.0,
Hardness:
45-50 mg CaCO3,
Alkalinity :
40-45 mg CaCO3/L
Temperature:
29 ± 1 °C













Reference3
(Published
since the
2006 Pb
Effects on Endpoint Effect Concentration AQCD)
Growth: Erickson et al.
Not significantly affected (201 0)
Survival:
Not significantly affected








Reproduction: Boyle et al.
No impairment observed to incidence (2P-1PJ
of spawning, numbers of eggs per
breeding pair or hatch rate of
embryos compared with pre-
exposure levels. Metal analyses
revealed significant increases in
whole-body Pb burdens of male fish
fed polychaetes from the
contaminated estuary.









November 2012
7-238
Draft - Do Not Cite or Quote

-------


Species
Tilapia
(Oreochromis
niloticus)








Channel catfish
(Ictalurus
Punctatus)









African catfish
(Clarias gariepinus)








Concentration
Mean
concentration of
Pb in food pellets:
100; 400; and 800
mg Pb/kg dry
weight (nominal)





Fish were fed
L. variegatus
exposed via water
to 576 |jg Pb/L









100; 300; and
500 |jg Pb/L
(nominal)
Pb was quantified
in tissues
following
exposure.





Exposure Method
Fish obtained from an
unpolluted fish farm in
Hangzhou, China were
held in tanks with
dechlorinated tap water
and were fed diets with
Pb nitrate twice daily
for 60 days.




Juvenile fish fed a live
diet of the oligochaete
L. variegatus for
30 days contaminated
with Pb.








24; 42; 90; 138; and
162 hour embryo
exposure to Pb nitrate
in dechlorinated tap
water.
These intervals
corresponded to 30;
48,; 96; 144; and 168
hours post-fertilization.



Modifying
factors
Temperature:
25 ± 1 °C
„. i_| .
pH.
7.1-7.5
DO:
7.5-7.8 mg/L
Alkalinity:
109 mg CaCOs/L
Hardness:
118 mg CaCO3/ L
Water from Lake
Superior that was
subsequently
filtered and UV-
treated.
Temperature: 25
°C
pH:
7.5-8.0,
Hardness:
45-50 mg CaCOs/L
Alkalinity:
40-45 mg CaCO3/L
Temperature:
24 °C
pH:
8.0
Conductivity:
700 us/cm

Oxygen:
90-95% saturation


Reference3
(Published
since the
2006 Pb
Effects on Endpoint Effect Concentration AQCD)
Growth: Dai et al.
No effects on growth were found. (2009b)

Survival:
Exposure to Pb-contaminated diets
did not result in mortality.




Growth: Erickson et al.
Not significantly affected (201 0)
Survival:
Not significantly affected









Growth: Osmanetal.
Malformations observed in exposed (2007b)
embryos (malformed embryos only
survived shortly after hatching), delay
in development.
Reproduction:
Concentration-dependent delay in
hatching, reduced percentage of
embryos completing egg stage
period from 75% in control to 40% in
500 ug Pb/L.
November 2012
7-239
Draft - Do Not Cite or Quote

-------


Species
Rainbow trout
(Oncorhynchus
mykiss)







Rainbow trout
(Oncorhynchus
mykiss)













Concentration
In diet:
6.9 mg Pb-(g dry
mass)
(L. variegatus
exposed via
sediments)






Control Pb-free
diet of 0.06 mg
Pb/kg dry weight,
and three different
diets of 7; 77; and
520 mg Pb dry
weight (0.02
(control), 3.7;
39.6; and 221 .5
mg Pb/day dry
weight calculated
for food
consumption)

Pb also quantified
in tissues


Exposure Method
Juvenile fish fed a live
diet of the oligochaete
L. variegatus for
30 days contaminated
with Pb of varying
concentrations of
Pb nitrate






21 -day exposure to
juvenile rainbow trout
via diet amended with
Pb nitrate. Fish were
held in aerated tanks
with dechlorinated
water









Modifying
factors
Water from Lake
Superior that was
subsequently
filtered and UV-
treated.
Temperature:
11 °C
pH:
7.5-8.0
Hardness:
45-50 mg CaCOs/L
Alkalinity :
40-45 mg CaCO3/L
Temperature:
11-13°C
pH
7.5-8.0

Hardness:
140 ppm
as CaCO3








Reference3
(Published
since the
2006 Pb
Effects on Endpoint Effect Concentration AQCD)
Growth: Erickson et al.
Not significantly affected (201 0)
Survival:
Not significantly affected







Growth: Alvesetal.
No effects on growth rates were (2006)
observed in rainbow trout
administered a diet containing three
concentrations of Pb.
Dietary Pb was poorly absorbed.
Comparison of dietary and water-
borne exposures suggest that toxicity
does not correlate with dietary
exposure, but does correlate with gill
accumulation from waterborne
exposure.

Survival:
Not significantly affected by dietary
Pb
November 2012
7-240
Draft - Do Not Cite or Quote

-------
Species
Rainbow trout
(Oncorhynchus
mykiss)





























Concentration Exposure Method
ELS (early life 96-hour static renewal
stage) test 1 : acute toxicity test with
1 2; 24; 54; 1 43; swim-up stage fry (2 to
and 384 ug Pb/L 4 weeks post-hatch)
(measured) reared from eggs used
ELS test 2- in the chronic studies.
8; 18; 37; 87; and Two 60+ day ELS
124 ug Pb/L exposures were
(measured) conducted in a flow-
through system using
temperature controlled
water from Little North
Fork of the South Fork
Coeurd'Alene river in
ELS 1 (69 days) and
water from the South
Fork in ELS 2 (62
days).























Modifying
factors
ELS 1:
Temperature
9.8 ± 0.6 °C
pH:
6.75 ±0.4
Hardness:
19.7± 1.5mg/L
as CaCO3
Conductivity:
45.8 ±2.2 us
Alkalinity:
19.6± 2.2mg/Las
PaPPl
OdO*— '3
DO:
10.2±0.7mg/L

ELS 2:

Temperature
\ O C 4. ri Q op
i z.o x u.y o
pH:
7.19 ± 0.3
Hardness:
29.4 ± 3.6 mg/L
as CaCO3
Conductivity:
69.1 ±7.4 us
Alkalinity:
27 mg/L ± 2.1 as
CaCO3
DO:
9.2 ± 0.9 mg/L







Effects on Endpoint
Growth:
In ELS 1 , growth generally
decreased as concentration
increased, with fish in the highest
surviving treatment (143 ug Pb/L )
exhibiting severely stunted growth
that was statistically different from
the control.
In ELS 2, growth increased in the
highest treatment with a slight
reduction in length and wet weight in
intermediate exposures.
Survival:
In ELS 1 Survival decreased as
concentration increased with
complete mortality before the end of
the test in the highest treatment
(384 ug Pb/L).
In ELS 2, survival decreased
significantly in the highest treatment
with high survival in intermediate
exposure. In both tests, the greatest
number of mortalities occurred
around or shortly after the swim-up
stage.














Reference3
(Published
since the
2006 Pb
Effect Concentration AQCD)
96-hour LC50: Mebane et al.
120ugPb/L (2008)
ELS 1: Survival
NOEC: 24 ug Pb/L
LOEC: 54 ug Pb/L
MATC: 36 ug Pb/L
EC10: 26 ug Pb/L
EC20 : 34 ug Pb/L
ELS 1: Weight
NOEC: 24 ug Pb/L
LOEC: 54 ug Pb/L
MATC: 36 ug Pb/L
EC10: 39 ug Pb/L
EC20: 55 ug Pb/L
ELS 1: Length
NOEC: 54 ug Pb/L
LOEC: 143ug Pb/L
MATC: 88 ug Pb/L
EC10: 64 ug Pb/L
EC20: 98 ug Pb/L

ELS 2: Survival
NOEC: 87 ug Pb/L
LOEC: 1 25 ug Pb/L
MATC: 104ug Pb/L
EC10: 108ug Pb/L
EC20: 113ug Pb/L
ELS 2: Weight
NOEC: 37 ug Pb/L
LOEC: 87 ug Pb/L
MATC: 57 ug Pb/L
EC10: 7 ug Pb/L
EC20: >87 ug Pb/L
ELS 2: Length
NOEC: 8 ug Pb/L
LOEC: 18 ug Pb/L
MATC: 12ugPb/L
EC10: >87 ug Pb/L
EC20: >87 ug Pb/L
November 2012
7-241
Draft - Do Not Cite or Quote

-------


Species
Southern leopard
frog
(Rana
sphenocephala)












Northern leopard
frog
(Rana pipiens)











Concentration
Sediment:
45; 75; 180; 540;
2,360; 3,940;
5,520; 7,580
mg Pb/kg
dry weight
Corresponding
sediment pore
water:
123; 227; 589;
1,833; 8,121;
13,579; 19,038;
24,427 ug Pb/L
(measured)







3; 10; and
100 ug Pb/L
(nominal)
(Pb was
measured in
tissues at the end
of the study. Pb
tissue
concentrations
ranged from 0.1 to
224.5 mg Pb/kg
dry mass and fell
within the range of
tissue
concentrations in
wild-caught
tadpoles).


Exposure Method
20; 40; 61 ; and 82-day
exposures in
Pb acetate spiked
sediment collected
from wetland, static
renewal twice per
week.












Northern leopard frogs
were exposed to Pb as
Pb nitrate in
dechlorinated water
from the embryonic
stage to
metamorphosis
(>66 days post-
hatching)









Modifying
factors
Mean (SD)
Temperature:
21.6°C
pH:
6.92
DOC:
6.08 mg/L
Conductivity:
168uS/L
Hardness:
7.30 mg Ca2+/L

Ammonia:
0.39 mg/L


Sediment:
Organic carbon:
8.25%
Sand: 22.4%
Silt: 38.4%
Clay: 39.1%
Temperature:
21 to 22 °C
PH
7.9
Hardness:
1 70 mg/L
as CaCO3










Effects on Endpoint
Growth:
Snout-vent length and body mass
increased through time in all
treatments; skeletal deformations
(spinal deformations, digits truncated
and twisted, long bones curved and
truncated) increased with Pb content
and length of exposure
Survival:
Exposure to a 3,940 mg/kg sediment
Pb (13,579 ug/L) pore water) killed all
tadpoles within 5 days; tadpoles that
reached climax stage (Gosner 42)
had no difference in survival among
treatments through the completion of
metamorphosis.






Growth:
Tadpole growth was significantly
slower in the early stages in
100 ug Pb/L treatment. More than
90% of tadpoles in the 100 ug Pb/L
treatment developed lateral spinal
curvature, whereas almost all the
tadpoles in the other groups were
morphologically normal. No
significant effect of Pb exposure was
found on percentage metamorphosis,
snout-vent length, mortality, or sex
ratio. Time to metamorphosis was
delayed in 100 mg/L treatment.



Reference3
(Published
since the
2006 Pb
Effect Concentration AQCD)
40 day EC50 Sparling et al.
(for deformed spinal (2006)
columns in sediment):
1 ,958 mg/kg Pb
Corresponding EC50
(for deformed spinal
columns in pore water):
6,734 ug/L
40 day EC50
(for deformed spinal
columns in sediment):
579 mg Pb/kg,
and
1 ,968 ug Pb/L (in pore
water).

Sediment: LC50:
3,728 mg Pb/kg
(Corresponding to a
Pore water LC50 of:
12,539 ug Pb/L).

Chen et al.
(2006b)









"References included are those which were published since the 2006 Pb AQCD.
November 2012
7-242
Draft - Do Not Cite or Quote

-------
Table 7-6 Saltwater plants, invertebrates, and vertebrates: growth, reproduction, and survival.


Species


Concentration

Exposure
Method

Modifying
factors


Effects on Endpoint


Effect Concentration
Reference3
(Published
since the
2006 Pb
AQCD)
Algae
Microalgae
(Tetraselmis chuii,
Rhodomonas salina,
Chaetocerossp.,
Isochrysis galbana,
Nannochloropsis
gaditana)





Microalgae
(Tetraselmis
suecica)





50; 100; 250; 500;
800; 1,000; 1,600;
3,000; and
6,000 |jg Pb/L,
(nominal). Five
nominal
concentrations
were analytically
verified: 51 ; 225;
824; 1 ,704;
6,348 |jg Pb/L
(measured)
20 ug Pb/L
(nominal)
T. suecica in this
study was then
fed to Artemia
franciscana
(mean Pb content
12 to 15 mg
Pb/kq)
y/
Populations of
each microalgal
species were
exposed for 72
hours to ten
progressively
increasing
nominal
concentrations of
Pb in filtered
seawater

72-hour
exposure to
Pb nitrate in
filtered natural
seawater from
Mazatlan Bay,
Mexico. This
was the first step
in a four-level
food chain.
Temperature:
20 ± 1 °C
pH
8.0





Temperature:
28 ± 2 °C
Salinity:
34.6 ± 1 .2 ppt
nU-l-
pH.
-T Q Q 0
/ .y-o.z
DO saturation:
90-95%
(>7 mg/L)
Growth:
Growth inhibition (as measured
by flow cytometry) was reported
for each species. Species cellular
size, sorption capacity, or
taxonomy did not explain
differences in sensitivity to Pb.





Growth:
Mean final cell concentrations,
growth rate and total dry biomass
were significantly reduced (40%
lower than control cultures).
Effects on primary, secondary
and tertiary consumers were
observed following Pb-exposure
via T. suecica at the base of a
simulated marine food chain.
EC50:
T. chuii:
2,640 ug Pb/L
R. salina:
900 ug Pb/L
Cftaefocerossp.:
105ug Pb/L
/. galbana:
1 ,340 ug Pb/L

N. gaditana:
740 ug Pb/L








Debeliuset al.
(2009)







Soto-Jimenez
et al. (201 1 b)






Invertebrates
Polychaete
(Hydroides elegans)





91 ; 245; 451 ;
4,443; 9,210; and
41 ,060 ug Pb/L
(measured)




24-hour
exposure of
fertilized eggs to
Pb chloride.
Assay was
stopped at
2 hours to
assess effects
on blastula.

Temperature:
27 ± 1 °C
DO
(86.5%)
Salinity
(34 ± 1 ppt)
pH(8.1 ±0.1)
Carbonate
24.5 mg/L.
Reproduction:
Exposure to Pb caused a
significant decrease in the
number of embryos developing
normally to blastula after 2 to 3
hours of exposure to Pb.



EC50
Fertilization
membrane stage :
30,370 ug Pb/L
Blastula
1 ,429 ug Pb/L
24 hour trochophore
larva:
231 ug Pb/L.
Gopalakrishnan
et al. (2007)





November 2012
7-243
Draft - Do Not Cite or Quote

-------



Species Concentration
Polychaete 48; 97; 201 ; 407;
(Hydroides elegans) 803;
' ' w ' 1,621 |jgPb/L
(measured)







Polychaete 85; 137; 251;
(Capitellasp.) 392; 487; 738;
871 mg Pb/kg
(measured)












Exposure
Method
A series of
experiments
were performed
from 20 minutes
to 4 days in
Pb chloride
using
polychaetes
collected from
seawater in
Chennai, India.



3 and 6-day
exposure for
growth
experiments,
96-hour
exposure to
Pb chloride
spiked sediment
from
Chi-kou Estuary,
Taiwan





Modifying
factors
Temperature:
27 ± 1 °C
DO:
7-9 mg/L
Salinity
34 ± 1 ppt
pH:
8.1 ±0.1
Carbonate:
22.5 mg/L



Aerating
circulating
seawater
Temperature:
20 ± 2 °C

Salinity:
30%









Effects on Endpoint
Reproduction:
Fertilization rate decreased by
70% in sperm pretreated with
97 ug Pb/L for 20 minutes.
Fertilization rate of eggs
pretreated in 48 ug Pb/L
decreased to 20% of control. Life
stages of H. elegans varied in
their sensitivity to Pb. Gametes,
embryo and larvae were more
sensitive than adults with the
larval settlement period being
most sensitive to Pb exposure.
Survival:
LC50 reported for adults
Growth:
Significant differences among
growth rates of Capitella sp. in
different levels of
Pb-contaminated sediments, with
the exception of 251 mg/kg
treatment in the 6-day
experiment. Growth rates
deceased significantly from the
control in the 3-day experiment
but changes were inconsistent
with increasing Pb concentration.
Survival:
No effect



Effect Concentration
EC50
Sperm toxicity:
380 ug Pb/L
Egg toxicity:
692 ug Pb/L
Embryo toxicity:
1,130 ug Pb/L
Blastula to trochophore:
261 ug Pb/L
Larval settlement:
100 ug Pb/L
Adult 96-hour LC50 :
946 ug/L
Growth LOAEL:
85 mg Pb/kg












Reference3
(Published
since the
2006 Pb
AQCD)
Gopalakrishnan
et al. (2008)








Horng et al.,
(2009)












November 2012
7-244
Draft - Do Not Cite or Quote

-------



Species
Amphipod
(Elasmopus laevis)







Amphipod
(Melita plumulosa)





















Concentration
30-mg Pb/kg
(control whole-
sediment),
58 mg Pb/kg;
118mg Pb/kg;
234 mg Pb/kg;
424 mg Pb/kg
(measured)





Water-only tests
ranged from 0 to
4,000 ug Pb/L
(analytically
verified)

Adult Sediment
Test:
500; 1,000; 2,000;
4,000 mg Pb/kg
dry weight
(analytically
verified)
Juvenile
Sediment Test:
500; 1,000; 2,000
mg Pb/kg dry
weight
(analytically
verified)




Exposure
Method
Multi-
generational
bioassay with
amphipods
collected in
Jamaica Bay,
New York
exposed
60+ days to
sediment spiked
with Pb acetate
in filtered
seawater. 10-
day and 28-day
bioassays were
also conducted.
Juveniles and
adults were
tested in 96-hour
seawater only or
1 0 day static-
non-renewal
exposure spiked
sediment
collected from
intertidal mud
flats, Woronora
River, New
South Wales,
Australia. Adults
were also tested
in 10-day
seawater only
exposures.








Modifying
factors
Temperature:
19-24°C
Salinity:
27-29 g/L
DO:
>6.57 mg/L





Temperature
21 ± 1 °C,
Salinity
30 ± 1 %,
PH
7.2-8.2,
Ammonia (total)
<3 mg N/L
















Effects on Endpoint
Reproduction:
Fecundity was reduced as
sediment Pb concentration
increased.
Onset of reproduction and
reproduction were delayed as Pb
concentration increased.
Survival:
No differences in adult survival
among the Pb concentrations
tested in 28-day and 60-day
exposures.



Survival:
Juvenile amphipods were more
sensitive to Pb than adults in
seawater and sediment
exposures.



















Effect Concentration
Fecundity and time of first
offspring was significantly
reduced with increasing
sediment concentration
above 118 mg Pb/kg. Onset
to reproduction significantly
delayed at 1 1 8 mg Pb/kg
and delayed further at
higher tested
concentrations.





96-hour seawater-only
Adults:
LC50 3,000 ug Pb/L
NOEC 850 ug Pb/L
LOEC 1 ,680 ug Pb/L
Juveniles:
LC50 1 ,530 ug Pb/L
NOEC 400 ug Pb/L
LOEC 600 ug Pb/L
Seawater-only 10 days:
Adults:
LC50 1 ,270 ug Pb/L
NOEC 1 90 ug Pb/L
LOEC 390 ug Pb/L
10 days Sediment-only
Adults:
LC50 NOEC, LOEC
>3,560, mg Pb/kg
Juveniles:
LC50 1 ,980 mg Pb/kg
NOEC 580 mg Pb/kg
LOEC 1 ,020 mg Pb/kg
Reference3
(Published
since the
2006 Pb
AQCD)
Ringenary et al.
(2007)







King et al.
(2006)


















November 2012
7-245
Draft - Do Not Cite or Quote

-------


Species
Brine shrimp
(Artemia
franciscana)




Shrimp
(Litopenaeus
vannamei)










Sea urchin
(Paracentrotus
lividus)









Concentration
Mean Pb content
12 to 15mg Pb/kg
from dietary
exposure to Pb.




Pb in exoskeleton,
hepatopancreas,
muscle, and
remaining tissues
was quantified on
daysO; 15; 28;
and 42 of the
dietary study





50 to
5,000 ug Pb/L
(nominal)








Exposure
Method
This was the
second step in a
four-level food
chain.
A. franciscana
feeding on
Tetraselmis
suecica cultured
in 20 ug/L Pb, as
Pb nitrate.
This was the
third step in a
four-level food
chain.
L. vannamei, fed
A. franciscana
(mean Pb
content 12 to 15
mg Pb/kg)
feeding on
T. suecica
cultured in
20 ug/L Pb as
Pb nitrate.
Gametes and
embryos
exposed 48 to
50 hours in
filtered seawater
to Pb nitrate
from adults
collected from
the Bay of
Naples, Italy.

Modifying
factors
Temperature:
28 ± 2°C
Salinity:
34.6 ± 1 .2 ppt
nU-l-
pH.
7.9-8.2
DO saturation:
90-95%
(>7 mg/L)
Temperature:
28 ± 2°C
Salinity:
34.6 ± 1 .2 ppt

pH:
7.9-8.2
DO saturation:
90-95%
(>7 mg/L)




Filtered sea
water
Temperature:
18 ± 1 °C

Salinity: 38%

pH: 8 ±0.2




Effects on Endpoint
Growth:
A tendency toward lower biomass
yields was reported (significant
only on day of final harvest).
Survival:
A tendency toward lower survival
was reported (significant only on
day of final harvest).


Growth:
Tendency toward decreased total
length and weight was reported
(significant only on day of final
harvest)


Survival:
Tendency toward lower survival
was reported (significant only on
day of final harvest)


Growth:
Up to LOEC concentration,
defects observed in plutei were
mainly reduction in size (20%);
above LOEC concentration,
developmental defects were
mainly larvae affected in skeletal
or gut differentiation up to
2,000 ug Pb/L, where arrest of
development started to increase.


Effect Concentration
Dry biomass was 1 95 mg/L
in control cultures and 153
mg/L in cultures fed Pb
exposed T. suecica. Mean
cell count (individuals/L) on
days 19-23 (harvest) was
320 in control and 255 in
A. franciscana cultures fed
Pb-exposed T. suecica.

Total mean length in the
shrimp fed the experimental
diet was 13 mm and wet
weight was 7.1 g compared
to mean total length (14.8
mm) and wet weight (8.5 g)
at day 42.
Mean survival in the shrimp
fed the experimental diet
ranged from 67 to 78%
compared to control survival
(84 to 90%) at day 42.

EC50
1 ,250 ug Pb/L
NOEL
250 ug Pb/L

LOEC
500 ug Pb/L



Reference3
(Published
since the
2006 Pb
AQCD)
Soto-Jimenez
et al. (201 1 b)





Soto-Jimenez
et al. (201 1 b)











Manzo et al.
(201 0)








November 2012
7-246
Draft - Do Not Cite or Quote

-------


Species
Mussel
(Mytilus
galloprovincialis)





Clam
(Meretrix meretrix)








Scallop
(Argopecten
ventricosus)













Concentration
3,500; 4,500;
5,500;
6,000 ug Pb/L
(nominal)





2; 20; 197; 1,016;
7,158ug Pb/L
(measured)








For growth:
10; 100; 1,000;
and 10,000 ug/L
(analytically
verified)

For survival:
280; 560; 1120;
2,250; and
5,000 ug/L
(analytically
verified)




Exposure
Method
24-hour static
aerated
exposure
in seawaterwith
Pb acetate with
mussels
collected from a
mussel farm in
Greece.
24-hour and 96-
hour toxicity test
with Pb nitrate in
filtered sea water
using gametes
from adults
collected from
Wenzhou, China
and held under
laboratory
conditions.




144 hour
(survival) or 30
day (growth)
exposure to
Pb nitrate (static
renewal every
48 hours) with
juvenile
A. ventricosus
hatched from
laboratory
cultures held at
Universidad
Autonoma de
Baja California
Sur, Mexico.

Modifying
factors
Temperature:
25 ± 2°C
Salinity:
35%

DO:
7-8 mg/L


Temperature:
28 ± 1 °C
Salinity:
ono/.
ZU fO
pH:
7.8






Temperature:
20°C
Salinity:
36 ± 1 %
HO-
uu.
>4 mg/L









Effects on Endpoint
Survival:
Mortality at high Pb
concentrations





Reproduction:
Embryo development inhibited by
increasing Pb concentrations

Growth:
Significant concentration-
dependent growth inhibition in
larvae. Larval metamorphosis
rate decreased, no adverse effect
on larval settlement at
20.4 ug Pb/L
Survival:
Significant concentration-
dependent survival inhibition of
embryos over time
Growth:
Juvenile growth rates and weight
were significantly reduced at high
concentrations of Pb


Survival:
Juvenile mortality was
significantly different than control
at 96 hour LC50







Effect Concentration
24-hour LC50
4,500 ug Pb/L






Embryogenesis EC50:
297 ug Pb/L

Growth:
EC50:
197ug Pb/L
Metamorphosis:
>7,160ugPb/L

48-hour LC50:
>7,160ugPb/L
96-hour LC50:
353 ug Pb/L
EC50 for growth:
4,210 ug Pb/L



72-hour LC50:
4,690 ug Pb/L
96-hour LC50:
830 ug Pb/L
120-hour LC50:
680 ug Pb/L
144-hour LC50:
680 ug Pb/L

Reference3
(Published
since the
2006 Pb
AQCD)
Vlahogianni
and Valavanidis
(2007)





Vlahogianni
and Valavanidis
(2007)









Sobrino-
Figueroa et al.
(2007)











November 2012
7-247
Draft - Do Not Cite or Quote

-------




Species
Bivalve
(Tellina deltoid alls)















Exposure
Concentration Method
1 ,000 mg Pb/kg 1 0 day direct
(analytically exposure to
verified) Pb nitrate spiked
sediment
collected from
Woronora River,
New South
Wales, Australia.
Adults used in
the test were
collected from
Lane Cove
River, Sydney,
Australia and
held in filtered
seawater.



Modifying
factors
Temperature:
21 ± 1 °C
Salinity
30 ± 1%,
PH
7.2-8.2
Ammonia (total)
<3 mg N/L











Effects on Endpoint Effect Concentration
Survival: 10DayNOEC
All individuals survived. Porewater: 1 5 ug Pb/L
dissolved Pb
Sediment: 1 ,000 mg Pb/kg











Reference3
(Published
since the
2006 Pb
AQCD)
King et al.
(201 0)












Vertebrates
Toadfish
(Tetractenos glaber)






Measured but not Field-collected
reported fish-Pb sampled
from sediments
collected in two
reference and
two metal
contaminated
estuaries near
Sydney,
Australia



Temperature :
15 to 16°C
Salinity:
29 to 31 %,
pH:
8.4 to 8.6,
amongst
estuaries.
Sediment pH:
7.0-7.5
Organic matter:
1.5-2.1%
Reproduction: N/A
Decreased oocyte diameter and
density in the toadfish were
associated with elevated levels of
Pb in the gonads of field collected
fish; authors state that this is
suggestive of a reduction in egg
size.




Alquezaret al.
(2006)






November 2012
7-248
Draft - Do Not Cite or Quote

-------


Species
Grunt fish
(Haemulon scudderi)











Concentration
Mean total Pb
body burden
increased from
0.55 to 3.32 mg
Pb/kg during the
feeding
experiment.









Exposure
Method
42-day dietary
exposure from
simulated marine
food chain-
shrimp,
L. vannamei, fed
A franciscana
(mean Pb
content
12-1 5 mg Pb/kg)
feeding on
T. suecica
cultured in
20 ug/L Pb as
Pb nitrate.

