United States
Environmental Protection
Agency
Environmental Criteria and
Assessment Office
Research Triangle Park NC 27711
EPA-600/8-83-028A
August 1983
External Review Draft
Research and Development
oEPA
Air Quality
Criteria for Lead
Volume III of IV
Review
Draft
(Do Not
Cite or Quote)
NOTICE
This document is a preliminary draft. It has not been formally
released by EPA and should not at this stage be construed to
represent Agency policy. It is being circulated for comment on its
technical accuracy and policy implications.
-------
EPA-600/8-83-028A
August 1983
External Review Draft No. 1
Draft
Do Not Quote or Cite
Air Quality Criteria
for Lead
Volume III of IV
NOTICE
This document is a preliminary draft. It has not been formally released by EPA and should not at this stage
be construed to represent Agency policy. It is being circulated for comment on its technical accuracy and
policy implications.
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Office of Health and Environmental Assessment
Environmental Criteria and Assessment Office
Research Triangle Park, NC 27711
-------
NOTICE
Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
11
-------
ABSTRACT
The document evaluates and assesses scientific Information on the health
and welfare effects associated with exposure to various concentrations of lead
in ambient air. The literature through 1983 has been reviewed thoroughly for
information relevant to air quality criteria, although the document is not
intended as a complete and detailed review of all literature pertaining to
lead. An attempt has been made to identify the major discrepancies in our
current knowledge and understanding of the effects of these pollutants.
Although this document is principally concerned with the health and
welfare effects of lead, other scientific data are presented and evaluated in
order to provide a better understanding of this pollutant in the environment.
To this end, the document includes chapters that discuss the chemistry and
physics of the pollutant; analytical techniques; sources, and types of
emissions; environmental concentrations and exposure levels; atmospheric
chemistry and dispersion modeling; effects on vegetation; and respiratory,
physiological, toxicological, clinical, and epidemiclogical aspects of human
exposure.
iii
-------
PRELIMINARY DRAFT
CONTENTS
Page
VOLUME I
Chapter 1.
VOLUME II
Chapter 2.
Chapter 3.
Chapter 4.
Chapter 5.
Chapter 6.
Chapter 7.
Chapter 8.
VOLUME III
Chapter 9.
Chapter 10.
Chapter 11.
Executive Summary and Conclusions
Introduction
Chemical and Physical Properties *.
Sampling and Analytical Methods for Environmental Lead
Sources and Emissions
Transport and Transformation
Environmental Concentrations and Potential Pathways to Human Exposure
Effects of Lead on Ecosystems
Quantitative Evaluation of Lead and Biochemical Indices of Lead
Exposure in Physiological Media
Metaboli sm of Lead
Assessment of Lead Exposures and Absorption in Human Populations
Volume IV
Chapter 12. Biological Effects of Lead Exposure
Chapter 13. Evaluation of Human Health Risk Associated with Exposure to Lead
and Its Compounds
1-1
2-1
3-1
4-1
5-1
6-1
7-1
8-1
9-1
10-1
11-1
12-1
13-1
TCPBA/H
iv
8/8/83
-------
PRELIMINARY DRAFT
TABLE OF CONTENTS
9. QUANTITATIVE EVALUATION OF LEAD AND BIOCHEMICAL INDICES OF LEAD EXPOSURE
IN PHYSIOLOGICAL MEDIA 9-1
9.1 INTRODUCTION 9-1
9.2 DETERMINATIONS OF LEAD IN BIOLOGICAL MEDIA 9-2
9.2.1 Sampling and Sample Handling Procedures for Lead
in Biological Media 9-2
9.2.1.1 Blood Sampling 9-3
9.2.1.2 Urine Sampling 9-4
9.2.1.3 Hair Sampling 9-4
9.2.1.4 Mineralized Tissue 9-4
9.2.1.5 Sampling Hand!ing in the Laboratory 9-5
9.2.2 Methods of Lead Analysis 9-6
9.2.2.1 Lead Analysis in Whole Blood 9-7
9.2.2.2 Lead in Plasma 9-10
9.2.2.3 Lead in Teeth 9-12
9.2.2.4 Lead in Hair 9-13
9.2.2.5 Lead in Urine 9-13
9.2.2..6 Lead in Other Tissues 9-14