Modifying
factors
Temperature:
28 ± 2 °C
Salinity:
34.6 ± 1 .2 ppt
nU-l- 7 Q R O
pn. /.y-o.z
DO saturation:
90-95%
(>7 mg/L)








Effects on Endpoint
Survival:
No significant differences
observed in intermediate and
final length, mean wet weight, or
Fulton's condition factor: Final
survival significantly lower; mean
total Pb body burden increased.










Effect


Concentration
Mean survival in the fish fed
the experimental diet ranged
from 65 to 75% compared to
control diet survival (88 to
91 %).


















Reference3
(Published
since the
2006 Pb
AQCD)
Soto-Jimenez
et al. (201 1 b)









References included are those which were published since the 2006 Pb AQCD.
November 2012
7-249
Draft - Do Not Cite or Quote

-------
References for Chapter 7
     Acosta. JA; Jansen. B; Kalbitz. K; Faz. A; Martinez-Martinez. S. (2011). Salinity increases mobility of
        heavy metals in soils. Chemosphere 85: 1318-1324.
        http://dx.doi.0rg/10.1016/i.chemosphere.2011.07.046

     Adevemo.  OK. (2007). Haematological profile of Clarias gariepinus (Burchell, 1822) exposed to lead.
        Turkish Journal of Fisheries and Aquatic Sciences 7: 163-169.

     Adevemo.  OK. (2008a). Histological alterations observed in the gills and ovaries of Clarias gariepinus
        exposed to environmentally relevant lead concentrations. J Environ Health 70: 48-51.

     Adevemo.  OK. (2008b). Histological alterations observed in the liver and brain of Clarias gariepinus
        exposed to chronic sublethal dose of lead. Bulletin of the European Association of Fish Pathologists
        28: 105-114.

     Ahlf. W: Drost. W: Heise. S. (2009). Incorporation of metal bioavailability into regulatory frameworks:
        Metal exposure in water and sediment. Journal of Soils and Sediments 9: 411-419.
        http://dx.doi.org/10.1007/sll368-009-0109-6

     Ahmad. MS: Ashraf. M; Tabassam. Q; Hussain. M; Firdous. H. (2011). Lead (Pb)-induced regulation of
        growth, photosynthesis, and mineral nutrition in maize (Zea mays L.) plants at early growth stages.
        Biol Trace Elem Res 144: 1229-1239. http://dx.doi.org/10.1007/sl2011-011-9099-5

     Ai-Khlaifat. AL; Al-Khashman. OA. (2007). Atmospheric heavy metal pollution in Aqaba City, Jordan,
        using Phoenix dactylifera L. leaves. Atmos Environ 41: 8891-8897.
        http://dx.doi.0rg/10.1016/i.atmosenv.2007.08.028

     Aisemberg. J; Nahabedian. DE; Wider. EA; Verrengia Guerrero. NR. (2005). Comparative study on two
        freshwater invertebrates for monitoring environmental lead exposure. Toxicology 210: 45-53.
        http://dx.doi.0rg/10.1016/i.tox.2005.01.005

     Akerblom. S: Baath. E; Bringmark. L; Bringmark. E. (2007). Experimentally induced effects of heavy
        metal on microbial activity and community structure of forest mor layers. Biol Fertil Soils 44: 79-91.
        http://dx.doi.org/10.1007/s00374-007-0181-2

     Albrecht. J: Abalos. M; Rice. TM. (2007). Heavy metal levels in ribbon snakes (Thamnophis sauritus)
        and anuran larvae from the Mobile-Tensaw River Delta, Alabama, USA. Arch Environ Contam
        Toxicol 53: 647-654. http://dx.doi.org/10.1007/s00244-006-0175-3

     Aldenberg. T; Slob. W. (1993). Confidence limits for hazardous concentrations based on logistically
        distributed NOEC toxicity data. Ecotoxicol Environ Saf 25: 48-63.
        http://dx.doi.org/10.1006/eesa.1993.1006

     Aldjain. I.. M.: Al-Whaibi. M.. H.; Al-Showiman. S.. S.: Siddiqui. M.. H. (2011). Determination of
        heavy metals in the fruit of date palm growing at different locations of Riyadh. Saudi J Biol Sci 18:
        175-180. http://dx.doi.0rg/10.1016/i.sibs.2010.12.001

     Allert. AL; Fairchild. JF; DiStefano. RJ; Schmitt. CJ: Brumbaugh. WG: Besser. JM. (2009a). Ecological
        effects of lead mining on Ozark streams: In-situ toxicity to woodland crayfish (Orconectes hylas).
        Ecotoxicol Environ Saf 72: 1207-1219. http://dx.doi.Org/10.1016/i.ecoenv.2008.08.005

     Allert. AL; Fairchild. JF; Schmitt. CJ: Besser. JM; Brumbaugh. WG: Olson. SJ. (2009b). Effects of
        mining-derived metals on riffle-dwelling benthic fishes in Southeast Missouri, USA. Ecotoxicol
        Environ Saf 72: 1642-1651. http://dx.doi.Org/10.1016/i.ecoenv.2009.02.014

     Alonso-Azcarate. J: Ruiz. E; Rodriguez. L. (2011).  Changes in heavy metal speciation induced by
        earthworm feeding activity. Fresen Environ Bull 20: 26-35.
November 2012                             7-250                   Draft - Do Not Cite or Quote

-------
     Alonso-Castro. AJ; Carranza-Alvarez. C: Alfaro-De la Torre. MC: Chavez-Guerrero. L; Garcia-De la
        Cruz. RF. (2009). Removal and accumulation of cadmium and lead by Typha latifolia exposed to
        single and mixed metal solutions. Arch Environ Contam Toxicol 57: 688-696.
        http://dx.doi.org/10.1007/s00244-009-9351-6

     Alquezar. R; Markich. SJ: Booth. DJ. (2006). Effects of metals on condition and reproductive output of
        the smooth toadfish in Sydney estuaries, south-eastern Australia. Environ Pollut 142: 116-122.
        http://dx.doi.0rg/10.1016/i.envpol.2005.09.009

     Alves Costa. JRM; Mela. M; da Silva de Assis. HC: Pelletier. E; Randi. MAP; de Oliveira Ribeiro. CA.
        (2007). Enzymatic inhibition and morphological changes in Hoplias malabaricus from dietary
        exposure to lead(II) or methylmercury. Ecotoxicol Environ Saf 67: 82-88.
        http://dx.doi.0rg/10.1016/i.ecoenv.2006.03.013

     Alves. LC: Glover. CN; Wood. CM. (2006). Dietary Pb accumulation in juvenile freshwater rainbow trout
        (Oncorhynchus mykiss). Arch Environ Contam Toxicol 51: 615-625.
        http://dx.doi.org/10.1007/s00244-005-0212-7

     Alves. LC: Wood. CM. (2006). The chronic effects of dietary lead in freshwater juvenile rainbow trout
        (Oncorhynchus mykiss) fed elevated calcium diets. Aquat Toxicol 78: 217-232.
        http://dx.doi.0rg/10.1016/i.aquatox.2006.03.005

     Amado. EM; Freire. CA; Souza. MM. (2006). Osmoregulation and tissue water regulation in the
        freshwater red crab Dilocarcinus pagei (Crustacea, Decapoda), and the effect of waterborne inorganic
        lead. Aquat Toxicol 79: 1-8. http://dx.doi.Org/10.1016/i.aquatox.2006.04.003

     Ancion. PY; Lear. G: Lewis. GD. (2010). Three common metal contaminants of urban runoff (Zn, Cu &
        Pb)  accumulate in freshwater biofilm and modify embedded bacterial communities. Environ Pollut
        158: 2738-2745. http://dx.doi.Org/10.1016/i.envpol.2010.04.013

     Anderson. MB; Preslan. JE; Jolibois. L; Bollinger. JE; George. WJ. (1997). Bioaccumulation of lead
        nitrate in red swamp crayfish (Procambrus clarkii). J Hazard Mater 54: 15-29.
        http://dx.doi.org/10.1016/S0304-3894(96)01852-3

     Anderson. WL; Havera. SP; Zercher. BW. (2000). Ingestion of lead and nontoxic shotgun pellets by ducks
        in the Mississippi flyway. J Wildl Manag 64: 848-857.

     Angelo. RT; Cringan. MS: Chamberlain. PL; Stahl. AJ: Haslouer. SG: Goodrich. CA. (2007). Residual
        effects of lead and zinc mining on freshwater mussels in the Spring River Basin (Kansas, Missouri,
        and  Oklahoma, USA). Sci Total Environ 384: 467-496.
        http://dx.doi.0rg/10.1016/i.scitotenv.2007.05.045

     Angelova. VR; Ivanova. RV: Todorov. JM; Ivanov. KI. (2010). Lead,  cadmium, zinc, and copper
        bioavailability in the soil-plant-animal system in a polluted area. ScientificWorldJournal 10: 273-285.
        http://dx.doi.org/10.1100/tsw.2010.33

     Anklev. GT; Bennett. RS: Erickson. RJ; Hoff. DJ: Hornung.  MW: Johnson. RD; Mount. PR; Nichols.
        JW: Russom. CL; Schmieder. PK; Serrrano. JA; Tietge. JE: Villeneuve. PL. (2010). Adverse outcome
        pathways: a conceptual framework to support ecotoxicology research and risk assessment [Review].
        Environ Toxicol Chem 29: 730-741. http://dx.doi.org/10.1002/etc.34

     Anklev. GT; Di Toro. DM; Hansen. DJ: Berry. WJ. (1996). Assessing the ecological risk of metals in
        sediments [Editorial]. Environ Toxicol Chem 15: 2053-2055.
        http://dx.doi.org/10.1002/etc.562Q151201

     Ansaldo. M; Nahabedian. DE; Di Fonzo. C: Wider. EA. (2009). Effect of cadmium, lead and arsenic on
        the oviposition, hatching and embryonic survival of Biomphalaria glabrata. Sci Total Environ 407:
        1923-1928.  http://dx.doi.0rg/10.1016/i.scitotenv.2008.12.001

     Ansaldo. M; Nahabedian. DE; Holmes-Brown. E; Agote. M; Ansav. CV: Guerrero. NR; Wider. EA.
        (2006). Potential use of glycogen level as biomarker of chemical stress in Biomphalaria glabrata.
        Toxicology 224: 119-127. http://dx.doi.Org/10.1016/i.tox.2006.04.037
November 2012                             7-251                  Draft - Do Not Cite or Quote

-------
     Antonious. G: Kochhar. T. (2009). Mobility of heavy metals from soil into hot pepper fruits: Afield study.
        Bull Environ Contam Toxicol 82: 59-63. http://dx.doi.org/10.1007/s00128-008-9512-8

     Antosiewicz. DM. (2005). Study of calcium-dependent lead-tolerance on plants differing in their level of
        Ca-deficiency tolerance. Environ Pollut 134: 23-34. http://dx.doi.Org/10.1016/i.envpol.2004.07.019

     Antunes. PM; Berkelaar. EJ; Boyle. D; Hale. BA; Hendershot. W: Voigt A. (2006). The biotic ligand
        model for plants and metals: Technical challenges for field application. Environ Toxicol Chem 25:
        875-882. http://dx.doi.Org/10.1897/04-586R.l

     Arai. T; Maeda. M; Yamakawa. H; Kamatani. A; Miyazaki. N. (2002). Growth effect on the uptake and
        elimination of trace metals in the abalones Haliotis. Fish Sci 68: 1094-1098.
        http://dx.doi.0rg/10.1046/i.1444-2906.2002.00537.x

     Arias-Almeida. JC: Rico-Martinez. R. (2011). Toxicity of cadmium, lead, Mercury and methyl parathion
        on euchlanis dilatata ehrenberg 1832 (Rotifera: Monogononta). Bull Environ Contam Toxicol 87:
        138-142. http://dx.doi.org/10.1007/s00128-011-0308-x

     Arunakumara. K; Zhang. X.  (2008). Heavy metal bioaccumulation and toxicity with special reference to
        microalgae. J Ocean Univ China 7: 60-64. http://dx.doi.org/10.1007/sll802-008-0060-y

     Arunakumara. K; Zhang. X;  Song. X. (2008). Bioaccumulation of Pb(2+) and its effects on growth,
        morphology and pigment contents of Spirulina (Arthrospira) platensis. J Ocean Univ China 7:
        397-403. http://dx.doi.org/10.1007/sll802-008-0397-2

     Ashraf. MY; Azhar. N; Ashraf. M; Hussain. M; Arshad. M. (2011). Influence of lead on growth and
        nutrient accumulation in canola (Brassica napus L.) cultivars. J Environ Biol 32: 659-666.

     Atici. T; Katircioglu. H; Akin. B. (2008). Sensitivity of freshwater microalgal strains (Chlorella vulgaris
        Beijerinck and Scenedesmus obliquus (Turpin) Kutzing) to heavy metals. Fresen Environ Bull 17:
        268-274.

     Atkinson. CA; Jollev. DF; Simpson. SL. (2007). Effect of overlying water pH, dissolved oxygen, salinity
        and sediment disturbances on metal release and sequestration from metal contaminated marine
        sediments. Chemosphere 69: 1428-1437. http://dx.doi.Org/10.1016/i.chemosphere.2007.04.068

     Atli. G: Canli. M. (2007). Enzymatic responses to metal exposures in a freshwater fish Oreochromis
        niloticus. Comp Biochem Physiol C Toxicol Pharmacol 145: 282-287.
        http://dx.doi.0rg/10.1016/i.cbpc.2006.12.012

     Atli. G: Canli. M. (2008). Responses of metallothionein and reduced glutathione in a freshwater fish
        Oreochromis niloticus following metal exposures. Environ Toxicol Pharmacol 25: 33-38.
        http://dx.doi.0rg/10.1016/i.etap.2007.08.007

     ATSDR (Agency for Toxic Substances and Disease Registry). (2004). Guidance manual for the
        assessment of joint toxic  action of chemical mixtures. Atlanta, GA: U.S. Department of Health and
        Human Services, Agency for Toxic Substances and Disease Registry.
        http://www.atsdr.cdc.gov/interactionproFiles/IP-ga/ipga.pdf

     Ayensu. E; van Claasen. PR; Collins. M; Dearing. A; Fresco. L; Gadgil. M; Gitav. H; Glaser. G: Juma. C:
        Krebs. J: Lenton. R; Lubchenco. J: McNeely. JA;  Moonev. HA; Pinstrup-Andersen. P; Ramos. M;
        Raven. P; Reid. WV: Samper. C: Sarukhan. J:  Schei. P; Tundisi. JG: Watson. RT; Guanhua. X; Zakri.
        AH. (1999). International ecosystem assessment. Science 286: 685-686.
        http://dx.doi.org/10.1126/science.286.5440.685

     Aznar. JC: Richer-Lafleche. M; Begin. C: Rodrigue. R. (2008a). Spatiotemporal reconstruction of lead
        contamination using tree rings  and organic soil layers. Sci Total Environ 407:  233-241.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.09.044

     Aznar. JC: Richer-Laflechea. M; Begin. C: Begin. Y.  (2009a). Lead exclusion and copper translocation in
        black spruce needles. Water Air Soil Pollut 203: 139-145.
        http://dx.doi.org/10.1007/sll270-009-9997-8
November 2012                              7-252                   Draft - Do Not Cite or Quote

-------
     Aznar. JC: Richer-Laflechea. M; Paucar-Munoza. H; Bordeleaua. M; Begina. Y. (2009b). Is tree growth
        reduction related to direct foliar injuries or soil chemistry modifications? Chemosphere 76:
        1366-1371. http://dx.doi.0rg/10.1016/i.chemosphere.2009.06.023

     Baath. E. (1989). Effects of heavy metals in soil on microbial processes and populations (a review)
        [Review]. Water Air Soil Pollut 47: 335-379. http://dx.doi.org/10.1007/BF00279331

     Baath. E; Diaz-Ravina. M; Bakken. LR. (2005). Microbial biomass, community structure and metal
        tolerance of a naturally Pb-enriched forest soil. Microb Ecol 50: 496-505.
        http://dx.doi.org/10.1007/s00248-005-0008-3

     Balistrieri. LS: Blank. RG. (2008). Dissolved and labile concentrations of Cd, Cu, Pb, and Zn in the South
        Fork Coeur d'Alene River, Idaho: Comparisons among chemical equilibrium models and implications
        for biotic ligand models. Appl Geochem 23: 3355-3371.
        http://dx.doi.0rg/10.1016/i.apgeochem.2008.06.031

     Bamborough. L; Cummings. SP (2009).  The impact of zinc and lead concentrations and seasonal
        variation on bacterial and actinobacterial community structure in a metallophytic grassland soil. Folia
        Microbiol  (Praha) 54: 327-334. http://dx.doi.org/10.1007/sl2223-009-0042-5

     Baos. R; Bias. J: Bortolotti. GR: Marchant. TA; Hiraldo. F. (2006). Adrenocortical response to stress and
        thyroid hormone status in free-living nestling white storks (Ciconia ciconia) exposed to heavy metal
        and arsenic contamination. Environ Health Perspect 114:  1497-1501.
        http://dx.doi.org/10.1289/ehp.9099

     Barnthouse. LW: Munns. RM. Jr; Sorensen. MT. (2008). Population-level ecological risk assessment.
        Pensacola, FL: Taylor & Francis, http://www.crcpress.com/product/isbn/9781420053326

     Bartell. SM. (2007). Ecosystem effects modeling. In GW Suterll (Ed.), Ecological risk assessment (2nd
        ed., pp. 413-432). Boca Raton, FL: CRC Press.

     Baumann. HA; Morrison. L; Stengel. DB. (2009). Metal accumulation and toxicity measured by PAM-
        Chlorophyll fluorescence in seven species of marine macroalgae. Ecotoxicol Environ Saf 72:
        1063-1075. http://dx.doi.0rg/10.1016/i.ecoenv.2008.10.010

     Bazar. MA; Quinn. MJ; Mozzachio. K; Bleiler. JA; Archer. CR; Phillips. CT; Johnson. MS. (2010).
        Toxicological responses of red-backed salamander (plethodon cinereus) exposed to aged and amended
        soils containing lead. Arch Environ Contam Toxicol 58: 1040-1047.
        http://dx.doi.org/10.1007/s00244-010-9471-z

     Bechard. KM; Gillis. PL; Wood. CM. (2008). Acute toxicity of waterborne Cd, Cu, Pb, Ni, and Zn to
        first-instar Chironomus riparius larvae. Arch Environ Contam Toxicol 54: 454-459.
        http://dx.doi.org/10.1007/s00244-007-9048-7

     Beebv. A; Richmond. L. (2010). Magnesium and the regulation of lead in three populations of the garden
        snail Cantareus aspersus. Environ Pollut 158: 2288-2293.
        http://dx.doi.0rg/10.1016/i.envpol.2010.02.002

     Beiras. R; Albentosa. M. (2003). Inhibition of embryo development of the commercial bivalves Ruditapes
        decussatus and Mytilus galloprovincialis by trace metals; implications for the implementation of
        seawater quality criteria. Aquaculture 230:  205-213.
        http://dx.doi.org/10.1016/S0044-8486(03)00432-0

     Bekesiova. B; S. H; Libantova.  J: Moravcikova. J: Matusikova. I. (2008). Heavy-metal stress induced
        accumulation of chitinase isoforms in plants. Mol Biol Rep 35: 579-588.
        http://dx.doi.org/10.1007/sll033-007-9127-x

     Bellrose. FC. (1959). Lead poisoning as a mortality  factor in waterfowl populations. Bull 111 Nat Hist
        Surv 27: 235-288.

     Benbrahim. M; Denaix. L; Thomas. AL;  Balet. J:  Carnus. JM. (2006). Metal concentrations in edible
        mushrooms following municipal sludge application on forest land. Environ Pollut 144:  847-854.
        http://dx.doi.0rg/10.1016/i.envpol.2006.02.014
November 2012                              7-253                   Draft - Do Not Cite or Quote

-------
     Bengtsson. G: Nordstrom. S: Rundgren. S. (1983). Population density and tissue metal concentration of
        lumbricids in forest soils near a brass mill. Environ Pollut Ecol Biol 30: 87-108.
        http://dx.doi.org/10.1016/0143-1471(83)90007-7

     Sergey. LL; Weis. JS. (2007). Molting as a mechanism of depuration of metals in the fiddler crab, Uca
        pugnax. Mar Environ Res 64: 556-562. http://dx.doi.Org/10.1016/i.marenvres.2007.04.009

     Berglund. AMM; Ingvarsson. PK; Danielsson. H; Nyholm. NEI. (2010). Lead exposure and biological
        effects in pied flycatchers (Ficedula hypoleuca) before and after the closure of a lead mine in northern
        Sweden. Environ Pollut 158: 1368-1375. http://dx.doi.Org/10.1016/j.envpol.2010.01.005

     Besser. JA; Brumbaugh. WG: Hardestv. DK; Hughes. JP; Ingersoll. CG. (2009). Assessment of metal-
        contaminated sediments from the Southeast Missouri (SEMO) mining district using sediment toxicity
        tests with amphipods and freshwater mussels. (08-NRDAR-02). Columbia, MO: U.S. Geological
        Survey, Columbia Environmental Research Center.

     Besser. JM; Brumbaugh. WG: Brunson. EL; Ingersoll. CG. (2005). Acute and chronic toxicity of lead in
        water and diet to the amphipod Hyalella azteca. Environ Toxicol Chem 24:  1807-1815.
        http://dx.doi.Org/10.1897/04-480R.l

     Besser. JM; Brumbaugh. WG: Kemble. NE; May. TW: Ingersoll. CG. (2004). Effects of sediment
        characteristics on the toxicity of chromium(III) and chromium(VI) to the amphipod, Hyalella azteca.
        Environ Sci Technol 38: 6210-6216. http://dx.doi.org/10.1021/es049715i

     Besser. JM: Brumbaugh. WG: May. TW: Schmitt. CJ. (2007). Biomonitoring of lead, zinc,  and cadmium
        in streams draining lead-mining and non-mining areas, Southeast Missouri, USA. Environ Monit
        Assess 129: 227-241. http://dx.doi.org/10.1007/sl0661-006-9356-9

     Beyer. WN; Dalgarn. J; Pudding. S: French. JB; Mateo. R; Miesner.  J; Sileo. L; Spann. J. (2004). Zinc
        and lead poisoning in wild birds in the Tri-State Mining District (Oklahoma, Kansas, and Missouri).
        Arch Environ Contam Toxicol 48: 108-117. http://dx.doi.org/10.1007/s00244-004-0010-7

     Beyer. WN; Pattee. OH; Sileo. L; Hoffman. DJ; Mulhern. BM. (1985). Metal contamination in wildlife
        living near two zinc smelters. Environ Pollut Ecol Biol 38: 63-86.
        http://dx.doi.org/10.1016/0143-1471(85)90094-7

     Bi. XY: Feng. XB: Yang. YG: Li. XD: Sin. GPY:  Oiu. GL: Oian. XL: Li. FL: He. TR: Li. P: Liu. TZ: Fu.
        ZY. (2007). Heavy metals in an impacted wetland system: Atypical case from Southwestern China.
        Sci Total Environ 387: 257-268. http://dx.doi.Org/10.1016/j.scitotenv.2007.07.059

     Bindler. R; Renberg. I; Klaminder. J. (2008). Bridging the gap between ancient metal pollution and
        contemporary biogeochemistry. J Paleolimnol 40: 755-770.
        http://dx.doi.org/10.1007/sl0933-008-9208-4

     Birceanu. O; Chowdhury. MJ; Gillis. PL; McGeer. JC: Wood. CM; Wilkie. MP (2008). Modes of metal
        toxicity and impaired branchial ionoregulation in rainbow trout exposed to mixtures of Pb and Cd in
        soft water. Aquat Toxicol 89: 222-231. http://dx.doi.Org/10.1016/i.aquatox.2008.07.007

     Bojarczuk. K; Kieliszewska-Rokicka. B. (2010). Effect of ectomycorrhiza on Cu  and Pb accumulation in
        leaves and roots of silver birch (Betula pendula Roth.) seedlings grown in metal-contaminated soil.
        Water Air Soil Pollut 207: 227-240. http://dx.doi.org/10.1007/sll270-009-0131-8

     Bonanno. G: Lo Giudice. R. (2010). Heavy metal bioaccumulation by the organs  of Phragmites australis
        (common reed) and their potential use as contamination indicators. Ecol Indicat 10: 639-645.
        http://dx.doi.0rg/10.1016/i.ecolind.2009.ll.002

     Borgmann. U; Couillard. Y; Doyle. P; Dixon. DG. (2005). Toxicity of sixty-three metals and metalloids to
        Hyalella azteca at two levels of water hardness. Environ Toxicol Chem 24: 641-652.
        http://dx.doi.Org/10.1897/04-177R.l

     Borgmann. U; Couillard. Y; Grapentine. LC. (2007). Relative contribution of food and water to 27 metals
        and metalloids accumulated by caged Hyalella azteca in two rivers affected by metal mining. Environ
        Pollut 145: 753-765. http://dx.doi.Org/10.1016/i.envpol.2006.05.020
November 2012                             7-254                   Draft - Do Not Cite or Quote

-------
     Borgmann. U; Kramar. O; Loveridge. C. (1978). Rates of mortality, growth, and biomass production of
        Lymnaea palustris during chronic exposure to lead. J Fish Res Board Can 35: 1109-1115.
        http://dx.doi.org/10.1139/f78-175

     Boughriet. A; Proix. N; Billon. G: Recourt. P; Ouddane. B. (2007). Environmental impacts of heavy
        metal discharges from a smelter in Deule-canal sediments (Northern France): Concentration levels and
        chemical fractionation. Water Air Soil Pollut 180: 83-95. http://dx.doi.org/10.1007/sll270-006-9252-5

     Boyle. D; Amlund. H; Lundebve. AK; Hogstrand. C: Bury. NR. (2010). Bioavailability of a natural lead-
        contaminated invertebrate diet to zebrafish. Environ Toxicol Chem 29: 708-714.
        http://dx.doi.org/10.1002/etc.61

     Bradham. KD; Dayton. EA; Basta. NT; Schroder. J; Pavton. M; Lanno. RP. (2006). Effect of soil
        properties on lead bioavailability and toxicity to earthworms. Environ Toxicol Chem  25: 769-775.
        http://dx.doi.Org/10.1897/04-552R.l

     Brix. KV; DeForest. DK; Burger. M; Adams. WJ. (2005). Assessing the relative sensitivity of aquatic
        organisms to divalent metals and their representation in toxicity datasets compared to natural aquatic
        communities. Hum Ecol Risk Assess 11: 1139-1156. http://dx.doi.org/10.1080/10807030500278628

     Brown. CS: Luebbert J: Mulcahy. D; Schamber. J: Rosenberg. PH. (2006). Blood lead levels of wild
        Steller's eiders  (Polysticta stelleri) and black scoters (Melanitta nigra) in Alaska using a portable blood
        lead analyzer. J Zoo Wildl Med 37: 361-365. http://dx.doi.Org/10.1638/05-092.l

     Brumbaugh. WG:  Schmitt. CJ: May. TW. (2005). Concentrations of cadmium, lead, and  zinc in fish from
        mining-influenced waters of northeastern Oklahoma: Sampling of blood, carcass, and liver for aquatic
        biomonitoring. Arch Environ Contam Toxicol 49: 76-88. http://dx.doi.org/10.1007/s00244-004-0172-3

     Bryan. CE; Christopher. SJ: Balmer. BC: Wells. RS. (2007). Establishing baseline levels of trace elements
        in blood and skin of bottlenose dolphins in Sarasota Bay,  Florida: Implications for non-invasive
        monitoring. Sci Total Environ 388: 325-342. http://dx.doi.Org/10.1016/i.scitotenv.2007.07.046

     Buekers. J: Redeker. ES:  Smolders. E. (2009). Lead toxicity to wildlife: Derivation of a critical blood
        concentration for wildlife monitoring based on literature data [Review]. Sci Total Environ 407:
        3431-3438. http://dx.doi.Org/10.1016/i.scitotenv.2009.01.044

     Burger. J: Campbell. KR; Murray. S: Campbell. TS: Gaines. KF; Jeitner. C: Shukla.  T; Burke. S:
        Gochfeld. M. (2007). Metal levels in blood, muscle and liver of water snakes (Nerodia spp.) from
        New Jersey, Tennessee and South Carolina.  Sci Total Environ 373: 556-563.
        http://dx.doi.0rg/10.1016/i.scitotenv.2006.06.018

     Burger. J: Gochfeld. M. (2005).  Effects  of lead on learning in herring  gulls: An avian wildlife model for
        neurobehavioral deficits. Neurotoxicology 26: 615-624. http://dx.doi.Org/10.1016/i.neuro.2005.01.005

     Burger. J: Murray. S: Gaines. KF; Novak. JM; Punshon. T; Dixon. C:  Gochfeld. M. (2006). Element
        levels in snakes in South Carolina: Differences between a control site and exposed site on the
        savannah river site. Environ  Monit Assess 112:  35-52. http://dx.doi.org/10.1007/sl0661-006-0695-3

     Businelli. D: Onofri. A; Massaccesi. L.  (2011). Factors involved in uptake of lead by some edible crops
        grown in agricultural soils of central Italy. Soil Sci 176: 472-478.
        http://dx.doi.org/10.1097/SS.Ob013e31822769fa

     Bustamante. P; Bertrand. M; Boucaud-Camou. E; Miramand. P. (2006). Subcellular distribution of Ag,
        Cd, Co, Cu, Fe, Mn, Pb, and Zn in the digestive gland of the common cuttlefish Sepia officinalis.
        Journal of Shellfish Research 25: 987-993.

     Bustamante. P: Miramand. P. (2005). Subcellular and body distributions of 17 trace elements in the
        variegated scallop Chlamys variafrom the French coast of the Bay of Biscay. Sci Total Environ 337:
        59-73. http://dx.doi.0rg/10.1016/j.scitotenv.2004.07.004

     Cade. TJ. (2007). Exposure of California condors to lead from spent ammunition. J Wildl Manag 71:
        2125-2133. http://dx.doi.org/10.2193/2007-084
November 2012                              7-255                  Draft - Do Not Cite or Quote

-------
     Caetano. M; Fonseca. N; Carlos Vale. RC. (2007). Mobility of Pb in salt marshes recorded by total
        content and stable isotopic signature. Sci Total Environ 380: 84-92.
        http://dx.doi.0rg/10.1016/i.scitotenv.2006.ll.026

     Camara Pellisso. S: Mufioz. MJ; Carballo. M; Sanchez-Vizcaino. JM. (2008). Determination of the
        immunotoxic potential of heavy metals on the functional activity of bottlenose dolphin leukocytes in
        vitro. Vet Immunol Immunopathol 121: 189-198. http://dx.doi.Org/10.1016/i.vetimm.2007.09.009

     Campos. JA; Tejera. NA. (2011). Bioconcentration factors and trace elements bioaccumulation in
        sporocarps of fungi collected from quartzite acidic soils. Biol Trace Elem Res 143: 540-554.
        http://dx.doi.org/10.1007/sl2011-010-8853-4

     Carine. F; Enrique. AG: Steven. C. (2009). Metal effects on phenol oxidase activities of soils. Ecotoxicol
        Environ Saf 72: 108-114. http://dx.doi.Org/10.1016/i.ecoenv.2008.03.008

     Carreras. HA; Pignata. ML. (2007). Effects of the heavy metals Cu2+, Ni2+, Pb2+, and Zn2+ on some
        physiological parameters of the lichen Usnea amblyoclada. Ecotoxicol Environ Saf 67: 59-66.
        http://dx.doi.0rg/10.1016/i.ecoenv.2006.05.005

     Carvalho. F; Oliveira. J: Alberto. G. (2011). Factors affecting 210Po and 210Pb activity concentrations in
        mussels and implications for environmental bio-monitoring programmes. J Environ Radioact 102:
        128-137. http://dx.doi.0rg/10.1016/i.jenvrad.2010.ll.003

     Casas. S: Gonzalez. JL; Andral. B; Cossa. D. (2008). Relation between metal concentration in water and
        metal content of marine mussels (Mytilus galloprovincialis): Impact of physiology. Environ Toxicol
        Chem 27: 1543-1552. http://dx.doi.Org/10.1897/07-418.l

     Casteel. SW: Weis. CP; Henningsen.  GM; Brattin. WL. (2006). Estimation of relative bioavailability of
        lead in soil and soil-like materials using young swine. Environ Health Perspect  114: 1162-1171.
        http://dx.doi.org/10.1289/ehp.8852

     Cenkci. S: Cioerci. IH; Yildiz. M; Oezav. C: Bozdao. A; Terzi. H. (2010). Lead contamination reduces
        chlorophyll biosynthesis and genomic template stability in Brassica rapa L. Environ Exp Bot 67:
        467-473. http://dx.doi.0rg/10.1016/i.envexpbot.2009.10.001

     Chadwick. EA; Simpson. VR; Nicholls. AE; Slater. FM. (2011). Lead levels in Eurasian otters decline
        with time and reveal interactions between sources, prevailing weather, and stream chemistry. Environ
        SciTechnol45: 1911-1916. http://dx.doi.org/10.1021/esl034602

     Chapin. FS. Ill; Sala. OE; Burke. 1C: Grime. JP; Hooper. DU; Lauenroth. WK; Lombard. A; Moonev.
        HA; Mosier. AR; Naeem.  S: Pacala. SW: Roy. J: Steffen. WL; Tilman. D. (1998). Ecosystem
        consequences of changing biodiversity. Bioscience 48: 45-52.

     Chapman. E.. E.; Dave. G..: Murimboh. J.. D. (2010). Ecotoxicological risk assessment of undisturbed
        metal contaminated soil at two remote lighthouse sites. Ecotoxicol Environ Saf 73: 961-969.
        http://dx.doi.0rg/10.1016/i.ecoenv.2010.02.014

     Chaube. R; Mishra. S: Singh. RK. (2010). In vitro effects of lead nitrate on steroid  profiles in the post-
        vitellogenic ovary of the catfish Heteropneustes fossilis. Toxicol In Vitro 24:  1899-1904.
        http://dx.doi.0rg/10.1016/i.tiv.2010.07.021

     Chen. TH; Gross. JA; Karasov. WH. (2006b).  Sublethal effects of lead on northern leopard frog (Rana
        pipiens) tadpoles. Environ Toxicol Chem 25: 1383-1389. http://dx.doi.Org/10.1897/05-356R.l

     Chen. WP: Li. LQ; Chang. AC: Wu. LS: Chanev. RL; Smith. R; Aiwa. H. (2009b). Characterizing the
        solid-solution partitioning coefficient and plant uptake factor of As, Cd, and Pb in California
        croplands. Agric Ecosyst Environ 129: 212-220. http://dx.doi.Org/10.1016/i.agee.2008.09.001

     Chen. ZZ; Zhu. L; Wilkinson. KJ. (2010c). Validation of the biotic ligand model in metal mixtures:
        Bioaccumulation of lead and copper. Environ Sci Technol 44: 3580-3586.
        http://dx.doi.org/10.1021/esl003457
November 2012                              7-256                  Draft - Do Not Cite or Quote

-------
     Chiesa. ME; Rosenberg. CE; Fink. NE; Salibian. A. (2006). Serum protein profile and blood cell counts
        in adult toads Bufo arenarum (Amphibia: Anura: Bufonidae): Effects of sublethal lead acetate. Arch
        Environ Contam Toxicol 50: 384-391. http://dx.doi.org/10.1007/s00244-004-0252-4

     Cho. YS: Bolick. JA; Butcher. DJ. (2009). Phytoremediation of lead with green onions (Allium
        fistulosum) and uptake of arsenic compounds by moonlight ferns (Pteris cretica cv Mayii). Microchem
        J 91: 6-8. http://dx.doi.0rg/10.1016/i.microc.2008.05.008

     Chrastny. V: Komarek. M; Haiek. T. (2010). Lead contamination of an agricultural soil in the vicinity of a
        shooting range.  Environ Monit Assess 162: 37-46. http://dx.doi.org/10.1007/sl0661-009-0774-3

     Church. ME; Gwiazda. R; Risebrough. RW: Sorenson. K; Chamberlain. CP; Parry. S: Heinrich. W:
        Rideout. BA; Smith. DR. (2006). Ammunition is the principal source of lead accumulated by
        California  Condors re-introduced to the wild. Environ Sci Technol 40:  6143-6150.
        http://dx.doi.org/10.1021/es060765s

     Ciardullo. S: Aureli. F; Coni. E; Guandalini. E; losi. F; Raggi. A; Rufo. G: Cubadda. F. (2008).
        Bioaccumulation potential of dietary arsenic, cadmium, lead, mercury, and selenium in organs and
        tissues of rainbow trout (Oncorhyncus mykiss) as a function offish growth. J Agric Food Chem 56:
        2442-2451. http://dx.doi.org/10.1021/if703572t

     Cid. N; Ibanez. C: Palanques. A; Prat. N. (2010). Patterns of metal bioaccumulation in two filter-feeding
        macroinvertebrates: Exposure distribution, inter-species differences and variability across
        developmental stages. Sci Total Environ 408: 2795-2806.
        http://dx.doi.0rg/10.1016/i.scitotenv.2010.03.030

     Clark. HF; Brabander. DJ: Erdil. RM. (2006). Sources, sinks, and exposure pathways of lead in urban
        garden soil. J Environ Qual 35: 2066-2074. http://dx.doi.org/10.2134/ieq2005.0464

     Coeurdassier.  M; Scheifler. R; de Vaufleury. A; Crini. N; Saccomani. C: Du Mont. LS: Badot.  PM.
        (2007). Earthworms influence metal transfer from soil to snails. Appl Soil Ecol 35: 302-310.
        http://dx.doi.0rg/10.1016/i.apsoil.2006.08.004

     Comino. E; Menegatti. S: Fiorucci. A; Schwitzguebel. JP (2011). Accumulation and translocation
        capacity of As, Co, Cr and Pb by forage plants. Agrochimica 55: 105-115.

     Company. R; Serafim. A; Lopes. B; Cravo. A; Kalman. J: Riba. I; DelValls. TA; Blasco. J: Delgado. J:
        Sarmiento. AM; Nieto. JM; Shepherd. TJ; Nowell. G: Bebianno. MJ. (2011). Source and impact of
        lead contamination on delta-aminolevulinic acid dehydratase activity in several marine bivalve species
        along the Gulf of Cadiz. Aquat Toxicol 101: 146-154. http://dx.doi.Org/10.1016/i.aquatox.2010.09.012

     Conkova. M; Kubiznakova. J. (2008). Lead isotope ratios in tree bark pockets: An indicator of past air
        pollution in the Czech Republic. Sci Total Environ 404: 440-445.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.04.025

     Costanza. R: d'Arge. R: De Groot. R: Farber. S: Grasso. M; Hannon. B; Limburg. K; Naeem. S: O'Neill.
        RV; Paruelo. J: Raskin. RG: Sutton. P; Van Den Belt. M. (1997). The value of the world's ecosystem
        services and natural capital. Nature 387: 253-259. http://dx.doi.org/10.1038/387253aO

     Couture. RM; J-F. C: Auger. D; Claisse. D; Gobeil. C: Cossa. D. (2010). Seasonal and decadal variations
        in lead sources to eastern North Atlantic mussels. Environ Sci Technol 44: 1211-1216.
        http://dx.doi.org/10.1021/es902352z

     Cravo. A; Lopes. B; Serafim. A; Company. R; Barreira. L; Gomes. T; Bebianno. MJ. (2009). A
        multibiomarker approach in Mytilus galloprovincialis to assess environmental quality. J Environ
        Monit 11:  1673-1686. http://dx.doi.org/10.1039/b909846a

     Croisetiere. L; Hare. L; Tessier. A. (2006). Afield experiment to determine the relative importance of prey
        and water as sources of As, Cd, Co, Cu, Pb, and Zn for the aquatic invertebrate Sialis velata. Environ
        Sci Technol 40: 873-879. http://dx.doi.org/10.1021/es0516209
November 2012                              7-257                  Draft - Do Not Cite or Quote

-------
     Cui. S: Zhou. QX; Chao. L. (2007). Potential hyperaccumulation of Pb, Zn, Cu and Cd in endurant plants
        distributed in an old smeltery, northeast China. Environ Geol 51: 1043-1048.
        http://dx.doi.org/10.1007/s00254-006-0373-3

     Currie. M; Hodson. ME; Arnold. RE; Langdon. CJ. (2005). Single versus multiple occupancy: Effects on
        toxicity parameters measured on Eisenia fetida in lead nitrate-treated soil. Environ Toxicol Chem 24:
        110-116. http://dx.doi.0rg/10.1897/03-686.l

     Cutter. BE; Guvette. RP (1993). Anatomical, chemical, and ecological factors affecting tree species
        choice in dendrochemistry studies. J Environ Qual 22: 611-619.
        httrj://dx.doi.org/10.2134/ieal993.00472425002200030028x

     Dai. W: Du. H; Fu. L; Jin. C: Xu. Z; Liu. H. (2009a). Effects of dietary Pb on accumulation,
        histopathology, and digestive enzyme activities in the digestive system of tilapia (Oreochromis
        niloticus). Biol Trace Elem Res 127: 124-131. http://dx.doi.org/10.1007/sl2011-008-8227-3

     Dai. W: Fu. L; Du. H; Jin. C: Xu. Z. (2009b). Changes in growth performance, metabolic enzyme
        activities, and content of Fe, Cu, and Zn in liver and kidney of tilapia (Oreochromis niloticus) exposed
        to dietary Pb. Biol Trace Elem Res 128: 176-183. http://dx.doi.org/10.1007/sl2011-008-8259-8

     Daily. GC. (1997). Introduction: What are ecosystem services? In GC Daily (Ed.), Nature's services:
        Societal dependence on natural ecosystems (pp. 1-10). Washington, DC: Island Press.

     Daily. GC: Ehrlich. PR. (1999). Managing earth's ecosystems: An interdisciplinary challenge. Ecosystems
        2: 277-280.

     Das. M; Maiti. SK. (2007). Metal accumulation in A. baccifera growing naturally on abandoned copper
        tailings pond. Environ Monit Assess 127: 119-125. http://dx.doi.org/10.1007/sl0661-006-9265-y

     Dauwe. T; Snoeijs. T; Bervoets. L; Blust. R; Eens. M. (2006). Calcium availability influences lead
        accumulation in a passerine bird. Animal Biology 56: 289-298.
        http://dx.doi.org/10.1163/157075606778441822

     Dayton. EA; Basta. NT; Pavton. ME; Bradham. KD; Schroder. JL; Lanno. RP. (2006). Evaluating the
        contribution  of soil properties to modifying lead phytoavailability and phytotoxicity. Environ Toxicol
        Chem 25: 719-725. http://dx.doi.Org/10.1897/05-307R.l

     Dazy. M; Masfaraud. JF; Ferard. JF. (2009). Induction of oxidative stress biomarkers  associated with
        heavy metal  stress in Fontinalis antipyretica Hedw. Chemosphere 75: 297-302.
        http://dx.doi.0rg/10.1016/i.chemosphere.2008.12.045

     De Boeck. G: Eyckmans. M; Lardon. I; Bobbaers. R; Sinha. AK; Blust. R. (2010). Metal accumulation
        and metallothionein induction in the spotted dogfish  Scyliorhinus canicula. Cornp Biochem Physiol A
        Mol Integr Physiol 155: 503-508. http://dx.doi.Org/10.1016/i.cbpa.2009.12.014

     de Groot. AC: Peijnenburg. WJG. M; van den Hoop. MAG. T; Ritsema. R: van Veen. RPM. (1998).
        Zware metalen in Nederlandse veldbodems: een experimentele en theoretische studie over
        evenwichtspartitie. (607220001). Bilthoven, The Netherlands: National Institute of Public Health and
        the Environment. http://www.rivm.n^ibliotheek/rapporten/607220001.html

     De Jonge. M..; Eyckmans. M..; Blust. R..: Bervoets. L.. (2011). Are accumulated sulfide-bound metals
        metabolically available in the benthic oligochaete Tubifex tubifex. Environ Sci Technol 45:
        3131-3137. http://dx.doi.org/10.1021/esl037595

     De Jonge. M; Blust.  R: Bervoets. L. (2010). The relation between acid volatile sulfides (AVS) and metal
        accumulation in aquatic invertebrates: Implications of feeding behavior and ecology. Environ Pollut
        158: 1381-1391. http://dx.doi.0rg/10.1016/i.envpol.2010.01.001

     De Jonge. M; Dreesen. F; De Paepe. J: Blust. R: Bervoets. L. (2009). Do acid volatile sulfides (AVS)
        influence the accumulation of sediment-bound metals to benthic invertebrates under natural field
        conditions? Environ Sci Technol 43: 4510-4516. http://dx.doi.org/10.1021/es8034945
November 2012                              7-258                  Draft - Do Not Cite or Quote

-------
     de Vaufleury. A; Coeurdassier. M; Pandard. P; Scheifler. R; Low. C: Crini. N; Badot PM. (2006). How
        terrestrial snails can be used in risk assessment of soils. Environ Toxicol Chem 25: 797-806.
        http://dx.doi.Org/10.1897/04-560R.l

     de Vries. W: Groenenberg. JE. (2009). Evaluation of approaches to calculate critical metal loads for forest
        ecosystems. Environ Pollut 157: 3422-3432. http://dx.doi.Org/10.1016/j.envpol.2009.06.021

     de Vries. W: Lofts. S: Tipping. E; Meili. M; Groenenberg. JE; Schiitze. G. (2007). Impact of soil
        properties on critical concentrations of cadmium, lead, copper, zinc, and mercury in soil and soil
        solution in view of ecotoxicological effects [Review]. Rev Environ Contam Toxicol 191: 47-89.
        http://dx.doi.org/10.1007/978-0-387-69163-3 3

     DeAngelis. PL; Bartell. SM; Brenkert. AL. (1989). Effects of nutrient recycling and food-chain length on
        resilience. Am Nat 134: 778-805.

     Debelius. B; Forja. JM; DelValls. A; Lubian. LM. (2009). Toxicity and bioaccumulation of copper and
        lead in five marine microalgae. Ecotoxicol Environ Saf 72: 1503-1513.
        http://dx.doi.0rg/10.1016/i.ecoenv.2009.04.006

     DeForest. DK; Brix. KV; Adams. WJ. (2007). Assessing metal bioaccumulation in aquatic environments:
        The inverse relationship between bioaccumulation factors, trophic transfer factors and exposure
        concentration. Aquat Toxicol 84: 236-246. http://dx.doi.Org/10.1016/i.aquatox.2007.02.022

     Del Toro. I; Floyd. K; Gardea-Torresdev. J: Borrok. D. (2010). Heavy metal distribution and
        bioaccumulation in Chihuahuan Desert Rough Harvester ant (Pogonomyrmex rugosus) populations.
        Environ Pollut 158: 1281-1287. http://dx.doi.Org/10.1016/i.envrjol.2010.01.024

     Delistratv. D; Yokel. J. (In Press) Ecotoxicological study of arsenic and lead contaminated soils in former
        orchards at the Hanford site, USA. Environ Toxicol. http://dx.doi.org/10.1002/tox.20768

     Deng. H; Ye. ZH; Wong. MH. (2009). Lead, zinc and iron (Fe2+) tolerances in wetland plants and relation
        to root anatomy and spatial pattern of ROL. Environ Exp Bot 65: 353-362.
        http://dx.doi.0rg/10.1016/i.envexpbot.2008.10.005

     Desrosiers. M; Gagnon. C: Masson. S: Martel. L; Babut. MP (2008). Relationships among total
        recoverable and reactive metals and metalloid in St. Lawrence River sediment: Bioaccumulation by
        chironomids and implications for ecological risk assessment. Sci Total Environ 389: 101-114.
        http://dx.doi.0rg/10.1016/i.scitotenv.2007.08.019

     Devall. MS: Thien. LB; Ellgaard. E; Flowers. G. (2006). Lead transport into Bayou Trepagnier wetlands
        in Louisiana, USA. J Environ Qua! 35: 758-765. http://dx.doi.org/10.2134/jeq2005.0217

     Pi Toro. DM; Allen. HE; Bergman. HL; Meyer. JS: Paquin. PR; Santore. RC. (2001). Biotic ligand model
        of the  acute toxicity of metals. 1. Technical basis. Environ Toxicol Chem 20: 2383-2396.
        http://dx.doi.org/10.1002/etc.5620201034

     Pi Toro. DM;  Mahonev. JD; Hansen. DJ; Scott. KJ; Carlson. AR. (1992). Acid volatile sulfide predicts
        the acute toxicity of cadmium and nickel in sediments. Environ Sci Technol 26: 96-101.
        http://dx.doi.org/10.1021/es00025a009

     Pi Toro. DM;  McGrath. JA; Hansen. DJ: Berry. WJ: Paquin. PR; Mathew. R;  Wu. KB; Santore. RC.
        (2005). Predicting sediment metal toxicity using a sediment biotic ligand model:  Methodology and
        initial  application. Environ Toxicol Chem 24: 2410-2427. http://dx.doi.Org/10.1897/04-413r.l

     Dickinson. NM; Baker. AJM; Doronila. A; Laidlaw. S: Reeves. RD. (2009). Phytoremediation of
        inorganics: Realism and synergies. Int J Phytoremediation 11: 97-114.
        http://dx.doi.org/10.1080/15226510802378368

     Dirilgen. N.. (2011). Mercury and lead: Assessing the toxic effects on growth and metal accumulation by
        Lemnaminor. Ecotoxicol Environ Saf 74: 48-54. http://dx.doi.Org/10.1016/i.ecoenv.2010.09.014

     Dogan. M; Savgideger. SD; Colak. U. (2009).  Effect of lead toxicity on aquatic macrophyte Elodea
        canadensis Michx. Bull Environ Contam Toxicol 83:  249-254.
        http://dx.doi.org/10.1007/s00128-009-9733-5
November 2012                              7-259                   Draft - Do Not Cite or Quote

-------
     Douav. F; Pruvot. C: Waterlot C: Fritsch. C: Fourrier. H; Loriette. A; Bidar. G: Grand. C: de Vaufleury.
        A; Scheifler. R. (2009). Contamination of woody habitat soils around a former lead smelter in the
        North of France. Sci Total Environ 407: 5564-5577. http://dx.doi.Org/10.1016/j.scitotenv.2009.06.015

     Douglas-Stroebel. E; Hoffman. DJ; Brewer. GL; Sileo. L. (2004). Effects of lead-contaminated sediment
        and nutrition on mallard duckling brain growth and biochemistry. Environ Pollut 131: 215-222.
        http://dx.doi.0rg/10.1016/i.envpol.2004.02.002

     Du. M; Wang. D. (2009). The neurotoxic effects of heavy metal exposure on GABAergic nervous system
        in nematode Caenorhabditis elegans. Environ Toxicol Pharmacol 27: 314-320.
        httrj://dx.doi.org/10.1016/i.etap.2008.11.011

     Duman. F; Cicek. M; Sezen. G. (2007). Seasonal changes of metal accumulation and distribution in
        common club rush (Schoenoplectus lacustris) and common reed (Phragmites australis). Ecotoxicology
        16: 457-463. http://dx.doi.org/10.1007/sl0646-007-0150-4

     Duman. F; Obali. O: Demirezen. D. (2006). Seasonal changes of metal accumulation and distribution in
        shining pondweed (Potamogeton lucens). Chemosphere 65: 2145-2151.
        http://dx.doi.0rg/10.1016/i.chemosphere.2006.06.036

     Dwivedi. S: Srivastava. S: Mishra.  S: Dixit B; Kumar. A; Tripathi. RD. (2008). Screening of native plants
        and algae growing on fly-ash  affected areas near National Thermal Power Corporation, Tanda, Uttar
        Pradesh, India for accumulation of toxic heavy metals. J Hazard Mater 158: 359-365.
        http://dx.doi.0rg/10.1016/i.jhazmat.2008.01.081

     Dzubai. A; Backor. M; Tomko. J; Peli. E; Tuba. Z. (2008). Tolerance of the lichen Xanthoria parietina (L.)
        Th. Fr. to metal stress. Ecotoxicol Environ Saf 70: 319-326.
        http://dx.doi.0rg/10.1016/i.ecoenv.2007.04.002

     Ebenso. IE; Ologhobo. AD. (2009a). Effects of lead pollution against juvenile Achatina achatina fed on
        contaminated artificial diet. Bull Environ Contam Toxicol 82: 583-585.
        httrj://dx.doi.org/10.1007/s00128-008-9625-0

     Ebenso. IE; Ologhobo. AD. (2009b). Effects of lead pollution at industrial contaminated sites on sentinel
        juvenile Achatina achatina. Bull Environ Contam Toxicol 82: 106-110.
        http://dx.doi.org/10.1007/s00128-008-9525-3

     Ebrahimpour. M; Mushrifah. I. (2009). Variation and correlations of selected heavy metals in sediment
        and aquatic plants in Tasik Chini, Malaysia. Environ Geol 57: 823-831.
        http://dx.doi.org/10.1007/s00254-008-1362-5

     Einsporn. S: Bressling. J: Koehler. A. (2009). Cellular localization of lead using an antibody-based
        detection system and enzyme activity changes in the gills and digestive gland of the blue mussel
        Mytilus edulis. Environ Toxicol Chem 28: 402-408. http://dx.doi.Org/10.1897/08-174.l

     Einsporn. S: Koehler. A. (2008).  Immune-localisations (GSSP) of subcellular accumulation sites of
        phenanthrene, aroclor 1254 and lead (Pb) in relation  to cytopathologies in the gills and digestive gland
        of the mussel Mytilus edulis.  Mar Environ Res 66: 185-186.
        http://dx.doi.0rg/10.1016/i.marenvres.2008.02.053

     Eisler. R. (2000). Handbook of chemical risk assessment: Health hazards to humans, plants, and  animals,
        Volume  1: Metals. Boca Raton,  FL: Lewis Publishers.

     El-Gendv. K..  S.: Radwan. M.. A.; Gad. A.. F. (2011). Feeding and growth responses of the snail Theba
        pisana to dietary metal exposure. Arch Environ Contam Toxicol 60: 272-280.
        http://dx.doi.org/10.1007/s00244-010-9542-l

     El-Rjoob. AWO: Massadeh. AM; Omari. MN. (2008). Evaluation of Pb, Cu, Zn, Cd, Ni and Fe levels in
        Rosmarinus officinalis labaiatae (Rosemary)  medicinal plant and soils in selected zones in  Jordan.
        Environ Monit Assess 140: 61-68. http://dx.doi.org/10.1007/sl0661-007-9847-3
November 2012                              7-260                  Draft - Do Not Cite or Quote

-------
     Epelde. L; Becerril. JM; Barrutia. O: Gonzalez-Oreja. JA; Garbisu. C. (2010). Interactions between plant
        and rhizosphere microbial communities in a metalliferous soil. Environ Pollut 158: 1576-1583.
        http://dx.doi.0rg/10.1016/i.envpol.2009.12.013

     ERG (Eastern Research Group Inc.). (2004). Issue paper on the bioavailability and bioaccumulation of
        metals. Lexington, MA.