9.2.3 Quality Assurance Procedures in Lead Analysis 9-15
9.3 DETERMINATION OF ERYTHROCYTE PORPHYRIN (FREE ERYTHROCYTE
PROTOPOPHYRIN, ZINC PROTOPORPHYRIN) 9-19
9.3.1 Methods of Erythrocyte Porphyrin Analysis 9-19
9.3.2 Interlaboratory Testing of Accuracy and Precision in
EP Measurement 9-23
9.4 MEASUREMENT OF URINARY COPROPORPHYRIN 9-24
9.5 MEASUREMENT OF DELTA-AMINOLEVULINIC ACID DEHYDRATASE ACTIVITY 9-24
9.6 MEASUREMENT OF DELTA-AMINOLEVULINIC ACID IN URINE AND OTHER MEDIA 9-26
9.7 MEASUREMENT OF PYRIMIDINE-51-NUCLEOTIDASE ACTIVITY 9-27
9.8 SUMMARY 9-29
9.8.1 Determinations of Lead in Biological Media 9-29
9.8.1.1 Measurements of Lead in Blood 9-29
9.8.1.2 Lead in Plasma 9-31
9.8.1.3 Lead in Teeth 9-31
9.8.1.4 Lead in Hair 9-31
9.8.1.5 Lead in Urine 9-31
9.8.1,6 Lead in Other Tissues 9-32
9.8.1.7 Qua!ity Assurance Procedures i n Lead Analyses 9-32
9.8.2 Determination of Erythrocyte Porphyrin (Free Erythrocyte
Protoporphyrin, Zi nc Protoporphyri n) 9-33
9.8.3 Measurement of Urinary Coproporphyrin 9-34
9.8.4 Measurement of Delta-Ami no!evullnic Acid Dehydratase Activity 9-34
9.8.5 Measurement of Delta-Aminolevulinic Acid in Urine and Other Media ... 9-35
9.8.6 Measurement of Pyrimidine-S'-Nucleotidase Activity 9-36
9.9 REFERENCES 9-37
10. METABOLISM OF LEAD 10-1
10.1 INTRODUCTION 10-1
10.2 LEAD ABSORPTION IN HUMANS AND ANIMALS 10-1
10.2.1 Respiratory Absorption of Lead 10-1
10.2.1.1 Human Studies 10-2
10.2.1.2 Animal Studies 10-5
v
TCPBVK 8/8/83
-------
PRELIMINARY DRAFT
TABLE OF CONTENTS (continued).
10.3
10.4
10.5
10.6
10.7
10.8
10.2.2 Gastrointestinal Absorption of Lead
10.2.2.1 Human Studies
10.2.2.2 Animal Studies
10.2. 3 Percutaneous Absorption of Lead
10.2.4 Transplacental Transfer of Lead
DISTRIBUTION OF LEAD IN HUMANS AND ANIMALS
10.3.1 Lead in Blood
10.3.2 Lead Levels in Tissues
10.3.2.1 Soft Tissues
10.3.2.2 Mineralizing Tissue
10.3.3 Chelatable Lead
10.3.4 Mathematical Descriptions of Physiological Lead Kinetics
10. 3. 5 Animal Studies
LEAD EXCRETION AND RETENTION IN HUMANS AND ANIMALS
10.4. 1 Human Studies
10.4.2 Animal Studies
INTERACTIONS OF LEAD WITH ESSENTIAL METALS AND OTHER FACTORS
10. 5. 1 Human Studies
10.5.2 Animal Studies
10.5.2.1 Interactions of Lead with Calcium
10. 5. 2. 2 Interactions of Lead with Iron
10. 5. 2. 3 Lead Interactions with Phosphate
10.5.2.4 Interactions of Lead with Vitamin D
10. 5. 2. 5 Interactions of Lead with Lipids
10.5.2.6 Lead Interaction with Protein
10.5.2.7 Interactions of Lead with Milk Components
10.5.2.8 Lead Interactions with Zinc and Copper
INTERRELATIONSHIPS OF LEAD EXPOSURE, EXPOSURE INDICATORS,
AND TISSUE LEAD BURDENS
10.6.1 Temporal Characteristics of Internal Indicators
of Lead Exposure
10.6.2 Biological Aspects of External Exposure- Internal
Indicator Relationships
10.6.3 Internal Indicator-Tissue Lead Relationships
METABOLISM OF LEAD ALKYLS
10.7.1 Absorption of Lead Alky Is in Humans and Animals
10.7.1.1 Gastrointestinal Absorption
10.7.1.2 Percutaneous Absorption of Lead Alkyls
10.7.2 Biotransformation and Tissue Distribution of Lead Alkyls
10.7.3 Excretion of Lead Alkyls
SUMMARY
10.8. 1 Lead Absorption in Humans and Animals
10.8. 1. 1 Respiratory Absorption of Lead
10.8.1.2 Gastrointestinal Absorption of Lead
10.8. 1. 3 Percutaneous Absorption of Lead
10.8.1.4 Transplacental Transfer of Lead
10.8.2 Distribution of Lead in Humans and Animals
10.8.2.1 Lead in Blood
10.8.2.2 Lead Levels in Tissues
10.8.3 Lead Excretion and Retention in Humans and Animals
10.8.3.1 Human Studies
10-6
10-6
10-10
10-12
10-12
10-13
10-14
10-15
10-16
10-19
10-20
10-22
10-23
10-24
10-24
10-28
10-31
10-31
10-33
10-34
10-38
10-38
10-39
10-39
10-39
10-40
10-40
10-41
10-41
10-42
10-43
10-45
10-46
10-46
10-46
10-46
10-48
10-49
10-49
10-49
10-50
10-51
10-51
10-51
10-51
10-52
10-54
10-54
TCPBA/K V1 8/8/83
-------
PRELIMINARY DRAFT
TABLE OF CONTENTS (continued).
Page
10.8.3.2 Animal Studies ........................................... 10-55
10.8.4 Interactions of Lead with Essential Metals and Other Factors ....... 10-56
10.8.4. 1 Human Studies .......... . ................................. 10-56
10.8.4.2 Animal Studies ........................................... 10-56
10.8.5 Interrelationships of Lead Exposure with Exposure Indicators
and Tissue Lead Burdens ............................................ 10-57
10.8.5.1 Temporal Characteristics of Internal Indicators of
Lead Exposure ............................................ 10-57
10.8.5.2 Biological Aspects of External Exposure- Internal
Indicator Relationships .................................. 10-58
10.8.5.3 Internal Indicator-Tissue Lead Relationships ............. 10-58
10.8.6 Metabolism of Lead Alkyls .......................................... 10-59
10.8.6.1 Absorption of Lead Alky Is in Humans and Animals .......... 10-59
10.8.6.2 Biotrans formation and Tissue Distribution of
Lead Alkyls .............................................. 10-59
10.8.6.3 Excretion of Lead Alklys ................................. 10-59
10. 9 REFERENCES [[[ 10-60
11. ASSESSMENT OF LEAD EXPOSURES AND ABSORPTION IN HUMAN POPULATIONS ................ 11-1
11. 1 INTRODUCTION [[[ 11-1
11.2 METHODOLOGICAL CONSIDERATIONS .............................................. 11-4
11.2.1 Analytical Problems ................................................ 11-4
11.2.2 Statistical Approaches ............................................. 11-5
11. 3 LEAD IN HUMAN POPULATIONS .................................................. 11-6
11. 3.1 Introduction [[[ 11-6
11. 3.2 Ancient and Remote Populations (Low Lead Exposures) ................ 11-6
11.3.2.1 Ancient Populations ...................................... 11-8
11.3.2.2 Remote Populations ................................ .. ...... 11-8
11.3.3 Levels of Lead and Demographic Covariates in U.S. Populations ...... 11-10
11.3.3. 1 The NHANES II Study ...................................... 11-10
11.3.3.2 The Childhood Blood Lead Screening Programs .............. 11-15
11.3.4 Time Trends ............ ............................................ 11-19
11.3.4.1 Time Trends in the Childhood Lead Poisoning Screening
Programs ................................................. 11-19
11.3.4.2 Newark [[[ 11-22
11.3.4.3 Boston [[[ 11-24
11.3.4.4 NHANES II ................................................ 11-24
11.3.4.5 Other Studies ............................................ 11-24
11.3.5 Distributional Aspects of Population Blood Lead Levels ............. 11-24
11.3.6 Exposure Covariates of Blood Lead Levels in Urban Children . .' ....... 11-31
11.3.6.1 Stark Study .............................................. 11-32
11.3.6.2 Charney Study ............................................ 11-33
11.3.6.3 Hammond Study ............................................ 11-34
11.3.6.4 Gilbert Study ............................................ 11-35
11. 4 STUDIES RELATING EXTERNAL DOSE TO INTERNAL EXPOSURE ........................ 11-36
11.4.1 Air Studies [[[ 11-37
11.4.1.1 The Griffin et al. Study ................................. 11-38
11.4.1.2 The Rablnowitz et al. Study .............................. 11-47
-------
PRELIMINARY DRAFT
TABLE OF CONTENTS (continued)
Page
11.4.1.6 Silver Valley/Kellogg, Idaho Study 11-58
11.4.1.7 Omaha, Nebraska Studies 11-65
11.4.1.8 Roels et al. Studies 11-67
11.4.1.9 Other Studies Relating Blood Lead Levels to
Air Exposure 11-70