     Erickson. RJ; Mount. PR; Highland. TL; Hockett. JR; Leonard. EN; Mattson. VR; Dawson. TD; Lott.
        KG. (2010). Effects of copper, cadmium, lead, and arsenic in a live diet on juvenile fish growth. Can J
        Fish Aquat Sci 67: 1816-1826. http://dx.doi.org/10.1139/F10-098

     Ernst. G: Zimmermann. S: Christie. P; Frev. B. (2008). Mercury, cadmium and lead concentrations in
        different ecophysiological groups of earthworms in forest soils. Environ Pollut 156: 1304-1313.
        http://dx.doi.0rg/10.1016/i.envpol.2008.03.002

     Esbaugh. A.. J.: Brix. K.. V.; Mager. E.. M.; Grosell. M.. (2011). Multi-linear regression models predict
        the effects of water chemistry on acute lead toxicity to Ceriodaphnia dubia and Pimephales promelas.
        Comp Biochem Physiol C Toxicol Pharmacol 154: 137-145.
        http://dx.doi.0rg/10.1016/i.cbpc.2011.04.006

     Ettler. V; Vanek. A; Mihaljevic. M; Bezdicka. P. (2005). Contrasting lead speciation in forest and tilled
        soils heavily polluted by lead metallurgy. Chemosphere 58: 1449-1459.
        http://dx.doi.0rg/10.1016/i.chemosphere.2004.09.084

     Evans. RD; Balch. GC: Evans. HE; Welbourn. PM. (2006). Uptake and elimination of lead, zinc, and
        copper by caddisfly larvae (Trichoptera: Hydropsychidae) using stable isotope tracers. Arch Environ
        Contain Toxicol 51: 35-42. http://dx.doi.org/10.1007/s00244-005-2080-6

     Fairbrother. A; Wenstel. R; Sappington. K; Wood. W. (2007). Framework for metals risk assessment.
        Ecotoxicol Environ Saf 68: 145-227. http://dx.doi.Org/10.1016/j.ecoenv.2007.03.015

     Farag. AM; Nimick. DA; Kimball. BA; Church. SE; Harper. DP; Brumbaugh. WG. (2007).
        Concentrations of metals in water, sediment, biofilm, benthic macroinvertebrates, and fish in the
        Boulder River watershed, Montana, and the role of colloids in metal uptake. Arch Environ Contam
        Toxicol 52: 397-409. http://dx.doi.org/10.1007/s00244-005-0021-z

     Farley. KJ: Carbonaro. RF: Fanelli. CJ: Costanzo. R: Rader.  KJ: Pi Toro. DM. (2011). TICKET-UWM: A
        coupled kinetic, equilibrium, and transport screening model for metals in lakes. Environ Toxicol Chem
        30: 1278-1287. http://dx.doi.org/10.1002/etc.518

     Fernandez-Leborans. G: Antonio-Garcia. MT. (1988). Effects of lead and cadmium in a community of
        protozoans. Acta Protozool 27: 141-159.

     Fernandez Severini.  MD; Botte. SE; Hoffmever. MS: Marcovecchio.  JE. (2011). Lead concentrations in
        zooplankton, water, and paniculate matter of a southwestern Atlantic temperate estuary (Argentina).
        Arch Environ Contam Toxicol 61: 243-260. http://dx.doi.org/10.1007/s00244-010-9613-3

     Field. PM; Sherrell.  RM. (2003). Direct determination of ultra-trace levels of metals in fresh water using
        desolvating micronebulization and HR-ICP-MS: Application to Lake Superior waters. J Anal At
        Spectrom 18:  254-259. http://dx.doi.org/10.1039/b210628k

     Firat. O..: Cogun. H.. Y;  Yuzereroghu T.. A.; Gok. G..: Firat. O..: Kargin.  F..; Kotemen. Y.. (2011). A
        comparative study on the effects of a pesticide (cypermethrin) and two metals (copper, lead) to serum
        biochemistry of Nile tilapia, Oreochromis niloticus. Fish Physiol Biochem 37: 657-666.
        http://dx.doi.org/10.1007/sl0695-011-9466-3

     Fritioff. A; Greger. M. (2006). Uptake and distribution of Zn, Cu, Cd, and Pb in an aquatic plant
        Potamogeton natans. Chemosphere 63: 220-227. http://dx.doi.Org/10.1016/i.chemosphere.2005.08.018

     Fritsch. C: Giraudoux. P; Coeurdassier. M; Douav. F; Raoul.  F; Pruvot. C: Waterlot. C: de Vaufleury. A;
        Scheifler. R. (2010).  Spatial distribution of metals in smelter-impacted soils of woody habitats:
        Influence of landscape and soil properties, and risk for wildlife. Chemosphere 81: 141-155.
        http://dx.doi.0rg/10.1016/i.chemosphere.2010.06.075
November 2012                              7-261                  Draft - Do Not Cite or Quote

-------
     Fujiwara. FG: Gomez. PR; Dawidowski. L; Perelman. P; Faggi. A. (2011). Metals associated with
        airborne paniculate matter in road dust and tree bark collected in a megacity (Buenos Aires,
        Argentina). Ecol Indicat 11: 240-247. http://dx.doi.Org/10.1016/i.ecolind.2010.04.007

     Fukunaga. A. .; Anderson. M.. J. (2011). Bioaccumulation of copper, lead and zinc by the bivalves
        Macomona liliana and Austrovenus stutchburyi. Exp Mar Bio Ecol 396: 244-252.
        http://dx.doi.0rg/10.1016/i.jembe.2010.10.029

     Fullmer. CS. (1997). Lead-calcium interactions: Involvement of 1,25-dihydroxyvitamin D. Environ Res
        72: 45-55. http://dx.doi.org/10.1006/enrs.1996.3689

     Furman. O; Strawn. DG: Heinz. GH; Williams. B. (2006). Risk assessment test for lead bioaccessibility to
        waterfowl in mine-impacted soils. J Environ Qua! 35:  450-458.
        http://dx.doi.org/10.2134/ieq2005.0316

     Gagnon. C: Fisher. NS. (1997). Bioavailability of sediment-bound methyl and inorganic mercury to
        marine bivalve. Environ Sci Technol 31: 993-998. http://dx.doi.org/10.1021/es960364k

     Gai. N; Yang. Y; Li. T; Yao. J: Wang. F; Chen. H. (2011). Effect of lead contamination on soil microbial
        activity measured by microcalorimetry.  Chinese Journal of Chemistry 29:  1541-1547.
        http://dx.doi.org/10.1002/cjoc.201180277

     Gal. J: Markiewicz-Patkowska. J: Hursthouse. A; Tamer. P. (2008). Metal uptake by woodlice in urban
        soils. Ecotoxicol Environ Saf 69: 139-149. http://dx.doi.Org/10.1016/i.ecoenv.2007.01.002

     Gallagher. FJ; Pechmann. I; Bogden. JD; Grabosky. J: Weis. P. (2008).  Soil metal concentrations and
        productivity  of Betula populifolia (gray birch) as measured by field spectrometry and incremental
        annual growth in an abandoned urban brownfield in New Jersey. Environ Pollut 156: 699-706.
        htrp://dx.doi.org/10.1016/i.envpol.20Q8.06.013

     Gambrell. RP. (1994). Trace and toxic metals in wetlands: Areview. J Environ Qual 23: 883-891.
        http://dx.doi.org/10.2134/ieql994.00472425002300050005x

     Gandhi, N,.; Bhavsar, S,. R; Diamond, M,. L. (2011).  Critical load analysis in hazard assessment of
        metals using a Unit World Model. Environ Toxicol Chem 30: 2157-2166.
        http://dx.doi.org/10.1002/etc.606

     Gandois.  L; Probst. A; Dumat C. (2010). Modelling trace metal extractability and solubility in French
        forest soils by using soil properties. Eur J Soil Sci 61:  271-286.
        http://dx.doi.0rg/10.llll/i.1365-2389.2009.01215.x

     Gao. Y; Zhou. P; Mao. L; Zhi. YE; Shi. WJ. (201 Ob). Assessment of effects of heavy metals combined
        pollution on  soil enzyme activities and microbial community structure: modified ecological
        doseresponse model and PCR-RAPD. Environmental Earth Sciences 60: 603-612.
        http://dx.doi.org/10.1007/sl2665-009-0200-8

     Garcia-Garcia. G: Nandini. S: Sarma. SSS. (2006). Turbidity mitigates lead toxicity to cladocerans
        (Cladocera).  Ecotoxicology 15: 425-436. http://dx.doi.org/10.1007/sl0646-006-0064-6

     Garcia-Garcia. G: Picazo-Paez. EA; Nandini. S: Sarma. SSS. (2007). Combined effects of sediment and
        lead (PbC12) on the demography of Brachionus patulus (Rotifera: Brachionidae). Hydrobiologia 593:
        209-218. http://dx.doi.org/10.1007/sl0750-007-9039-8

     Garcia-Rico. L..; Tejeda-Valenzuela. L..; Burgos-Hernandez. A..  (2010). Seasonal variations in the
        concentrations of metals in Crassostrea corteziensis from  Sonora, Mexico. Bull Environ Contam
        Toxicol 85: 209-213. http://dx.doi.org/10.1007/s00128-010-0055-4

     Garg. P; Satva. S: Sharma.  S. (2009). Effect of heavy metal supplementation on local  (Allolobophora
        parva) and exotic (Eisenia fetida) earthworm species: A comparative study. J Environ Sci Health A
        Tox Hazard Subst Environ Eng 44: 1025-1032. http://dx.doi.org/10.1080/10934520902996997

     Gaur. JP; Noraho. N; Chauhan. YS. (1994). Relationship between heavy metal accumulation and toxicity
        in Spirodela  polyrhiza (L) Schleid and Azolla pinnata  R Br. Aquat Bot 49: 183-192.
November 2012                              7-262                   Draft - Do Not Cite or Quote

-------
     Gebologlu. N: Cetin. SC: Ece. A; Yilmaz. E; Elmastas. M. (2005). Assessment of lead and cadmium
        contents of tomatoes and beans grown in the vicinity of highway of Tokat, Turkey. Asian J Chem 17:
        730-736.

     Gedan. K; Silliman. BR; Bertness. MD. (2009). Centuries of human-driven change in salt marsh
        ecosystems [Review]. Ann Rev Mar Sci 1: 117-141.
        http://dx.doi.org/10.1146/annurev.marine.010908.163930

     Gitav. H; Brown. S: Easterling. W: Jallow. B. (2001). Ecosystems and their goods and services.  In
        Climate change 2001: Impacts, adaptation and vulnerability: Contribution of Working Group II to the
        third assessment report of the Intergovernmental Panel on Climate Change. Cambridge, United
        Kingdom: Cambridge University Press.

     Giusi. G: Alo. R; Crudo. M; Facciolo. RM; Canonaco. M. (2008). Specific cerebral heat shock proteins
        and histamine receptor cross-talking mechanisms promote distinct lead-dependent neurotoxic
        responses in teleosts. Toxicol Appl Pharmacol 227: 248-256.
        http://dx.doi.0rg/10.1016/i.taap.2007.10.018

     Gnandi. K; Tchangbedji. G: Killi. K; Baba. G: Abbe. K. (2006). The impact of phosphate mine tailings on
        the bioaccumulation of heavy metals in marine fish and crustaceans from  the coastal zone of Togo.
        Mine Water Environ 25: 56-62. http://dx.doi.org/10.1007/sl0230-006-0108-4

     Gomez-Ramirez. P..; Martinez-Lopez. E..; Maria-Mojica. P..: Leon-Ortega. M..; Garcia-Fernandez. A..
        1 (2011). Blood lead levels and 5-ALAD inhibition in nestlings of Eurasian Eagle Owl (Bubo bubo)
        to assess lead exposure associated to an abandoned mining area. Ecotoxicology 20: 131-138.
        http://dx.doi.org/10.1007/sl0646-010-0563-3

     Gopal. R; Rizvi. AH. (2008). Excess lead alters growth, metabolism and translocation of certain nutrients
        in radish. Chemosphere 70: 1539-1544. http://dx.doi.Org/10.1016/i.cheniosphere.2007.08.043

     Gopalakrishnan. S: Thilagam. H; Raja. PV. (2007). Toxicity of heavy metals on embryogenesis and larvae
        of the marine sedentary polychaete Hydroides elegans. Arch Environ Contam Toxicol 52: 171-178.
        http://dx.doi.org/10.1007/s00244-006-0038-v

     Gopalakrishnan. S: Thilagam. H; Raja. PV. (2008). Comparison of heavy metal toxicity in life stages
        (spermiotoxicity, egg toxicity, embryotoxicity and larval toxicity) of Hydroides elegans. Chemosphere
        71: 515-528. http://dx.doi.Org/10.1016/i.chemosphere.2007.09.062

     Gorski. J; Nugegoda. D. (2006). Sublethal toxicity of trace metals to larvae of the blacklip abalone,
        Haliotis rubra. Environ Toxicol Chem 25: 1360-1367. http://dx.doi.Org/10.1897/05-060R.l

     Goriir. G. (2007). Effects of host plant contaminated with heavy metals on the life history traits of aphids
        (Brevicoryne brassicae L.). Polish Journal of Ecology 55: 113-120.

     Goto.  D; Wallace. WG. (2010). Metal intracellular partitioning as a detoxification mechanism for
        mummichogs (Fundulus heteroclitus) living in metal-polluted salt marshes. Mar Environ Res 69:
        163-171. http://dx.doi.0rg/10.1016/i.marenvres.2009.09.008

     Goulder. LH; Kennedy. D. (1997). Valuing ecosystem services: Philosophical bases and empirical
        methods. In GC Daily (Ed.), Nature's Services: Societal Dependence  on Natural Ecosystems (pp.
        23-47). Washington, DC: Island Press.

     Groenenberg. JE; Koopmans. GF; Comans. RNJ. (2010). Uncertainty analysis of the nonideal competitive
        adsorption-Donnan model: Effects of dissolved organic matter variability on predicted metal
        speciation in soil solution. Environ Sci Technol 44: 1340-1346. http://dx.doi.org/10.1021/es902615w

     Grosell. M; Brix. KV  (2009). High net calcium uptake explains the hypersensitivity of the freshwater
        pulmonate snail, Lymnaea stagnalis, to chronic lead exposure. Aquat  Toxicol 91: 302-311.
        http://dx.doi.0rg/10.1016/i.aquatox.2008.10.012

     Grosell. M; Gerdes. R: Brix. KV. (2006a). Influence of Ca, humic acid and pH on lead accumulation and
        toxicity in the fathead minnow during prolonged water-borne lead exposure. Comp Biochem Physiol
        C Toxicol Pharmacol 143: 473-483.  http://dx.doi.Org/10.1016/i.cbpc.2006.04.014
November 2012                             7-263                   Draft - Do Not Cite or Quote

-------
     Grosell. M; Gerdes. RM; Brix. KV. (2006b). Chronic toxicity of lead to three freshwater invertebrates -
        Brachionus calyciflorus, Chironomus tentans, and Lymnaea stagnalis. Environ Toxicol Chem 25:
        97-104. http://dx.doi.Org/10.1897/04-654R.l

     Guillen. J; Baeza. A; Ontalba. MA; Miguez. MR (2009). Pb-210 and stable lead content in fungi: Its
        transfer from soil. Sci Total Environ 407: 4320-4326.
        http://dx.doi.0rg/10.1016/i.scitotenv.2009.03.025

     Quo. XY; Zhang. SZ; Shan. XQ; Luo. L; Pel ZG: Zhu. YG: Liu. T; Xie. YN; Gault. A. (2006b).
        Characterization of Pb, Cu, and Cd adsorption on paniculate organic matter in soil. Environ Toxicol
        Chem 25: 2366-2373. http://dx.doi.Org/10.1897/05-636R.l

     Guo. YL; Yang. YC: Wang. DY. (2009). Induction of reproductive deficits in nematode Caenorhabditis
        elegans exposed to metals at different developmental stages. Reprod Toxicol 28: 90-95.
        http://dx.doi.0rg/10.1016/i.reprotox.2009.03.007

     Gupta. DK; Huang. HG: Yang. XE; Razafindrabe. BHN; Inouhe. M. (2010). The detoxification of lead in
        Sedum alfredii H. is not related to phytochelatins but the glutathione. J Hazard Mater 177: 437-444.
        http://dx.doi.0rg/10.1016/i.jhazmat.2009.12.052

     Guvette. RP; Cutter. BE; Henderson. GS. (1991). Long-term correlations between mining activity and
        levels of lead and cadmium in tree-rings of eastern red-cedar. J Environ Qua! 20: 146-150.
        http://dx.doi.org/10.2134/ieql991.00472425002000010022x

     Gvedu-Ababio. TK;  Baird. D. (2006). Response of meiofauna and nematode communities to increased
        levels of contaminants in a laboratory microcosm experiment. Ecotoxicol Environ Saf 63: 443-450.
        http://dx.doi.0rg/10.1016/i.ecoenv.2005.01.010

     Ha. MH; Choi. J.  (2009). Effects of environmental contaminants on hemoglobin gene expression in
        Daphnia magna: A potential biomarker for freshwater quality monitoring. Arch Environ Contain
        Toxicol 57: 330-337. http://dx.doi.org/10.1007/s00244-007-9079-0

     Ha. NTH: Sakakibara. M; Sano. S: Hori. RS: Sera. K. (2009). The potential of Eleocharis acicularis for
        phytoremediation: Case study at an abandoned mine site. CLEAN - Soil, Air, Water 37: 203-208.
        http://dx.doi.org/10.1002/clen.200900009

     Hall. JR; Ashmore. M; Fawehinmi. J: Jordan. C: Lofts. S: Shotbolt.  L; Spurgeon. DJ;  Svendsen. C:
        Tipping. E. (2006). Developing a critical load approach for national risk assessments of atmospheric
        metal deposition. Environ Toxicol Chem 25: 883-890. http://dx.doi.Org/10.1897/04-571R.l

     Han. YL; Huang. SZ; Gu. JG: Qiu. S: Chen. JM. (2008). Tolerance and accumulation  of lead by species
        of Iris L. Ecotoxicology 17: 853-859. http://dx.doi.org/10.1007/sl0646-008-0248-3

     Hansen. JA; Audet. D; Spears. BL; Healy. KA; Brazzle. RE; Hoffman. DJ: Dailev. A; Beyer. WN.
        (2011 a). Lead exposure and poisoning of songbirds using the Coeur d'Alene River Basin, Idaho, USA.
        Integr Environ Assess Manag 7: 587-595. http://dx.doi.org/10.1002/ieam.201

     Hassan. R; Scholes. R; Ash. N. (2005). Ecosystems and human well-being: Current state and trends.
        Washington, DC: Island Press.

     Heier. LS: Lien. IB;  Stromseng. AE; Ljones. M; Rosseland. BO; Tollefsen. KE; Salbu. B. (2009).
        Speciation of lead, copper, zinc and antimony in water draining a shooting range: Time dependant
        metal accumulation and biomarker responses in brown trout (Salmo trutta L.). Sci  Total Environ 407:
        4047-4055. http://dx.doi.Org/10.1016/i.scitotenv.2009.03.002

     Henczova. M; Deer. AK; Filla. A; Komlosi. V: Mink. J. (2008). Effects of Cu(2+) and Pb(2+) on different
        fish species: Liver cytochrome P450-dependent monooxygenase activities and FTIR spectra. Comp
        Biochem Physiol C Toxicol Pharmacol 148:  53-60. http://dx.doi.Org/10.1016/j.cbpc.2008.03.010

     Hirsch. HVB; Possidente. D: Possidente. B. (2010). Pb2+: An endocrine disrupter in Drosophila?
        [Review]. Physiol Behav 99: 254-259. http://dx.doi.Org/10.1016/i.physbeh.2009.09.014
November 2012                             7-264                   Draft - Do Not Cite or Quote

-------
     Hong. FS: Wu. C: Liu. C: Wang. L; Gao. FQ; Yang. F; Xu. JH; Liu. T; Xie. YN; Li. XZ. (2007a). Direct
        evidence for interaction between lead ions and kidney DNAfrom silver crucian carp. Chemosphere
        68: 1442-1446. http://dx.doi.Org/10.1016/i.chemosphere.2007.04.020

     Hooper. DU; Vitousek. PM. (1997). The effects of plant composition and diversity on ecosystem
        processes. Science 277: 1302-1305. http://dx.doi.org/10.1126/science.277.5330.1302

     Horng. CY; Wang. SL; Cheng. IJ.  (2009). Effects of sediment-bound Cd, Pb, and Ni on the growth,
        feeding, and survival of Capitella sp I. Exp Mar Bio Ecol 371: 68-76.
        http://dx.doi.0rg/10.1016/i.jembe.2009.01.008

     Horvat. T; Vidakovic-Cifrek. Z; Orescanin. V; Tkalec. M; Pevalek-Kozlina. B. (2007). Toxicity
        assessment of heavy metal mixtures by Lemna minor L. Sci Total Environ 384: 229-238.
        http://dx.doi.0rg/10.1016/i.scitotenv.2007.06.007

     Hou. JL; Zhuang. P; Zhang. LZ; Feng. L; Zhang. T; Liu. JY; Feng. GP (2011). Morphological
        deformities and recovery, accumulation and elimination of lead in body tissues of Chinese sturgeon,
        Acipenser sinensis, early life stages: A laboratory study. Journal of Applied Ichthyology (Print) 27:
        514-519. http://dx.doi.0rg/10.llll/i.1439-0426.2011.01703.x

     Hovmand. MF; Nielsen. SP: Johnsen. I. (2009). Root uptake of lead by Norway spruce grown on 210Pb
        spiked soils. Environ Pollut 157: 404-409. http://dx.doi.Org/10.1016/i.envpol.2008.09.038

     Hu. Q; Qi. HY; Zeng. JH; Zhang.  HX. (2007b). Bacterial diversity in soils around a lead and zinc mine. J
        Environ Sci 19: 74-79.  http://dx.doi.org/10.1016/S1001-0742(07)60012-6

     Hu. X; Ding. Z. (2009). Lead/cadmium contamination and lead isotopic ratios in vegetables grown in
        peri-urban and mining/smelting contaminated sites in Nanjing, China. Bull Environ Contam Toxicol
        82: 80-84. http://dx.doi.org/10.1007/s00128-008-9562-y

     Hu. X; Hu. C: Sun. X; Lu. M; Su.  B; Cao. A. (2009b). Effects of simulated acid rain on soil acidification,
        availabilities and temporal and spatial variations of Cu and Pb in a vegetable field under natural
        conditions. J Food Agr Environ 7: 92-96.

     Huang. CC: Xu. Y; Briggler. JT; McKee.  M; Nam. P; Huang. YW. (2010). Heavy metals, hematology,
        plasma chemistry, and parasites in adult hellbenders (Cryptobranchus alleganiensis). Environ Toxicol
        Chem 29:  1132-1137. http://dx.doi.org/10.1002/etc.148

     Huang. JH;  Kalbitz. K; Matzner. E. (2008). Mobility of trimethyllead and total lead in the forest floor.
        Soil Sci  Soc Am J 72: 978-984. http://dx.doi.org/10.2136/sssai2006.0118

     Hui. N; Selonen. S: Hanzel. J: Tuomela. M; Rainio. A; Kontio. H; Hakala. K; Lankinen. P; Steffen. K;
        Fingerroos. T; Strommer. R; Setala. H; Hatakka. A; Romantschuk. M. (2009). Influence of lead on
        organisms within the detritus food web of a contaminated pine forest soil. Boreal Environ Res  14:
        70-85.

     Hurd. NA; Sternberg. SPK. (2008). Bioremoval of aqueous lead using Lemna minor. Int J
        Phytoremediation 10: 278-288. http://dx.doi.org/10.1080/15226510802096036

     Inouve. LS: Yoo. LJ; Talent. LG: Clarke. JU: Jones. RP; Steevens. JA; Boyd. RE. (2007). Assessment of
        lead uptake in reptilian prey species. Chemosphere 68: 1591-1596.
        http://dx.doi.0rg/10.1016/i.chemosphere.2006.12.045

     IPCC (Intergovernmental Panel on Climate Change). (2007). Climate change 2007: Impacts, adaptation
        and vulnerability. Cambridge, U.K.: Cambridge University Press.

     Isidori. M..; Cangiano. M..; Palermo. F. .A.; Parrella. A.. (2010). E-screen and vitellogenin assay for the
        detection of the estrogenic activity of alkylphenols and trace elements. Comp Biochem Physiol C
        Toxicol Pharmacol  152: 51-56. http://dx.doi.Org/10.1016/i.cbpc.2010.02.011

     Islam. E;  Liu. D; Li. T; Yang. X; Jin. X; Mahmood. Q: Tian. S: Li. J. (2008). Effect of Pb toxicity on leaf
        growth, physiology and ultrastructure in the two ecotypes of Elsholtzia argyi. J Hazard Mater 154:
        1-3. http://dx.doi.0rg/10.1016/i.ihazmat.2007.10.121
November 2012                              7-265                   Draft - Do Not Cite or Quote

-------
     Jakl. M; Dytrtova. JJ: Miholova. D; Kolihova. D; Szakova. J: Tlustos. P. (2009). Passive diffusion
        assessment of cadmium and lead accumulation by plants in hydroponic systems. Chem Speciation
        Bioavailability 21: 111-120. http://dx.doi.org/10.3184/095422909x456870

     Jara-Marini. ME; Soto-Jimenez. MF; Paez-Osuna. F. (2009). Trophic relationships and transference of
        cadmium, copper, lead and zinc in a subtropical coastal lagoon food web from SE Gulf of California.
        Chemosphere 77: 1366-1373. http://dx.doi.Org/10.1016/i.chemosphere.2009.09.025

     Jeffree. RA; Oberhansli. F; Tevssie. JL. (2008). The accumulation of lead and mercury from seawater and
        their depuration by eggs of the spotted dogfish Scyliorhinus canicula (Chondrichthys). Arch Environ
        Contam Toxicol 55: 451-461. http://dx.doi.org/l0.1007/s00244-007-9103-4

     Jezierska. B; Lugowska. K; Witeska. M. (2009). The effects of heavy metals on embryonic development
        offish (a review) [Review]. Fish Physiol Biochem 35: 625-640.
        http://dx.doi.org/10.1007/sl0695-008-9284-4

     Jing. G: Li. Y; Xie. L; Zhang. RQ. (2007). Different effects of Pb2+ and Cu2+ on immune and
        antioxidant enzyme activities in the mantle of Pinctada fucata. Environ Toxicol Pharmacol 24:
        122-128. http://dx.doi.0rg/10.1016/i.etap.2007.04.002

     John. R; Ahmad. P; Gadgil. K; Sharma. S. (2009). Heavy metal toxicity: Effect on plant growth,
        biochemical parameters and metal accumulation by Brassica juncea L. International Journal of Plant
        Production 3: 65-75.

     Jones. JC. (1939). On the occurrence of lead shot in stomachs of North American gruiformes. J Wildl
        Manag 3: 353-357.

     Jones. RP; Bednar. AJ; Inouye. LS. (2009b). Subcellular compartmentalization of lead in the earthworm,
        Eisenia fetida: Relationship to survival and reproduction. Ecotoxicol Environ Saf 72:  1045-1052.
        http://dx.doi.0rg/10.1016/i.ecoenv.2008.12.011

     Jordaens. K; De Wolf. H; Vandecasteele. B; Blust. R; Backeljau. T. (2006). Associations between shell
        strength, shell morphology and heavy metals in the land snail Cepaea nemoralis (Gastropoda,
        Helicidae). Sci Total Environ 363: 285-293. http://dx.doi.Org/10.1016/j.scitotenv.2005.12.002

     Joshi.  SR. (2008). Influence of roadside pollution on the phylloplane microbial community of Alnus
        nepalensis  (Betulaceae). Rev Biol Trop 56: 1521-1529.

     Jothinavagi. N; Anbazhagan. C. (2009). Heavy metal monitoring of Rameswaram coast by some
        Sargassum species. Am-Eurasian J Sci Res 4: 73-80.

     Kalis.  EJJ; Temminghoff. EJM; Visser. A; van Riemsdijk. WH. (2007). Metal uptake by Lolium perenne
        in contaminated soils using a four-step approach. Environ Toxicol Chem 26: 335-345.
        http://dx.doi.Org/10.1897/06-173R.l

     Kalis.  EJJ: Temminghoff. EJM: Weng. LP; van Riemsdijk. WH. (2006). Effects of humic acid and
        competing cations on metal uptake by Lolium perenne. Environ Toxicol Chem  25: 702-711.
        http://dx.doi.Org/10.1897/04-576R.l

     Kalman. J: Riba. I; Blasco. J:  Delvalls. TA. (2008). Is delta-aminolevulinic acid dehydratase activity in
        bivalves from south-west Iberian Peninsula a good biomarker of lead exposure? Mar Environ Res 66:
        38-40. http://dx.doi.0rg/10.1016/j.marenvres.2008.02.016

     Kamaruzzaman. BY; Akbar. B; Jalal. K; Shahbudin. S. (2010). Accumulation of metals in the gills of
        tilapia fingerlings (Oreochromis niloticus) from in vitro toxicology study. J Fish Aquat Sci 5: 503-509.
        http://dx.doi.org/10.3923/ifas.2010.503.509

     Kannan. K; Agusa. T; Perrotta. E; Thomas. NJ; Tanabe. S. (2006). Comparison of trace element
        concentrations in livers of diseased, emaciated and non-diseased southern sea otters from the
        California coast. Chemosphere 65: 2160-2167. http://dx.doi.Org/10.1016/i.chemosphere.2006.06.003

     Kaufman. CA; Bennett. JR; Koch. I; Reimer. KJ. (2007). Lead bioaccessibility in food web intermediates
        and the influence on ecological risk characterization. Environ Sci Technol 41: 5902-5907.
        http://dx.doi.org/10.1021/es062443u
November 2012                             7-266                   Draft - Do Not Cite or Quote

-------
     Kawakami. SK; Gledhill. M; Achterberg. EP. (2006). Production of phytochelatins and glutathione by
        marine phytoplankton in response to metal stress. J Phycol 42: 975-989.
        http://dx.doi.0rg/10.llll/j.1529-8817.2006.00265.x

     Kay. SH; Haller. WT. (1986). Heavy metal bioaccumulation and effects on waterhyacinth weevils,
        Neochetina eichhorniae, feeding on waterhyacinth, Eichhornia crassipes. Bull Environ Contam
        Toxicol 37: 239-245. http://dx.doi.org/10.1007/BF01607756

     Kempter. H; Krachler. M; Shotyk. W. (2010). Atmospheric Pb and Ti accumulation rates from Sphagnum
        moss: Dependence upon plant productivity. Environ Sci Technol 44: 5509-5515.
        http://dx.doi.org/10.1021/esl00366d

     Kertesz. V; Bakonyi. G: Farkas. B. (2006). Water pollution by Cu and Pb can adversely affect mallard
        embryonic development. Ecotoxicol Environ Saf 65: 67-73.
        http://dx.doi.0rg/10.1016/i.ecoenv.2005.05.016

     Kertesz. V; Fancsi. T. (2003). Adverse effects of (surface water pollutants) Cd, Cr and Pb on the
        embryogenesis of the mallard. Aquat Toxicol 65: 425-433.
        http://dx.doi.org/10.1016/S0166-445X(03)00155-3

     Khan. MAQ; Ahmed. SA; Catalin. B; Khodadoust. A; Ajayi. O: Vaughn. M. (2006). Effect of temperature
        on heavy metal toxicity to juvenile crayfish, Orconectes immunis (Hagen). Environ Toxicol 21:
        513-520. http://dx.doi.org/10.1002/tox.20213

     Khan. S: Hesham. AEL; Qiao. M; Rehman. S: He. JZ.  (2010c). Effects of Cd and Pb on soil microbial
        community structure and activities.  Environ Sci Pollut Res Int 17: 288-296.
        http://dx.doi.org/10.1007/sll356-009-0134-4

     Khangarot. BS. (1991). Toxicity of metals to a freshwater tubificid worm, Tubifex tubifex (Muller). Bull
        Environ Contam Toxicol 46: 906-912.

     Khozhina. El; Sherriff. BL. (2008). The accumulation of Cu, Zn, Cd, and Pb in the aquatic biomass of
        sulphide tailing ponds. Geochem Int 46: 897-911. http://dx.doi.org/10.1134/s0016702908090048

     Kibblewhite. MG: Ritz. K; Swift. MJ. (2008). Soil health in agricultural systems. Philos Trans R Soc
        Lond B Biol Sci 363: 685-701. http://dx.doi.org/10.1098/rstb.2007.2178

     Kimbrough. KL; Lauenstein. GG: Christensen. JD; Apeti. DA. (2008). An assessment of two decades of
        contaminant monitoring in the nation's coastal zone. Silver Spring, MD: National Centers for Coastal
        Ocean Science, http://aquaticcommons.org/2232/

     King. CK; Dowse. MC: Simpson. SL. (2010). Toxicity of metals to the Bivalve tellina deltoidalis and
        relationships between metal bioaccumulation and metal partitioning between seawater and marine
        sediments. Arch Environ Contam Toxicol 58: 657-665. http://dx.doi.org/10.1007/s00244-009-9413-9

     King. CK; Gale. SA; Stauber. JL. (2006). Acute toxicity and bioaccumulation of aqueous and sediment-
        bound metals in the estuarine amphipod Melita plumulosa. Environ Toxicol 21: 489-504.
        http://dx.doi.org/10.1002/tox.20211

     Kitvatanachai. S: Apiwathnasorn. C: Leemingsawat. S: Wongwit. W: Tornee. S. (2005). Determination of
        lead toxicity in Culex quinquefasciatus mosquitoes in the laboratory. Southeast Asian J Trop Med
        Public Health 36: 862-874.

     Klaminder. J: Bindler. R; Emteryd. O: Renberg. I. (2005). Uptake and recycling of lead by boreal forest
        plants: Quantitative estimates from a site in northern Sweden. Geochim Cosmo Act 69: 2485-2496.
        http://dx.doi.0rg/10.1016/i.gca.2004.ll.013

     Klok. C: Van der Hout. A; Bodt. J. (2006). Population growth and development of the  earthworm
        Lumbricus rubellus in a polluted field soil: Possible consequences for the godwit (Limosa limosa).
        Environ Toxicol Chem 25: 213-219. http://dx.doi.Org/10.1897/05-286R.l

     Knowlton. MF; Boyle. TP; Jones. JR. (1983). Uptake  of lead from  aquatic sediment by submersed
        macrophytes and crayfish. Arch Environ Contam Toxicol 12: 535-541.
        http://dx.doi.org/10.1007/BF01056549
November 2012                             7-267                  Draft - Do Not Cite or Quote

-------
     Kobler. J; Fitz. WJ: Dirnbock. T; Mirtl. M. (2010). Soil type affects migration pattern of airborne Pb and
        Cd under a spruce-beech forest of the UN-ECE integrated monitoring site Zobelboden, Austria.
        Environ Pollut 158: 849-854. http://dx.doi.Org/10.1016/i.envpol.2009.09.026

     Kohler. HR; Alberti. G: Seniczak. S: Seniczak. A. (2005). Lead-induced hsp70 and hsp60 pattern
        transformation and leg malformation during postembryonic development in the oribatid mite,
        Archegozetes longisetosus Aoki. Comp Biochem Physiol C Toxicol Pharmacol 141: 398-405.
        http://dx.doi.0rg/10.1016/i.cbpc.2005.09.003

     Koivula. MJ; Eeva. T. (2010). Metal-related oxidative stress in birds. Environ Pollut 158:  2359-2370.
        http://dx.doi.0rg/10.1016/i.envpol.2010.03.013

     Komarek. M; Ettler. V; Szakova. J: Sebek. O; Tlustos. P. (2009). Bioavailability of lead and cadmium in
        soils artificially contaminated with smelter fly ash. Bull Environ Contam Toxicol 83: 286-290.
        http://dx.doi.org/10.1007/s00128-009-9742-4

     Kominkova. D; Nabelkova. J. (2005). The risk assessment of heavy metals in the ecosystem of urban
        creeks. Water Sci Technol 53: 10. http://dx.doi.org/10.2166/wst.2006.298

     Komjarova. I; Blust. R. (2008). Multi-metal interactions between Cd, Cu, Ni, Pb and Zn in water flea
        Daphnia magna, a stable isotope experiment. Aquat Toxicol 90: 138-144.
        http://dx.doi.0rg/10.1016/i.aquatox.2008.08.007

     Komjarova. I; Blust. R. (2009a). Application of a stable isotope technique to determine the simultaneous
        uptake of cadmium, copper, nickel, lead, and zinc by the water flea Daphnia magna from water and
        the green algae Pseudokirchneriella subcapitata. Environ Toxicol Chem 28: 1739-1748.
        http://dx.doi.0rg/10.1897/08-437.l

     Komjarova. I; Blust. R. (2009b). Effect of Na, Ca and pH on simultaneous uptake of Cd, Cu, Ni, Pb, and
        Zn in the water flea Daphnia magna measured using stable isotopes. Aquat Toxicol 94: 81-86.
        http://dx.doi.0rg/10.1016/i.aquatox.2009.05.018

     Komjarova. I; Blust. R. (2009c). Effects of Na, Ca, and pH on the simultaneous uptake of Cd, Cu, Ni, Pb,
        and Zn in the zebrafish Danio rerio: A stable isotope experiment. Environ Sci Technol 43: 7958-7963.
        http://dx.doi.org/10.1021/es9016987

     Kools. SAE: Boivin. MEY: Van Per Wurff. AWG: Berg. MP: Van Gestel. CAM: Van Straalen. NM.
        (2009). Assessment of structure and function in metal polluted grasslands using Terrestrial Model
        Ecosystems.  Ecotoxicol Environ Saf 72: 51-59. http://dx.doi.Org/10.1016/j.ecoenv.2008.03.016

     Korean. SE; Cigerci. IH;  Konuk. M. (2007). Screening of delta-aminolevulinic acid dehydratase from
        Pseudomonas strains as biosensor for lead and some other metals contamination. Environ Monit
        Assess  134: 263-269.  http://dx.doi.org/10.1007/sl0661-007-9615-4

     Kramer. JL; Redig. PT. (1997). Sixteen years of lead poisoning in eagles, 1980-95: An epizootiologic
        view. Journal of Raptor Research 31: 327-332.

     Kuang. YW: Zhou. GY; Wen. DZ; Liu. SZ. (2007). Heavy metals in bark of Pinus massoniana (Lamb.) as
        an indicator of atmospheric deposition near a smeltery at Qujiang, China. Environ Sci Pollut Res Int
        14: 270-275. http://dx.doi.org/10.1065/espr2006.09.344

     Kudirat. AO. (2008). Bioconcentration of lead in the tissues of feral and laboratory exposed Clarias
        gariepinus. Journal of Medical Sciences 8: 281-286.

     Kuklova. M..; Kukla. J..: Hnilicka. F.. (2010). The soil-to-herbs transfer of heavy metals in spruce
        ecosystems. Pol J Environ Stud 19: 1263-1268.

     Kumar. G: Tripathi. R. (2008). Lead-induced cytotoxicity and mutagenicity in grass pea. Turkish Journal
        of Biology 32: 73-78.

     Kurilenko. VV;  Osmolovskava. NG. (2006). Ecological-biogeochemical role of macrophytes in aquatic
        ecosystems of urbanized territories (an example of small water bodies of St.  Petersburg). Russian
        Journal of Ecology 37: 147-151. http://dx.doi.org/10.1134/sl067413606030015
November 2012                              7-268                  Draft - Do Not Cite or Quote

-------
     Kwartirnikov. MA; Lavchieva-Nacheva. GV: Lavchiev. VI. (1999). The effects of lead intoxication on the
        survival and behavior of Leptinotarsa decemlineata Say (Coloeptera: Chrysomelidae). Acta Zoolog
        BulgSl: 103-108.

     Lacoue-Labarthe. T; Warnau. M; Metian. M; Oberhansli. F; Rouleau. C: Bustamante. P. (2009).
        Biokinetics of Hg and Pb accumulation in the encapsulated egg of the common cuttlefish Sepia
        officinalis: Radiotracer experiments.  Sci Total Environ 407: 6188-6195.
        http://dx.doi.0rg/10.1016/i.scitotenv.2009.09.003

     Lagisz. M; Laskowski. R. (2008). Evidence for between-generation effects in carabids exposed to heavy
        metals pollution. Ecotoxicology 17: 59-66. http://dx.doi.org/10.1007/sl0646-007-0176-7

     Lamelas. C: Pinheiro. JP; Slaveykova. VI. (2009). Effect of humic acid on Cd(ll), Cu(ll), and Pb(ll)
        uptake by freshwater algae: Kinetic and cell wall speciation considerations. Environ Sci Technol 43:
        730-735. http://dx.doi.org/10.1021/es802557r

     Lamelas. C: Slaveykova. VI. (2007). Comparison of Cd(II), Cu(II), and Pb(II) biouptake by green algae
        in the presence of humic acid. Environ Sci Technol 41: 4172-4178.
        http://dx.doi.org/10.1021/es063102i

     Lamelas. C: Slaveykova. VI. (2008). Pb uptake by the freshwater alga Chlorella kesslerii in the presence
        of dissolved organic matter of variable composition. Environ Chem 5: 366-372.
        http://dx.doi.org/10.1071/en08043

     Lamelas. C: Wilkinson. KJ; Slaveykova. VI. (2005). Influence of the composition of natural organic
        matter on Pb bioavailability to microalgae. Environ Sci Technol 39: 6109-6116.
        http://dx.doi.org/10.1021/es050445t

     Landers. DH; Simonich. SL; Jaffe. DA; Geiser. LH; Campbell. DH; Schwindt. AR; Schreck. CB; Kent.
        ML; Hafner. WD; Taylor. HE; Hageman. KJ: Usenko. S: Ackerman. LK; Schrlau. JE; Rose. NL; Blett.
        TF; Erwav. MM. (2008). The fate, transport, and ecological impacts of airborne contaminants in
        western national parks (USA). (EPA/600/R-07/138). Corvallis, Oregon: U.S. Environmental
        Protection Agency, NHEERL, Western Ecology Division.
        http://www.nature.nps.gov/air/studies/air toxics/WACAPreport.cfm

     Landre. AL; Watmough. SA; Dillon. PJ.  (2010). Metal pools, fluxes, and budgets in an acidified  forested
        catchment on the Precambrian Shield, Central Ontario, Canada. Water Air Soil Pollut 209: 209-228.
        http://dx.doi.org/10.1007/sll270-009-0193-7

     Langdon. CJ: Hodson. ME; Arnold. RE;  Black. S. (2005).  Survival, Pb-uptake and behaviour of three
        species of earthworm in Pb treated soils determined using an OECD-style toxicity test and a soil
        avoidance test. Environ Pollut 138: 368-375. http://dx.doi.Org/10.1016/i.envpol.2005.03.002

     Larsson. P; Holmqvist. N; Stenroth. P; Berglund. O: Nystrom. P; Graneli. W (2007). Heavy metals and
        stable isotopes in a benthic omnivore in a trophic gradient of lakes. Environ Sci Technol 41:
        5973-5979. http://dx.doi.org/10.1021/es0704838

     Lazzaro. A; Schulin. R; Widmer. F; Frev. B. (2006). Changes in lead availability affect bacterial
        community structure but not basal respiration in a microcosm study with forest soils. Sci Total
        Environ 371: 110-124. http://dx.doi.Org/10.1016/i.scitotenv.2006.08.033

     Le Faucheur. S: Schildknecht. F; Behra. R: Sigg. L. (2006). Thiols in Scenedesmus vacuolatus upon
        exposure to metals and metalloids. Aquat Toxicol 80: 355-361.
        http://dx.doi.0rg/10.1016/i.aquatox.2006.10.002

     Leal. RB; Ribeiro. SJ: Posser. T; Cordova. FM; Rigon. AP; Filho. EZ; Bainy. ACD. (2006). Modulation
        of ERK1/2 and p38(MAPK) by lead in the cerebellum of Brazilian catfish Rhamdia  quelen. Aquat
        Toxicol 77: 98-104. http://dx.doi.Org/10.1016/i.aquatox.2005.ll.002

     Leland. HV; Kuwabara. JS. (1985). Trace metals. In GM Rand; SR Petrocelli (Eds.), Fundamentals of
        aquatic toxicology: Methods and applications (pp. 374-415). Washington, DC: Hemisphere
        Publishing.
November 2012                              7-269                  Draft - Do Not Cite or Quote

-------
     Levin. SA. (1998). Ecosystems and the biosphere as complex adaptive systems. Ecosystems 1: 431-436.
        http://dx.doi.org/10.1007/sl00219900037

     Li. LZ; Zhou. DM; Wang. P; Luo. XS. (2008b). Subcellular distribution of Cd and Pb in earthworm
        Eisenia fetida as affected by Ca2+ ions and Cd-Pb interaction. Ecotoxicol Environ Saf 71: 632-637.
        http://dx.doi.0rg/10.1016/i.ecoenv.2008.04.001

     Li. M; Liu. ZT; Xu. Y; Cui. YB; Li. PS: Kong. ZM. (2009b). Comparative effects of Cd and Pb on
        biochemical response and DNA damage in the earthworm Eisenia fetida (Annelida, Oligochaeta).
        Chemosphere 74: 621-625. http://dx.doi.Org/10.1016/i.chemosphere.2008.10.048

     Li. Y; Zu. Y; Fang. Q; Gao. Z; Schvartz. C. (2009d). Relationship between heavy metal concentrations of
        herbaceous plants and soils at four Pb-Zn mining sites in Yunnan, China. Front Environ Sci Eng China
        3: 325-333. http://dx.doi.org/10.1007/sll783-009-0024-5

     Ling. Q; Hong. F. (2009). Effects of Pb2+ on the structure and function of photosystem II of Spirodela
        polyrrhiza. Biol Trace Elem Res 129: 251-260. http://dx.doi.org/10.1007/sl2011-008-8283-8

     Ling. Q; Hong. F. (2010). Antioxidative role of cerium against the toxicity of lead in the liver of silver
        crucian carp. Fish Physiol Biochem 36: 367-376. http://dx.doi.org/10.1007/sl0695-008-9301-7

     Liu. DW: Liu. XM; Chen. ZW: Xu. HZ; Ding. XF (2010a). Bioaccumulation of lead and the effects of
        lead on catalase activity, glutathione levels, and chlorophyll content in the leaves of wheat. Commun
        Soil Sci Plant Anal 41: 935-944. http://dx.doi.org/10.1080/00103621003646022

     Lock. K; Waegeneers. N; Smolders. E;  Criel. P; Van Eeckhout H; Janssen. CR. (2006). Effect of leaching
        and aging on the bioavailability of lead to the springtail Folsomia Candida. Environ Toxicol Chem 25:
        2006-2010. http://dx.doi.Org/10.1897/05-612R.l

     Locke. LN; Thomas. NJ. (1996). Lead poisoning of waterfowl and raptors. In A Fairbrother; LN Locke;
        GL Hoff (Eds.), Noninfectious diseases of wildlife (2nd ed., pp.  108-117). Ames, IA: Iowa State Press.

     Lofts. S: Chapan. PM; Dwver. R; Mclaughlin. MJ; Choeters. I; Sheppard. SC: Adams. WJ: Allowav. BJ;
        Antunes. PM; Campbell. PG: Davies. B; Degryse. F; De Vries. W: Farley. K; Garrett. RG: Green. A;
        Jan Groenenberg. B; Hale. B; Harrass. M; Hendershot. WH; Keller. A; Lanno. R; Liang. T; Liu. WX;
        Ma. Y; Menzie. C: Moolenaar.  SW: Piatkiewicz. W: Reimann. C: Rieuwerts. JS: Santore. RC: Sauve.
        S: Schuetze. G: Schlekat. C: Skeaff. J: Smolders. E; Tao. S: Wilkins. J: Zhao. FJ. (2007). Critical loads
        of metals and other trace elements to terrestrial environments. Environ Sci Technol 41: 6326-6331.
        http://dx.doi.org/10.1021/es0726106

     Luoma. SN; Rainbow. PS. (2005).  Why is metal bioaccumulation so variable? Biodynamics as a unifying
        concept [Review]. Environ Sci  Technol 39: 1921-1931. http://dx.doi.org/10.1021/es048947e

     Ma. L; Xu. R; Jiang. J. (2010). Adsorption and desorption of Cu(II) and Pb(II) in paddy soils cultivated
        for various years in the subtropical China. J Environ Sci 22: 689-695.
        http://dx.doi.org/10.1016/S1001-0742(09)60164-9

     Ma. Y; Dickinson. NM; Wong. MH. (2006). Beneficial effects of earthworms and arbuscular mycorrhizal
        fungi on establishment of leguminous trees on Pb/Zn mine tailings. Soil Biol Biochem 38: 1403-1412.
        http://dx.doi.0rg/10.1016/i.soilbio.2005.10.016

     MacDonald. DP; Ingersoll.  CG: Berger. TA. (2000). Development and evaluation of consensus-based
        sediment quality guidelines for freshwater ecosystems. Arch Environ Contam Toxicol 39: 20-31.
        http://dx.doi.org/10.1007/s002440010075

     MacDonald. DP; Smorong. DE; Ingersoll. CG: Besser. JM; Brumbaugh. WG: Kemble. NE; May. TE;
        Ivev. CD; Irving. S: O'Hare. M. (2009). Development and evaluation of sediment and pore-water
        toxicity thresholds to support sediment quality assessments in the Tri-state Mining District (TSMD),
        Missouri, Oklahoma and Kansas. Nanaimo, Canada: MacDonald Environmental Sciences.
        http://www.fws.gov/soutfawest/es/Oklahoma/Documents/Contaminants/01TSMD%20Biological%20D
        ata%20Evaluation%202008 Volume%201 Vers4 Feb09.pdf
November 2012                             7-270                  Draft - Do Not Cite or Quote

-------
     MacFarlane. GR: Koller. CE; Blomberg. SP. (2007). Accumulation and partitioning of heavy metals in
        mangroves: A synthesis of field-based studies. Chemosphere 69: 1454-1464.
        http://dx.doi.0rg/10.1016/i.chemosphere.2007.04.059

     MacFarlane. GR: Schreider. M; McLennan. B. (2006). Biomarkers of heavy metal contamination in the
        red fingered marsh crab, Parasesarma erythodactyla. Arch Environ Contam Toxicol 51: 584-593.
        http://dx.doi.org/10.1007/s00244-005-5067-4

     Mackie. GL. (1989). Tolerances of five benthic invertebrates to hydrogen ions and metals (Cd, Pb, Al).
        Arch Environ Contam Toxicol 18: 215-223. http://dx.doi.org/10.1007/BF01056206

     Mager.  EM. (2012). Lead. In Homeostasis and toxicology of non-essential metals. U.K.: Academic Press.
        http://www.elsevierdirect.com/ISBN/9780123786340/Fish-Phvsiologv-Homeostasis-and-Toxicologv-
        of-NonEssential-Metals

     Mager.  EM; Brix. KV; Gerdes. RM; Ryan. AC: Grosell. M. (2011 a). Effects of water chemistry on the
        chronic toxicity of lead to the  cladoceran, Ceriodaphnia dubia. Ecotoxicol Environ Saf 74: 238-243.
        http://dx.doi.0rg/10.1016/i.ecoenv.2010.ll.005

     Mager.  EM; Brix. KV: Grosell. M. (2010). Influence of bicarbonate and humic acid on effects of chronic
        waterborne lead exposure to the fathead minnow (Pimephales promelas). Aquat Toxicol 96: 135-144.
        http://dx.doi.0rg/10.1016/i.aquatox.2009.10.012

     Mager.  EM; Esbaugh. AJ; Brix. KV: Ryan. AC: Grosell. M. (20 lib). Influences of water chemistry on the
        acute toxicity of lead to Pimephales promelas and Ceriodaphnia dubia. Comp Biochem Physiol C
        Toxicol Pharmacol 153: 82-90. http://dx.doi.Org/10.1016/i.cbpc.2010.09.004

     Mager.  EM; Grosell. M. (2011). Effects of acute and chronic waterborne lead exposure on the swimming
        performance and aerobic  scope of fathead minnows (Pimephales promelas). Comp Biochem Physiol C
        Toxicol Pharmacol 154: 7-13.  http://dx.doi.Org/10.1016/i.cbpc.2011.03.002

     Mager.  EM; Wintz. H; Vulpe. CD; Brix. KV: Grosell. M. (2008). Toxicogenomics of water chemistry
        influence on chronic lead exposure to the fathead minnow (Pimephales promelas). Aquat Toxicol 87:
        200-209. http://dx.doi.0rg/10.1016/i.aquatox.2008.02.001

     Magrisso. S: Belkin. S: Erel. Y. (2009). Lead bioavailability in soil and soil components. Water Air Soil
        Pollut 202: 315-323. http://dx.doi.org/10.1007/sll270-009-9978-v

     Mahler. BJ; Van Metre. PC: Callender. E. (2006). Trends in metals in urban and reference lake sediments
        across the United States, 1970 to 2001. Environ Toxicol Chem 25: 1698-1709.

     Mahmoudi. E; Essid. N; Beyrem. H; Hedfi. A; Boufahja. F; Vitiello. P: Aissa. P. (2007).  Individual and
        combined effects of lead and zinc on a free-living marine nematode community: Results from
        microcosm experiments. Exp Mar Bio Ecol 343: 217-226.
        http://dx.doi.0rg/10.1016/i.jembe.2006.12.017

     Maitv. S: Roy. S: Bhattacharya. S: Chaudhury. S. (2011). Metallothionein responses in the earthworm
        Lampito mauritii (Kinberg) following lead  and zinc exposure: A promising tool for monitoring metal
        contamination. European Journal of Soil Biology 47: 69-71.
        http://dx.doi.0rg/10.1016/i.eisobi.2010.10.001

     Maitv. S: Roy. S: Chaudhury. S: Bhattacharya. S. (2008). Antioxidant responses of the earthworm
        Lampito mauritii exposed to Pb and Zn contaminated soil. Environ Pollut 151: 1-7.
        http://dx.doi.0rg/10.1016/i.envpol.2007.03.005

     Malik. N; Biswas. AK; Qureshi. TA; Borana. K; Virha. R. (2010). Bioaccumulation of heavy metals in
        fish tissues of a freshwater lake of Bhopal.  Environ Monit Assess 160: 267-276.
        http://dx.doi.org/10.1007/sl0661-008-0693-8

     Mander. U; Mitsch. WJ. (2009). Pollution control by wetlands. Ecol Eng 35: 153-158.
        http://dx.doi.0rg/10.1016/i.ecoleng.2008.10.005
November 2012                             7-271                  Draft - Do Not Cite or Quote

-------
     Manzo. S: Buono. S: Cremisini. C. (2010). Cadmium, lead and their mixtures with copper: Paracentrotus
        lividus embryotoxicity assessment, prediction, and offspring quality evaluation. Ecotoxicology 19:
        1209-1223. http://dx.doi.org/10.1007/sl0646-010-0506-z

     Marchand. L; Mench. M; Jacob. PL; Otte. ML. (2010). Metal and metalloid removal in constructed
        wetlands, with emphasis on the importance of plants and standardized measurements: A review
        [Review]. Environ Pollut 158:  3447-3467. http://dx.doi.Org/10.1016/i.envpol.2010.08.018

     Maret. TR; Cain. DJ; MacCov. DE; Short. TM. (2003).  Response of benthic invertebrate assemblages to
        metal exposure and bioaccumulation associated with hard-rock mining in northwestern streams, USA.
        J North Am Benthol Soc 22: 598-620.

     Marin-Guirao. L; Atucha. AM; Barba. JL; Lopez. EM; Fernandez. AJG. (2005). Effects of mining wastes
        on a seagrass ecosystem: Metal accumulation and bioavailability, seagrass dynamics and  associated
        community structure. Mar Environ Res 60: 317-337.
        http://dx.doi.0rg/10.1016/i.marenvres.2004.ll.002

     Martin. M; Osborn. KE; Billig. P; Glickstein. N. (1981). Toxicities often metals to Crassostrea gigas and
        Mytilus edulis embryos and Cancer magister larvae. Mar Pollut Bull 12: 305-308.
        http://dx.doi.org/10.1016/0025-326X(81)90081-3

     Martinez-Haro. M; Green. AJ; Mateo. R. (2011). Effects of lead exposure on oxidative stress biomarkers
        and plasma biochemistry in waterbirds in the field. Environ Res  111: 530-538.
        http://dx.doi.0rg/10.1016/i.envres.2011.02.012

     McGlothan. JL; Karcz-Kubicha. M; Guilarte. TR. (2008). Developmental lead exposure impairs
        extinction of conditioned fear in young adult rats. Neurotoxicology 29: 1127-1130.
        http://dx.doi.0rg/10.1016/i.neuro.2008.06.010

     McLaughlin. MJ; Lofts. S: Warne. M; Amorim. MJB; Fairbrother. A; Lanno. R; Hendershot. W: Schlekat.
        CE; Ma. Y; Paton. GI.  (2010).  Derivation of ecologically-based soil standards for trace elements. In G
        Merrington; I Schoeters (Eds.), Soil quality standards for trace elements: Derivation, implementation,
        and interpretation (pp.  7-80). Boca Raton, FL: CRC Press, http://dx.doi.org/10.1201/bl0293-3

     Mebane. CA; Hennessv. DP; Dillon. FS. (2008). Developing  acute-to-chronic toxicity ratios  for lead,
        cadmium, and zinc using rainbow trout, a mayfly, and a midge. Water Air Soil Pollut 188: 41-66.
        http://dx.doi.org/10.1007/sll270-007-9524-8

     Mehrotra. V: Saxena. VL; Saxena. AK. (2008). Impact of different doses of lead on internal organs of
        quails. J Environ Biol 29:  147-149.

     Mellem. JJ; Baijnath. H; Odhav. B. (2009). Translocation and accumulation of Cr, Hg, As, Pb, Cu and Ni
        by Amaranthus dubius (Amaranthaceae) from contaminated sites. J Environ Sci Health A Tox Hazard
        Subst Environ Eng 44: 568-575. http://dx.doi.org/10.1080/10934520902784583

     Menta. C: Maggiani. A; Vattuone. Z. (2006). Effects of Cd and Pb on the survival and juvenile production
        of Sinella coeca and Folsomia  Candida. European Journal of Soil Biology 42: 181-189.
        http://dx.doi.0rg/10.1016/i.ejsobi.2006.01.001

     Metian. M; Warnau. M; Oberhansli. F; Bustamante. P. (2009). Delineation of Pb contamination pathways
        in two Pectinidae: The variegated scallop Chlamys varia and the king scallop Pecten maximus. Sci
        Total Environ 407: 3503-3509. http://dx.doi.Org/10.1016/i.scitotenv.2009.02.010

     Meyer. C: Gilbert. D;  Gaudry. A; Franchi. M; Nguyen-Viet. H; Fabure. J: Bernard. N. (2010).
        Relationship  of atmospheric pollution characterized by gas (NO2) and particles (PM10) to microbial
        communities living in bryophytes at three differently polluted sites (rural, urban, and industrial).
        Microb Ecol  59: 324-334.  http://dx.doi.org/10.1007/s00248-009-9580-2

     Meyer. JS: Adams. WJ: Brix.  KV; Luoma. SN; Mount.  PR; Stubblefield. WA; Wood. CM. (2005).
        Toxicity of dietborne metals to aquatic organisms. In JS Meyer; WJ Adams; KV Brix; SN Luoma; DR
        Mount; WA Stubblefield; CM Wood (Eds.). Pensacola, FL: Society of Environmental  Toxicology and
        Chemistry.
November 2012                             7-272                   Draft - Do Not Cite or Quote

-------
     Migula. P; Binkowska. K. (1993). Feeding strategies of grasshoppers (Chorthippus spp) on heavy metal
        contaminated plants. Sci Total Environ 134: 1071-1083.
        http://dx.doi.org/10.1016/S0048-9697(05)80112-3

     Miller. JR; Anderson. JB; Lechler. PJ; Kondrad. SL; Galbreath. PF; Salter. EB. (2005). Influence of
        temporal variations in water chemistry on the Pb isotopic composition of rainbow trout
        (Oncorhynchus mykiss). Sci Total Environ 350: 204-224.
        http://dx.doi.0rg/10.1016/i.scitotenv.2005.01.030

     Miramand. P; Bustamante. P; Bentlev. D; Koueta. N. (2006). Variation of heavy metal concentrations
        (Ag, Cd, Co, Cu, Fe, Pb, V, and Zn) during the life cycle of the common cuttlefish Sepia officinalis.
        Sci Total Environ 361: 132-143. http://dx.doi.Org/10.1016/j.scitotenv.2005.10.018

     Miretzky. P; Munoz. C: Carrillo-Chavez. A.  (2007). A sandy loam soil as a natural control for Pb
        contamination. Environ Chem Lett 5: 131-136. http://dx.doi.org/10.1007/sl0311-007-0093-2

     Mishra. S: Srivastava. S: Tripathi. RD; Kumar. R; Seth. CS: Gupta. DK. (2006b). Lead detoxification by
        coontail (Ceratophyllum demersum L.) involves induction of phytochelatins and antioxidant system in
        response to its accumulation.  Chemosphere 65: 1027-1039.
        http://dx.doi.0rg/10.1016/i.chemosphere.2006.03.033

     Mishra. VK; Upadhyav. AR; Pandev. SK; Tripathi. BD. (2008). Concentrations of heavy metals and
        aquatic macrophytes of Govind Ballabh Pant Sagar an anthropogenic lake affected by coal mining
        effluent. Environ Monit Assess  141: 49-58. http://dx.doi.org/10.1007/sl0661-007-9877-x

     Mo. MH; Chen. WM; Yang. HR; Zhang. KQ. (2008). Diversity and metal tolerance of nematode-trapping
        fungi in Pb-polluted soils. J Microbiol 46: 16-22. http://dx.doi.org/10.1007/sl2275-007-0174-8

     Mogren. CL; Trumble. JT. (2010). The impacts of metals and metalloids on insect behavior. Entomol Exp
        Appl 135: 1-17.  http://dx.doi.0rg/10.llll/i.1570-7458.2010.00967.x

     Mohapatra. A; Rautrav. TR; Patra. AK; Vijavan. V; Mohantv. RK. (2009). Trace element-based food
        value evaluation in soft and hard shelled  mud crabs. Food Chem Toxicol 47: 2730-2734.
        http://dx.doi.0rg/10.1016/i.fct.2009.07.037

     Monteiro. V; Cavalcante. DG: Vilela. MB; Sofia. SH; Martinez. CB. (2011). In vivo and in vitro
        exposures for the evaluation of the genotoxic effects of lead on the Neotropical freshwater fish
        Prochilodus lineatus. Aquat Toxicol 104: 291-298. http://dx.doi.Org/10.1016/j.aquatox.2011.05.002

     Montufar-Melendez. AI; Sanchez-Ortiz. JR; Sarma. SSS: Nandini. S. (2007). Combined effects of
        temperature and lead concentration on the competition between the rotifers Brachionus havanaensis
        and Brachionus rubens (Rotifera: Brachionidae). J Environ Sci Health A Tox Hazard Subst Environ
        Eng 42:  393-398. http://dx.doi.org/10.1080/10934520601144717

     Morris. S: van Aardt. WJ: Ahern. MD. (2005). The effect of lead on the metabolic and energetic status of
        the Yabby, Cherax destructor, during environmental hypoxia. Aquat Toxicol 75:  16-31.
        http://dx.doi.0rg/10.1016/i.aquatox.2005.07.001

     Morselli. L; Bernardi. E; Passarini. F; Tesini. E. (2006). Critical loads for Cd and Pb in the province of
        Bologna. Ann Chim 96: 697-705. http://dx.doi.org/10.1002/adic.200690072

     Mosher. S: Cope. WG: Weber. FX; Shea. D; Kwak. TJ. (2012). Effects of lead on Na(+), K(+)ATPase and
        hemolymph ion  concentrations in the freshwater mussel Elliptic complanata. Environ Toxicol  27:
        268-276. http://dx.doi.org/10.1002/tox.20639

     Mouchet. F; Cren. S: Cunienq. C: Devdier. E; Guilet. R; Gauthier. L. (2007). Assessment of lead
        ecotoxicity in water using the amphibian larvae (Xenopus laevis) and preliminary study of its
        immobilization in meat and bone meal combustion residues. Biometals 20: 113-127.
        http://dx.doi.org/10.1007/sl0534-006-9019-x

     Mulhern. BM; Reichel. WL; Locke. LN; Lamont. TG: Belisle. A; Cromartie. E; Baglev. GE; Proutv. RM.
        (1970). Organochlorine residues and autopsy data from bald eagles 1966-68. Pestic Monit J 4:
        141-144.
November 2012                              7-273                   Draft - Do Not Cite or Quote

-------
     Murray. H; Thompson. K; Macfie. SM. (2009). Site- and species-specific patterns of metal bioavailability
        in edible plants. Botany 87: 702-711. http://dx.doi.org/10.1139/b09-019

     Nahmani. J; Hodson. ME; Black.  S. (2007). Effects of metals on life cycle parameters of the earthworm
        Eisenia fetida exposed to field-contaminated, metal-polluted soils. Environ Pollut 149: 44-58.
        http://dx.doi.0rg/10.1016/i.envpol.2006.12.018

     Nahmani. J; Hodson. ME; Devin.  S: Vijver. MG. (2009). Uptake kinetics of metals by the earthworm
        Eisenia fetida exposed to field-contaminated soils. Environ Pollut 157: 2622-2628.
        http://dx.doi.0rg/10.1016/i.envpol.2009.05.002

     Nain. S: Smits. JE. (2011). Subchronic lead exposure, immunotoxicology and increased disease resistance
        in Japanese quail (Corturnix coturnix japonica). Ecotoxicol Environ Saf 74: 787-792.
        http://dx.doi.0rg/10.1016/i.ecoenv.2010.10.045

     Nakavama. SM; Ikenaka. Y; Hamada. K; Muzandu. K; Choongo. K; Teraoka. H; Mizuno. N; Ishizuka. M.
        (2011). Metal and metalloid contamination in roadside soil  and wild rats around a Pb-Zn mine in
        Kabwe, Zambia. Environ Pollut 159: 175-181. http://dx.doi.Org/10.1016/i.envpol.2010.09.007

     Nakhle. KF; Cossa. D; Khalaf. G: Beliaeff. B. (2006). Brachidontes variabilis and Patella sp as
        quantitative biological indicators for cadmium, lead and mercury in the Lebanese coastal waters.
        Environ Pollut 142: 73-82. http://dx.doi.Org/10.1016/i.envpol.2005.09.016

     Nam. DH; Lee. DP. (2006). Monitoring for Pb and Cd pollution using feral pigeons in rural, urban,  and
        industrial environments of Korea. Sci Total Environ 357: 288-295.
        http://dx.doi.0rg/10.1016/i.scitotenv.2005.08.017

     Nandi.  S: Gupta. PS: Selvaraiu. S: Roy. SC: Ravindra. JP (2010). Effects of exposure to heavy metals on
        viability, maturation, fertilization, and embryonic development of buffalo  (Bubalus bubalis) oocytes in
        vitro. Arch Environ Contam Toxicol 58:  194-204. http://dx.doi.org/10.1007/s00244-009-9342-7

     Nguyen-Viet. H; Bernard. N; Mitchell. BAD; Badot. PM; Gilbert. D. (2008).  Effect of lead pollution on
        testate amoebae communities living in Sphagnum fallax: An experimental study. Ecotoxicol Environ
        Saf 69: 130-138. http://dx.doi.Org/10.1016/i.ecoenv.2007.02.007

     Nilsson. J; Grennfelt. P. (1988). Critical loads for sulphur and nitrogen: Report from a workshop held at
        Skokloster, Sweden, 19-24 March 1988. Skokloster, Sweden: Nordic Council of Ministers:
        Copenhagen.

     Nimptsch. J: Wunderlin. DA; Dollan. A; Pflugmacher. S. (2005). Antioxidant and biotransformation
        enzymes in Myriophyllum quitense as biomarkers of heavy metal exposure and eutrophication in
        Suquia River basin (Cordoba, Argentina). Chemosphere 61: 147-157.
        http://dx.doi.0rg/10.1016/i.chemosphere.2005.02.079

     Niyogi. S: Wood. CM. (2004). Biotic ligand model, a flexible tool for developing site-specific water
        quality guidelines for metals [Review]. Environ Sci Technol 38: 6177-6192.
        http://dx.doi.org/10.1021/es0496524

     NOAA (National Oceanic and Atmospheric Administration). (1999). Sediment quality guidelines
        developed for the national status and trends program. Silver Spring, MD: National Ocean Service.

     Notten. MJM; Oosthoek. AJP; Rozema. J: Aerts. R. (2005). Heavy metal concentrations in a soil-plant-
        snail food chain along  a terrestrial soil pollution gradient. Environ Pollut 138: 178-190.
        http://dx.doi.0rg/10.1016/i.envpol.2005.01.011

     Notten. MJM: Oosthoek. AJP: Rozema. J: Aerts. R. (2006). Heavy metal pollution affects consumption
        and reproduction of the landsnail Cepaea nemoralis fed on naturally polluted Urtica dioica leaves.
        Ecotoxicology 15: 295-304.  http://dx.doi.org/10.1007/sl0646-006-0059-3

     Notten. MJM: Walraven. N; Beets. CJ: Vroon. P; Rozema. J: Aerts. R.  (2008). Investigating the origin of
        Pb pollution in a terrestrial soil-plant-snail food chain by means of Pb isotope ratios. Appl Geochem
        23:  1581-1593. http://dx.doi.0rg/10.1016/i.apgeochem.2008.01.010
November 2012                              7-274                  Draft - Do Not Cite or Quote

-------
     NFS (U.S. National Park Service). (2011). WACAP database.
        http://www.nature.nps.gov/air/studies/air toxics/WACAPreport.cfm

     Nwachukwu. OI; Pulford. ID. (2011). Microbial respiration as an indication of metal toxicity in
        contaminated organic materials and soil. J Hazard Mater 185: 1140-1147.
        http://dx.doi.0rg/10.1016/i.jhazmat.2010.10.024

     Odjegba. VJ: Fasidi. IP. (2006). Effects of heavy metals on some proximate composition of Eichhornia
        crassipes. J Appl Sci Environ Manag 10: 83-87.

     Odum. HT. (2000). Heavy metals in the environment: Using wetlands for their removal. Boca Raton, FL:
        Lewis Publishers.

     Ojo. AA; Wood. CM. (2007). In vitro analysis of the bioavailability of six metals via the gastro-intestinal
        tract of the rainbow trout (Oncorhynchus mykiss). Aquat Toxicol 83: 10-23.
        http://dx.doi.0rg/10.1016/i.aquatox.2007.03.006

     Oliveira Ribeiro. CA; Filipak Neto. F; Mela. M; Silva. PH; Randi. MAP; Rabitto. IS: Alves Costa. JRM;
        Pelletier. E. (2006). Hematological findings in neotropical fish Hoplias malabaricus exposed to
        subchronic and dietary doses of methylmercury, inorganic lead, and tributyltin chloride. Environ Res
        101: 74-80. http://dx.doi.0rg/10.1016/j.envres.2005.ll.005

     Osman. AGM; McKkawv. IA; Verreth. J: Kirschbaum. F. (2007a). Effects of lead nitrate on the activity of
        metabolic enzymes during early developmental stages of the African catfish, Clarias gariepinus
        (Burchell, 1822). Fish Physiol Biochem 33: 1-13. http://dx.doi.org/10.1007/sl0695-006-9111-8

     Osman. AGM: Wuertz. S: Mekkawv. IA; Exner. HJ; Kirschbaum. F. (2007b). Lead induced
        malformations in embryos of the African catfish Clarias gariepinus (Burchell, 1822). Environ Toxicol
        22: 375-389. http://dx.doi.org/10.1002/tox.20272

     Owojori. OJ; Healev. J: Princz. J: Siciliano. SD. (2011). Can avoidance behavior of the mite Oppia nitens
        be used as a rapid toxicity test for soils contaminated with metals or organic chemicals. Environ
        Toxicol Chem 30: 2594-2601. http://dx.doi.org/10.1002/etc.658

     Paczkowska. M; Kozlowska. M; Golinski. P. (2007). Oxidative stress enzyme activity  in Lemna minor
        L. exposed to cadmium and lead. Acta Biol Cracov Bot 49: 33-37.

     Palaniappan. PR; Krishnakumar. N; Vadivelu. M. (2009). Bioaccumulation of lead and the influence of
        chelating agents in Catla catla fmgerlings. Environ Chem Lett 7: 51-54.
        http://dx.doi.org/10.1007/sl0311-008-0134-5

     Paauin. PR: Gorsuch. JW: Apte. S: Batlev. GE: Bowles. KG: Campbell. PGC: Delos. CG: Pi Toro. DM:
        Dwver. RL; Galvez. F; Gensemer. RW: Goss. GG: Hogstrand. C: Janssen. CR; McGeer. JC: Naddv.
        RB; Plavle. RC: Santore. RC: Schneider. U; Stubblefield. WA; Wood. CM; Wu. KB. (2002). The
        biotic ligand model: A historical overview [Review]. Comp Biochem Physiol C Toxicol Pharmacol
        133: 3-35. http://dx.doi.org/10.1016/S1532-0456(02)00112-6

     Patel. M; Rogers. JT; Pane. EF; Wood. CM. (2006). Renal responses to acute lead waterborne exposure in
        the freshwater rainbow trout (Oncorhynchus mykiss). Aquat Toxicol 80: 362-371.
        http://dx.doi.0rg/10.1016/i.aquatox.2006.09.012

     Patfaak. SP; Gopal. K. (2009). Bacterial density and antibiotic resistance of Aeromonas sp. in organs of
        metal-stressed freshwater fish Channa punctatus. Toxicol Environ Chem 91: 331-337.
        http://dx.doi.org/10.1080/02772240802098222

     Patrick. GJ: Farmer. JG. (2006). A stable lead isotopic investigation of the use of sycamore tree rings as a
        historical biomonitor of environmental lead contamination. Sci Total Environ 362:  278-291.
        http://dx.doi.0rg/10.1016/i.scitotenv.2005.12.004

     Patrick. L. (2006). Lead toxicity, a review of the literature. Part I: Exposure, evaluation, and treatment
        [Review]. Altern Med Rev 11: 2-22.
November 2012                             7-275                  Draft - Do Not Cite or Quote

-------
     Pauget B; Gimbert. F; Coeurdassier. M; Scheifler. R; de Vaufleury. A. (2011). Use of chemical methods
        to assess Cd and Pb bioavailability to the snail Cantareus aspersus: A first attempt taking into account
        soil characteristics. J Hazard Mater 192: 1804-1811. http://dx.doi.Org/10.1016/i.jhazmat.2011.07.016

     Pearce. NJG: Mann. VL. (2006). Trace metal variations in the shells of Ensis siliqua record pollution and
        environmental conditions in the sea to the west of mainland Britain. Mar Pollut Bull 52: 739-755.
        http://dx.doi.0rg/10.1016/i.marpolbul.2005.ll.003

     Penttinen. OP; Kilpi-Koski. J; Jokela. M; Toivainen. K; Vaisanen. A. (2008). Importance of dose metrics
        for lethal and sublethal sediment metal toxicity in the oligochaete worm Lumbriculus variegates.
        Journal of Soils and Sediments 8: 59-66. http://dx.doi.org/10.1065/iss2007.12.267

     Peralta-Videa. JR; Lopez. ML; Naravan. M; Saupe. G: Gardea-Torresdev. J. (2009). The biochemistry of
        environmental heavy metal uptake by plants: Implications for the food chain. Int J Biochem Cell Biol
        41: 1665-1677. http://dx.doi.Org/10.1016/i.biocel.2009.03.005

     Pereira. AA; van Hattum. B; de Boer. J: van Bodegom. PM; Rezende. CE; Salomons. W. (2010). Trace
        elements and carbon and nitrogen stable isotopes in organisms from a tropical coastal lagoon. Arch
        Environ Contam Toxicol 59: 464-477. http://dx.doi.org/10.1007/s00244-010-9489-2

     Perez-Legaspi. IA; Rico-Martinez. R. (2001). Acute toxicity tests on three species of the genus Lecane
        (Rotifera: Monogononta). Hydrobiologia 446/447:  375-381.
        http://dx.doi.Org/10.1023/A:1017531712808

     Pernet-Coudrier. B; Companys. E; Galceran. J: Morev. M; Mouchel. JM; Puy. J: Ruiz. N; Varrault. G.
        (2011). Pb-binding to various dissolved organic matter in urban aquatic systems: Key role of the most
        hydrophilic fraction. Geochim Cosmo Act 75: 4005-4019. http://dx.doi.Org/10.1016/j.gca.2011.04.030

     Perugini. M; Manera. M; Grotta. L; Abete. MC: Tarasco. R: Amorena. M. (2011). Heavy metal (Hg, Cr,
        Cd, and Pb) contamination in urban areas and wildlife reserves: Honeybees as bioindicators. Biol
        Trace Elem Res 140: 170-176. http://dx.doi.org/10.1007/sl2011-010-8688-z

     Peterson. G: Allen. CR: Holling. CS. (1998). Ecological resilience, biodiversity, and scale. Ecosystems  1:
        6-18. http://dx.doi.org/10.1007/sl00219900002

     Petraitis. PS: Latham. RE. (1999). The importance of scale in testing the origins of alternative community
        states. Ecology 80: 429-442. http://dx.doi.org/10.1890/0012-9658(1999)080r0429:TIOSIT12.0.CO:2

     Pimentel. D; Wilson. C: McCullum. C: Huang. R; Dwen. P; Flack. J: Tran. Q; Saltman. T; Cliff. B.
        (1997). Economic and environmental benefits of biodiversity. Bioscience 47: 747-757.

     Pimm. SL. (1984). The complexity and stability of ecosystems [Review]. Nature 307: 321-326.
        http://dx.doi.org/10.1038/307321aO

     Piotrowska. A; Bajguz. A; Godlewska-Zylkiewicz . B; Zambrzycka. E. (2010). Changes in growth,
        biochemical components, and antioxidant activity in aquatic plant Wolffia arrhiza (Lemnaceae)
        exposed to cadmium and lead. Arch Environ Contam Toxicol 58: 594-604.
        http://dx.doi.org/10.1007/s00244-009-9408-6

     Poulton. BC: Allert. AL; Besser. JM; Schmitt. CJ: Brumbaugh. WG: Fairchild. JF. (2010). A
        macroinvertebrate assessment of Ozark streams located in lead-zinc mining areas of the Viburnum
        Trend in southeastern Missouri, USA. Environ Monit Assess 163: 619-641.
        http://dx.doi.org/10.1007/sl0661-009-0864-2

     Probst. A; Liu. H; Fanjul. M; Liao. B; Hollande. E. (2009). Response of Vicia faba L. to metal toxicity on
        mine tailing substrate: Geochemical and morphological changes in leaf and root. Environ Exp Bot 66:
        297-308. http://dx.doi.0rg/10.1016/i.envexpbot.2009.02.003

     Quan. W: Zhu. J: Ni. Y; Shi. L; Chen. Y. (2009). Fauna! utilization of constructed intertidal oyster
        (Crassostrea rivularis) reef in the Yangtze River estuary, China. Ecol Eng 35: 1466-1475.
        http://dx.doi.0rg/10.1016/i.ecoleng.2009.06.001
November 2012                              7-276                   Draft - Do Not Cite or Quote

-------
     Rabitto. IS: Alves Costa. JRM; Silva de Assis. HC: Pelletier. E; Akaishi. FM; Anjos. A; Randi. MAP;
        Oliveira Ribeiro. CA. (2005). Effects of dietary Pb(II) and tributyltin on neotropical fish, Hoplias
        malabaricus: Histopathological and biochemical findings. Ecotoxicol Environ Saf 60: 147-156.
        http://dx.doi.0rg/10.1016/i.ecoenv.2004.03.002

     Radwan. MA; El-Gendv. KS: Gad. AF. (2010). Oxidative stress biomarkers in the digestive gland of
        Theba pisana exposed to heavy metals. Arch Environ Contam Toxicol 58: 828-835.
        http://dx.doi.org/10.1007/s00244-009-9380-l

     Rai. PK. (2008). Heavy metal pollution in aquatic ecosystems and its phytoremediation using wetland
        plants: An ecosustainable approach [Review]. Int J Phytoremediation 10: 133-160.
        http://dx.doi.org/10.1080/15226510801913918

     Raimundo. J: Vale. C: Duarte. R; Moura. I. (2008).  Sub-cellular partitioning of Zn, Cu, Cd and Pb in the
        digestive gland of native Octopus vulgaris exposed to different metal concentrations (Portugal). Sci
        Total Environ 390: 410-416. http://dx.doi.Org/10.1016/i.scitotenv.2007.10.029

     Rainbow. PS. (1996). Heavy metals in aquatic invertebrates. In WN Beyer; GH Heinz; AW Redmon-
        Norwood (Eds.), Environmental contaminants in wildlife: Interpreting tissue concentrations (pp.
        405-425). Boca Raton, FL: CRC Press.

     Rainbow. PS; Poirier. L; Smith. BD; Brix. KV; Luoma.  SN. (2006). Trophic transfer of trace metals from
        the polychaete worm Nereis diversicolor to the polychaete N-virens and the decapod crustacean
        Palaemonetes varians. Mar Ecol Prog Ser 321: 167-181. http://dx.doi.org/10.3354/meps321167

     Ramesh. M; Saravanan. M; Kavitha. C. (2009). Hormonal responses of the fish, Cyprinus carpio, to
        environmental lead exposure. Afr J Biotechnol 8: 4154-4158.

     Rau. S; Miersch. J; Neumann. D; Weber. E; Krauss. GJ. (2007). Biochemical responses of the aquatic
        moss Fontinalis  antipyretica to Cd, Cu, Pb and Zn determined by chlorophyll fluorescence and protein
        levels. Environ Exp Bot 59: 299-306. http://dx.doi.Org/10.1016/j.envexpbot.2006.03.001

     Reddv. AM; Kumar. SG; Jvothsnakumari.  G; Thimmanaik. S; Sudhakar. C. (2005). Lead induced changes
        in antioxidant metabolism of horsegram (Macrotyloma uniflorum (Lam.) Verde.) and bengalgram
        (Cicer arietinum L.). Chemosphere 60: 97-104. http://dx.doi.Org/10.1016/i.chemosphere.2004.ll.092

     Reglero. MM; Monsalve-Gonzalez. L; Taggart. MA; Mateo. R. (2008). Transfer of metals to plants and
        red deer in an old lead mining area in Spain.  Sci Total Environ 406: 287-297.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.06.001

     Reglero. MM; Taggart. MA;  Castellanos. P; Mateo. R. (2009a). Reduced sperm quality in relation to
        oxidative stress in red deer from a lead mining area.  Environ Pollut 157: 2209-2215.
        http://dx.doi.0rg/10.1016/i.envpol.2009.04.017

     Reichmuth. JM; Weis. P; Weis. JS. (2010). Bioaccumulation and depuration of metals in blue crabs
        (Callinectes sapidus Rathbun) from a contaminated and clean estuary. Environ Pollut 158: 361-368.
        http://dx.doi.0rg/10.1016/i.envpol.2009.09.009

     Reinfelder. JR; Fisher. NS; Luoma. SN; Nichols. JW; Wang. WX. (1998). Trace element trophic transfer
        in aquatic organisms: A critique of the kinetic model approach. Sci Total Environ 219: 117-135.
        http://dx.doi.org/10.1016/S0048-9697(98)00225-3

     Reynolds. KD; Schwarz. MS; Mcfarland. CA; Mcbride. T; Adair. B. (2006). Northern pocket gophers
        (Thomomys talpoides) as biomonitors of environmental metal contamination. Environ Toxicol Chem
        25: 458-469. http://dx.doi.Org/10.1897/05-130Rl.l

     Rhea. DT; Harper. DP; Farag. AM; Brumbaugh. WG. (2006). Biomonitoring in the Boulder River
        watershed, Montana, USA: Metal concentrations in biofilm and macroinvertebrates, and relations with
        macroinvertebrate assemblage. Environ Monit Assess 115: 381-393.
        http://dx.doi.org/10.1007/sl0661-006-7086-7
November 2012                             7-277                   Draft - Do Not Cite or Quote

-------
     Rice. C: Ghorai. JK; Zalewski. K; Weber. DN. (2011). Developmental lead exposure causes startle
        response deficits in zebrafish. Aquat Toxicol 105: 600-608.
        http://dx.doi.0rg/10.1016/i.aquatox.2011.08.014

     Ringenary. MJ; Molof. AH; Tanacredi. JT; Schreibman. MP; Kostarelos. K. (2007). Long-term sediment
        bioassay of lead toxicity in two generations of the marine amphipod Elasmopus laevis, SI Smith,
        1873. Environ Toxicol Chem 26: 1700-1710. http://dx.doi.Org/10.1897/06-303Rl.l

     Robinson. GR; Sibrell. PL; Boughton. CJ: Yang. LH. (2007). Influence of soil chemistry on metal and
        bioessential element concentrations in nymphal and adult periodical cicadas (Magicicada spp.). Sci
        Total Environ 374: 367-378. http://dx.doi.Org/10.1016/i.scitotenv.2006.12.031

     Rodriguez-Estival. J: Martinez-Haro. M; Monsalve-Gonzalez. L; Mateo. R. (2011). Interactions between
        endogenous and dietary antioxidants against Pb-induced oxidative stress in wild ungulates from a Pb
        polluted mining area. Sci Total Environ 409: 2725-2733.
        http://dx.doi.0rg/10.1016/i.scitotenv.2011.04.010

     Rogers. JT; Patel. M; Gilmour. KM; Wood. CM. (2005). Mechanisms behind Pb-induced disruption of
        Na+ and Cl- balance in rainbow trout (Oncorhynchus mykiss). Am J Physiol Regul Integr Comp
        Physiol 58: R463-R472. http://dx.doi.org/10.1152/aipregu.00362.2004

     Rogers. JT: Wood. CM. (2004). Characterization of branchial lead-calcium interaction in the freshwater
        rainbow trout (Oncorhynchus mykiss). J Exp Biol 207: 813-825. http://dx.doi.org/10.1242/jeb.00826

     Rogers. JT: Wood. JG. (2003). lonoregulatory disruption as the acute toxic mechanism for lead in the
        rainbow trout (Oncorhynchus mykiss). Aquat Toxicol 64: 215-234.
        http://dx.doi.org/10.1016/S0166-445X(03)00053-5

     Rogival. D; Scheirs. J: Blust. R. (2007). Transfer and accumulation of metals in a soil-diet-wood mouse
        food chain along a metal pollution gradient. Environ Pollut 145: 516-528.
        http://dx.doi.0rg/10.1016/i.envpol.2006.04.019

     Rokytova. L; Kula. E; Kodarova. L; Peslova. A. (2004). Feeding of the willow leaf beetle Lochmaea
        capreae L. (Coleoptera, Chrysomelidae) on leaves of birch (Betula pendula Roth) contaminated by
        heavy metals. J Forest Sci 50: 109-117.

     Roodbergen. M; Klok. C: van der Hout. A. (2008).  Transfer of heavy metals in the food chain earthworm
        Black-tailed godwit (Limosa limosa): Comparison of a polluted and a reference site in The
        Netherlands. Sci Total Environ 406: 407-412. http://dx.doi.Org/10.1016/i.scitotenv.2008.06.051

     Roulier. JL; Tusseau-Nuillemin. MH; Coquerv. M;  Geffard. O: Garric. J. (2008a). Measurement of
        dynamic mobilization of trace metals in sediments using DOT and comparison with bioaccumulation
        in Chironomus riparius: First results of an experimental study. Chemosphere 70: 925-932.
        http://dx.doi.0rg/10.1016/i.chemosphere.2007.06.061

     Rozon-Ramilo. LD; Dube. MG: Squires. AJ; Niyogi. S. (2011). Examining waterborne and dietborne
        routes of exposure and their contribution to biological response patterns in fathead minnow
        (Pimephales promelas). Aquat Toxicol 105: 466-481. http://dx.doi.Org/10.1016/i.aquatox.2011.07.006

     Rubio-Franchini. I; Saavedra. JM; Rico-Martinez. R. (2008). Determination of lead in samples of
        zooplankton, water, and sediments in a Mexican reservoir: Evidence for lead biomagnification in
        lower/intermediate trophic levels? Environ Toxicol 23: 459-465. http://dx.doi.org/10.1002/tox.20357

     Ruelas-Inzunza. J: Paez-Osuna. F. (2008). Trophic distribution of Cd, Pb, and Zn in a food web from
        Altata-Ensenada del Pabellon subtropical lagoon, SE Gulf of California. Arch Environ Contam
        Toxicol 54: 584-596. http://dx.doi.org/10.1007/s00244-007-9075-4

     Ruiz. E; Alonso-Azcarate. J: Rodriguez. L. (2011).  Lumbricus terrestris L. activity increases the
        availability of metals and their accumulation in  maize and barley. Environ Pollut 159: 722-728.
        http://dx.doi.0rg/10.1016/i.envpol.2010.ll.032

     Sadiq. M. (1992). Lead in marine environments. In Toxic metal chemistry in marine environments. New
        York, NY: Marcel Dekker, Inc.
November 2012                              7-278                   Draft - Do Not Cite or Quote

-------
     Salice. CJ: Suski. JG: Bazar. MA; Talent. LG. (2009). Effects of inorganic lead on Western fence lizards
        (Sceloporus occidentalis). Environ Pollut 157: 3457-3464.
        http://dx.doi.0rg/10.1016/i.envpol.2009.06.013

     Samecka-Cymerman. A; Kempers. AJ. (2007). Heavy metals in aquatic macrophytes from two small
        rivers polluted by urban, agricultural and textile industry sewages SW Poland. Arch Environ Contam
        Toxicol 53: 198-206. http://dx.doi.org/10.1007/s00244-006-0059-6

     Sanchez-Marin. P..; Bellas. J..: Mubiana. V.. K.; Lorenzo. J.. I.; Blust. R..: Beiras. R.. (2011). Pb
        uptake by the marine mussel Mytilus sp. Interactions with dissolved organic matter. Aquat Toxicol
        102: 48-57. http://dx.doi.Org/10.1016/j.aquatox.2010.12.012

     Sanchez-Marin. P; Beiras. R.  (2011). Adsorption of different types of dissolved organic matter to marine
        phytoplankton and implications for phytoplankton growth and Pb bioavailability. Journal of Plankton
        Research 33: 1396-1409. http://dx.doi.org/10.1093/plankt/fbr039

     Sanchez-Marin. P; Lorenzo. JI; Blust.  R: Beiras. R. (2007). Humic acids increase dissolved lead
        bioavailability for marine  invertebrates. Environ Sci Technol 41: 5679-5684.
        http://dx.doi.org/10.1021/es070088h

     Sanchez-Marin. P; Santos-Echeandia.  J: Nieto-Cid. M; Alvarez-Salgado. XA; Beiras. R. (2010a). Effect
        of dissolved organic matter (DOM) of contrasting origins on Cu and Pb speciation and toxicity to
        Paracentrotus lividus larvae. Aquat Toxicol 96: 90-102.
        http://dx.doi.0rg/10.1016/i.aquatox.2009.10.005

     Sanchez-Marin. P; Slaveykova. VI; Beiras. R. (2010b). Cu and Pb accumulation by the marine diatom
        Thalassiosira weissflogii in the presence of humic acids.  Environ Chem 7: 309-317.
        http://dx.doi.org/10.1071/enl0015

     Savard. MM; Begin. C: Parent. M; Marion. J: Smirnoff. A. (2006). Dendrogeochemical distinction
        between geogenic and anthropogenic emissions of metals and gases near a copper smelter.
        Geochemistry: Exploration, Environment, Analysis 6: 237-247.
        http://dx.doi.org/10.1144/1467-7873/05-096

     Sawasdee. B; Kohler. HR. (2010). Metal sensitivity of the embryonic development of the ramshorn snail
        Marisa cornuarietis (Prosobranchia). Ecotoxicology 19:  1487-1495.
        http://dx.doi.org/10.1007/sl0646-010-0534-8

     Scheidegger. C: Behra. R: Sigg. L. (2011). Phytochelatin formation kinetics and toxic effects in the
        freshwater alga Chlamydomonas reinhardtii upon short- and long-term exposure to lead(II). Aquat
        Toxicol 101: 423-429. http://dx.doi.Org/10.1016/i.aauatox.2010.ll.016

     Scheifler. R; Coeurdassier.  M; Morilhat. C: Bernard. N; Faivre. B; Flicoteaux. P; Giraudoux. P; Noel. M;
        Piotte. P: Rieffel. D; de  Vaufleury. A; Badot. PM. (2006a). Lead concentrations in feathers and blood
        of common blackbirds (Turdus merula) and in earthworms inhabiting unpolluted and moderately
        polluted urban areas. Sci Total Environ 371:  197-205.
        http://dx.doi.0rg/10.1016/i.scitotenv.2006.09.011

     Scheifler. R; De Vaufleury. A; Coeurdassier. M;  Crini. N; Badot. PM. (2006b). Transfer of Cd, Cu, Ni, Pb,
        and Zn in a soil-plant-invertebrate food chain: A microcosm study. Environ Toxicol Chem 25:
        815-822. http://dx.doi.Org/10.1897/04-675R.l

     Schipper. AM; Wiinhoven.  S: Leuven. R; Ragas. AMJ; Hendriks. AJ. (2008). Spatial distribution and
        internal metal concentrations of terrestrial arthropods  in a moderately contaminated lowland
        floodplain along the Rhine River. Environ Pollut 151: 17-26.
        http://dx.doi.0rg/10.1016/i.envpol.2007.03.007

     Schmitt. CJ: Whyte. JJ: Brumbaugh. WG: Tillitt. DE. (2005). Biochemical effects of lead, zinc, and
        cadmium from mining on  fish  in the Tri-States District of northeastern Oklahoma, USA. Environ
        Toxicol Chem 24: 1483-1495.  http://dx.doi.Org/10.1897/04-332R.l
November 2012                              7-279                   Draft - Do Not Cite or Quote

-------
     Schmitt. CJ: Whyte. JJ: Roberts. AP; Annis. ML; May. TW: Tilitt. DE. (2007b). Biomarkers of metals
        exposure in fish from lead-zinc mining areas of southeastern Missouri, USA. Ecotoxicol Environ Saf
        67: 31-47. http://dx.doi.Org/10.1016/i.ecoenv.2006.12.011

     Schroth. AW: Bostick. BC: Kaste. JM; Friedland. AJ. (2008). Lead sequestration and species
        redistribution during soil organic matter decomposition. Environ Sci Technol 42: 3627-3633.
        http://dx.doi.org/10.1021/es703002b

     Semreen. MH; Aboul-Enein. HY. (2011). Determination of heavy metal content in wild-edible mushroom
        from Jordan. Anal Lett 44: 932-941. http://dx.doi.org/10.1080/00032711003790072

     Serafim. A; Lopes. B; Company. R; Cravo. A; Gomes. T; Sousa. V: Bebianno. MJ. (2011). Amulti-
        biomarker approach in cross-transplanted mussels Mytilus galloprovincialis. Ecotoxicology 20:
        1959-1974.  http://dx.doi.org/10.1007/sl0646-011-0737-7

     Shacklette. HT; Boerngen. JG. (1984). Element concentrations in soils and other surficial materials of the
        conterminous United States. (1270). Washington D.C.: Government Printing Office.
        http://pubs.usgs.gov/pp/1270/

     Shaheen. SM; Tsadilas. CD. (2009). Concentration of lead in soils and some vegetable plants in north
        Nile Delta as affected by soil type and irrigation water. Commun Soil Sci Plant Anal 40: 327-344.
        http://dx.doi.org/10.1080/00103620802649237

     Shakya. K; Chettri. MK; Sawidis. T. (2008). Impact of heavy metals (copper, zinc, and lead) on the
        chlorophyll content of  some mosses. Arch Environ Contam Toxicol 54: 412-421.
        http://dx.doi.org/10.1007/s00244-007-9060-y

     Shore. RF; Rattner. BA. (2001). Ecotoxicology of wild mammals. In RF Shore; BA Rattner (Eds.).
        Hoboken, NJ: John Wiley & Sons.

     Shu. YH; Zhou. JL; Tang.  WC: Lu. K; Zhou. Q; Zhang. GR. (2009). Molecular characterization and
        expression pattern of Spodoptera litura (Lepidoptera: Noctuidae) vitellogenin, and its response to lead
        stress. J Insect Physiol  55: 608-616. http://dx.doi.Org/10.1016/i.jinsphys.2009.03.005

     Shuhaimi-Othman. M; Yakub. N; Ramie. NA; Abas. A.  (2011 a). Sensitivity of the freshwater prawn,
        Macrobrachium lanchesteri (Crustacea: Decapoda), to heavy metals. Toxicol Ind Health 27: 523-530.
        http://dx.doi.org/10.1177/0748233710391993

     Shuhaimi-Othman. M; Yakub. N; Ramie. NA; Abas. A.  (20 lib). Toxicity of metals to a freshwater
        ostracod: Stenocypris major. Journal of Toxicology 2011: 136104.
        http://dx.doi.org/10.1155/2011/136104

     Shuhaimi-Othman. M; Yakub. N; Umirah. NS: Abas. A. (201 Ic). Toxicity of eight metals to Malaysian
        freshwater midge larvae Chironomus javanus (Diptera, Chironomidae). Toxicol Ind Health 27:
        879-886. http://dx.doi.org/10.1177/0748233711399318

     Simkiss. K; Jenkins. KGA; Mclellan.  J; Wheeler. E. (1982). Methods of metal incorporation into
        intracellular granules. Experientia 38: 333-335. http://dx.doi.org/10.1007/BF01949375

     Simpson. SL; Batlev. GE.  (2007). Predicting metal toxicity in sediments: A critique of current approaches
        [Review]. Integr Environ Assess Manag 3: 18-31. http://dx.doi.org/10.1002/ieam.5630030103

     Singer. C: Zimmermann. S: Sures. B.  (2005).  Induction of heat  shock proteins (hsp70) in the zebra mussel
        (Dreissena polymorpha) following exposure to platinum group metals (platinum, palladium and
        rhodium): Comparison with lead and cadmium exposures. Aquat Toxicol 75: 65-75.
        http://dx.doi.0rg/10.1016/i.aquatox.2005.07.004

     Singh. R; Tripathi. RD; Dwivedi. S: Kumar. A; Trivedi. PK; Chakrabartv. D. (2010). Lead
        bioaccumulation potential of an aquatic macrophyte  Najas indica are related to antioxidant system.
        Bioresour Technol 101: 3025-3032. http://dx.doi.Org/10.1016/i.biortech.2009.12.031

     Sizmur. T; Hodson. ME. (2009). Do earthworms impact metal mobility and availability in soil? A review
        [Review]. Environ Pollut 157: 1981-1989. http://dx.doi.Org/10.1016/i.envpol.2009.02.029
November 2012                             7-280                   Draft - Do Not Cite or Quote

-------
     Sizmur. T; Hodson. MF. (2008). The impact of Eisenia veneta on As, Cu, Pb and Zn uptake by ryegrass
        (Lolium perenne L.). Mineral Mag 72: 495-499. http://dx.doi.Org/10.1180/minmag.2008.072.l.495

     Sizmur. T; Tilston. EL; Charnock. J; Palumbo-Roe. B; Watts. MJ; Hodson. ME. (2011). Impacts of
        epigeic, anecic and endogeic earthworms on metal and metalloid mobility and availability. J Environ
        Monit 13: 266-273. http://dx.doi.org/10.1039/cOem00519c

     Skorbiowicz. E. (2006).  Cadmium and lead in bottom sediments and reed sweet grass (Glycerii aquatica)
        roots in selected rivers of upper Narew river basins. Pol J Environ Stud 15: 482-489.

     Slaveykova. VI; Wilkinson. KJ. (2005). Predicting the bioavailability of metals and metal complexes:
        Critical review of the biotic ligand model [Review]. Environ Chem 2: 9-24.
        http://dx.doi.org/10.1071/EN04076

     Sloman. KA; Lepage. O: Rogers. JT; Wood. CM; Winberg. S. (2005). Socially-mediated differences in
        brain monoamines in rainbow trout: Effects of trace metal contaminants. Aquat Toxicol 71: 237-247.
        http://dx.doi.0rg/10.1016/i.aquatox.2004.ll.008

     Smith. DM; Mielke. HW: Heneghan. JB. (2009a). Subchronic lead feeding study in male rats and
        micropigs. Environ Toxicol 24: 453-461. http://dx.doi.org/10.1002/tox.20448

     Smith. JT; Walker. LA; Shore. RF; Durell. S: Howe. PD; Taylor. M.  (2009b). Do estuaries pose a toxic
        contamination risk for wading birds? Ecotoxicology 18: 906-917.
        http://dx.doi.org/10.1007/sl0646-009-0352-z

     Smolders. E; McGrath. S: Fairbrother. A; Hale. B; Lombi. E; McLaughlin. M; Rutgers. M; Van der Vliet.
        L_. (2007). Hazard assessment of inorganic metals and metal substances in terrestrial systems. In
        Assessing the hazard of metals and inorganic metal substances in aquatic and terrestrial systems. Boca
        Raton, FL: CRC Press, http://dx.doi.org/10.1201/9781420044416.ch6

     Smolders. E; Ports. K; van Sprang. P; Schoeters. I; Janssen. CR; McGrath. SP; McLaughlin. MJ. (2009).
        Toxicity of trace metals in soil as affected by soil type and aging  after contamination: Using calibrated
        bioavailability models to set ecological soil standards. Environ Toxicol Chem 28: 1633-1642.
        http://dx.doi.0rg/10.1897/08-592.l

     Snoeijs. T; Dauwe. T; Pinxten. R; Darras. VM; Arckens. L; Eens. M. (2005). The combined effect of lead
        exposure and high or low dietary calcium on health and immunocompetence in the zebra finch
        (Taeniopygia guttata). Environ Pollut 134: 123-132. http://dx.doi.Org/10.1016/i.envpol.2004.07.009

     Sobrino-Figueroa. A; Caceres-Martinez. C.  (2009). Alterations of valve closing behavior in juvenile
        Catarina scallops (Argopecten ventricosus Sowerby, 1842) exposed to toxic metals. Ecotoxicology 18:
        983-987. http://dx.doi.org/10.1007/sl0646-009-0358-6

     Sobrino-Figueroa. AS: Caceres-Martinez. C: Botello. AV; Nunez-Nogueira. G. (2007). Effect of
        cadmium, chromium, lead and metal mixtures on survival and growth of juveniles of the scallop
        Argopecten ventricosus (Sowerby II, 1842). J Environ Sci Health A Tox Hazard Subst Environ Eng
        42: 1443-1447. http://dx.doi.org/10.1080/10934520701480821

     Sobrino. AS: Miranda. MG: Alvarez. C: Quiroz. A. (2010). Bio-accumulation and toxicity of lead (Pb) in
        Lemnagibba L (duckweed). J Environ Sci Health A Tox Hazard Subst Environ Eng 45: 107-110.
        http://dx.doi.org/10.1080/10934520903389267

     Son. J; Ryoo. MI; Jung. J: Cho. K. (2007). Effects of cadmium, mercury and lead on the  survival and
        instantaneous rate of increase of Paronychiurus kimi (Lee) (Collembola). Appl Soil Ecol 35: 404-411.
        http://dx.doi.0rg/10.1016/i.apsoil.2006.07.002

     Sonmez. O; Bukun. B; Kava. C: Aydemir. S. (2008). The assessment of tolerance to heavy metals (Cd,Pb
        and Zn) and their accumulation in three weed species. Pakistan J Bot 40: 747-754.

     Soto-Jimenez. MF; Arellano-Fiore. C: Rocha-Velarde. R; Jara-Marini. ME; Ruelas-Inzunza. J: Paez-
        Osuna. F (2011 a). Trophic transfer of lead through a model marine four-level food chain: Tetraselmis
        suecica, Artemia franciscana, Litopenaeus vannamei, and Haemulon scudderi. Arch Environ Contam
        Toxicol 61: 280-291. http://dx.doi.org/10.1007/s00244-010-9620-4
November 2012                             7-281                  Draft - Do Not Cite or Quote

-------
     Soto-Jimenez. MF; Arellano-Fiore. C: Rocha-Velarde. R; Jara-Marini. ME; Ruelas-Inzunza. J: Voltolina.
        D; Frias-Espericueta. MG: Quintero-Alvarez. JM; Paez-Osuna. F. (201 Ib). Biological responses of a
        simulated marine food chain to lead addition. Environ Toxicol Chem 30: 1611-1617.
        http://dx.doi.org/10.1002/etc.537

     Southward Hogan. L; Marschall. E; Folt. C: Stein. RA. (2007). How non-native species in Lake Erie
        influence trophic transfer of mercury and lead to top predators. J Great Lakes Res 33: 46-61.
        http://dx.doi.org/10.3394/0380-1330(2007)33r46:HNSrLE12.0.CO:2

     Sparling. DW: Krest S: Ortiz-Santaliestra. M. (2006). Effects of lead-contaminated sediment on Rana
        sphenocephala tadpoles. Arch Environ Contam Toxicol 51: 458-466.
        http://dx.doi.org/10.1007/s00244-005-0243-0

     Spears. BL; Hansen. JA; Audet. DJ. (2007). Blood lead concentrations in waterfowl utilizing Lake Coeur
        d'Alene, Idaho. Arch Environ Contam Toxicol 52: 121-128.
        http://dx.doi.org/10.1007/s00244-006-0061-z

     Spokas. EG: Spur. BW: Smith. H; Kemp. FW: Bogden. JD. (2006). Tissue lead concentration during
        chronic exposure of Pimephales promelas (fathead minnow) to lead nitrate in aquarium water. Environ
        Sci Technol 40: 6852-6858. http://dx.doi.org/10.1021/es060811o

     Spry. DJ; Wiener. JG. (1991). Metal bioavailability and toxicity to fish in low-alkalinity lakes: a critical
        review [Review]. Environ Pollut 71: 243-304. http://dx.doi.org/10.1016/0269-7491(91)90034-T

     Spurgeon. DJ: Hopkin. SP; Jones. DT. (1994). Effects of cadmium, copper, lead and zinc on growth,
        reproduction and survival of the earthworm Eisenia fetida (Savigny): Assessing the environmental
        impact of point-source metal contamination in terrestrial ecosystems. Environ Pollut 84:  123-130.
        http://dx.doi.org/10.1016/0269-7491(94)90094-9

     Stavros. HCW: Bossart. GD; Hulsev.  TC: Fair. PA. (2007). Trace element concentrations in skin of free-
        ranging bottlenose dolphins (Tursiops truncatus) from the southeast Atlantic coast. Sci Total Environ
        388: 300-315. http://dx.doi.Org/10.1016/i.scitotenv.2007.07.030

     Stefanowicz. AM; Niklinska. M; Laskowski. R. (2008). Metals affect soil bacterial and fungal functional
        diversity differently. Environ Toxicol Chem 27: 591-598. http://dx.doi.Org/10.1897/07-288.l

     Stobrawa. K; Lorenc-Plucinska. G. (2008). Thresholds of heavy-metal toxicity in cuttings of European
        black poplar (Populus nigra L.) determined according to antioxidant status of fine roots and
        morphometrical disorders. Sci Total Environ 390: 86-96.
        http://dx.doi.0rg/10.1016/i.scitotenv.2007.09.024

     Strom. SM; Patnode. KA; Langenberg. JA; Bodenstein. BL;  Scheuhammer. AM. (2005). Lead
        contamination in American woodcock (Scolopax minor) from Wisconsin. Arch Environ Contam
        Toxicol 49: 396-402. http://dx.doi.org/10.1007/s00244-005-7063-0

     Sudova. R: Vosatka. M. (2007). Differences in the effects of three arbuscular mycorrhizal fungal  strains
        on P and Pb accumulation by maize plants. Plant Soil 296: 77-83.
        http://dx.doi.org/10.1007/slll04-007-9291-8

     Suicmez. M; Kayim. M; Koseoglu. D; Hasdemir. E. (2006). Toxic effects of lead on the liver and gills of
        Oncorhynchus mykiss Walbaum 1792. Bull Environ Contam Toxicol 77: 551-558.
        http://dx.doi.org/10.1007/s00128-006-1099-3

     Sun. SQ: Wang. DY; He. M; Zhang. C. (2009). Monitoring of atmospheric heavy metal deposition in
        Chongqing, China: Based on moss bag technique. Environ Monit Assess 148: 1-9.
        http://dx.doi.org/10.1007/sl0661-007-0133-l

     Sun. SQ: Wang. GX; He. M; Cao. T. (2011). Effects of Pb and Ni stress on oxidative stress parameters in
        three moss species. Ecotoxicol Environ Saf 74: 1630-1635.
        http://dx.doi.0rg/10.1016/i.ecoenv.2011.04.002
November 2012                              7-282                   Draft - Do Not Cite or Quote

-------
     Sundbv. B; Caetano. M; Vale. C: Gobeil. C: Luther. GW. Ill; Nuzzio. DB. (2005). Root-induced cycling
        of lead in salt marsh sediments. Environ Sci Technol 39: 2080-2086.
        http://dx.doi.org/10.1021/es048749n

     Suter. GW: Norton. SB; Fairbrother. A. (2005). Individuals versus organisms versus populations in the
        definition of ecological assessment endpoints. Integr Environ Assess Manag 1: 397-400.
        http://dx.doi.org/10.1002/ieam.5630010409

     Suter. GW: Rodier. DJ; Schwenk. S: Trover. ME; Tyler. PL; Urban. DJ; Wellman. MC: Wharton. S.
        (2004). The U.S. Environmental Protection Agency's Generic Ecological Assessment Endpoints. Hum
        Ecol Risk Assess 10: 967-981. http://dx.doi.org/10.1080/10807030490887104

     Suthar. S: Singh. S: Dhawan. S. (2008). Earthworms as bioindicator of metals (Zn, Fe, Mn, Cu, Pb and
        Cd) in soils: Is metal bioaccumulation affected by their ecological category? Ecol Eng 32: 99-107.
        http://dx.doi.0rg/10.1016/i.ecoleng.2007.10.003

     Svendsen. C: Hankard. PK; Lister. LJ; Fishwick. SK; Jonker. MJ; Spurgeon. DJ. (2007). Effect of
        temperature and season on reproduction, neutral red retention and metallothionein responses of
        earthworms exposed to metals in field soils. Environ Pollut 147:  83-93.
        http://dx.doi.0rg/10.1016/i.envpol.2006.08.012

     Swarup. D: Naresh. R; Varshnev. VP; Balagangatharathilagar. M; Kumar. P; Nandi. D; Patra. RC. (2007).
        Changes in plasma hormones profile and liver function in cows naturally exposed to lead and
        cadmium around different industrial areas. Res Vet Sci 82: 16-21.
        http://dx.doi.0rg/10.1016/i.rvsc.2006.05.002

     Szivak. I; Behra. R; Sigg. L. (2009). Metal-induced reactive oxygen species production in
        Chlamydomonas reinhardtii (chlorophyceae). J Phycol 45: 427-435.
        http://dx.doi.0rg/10.llll/i.1529-8817.2009.00663.x

     Tamura. H; Honda. M; Sato. T; Kamachi. H. (2005). Pb hyperaccumulation and tolerance in  common
        buckwheat (Fagopyrum esculentum Moench). J Plant Res 118: 355-359.
        http://dx.doi.org/10.1007/sl0265-005-0229-z

     Theegala. CS: Suleiman. AA; Carriere. PA. (2007). Toxicity and biouptake of lead and arsenic by
        Daphnia pulex. J Environ Sci Health A Tox Hazard Subst Environ Eng 42: 27-31.
        http://dx.doi.org/10.1080/10934520601015404

     Tilman. D. (1996). Biodiversity: Population versus ecosystem stability. Ecology 77: 350-363.
        http://dx.doi.org/10.2307/2265614

     Tilman. D. (2000). Causes, consequences and ethics of biodiversity.  Nature 405: 208-211.
        http://dx.doi.org/10.1038/35012217

     Tilman. D; Downing.  JA. (1994). Biodiversity and stability in grasslands. Nature 367: 363-365.
        http://dx.doi.org/10.1038/367363aO

     Tollett. VD: Benvenutti. EL; Deer. LA; Rice. TM. (2009). Differential toxicity to Cd, Pb, and Cu in
        dragonfly larvae (Insecta: Odonata). Arch Environ Contam Toxicol 56: 77-84.
        http://dx.doi.org/10.1007/s00244-008-9170-l

     U.S. EPA (U.S. Environmental Protection Agency). (1977). Air quality criteria for lead [EPA Report].
        (EPA-600/8-77-017). Washington, D.C.
        http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=20013GWR.txt

     U.S. EPA (U.S. Environmental Protection Agency). (1985). Ambient water quality criteria for lead - 1984
        [EPA Report]. (EPA 440/5-84-027). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (1986a). Air quality criteria for lead [EPA Report].
        (EPA/600/8-83/028aF-dF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647
November 2012                             7-283                   Draft - Do Not Cite or Quote

-------
     U.S. EPA (U.S. Environmental Protection Agency). (1986b). Air quality criteria for lead: Volume I of IV
        [EPA Report]. (EPA-600/8-83/028aF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=32647

     U.S. EPA (U.S. Environmental Protection Agency). (1998). Guidelines for ecological risk assessment
        [EPA Report]. (EPA/630/R-95/002F). Washington, D.C.
        http://www.epa.gov/raf/publications/guidelines-ecological-risk-assessment.htm

     U.S. EPA (U.S. Environmental Protection Agency). (2002b). National recommended water quality
        criteria: 2002 [EPA Report]. (EPA-822-R-02-047). Washington, D.C.
        http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=P1005EYQ.txt

     U.S. EPA (U.S. Environmental Protection Agency). (2003a). Generic ecological assessment endpoints
        (GEAEs) for ecological risk assessment [EPA Report]. (EPA-630/P-02/004F). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2003b). Guidance for developing ecological soil
        screening levels (Eco-SSLs): Review of background concentration for metals  - Attachment 1-4 [EPA
        Report].  (OSWER Directive 92857-55). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2004). Air quality criteria for paniculate matter [EPA
        Report].  (EPA/600/P-99/002aF-bF). Research Triangle Park, NC.
        http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=87903

     U.S. EPA (U.S. Environmental Protection Agency). (2005b). Ecological soil screening levels for lead:
        Interim final [EPA Report]. (OSWER Directive 9285.7-70). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2005d). Procedures for the derivation of equilibrium
        partitioning sediment benchmarks (ESBs) for the protection of benthic organisms: Metal mixtures
        (cadmium, copper, lead, nickel, silver and zinc) [EPA Report]. (EPA-600-R-02-011).
        Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2006b). Air quality criteria for lead: Volume I of II
        [EPA Report]. (EPA/600/R-05/144aF). Research Triangle Park, NC.
        http ://cfpub. epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823

     U.S. EPA (U.S. Environmental Protection Agency). (2006c). Air quality criteria for lead: Volume II of II
        [EPA Report]. (EPA/600/R-05/144bF). Research Triangle Park, NC.
        http ://cfpub .epa. gov/ncea/CFM/recordisplav. cfm?deid= 158823

     U.S. EPA (U.S. Environmental Protection Agency). (2007a). Aquatic life ambient freshwater quality
        criteria - Copper: 2007 Revision [EPA Report]. (EPA-822/R-07-001). Washington, D.C.
        http://water.epa.gov/scitech/swguidance/standards/criteria/aqlife/pollutants/copper/2007 index.cfm

     U.S. EPA (U.S. Environmental Protection Agency). (2007c). Framework for metals risk assessment [EPA
        Report].  (EPA 120/R-07/001). Washington, D.C. http://www.epa.gov/raf/metalsframework/index.htm

     U.S. EPA (U.S. Environmental Protection Agency). (2007d). Guidance for developing ecological soil
        screening levels [EPA Report]. (OSWER Directive 92857-55). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2008b). Ambient aquatic life water quality: Lead
        (draft) [EPA Report]. (EPA Contract No. 68-C-04-006). Washington, D.C.

     U.S. EPA (U.S. Environmental Protection Agency). (2008e). Integrated science assessment for oxides of
        nitrogen and sulfur: Ecological criteria [EPA Report]. (EPA/600/R-08/082F). Research Triangle Park,
        NC. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=201485

     U.S. EPA (U.S. Environmental Protection Agency). (2009b). National recommended water quality criteria
        [EPA Report]. Washington, D.C. http://www.epa.gov/waterscience/criteria/wqctable/nrwqc-2009.pdf

     UNEP (United Nations Environment Programme). (2003). Ecosystems and human well-being: A
        framework for assessment. Washington, DC: Island Press.
November 2012                              7-284                  Draft - Do Not Cite or Quote

-------
     Uvsal. Y; Taner. F. (2009). Effect of pH, temperature, and lead concentration on the bioremoval of lead
        from water using Lemna minor. Int J Phytoremediation 11: 591-608.
        http://dx.doi.org/10.1080/15226510902717648

     Vaisvalavicius. R; Motuzas. A; Prosvcevas. I; Levinskaite. L; Zakarauskaite. D; Grigaliuniene. K; Butkus.
        V (2006). Effect of heavy metals on microbial communities and enzymatic activity in soil column
        experiment. Arch Agron Soil Sci 52: 161-169. http://dx.doi.org/10.1080/03650340600566431

     van Capelleveen. HE; van Straalen. NM; van den Berg. M; van Wachem. E. (1986). Avoidance as a
        mechanism of tolerance for lead in terrestrial arthropods. In HWH Veldthuis (Ed.), Proceedings of the
        3rd European Congress of Entomology (pp. 251-254). Amsterdam, The Netherlands: Nederlandse
        Entomologische Vereniging.

     Vandenhove. H; Olyslaegers. G: Sanzharova. N; Shubina. O: Reed. E; Shang. Z; Velasco. H. (2009).
        Proposal for new best estimates of the soil-to-plant transfer factor of U, Th, Ra, Pb and Po. J Environ
        Radioact 100: 721-732. http://dx.doi.Org/10.1016/i.jenvrad.2008.10.014

     Vedamanikam. VJ; Shazilli. NAM. (2008a). Comparative toxicity of nine metals to two Malaysian
        aquatic dipterian larvae with reference to temperature variation. Bull Environ Contam Toxicol 80:
        516-520. http://dx.doi.org/10.1007/s00128-008-9413-x

     Vedamanikam. VJ: Shazilli. NAM. (2008b). The effect of multi-generational exposure to metals and
        resultant change in median lethal toxicity tests values over subsequent generations. Bull Environ
        Contam Toxicol 80: 63-67. http://dx.doi.org/10.1007/s00128-007-9317-l

     Veltman. K; Huijbregts. MAJ; Hendriks. AJ. (2010). Integration of biotic ligand models (BLM) and
        bioaccumulation kinetics into a mechanistic framework for metal uptake in aquatic organisms.
        Environ Sci Technol 44: 5022-5028. http://dx.doi.org/10.1021/es903697c

     Vermeulen. F; Van Den Brink. NW: D'Have. H; Mubiana. VK; Blust. R; Bervoets. L; De Coen. W.
        (2009). Habitat type-based bioaccumulation and risk assessment of metal and As contamination in
        earthworms, beetles and woodlice. Environ Pollut 157: 3098-3105.
        http://dx.doi.0rg/10.1016/i.envpol.2009.05.017

     Vigneshkumar. B; Pandian. SK; Balamurugan. K. (In Press) Catalase activity and innate immune
        response of Caenorhabditis elegans against the heavy metal toxin lead. Environ Toxicol.
        http://dx.doi.org/10.1002/tox.20722

     Viiver. MG: Van gestel. CAM; Lanno. RP; Van straalen. NM; Peijnenburg. WJG. M. (2004). Internal
        Metal Sequestration and Its Ecotoxicological Relevance: A Review. Environ Sci Technol 38:
        4705-4712. http://dx.doi.org/10.1021/es040354g

     Viiver. MG: van Gestel. CAM; van Straalen. NM; Lanno. RP; Peijnenburg. W. (2006). Biological
        significance of metals partitioned to subcellular fractions within earthworms (Aporrectodea
        caliginosa). Environ Toxicol Chem 25: 807-814. http://dx.doi.Org/10.1897/05-128R.l

     Vink. JPM. (2009). The origin of speciation: Trace metal kinetics over natural water/sediment interfaces
        and the consequences for bioaccumulation. Environ Pollut 157: 519-527.
        http://dx.doi.0rg/10.1016/i.envpol.2008.09.037

     Vitousek. PM; Aber. JD; Howarth. RW: Likens. GE; Matson. PA; Schindler. DW:  Schlesinger. WH;
        Tilman. DG. (1997). Human alteration of the global nitrogen cycle: Sources and consequences. Ecol
        Appl 7: 737-750. http://dx.doi.org/10.1890/1051-0761(1997)007r0737:HAOTGN12.0.CO:2

     Vlahogianni. TH; Valavanidis. A. (2007).  Heavy-metal effects on lipid peroxidation and antioxidant
        defence enzymes in mussels Mytilus galloprovincialis. Chem Ecol 23: 361-371.
        http://dx.doi.org/10.1080/02757540701653285

     Wall. DH. (1999). Biodiversity and ecosystem functioning. Bioscience 49: 107-108.

     Wang. C: Gu. X; Wang. X; Quo. H; Geng. J: Yu. H; Sun. J. (2011 a). Stress response and potential
        biomarkers in spinach (Spinacia oleracea L.)  seedlings exposed to soil lead. Ecotoxicol Environ Saf
        74: 41-47. http://dx.doi.Org/10.1016/i.ecoenv.2009.02.009
November 2012                              7-285                   Draft - Do Not Cite or Quote

-------
     Wang. C: Tian. Y; Wang. X; Geng. J; Jiang. J; Yu. H; Wang. C. (2010a). Lead-contaminated soil induced
        oxidative stress, defense response and its indicative biomarkers in roots of Vicia faba seedlings.
        Ecotoxicology 19: 1130-1139. http://dx.doi.org/l0.1007/s 10646-010-0496-x

     Wang. C: Yan. X; Wang. P; Hou. J: Hui. Z. (2010b). Effects of salinity and pH on the uptake and
        distribution of Pb and Cd in the aquatic plant Artemisia selengensis in different seasons. Agrochimica
        54: 65-78.

     Wang. CR: Tian. Y; X-R. W: Yu. HX; Lu.  XW: Wang. C: Wang. H. (2010c). Hormesis effects and
        implicative application in assessment of lead-contaminated soils in roots of Vicia faba seedlings.
        Chemosphere 80: 965-971. http://dx.doi.Org/10.1016/i.chemosphere.2010.05.049

     Wang. CR: Wang. XR: Tian. Y: Xue. YG:  Xu. XH: Sui. YX: Yu. HX. (2008a). Oxidative stress and
        potential biomarkers in tomato seedlings subjected to soil lead contamination. Ecotoxicol Environ Saf
        71: 685-691. http://dx.doi.Org/10.1016/i.ecoenv.2008.01.002

     Wang. CR: Wang. XR: Tian. Y: Yu. HX: Gu. XY: Du. WC: Zhou. H.  (2008b). Oxidative stress, defense
        response, and  early biomarkers for lead-contaminated soil in Vicia faba seedlings. Environ Toxicol
        Chem 27: 970-977. http://dx.doi.Org/10.1897/07-344.l

     Wang. DY; Peng. Y. (2007). Multi-biological defects caused by lead exposure exhibit transferable
        properties from exposed parents to their progeny in Caenorhabditis elegans. J Environ Sci 19:
        1367-1372. http://dx.doi.org/10.1016/S1001-0742(07)60223-X

     Wang. DY; Xing. XJ. (2008). Assessment  of locomotion behavioral defects induced by acute toxicity
        from heavy metal exposure in nematode Caenorhabditis elegans. J Environ Sci 20: 1132-1137.
        http://dx.doi.org/10.1016/S1001-0742(08)62160-9

     Wang. G: Su. MY; Chen. YH; Lin. FF; Luo. D; Gao. SF. (2006b). Transfer characteristics of cadmium
        and lead from soil to the edible parts of six vegetable species in southeastern China. Environ Pollut
        144:  127-135. http://dx.doi.0rg/10.1016/i.envpol.2005.12.023

     Wang. MZ: Jia. XY. (2009). Low levels of lead exposure induce oxidative damage and DNA damage in
        the testes of the frog Rana nigromaculata. Ecotoxicology 18: 94-99.
        http://dx.doi.org/10.1007/sl0646-008-0262-5

     Wang. N; Ingersoll. CG: Ivev. CD; Hardestv. DK; May. TW: Augspurger. T; Roberts. AD; van Genderen.
        E; Barnhart. MC. (2010e). Sensitivity of early life stages of freshwater mussels (Unionidae) to acute
        and chronic toxicity of lead, cadmium, and  zinc in water. Environ Toxicol Chem 29: 2053-2063.
        http://dx.doi.org/10.1002/etc.250

     Wang. Q; Liu. B; Yang. H; Wang. X; Lin. Z. (2009d). Toxicity of lead, cadmium and mercury on
        embryogenesis, survival, growth and metamorphosis of Meretrix meretrix larvae. Ecotoxicology 18:
        829-837. http://dx.doi.org/10.1007/sl0646-009-0326-l

     Wang. WX; Rainbow. PS. (2008). Comparative approaches to understand metal bioaccumulation in
        aquatic animals [Review]. Comp Biochem Physiol C Toxicol Pharmacol 148: 315-323.
        http://dx.doi.0rg/10.1016/i.cbpc.2008.04.003

     Wang. ZW: Nan.  ZR; Wang. SL; Zhao. ZJ. (20 lie). Accumulation and distribution of cadmium and lead
        in wheat (Triticum aestivum L.) grown in contaminated soils from the oasis, north-west China. J Sci
        Food Agric 91: 377-384. http://dx.doi.org/10.1002/isfa.4196

     Ward. DM; Nislow. KH; Chen. CY; Folt. CL. (2010). Reduced trace element concentrations in fast-
        growing juvenile atlantic salmon in natural  streams.  Environ Sci Technol 44: 3245-3251.
        http://dx.doi.org/10.1021/es902639a

     Watanabe. K; Monaghan. MT; Takemon. Y; Omura. T. (2008). Biodilution of heavy metals in a stream
        macroinvertebrate food web: Evidence from stable isotope analysis.  Sci Total Environ 394: 57-67.
        http://dx.doi.0rg/10.1016/i.scitotenv.2008.01.006
November 2012                             7-286                   Draft - Do Not Cite or Quote

-------
     Watmough. SA. (1999). Monitoring historical changes in soil and atmospheric trace metal levels by
        dendrochemical analysis. Environ Pollut 106: 391-403.
        http://dx.doi.org/10.1016/S0269-7491(99)00102-5

     Watmough. SA; Dillon. PJ. (2007). Lead biogeochemistry in a central Ontario forested watershed.
        Biogeochemistry 84: 143-159. http://dx.doi.org/10.1007/sl0533-007-9110-6

     Watson. AP; Van Hook. RI; Jackson. PR; Reichle. DE. (1976). Impact of a lead mining smelting complex
        on the forest- floor litter arthropod fauna in the new lead belt region of southeast Missouri.
        (ORNL/NSF/EATC-30). Oak Pudge, Tennessee: Oak Pudge National Laboratory.

     Weber.  DN. (1993). Exposure to sublethal levels of waterborne lead alters reproductive behavior patterns
        in fathead minnows (Pimephales promelas). Neurotoxicology 14: 347-358.

     Weber.  DN. (1996). Lead-induced metabolic imbalances and feeding alterations in juvenile fathead
        minnows (Pimephales promelas). Environ Toxicol 11: 45-51.

     Weir. PA; Hine. CH. (1970). Effects of various metals on behavior of conditioned goldfish. Arch Environ
        Health 20:  45-51.

     Weis. JS: Weis. P. (1998). Effects of exposure to lead on behavior of mummichog (Fundulus heteroclitus
        L) larvae. J Exp Zool Part A Ecol Genet Physiol 222: 1-10.

     Westman. WE. (1977). How much are nature's services worth? Measuring the social benefits of
        ecosystem  functioning is both controversial and illuminating. Science 197: 960-964.
        http://dx.doi.org/10.1126/science.197.4307.960

     Wetmore. A. (1919). Lead poisoning in waterfowl. US Dep Agric Bull 793: 1-12.

     Witte. KM; Wantv. RB; Ridley. WI. (2004). Engelmann Spruce (Picea engelmannii) as a biological
        monitor of changes in soil metal loading related to past mining activity. Appl Geochem 19:
        1367-1376. http://dx.doi.0rg/10.1016/i.apgeochem.2004.01.022

     Wong. CC: Wu. SC: Kuek. C: Khan. AG: Wong. MH. (2007). The role of mycorrhizae associated with
        vetiver grown in Pb-/Zn-contaminated soils:  Greenhouse study.  Restor Ecol 15: 60-67.
        http://dx.doi.Org/10.llll/i.1526-100X.2006.00190.x

     Wong. LC: Kwok.  KWH; Leung. KMY; Wong.  CK. (2009). Relative sensitivity distribution of freshwater
        planktonic  crustaceans to trace metals. Hum Ecol Risk Assess 15: 1335-1345.
        http://dx.doi.org/10.1080/10807030903307115

     Worms. I; Simon. DF; Hassler. CS: Wilkinson. KJ. (2006). Bioavailability of trace metals to aquatic
        microorganisms: Importance of chemical, biological and physical processes on biouptake [Review].
        Biochimie  88: 1721-1731. http://dx.doi.Org/10.1016/i.biochi.2006.09.008

     WRI (World Resources Institute). (2000). World resources 2000-2001: People and ecosystems: The
        fraying web of life. Washington, D.C.

     Wu. X; Hong.  F: Liu. C: Su. MY; Zheng. L; Gao. FQ; Yang. F. (2008c). Effects of Pb2+ on energy
        distribution and photochemical activity of spinach chloroplast. Spectrochim Acta A Mol Biomol
        Spectrosc 69: 738-742. http://dx.doi.Org/10.1016/i.saa.2007.05.047

     Xie. W: Zhou. J; Wang. H; Liu. Q; Xia.  J: Lv. X. (2011). Cu and Pb accumulation in maize (Zea mays L.)
        and  soybean (Glycine max L.) as affected by N, P and K application. African Journal of Agricultural
        Research 6: 1469-1476.

     Xing. X; Du. M; Zhang. Y; Wang. D. (2009a). Adverse effects of metal exposure on chemotaxis towards
        water-soluble attractants regulated mainly by ASE sensory neuron in nematode Caenorhabditis
        elegans. J Environ Sci 21:  1684-1694. http://dx.doi.org/10.1016/sl001-0742(08)62474-2

     Xing. X; Guo. Y; Wang. D. (2009b).  Using  the larvae nematode Caenorhabditis elegans to evaluate
        neurobehavioral toxicity to metallic salts. Ecotoxicol Environ Saf 72: 1819-1823.
        http://dx.doi.0rg/10.1016/i.ecoenv.2009.06.006
November 2012                             7-287                  Draft - Do Not Cite or Quote

-------
     Xing. X; Rui. Q; Du. M; Wang. D. (2009c). Exposure to lead and mercury in young larvae induces more
        severe deficits in neuronal survival and synaptic function than in adult nematodes. Arch Environ
        Contam Toxicol 56: 732-741. http://dx.doi.org/10.1007/s00244-009-9307-x

     Xiong. ZT; Zhao. F; Li. MJ. (2006). Lead toxicity in Brassica pekinensis Rupr.:  Effect on nitrate
        assimilation and growth. Environ Toxicol 21: 147-153. http://dx.doi.org/10.1002/tox.20167

     Xu. J; Ke. X; Krogh. PH; Wang. Y; Luo. YM; Song. J. (2009a). Evaluation of growth and reproduction as
        indicators of soil metal toxicity to the Collembolan, Sinella curviseta. Insect  Sci 16: 57-63.
        http://dx.doi.0rg/10.llll/i.1744-7917.2009.00254.x

     Xu. J; Wang. Y; Luo. YM; Song. J: Ke. X. (2009b). Effects of copper, lead and zinc in soil on egg
        development and hatching of Folsomia Candida. Insect Sci 16: 51-55.
        http://dx.doi.0rg/10.llll/i.1744-7917.2009.00253.x

     Yang. H; Shen. Z; Zhu. S: Wang. W. (2008a). Heavy metals in wetland plants and soil of Lake Taihu,
        China. Environ Toxicol  Chem 27: 38-42. http://dx.doi.Org/10.1897/07-089.l

     Yang. R; Tang. J: Chen. X;  Hu. S. (2007). Effects of coexisting plant species on  soil microbes and soil
        enzymes in metal lead contaminated soils. Appl Soil Ecol 37: 240-246.
        http://dx.doi.0rg/10.1016/i.apsoil.2007.07.004

     Yang. R; Yu. G: Tang. J: Chen. X. (2008b). Effects of metal lead on growth and  mycorrhizae of an
        invasive plant species (Solidago canadensis L.). J Environ Sci 20: 739-744.
        http://dx.doi.org/10.1016/S1001-0742(08)62121-X

     Yang. Y; Campbell. CD; Clark. L; Cameron. CM; Paterson. E. (2006). Microbial indicators of heavy
        metal contamination in urban and rural soils.  Chemosphere 63: 1942-1952.
        http://dx.doi.0rg/10.1016/i.chemosphere.2005.10.009

     Yu. DY; Xu. ZR; Yang. XG. (2005). Effects of lead and particulate montmorillonite on growth
        performance, hormone and organ weight in pigs. Asian - Australasian Journal of Animal Sciences 18:
        1775-1779.

     Zaprjanova. P; Dospatliev. L; Angelova. V: Ivanov. K. (2010). Correlation between soil characteristics
        and lead and cadmium content in the aboveground biomass of Virginia tobacco. Environ Monit Assess
        163: 253-261. http://dx.doi.org/10.1007/sl0661-009-0831-v

     Zeng. LS: Liao. M; Chen. CL; Huang. CY. (2007). Effects of lead contamination on soil enzymatic
        activities, microbial biomass, and rice physiological indices in soil-lead-rice  (Oryza sativa L.) system.
        Ecotoxicol Environ Saf 67: 67-74. http://dx.doi.Org/10.1016/i.ecoenv.2006.05.001

     Zhang. CB; Huang. LN; Luan. TG: Jin. J: Lan. CY. (2006). Structure and function of microbial
        communities during the early stages of revegetation of barren soils in the vicinity of a Pb/Zn smelter.
        Geoderma 136: 555-565. http://dx.doi.Org/10.1016/i.geoderma.2006.04.011

     Zhang. H; Luo. Y; Song. J:  Zhang. H; Xia. J: Zhao. Q. (20lib). Predicting As, Cd and Pb uptake by rice
        and vegetables using field data from China. J Environ Sci 23: 70-78.
        http://dx.doi.org/10.1016/sl001-0742(10)60375-0

     Zhang. Y; Song. JM; Yuan. HM; Xu. YY; He. ZP; Duan. LQ. (201 Ob). Biomarker responses in the
        bivalve (Chlamys farreri) to exposure of the environmentally relevant concentrations of lead, mercury,
        copper. Environ Toxicol Pharmacol 30: 19-25. http://dx.doi.Org/10.1016/i.etap.2010.03.008

     Zhang. YM; Huang. DJ; Zhao. DQ: Long. J: Song. G: Li. A. (2007b). Long-term toxicity effects of
        cadmium and lead on Bufo raddei tadpoles. Bull Environ Contam Toxicol 79: 178-183.
        http://dx.doi.org/10.1007/s00128-007-9152-4

     Zhang. Z; Song. X; Wang. Q: Lu. X. (2012). Cd and Pb contents in soil, plants, and grasshoppers along a
        pollution gradient in Huludao City, northeast  China. Biol Trace Elem Res 145: 403-410.
        http://dx.doi.org/10.1007/sl2011-011-9199-2
November 2012                              7-288                   Draft - Do Not Cite or Quote

-------
     Zheng. RQ; Li. CY. (2009). Effect of lead on survival, locomotion and sperm morphology of Asian
        earthworm, Pheretima guillelmi. J Environ Sci 21: 691-695.
        http://dx.doi.org/10.1016/sl001-0742(08)62325-6

     Zhuang. P; Zou. HL; Shu. WS. (2009). Biotransfer of heavy metals along a soil-plant-insect-chicken food
        chain: Field study. J Environ Sci 21: 849-853. http://dx.doi.org/10.1016/sl001-0742(08)62351-7
November 2012                             7-289                   Draft - Do Not Cite or Quote

-------