11.4.1.10 Summary of Blood Lead vs. Inhaled Air Lead Relations ..... 11-74
11.4.2 Dietary Lead Exposures Including Water 11-80
11.4.2.1 Lead Ingestion from Typical Diets 11-81
11.4.2.2 Lead Ingestion from Experimental Dietary Supplements 11-90
11.4.2.3 Inadvertent Lead Ingestion From Lead Plumbing 11-93
11.4.2.4 Summary of Dietary Lead Exposures Including Water 11-97
11.4.3 Studies Relating Lead in Soil and Dust to Blood Lead 11-105
11.4.3.1 Omaha Nebraska Studies 11-105
11.4.3.2 The Stark Study 11-106
11.4.3.3 The Silver Valley/Kellogg Idaho Study JJ~JSf
11.4.3.4 Charleston Studies }}'¥£
11.4.3.5 Barltrop Studies 11-107
11.4.3.6 The British Columbia Studies 11-108
11.4.3.7 Other Studies of Soil and Dusts 11-109
11.4.3.8 Summary of Soi 1 and Dust Lead 11-113
11.4.4 Paint Lead Exposures 11-115
11.5 SPECIFIC SOURCE STUDIES 11-121
11.5.1 Combustion of Gasoline Antiknock Compounds 11-121
11.5.1.1 Isotope Studies 11-121
11.5.1.2 Studies of Childhood Blood Lead Poisoning
Control Programs 11-130
11.5.1.3 NHANES II 11-133
11.5.1.4 Frankfurt, West Germany 11-136
11.5.2 Primary Smelters Populations 11-137
11.5.2.1 El Paso, Texas 11-137
11.5.2.2 CDC-EPA Study 11-139
11.5.2.3 Meza Valley, Yugoslavia 11-139
11.5.2.4 Kosovo Province, Yugoslavia 11-140
11.5.2.5 The Cavalleri Study 11-141
11.5.3 Battery Plants 11-142
11.5.4 Secondary Smelters 11-145
11.5.5 Secondary Exposure of Chi 1dren 11-145
11.5.6 Miscellaneous Studies 11-152
11.5.6.1 Studies Using Indirect Measures of Air Exposure n-152
11.5.6.2 Miscellaneous Sources of Lead 11-156
11.6 SUMMARY 11-158
11.7 REFERENCES 11-166
APPENDIX 11A 11A-1
APPENDIX 11B 11B-1
APPENDIX 11C 11C-1
APPENDIX 110 11D-1
TCPBA/K
viii
8/8/83
-------
PRELIMINARY DRAFT
LIST OF FIGURES
Figure Page
10-1 Effect of particle size on lead deposition rate in the lung 10-4
11-1 Pathways of 1ead from the envi ronment to man 11-3
11-2 Estimate of world-wide lead production and lead concentrations in
bones (pg/gm) from 5500 years before present to the present time 11-7
11-3 Geometric mean blood lead levels by race and age for younger children
in the NHANES II study 11-16
11-4 Geometric means for blood lead values by race and age for younger
children in the New York City screening program (1970-1976) 11-20
11-5 Time dependence of blood lead for blacks, aged 24 to 35 months,
in New York City and Chicago 11-23
11-6 Modeled umbilical cord blood lead levels by date of sample collection
for i nfants i n Boston 11-25
11-7 Average blood lead levels of U.S. population 6 months - 74 years,
United States, February 1976 - February 1980, based on dates of
examination of NHANES II examinees with blood lead determinations 11-26
11-8 Histograms of blood lead levels with fitted lognormal curves for
the NHANES II study 11-30
11-9 Graph of the average normalized increase in blood lead for subjects
exposed to 10.9 g/m3 of lead in the Griffin et al. study 11-41
11-10 Control subjects in Griffin experiment at 3.2 pg/m3 11-42
11-11 Data plots for individual subjects with time for Kehoe data as
presented by Gross 11-54
11-12 Blood lead vs. air lead relationships for Kehoe inhalation studies:
linear relation for low exposures, quadratic for high exposures, with
95 percent conf i dence bands 11-55
11-13 Monthly ambient air lead concentrations in Kellogg, Idaho,
1971 through 1975 11-59
11-14 Fitted equations to the Kellogg, Idaho/Silver Valley adjusted
blood lead data 11-64
11-15 Blood-lead concentrations vs. weekly lead intake for bottle-
fed infants 11-87
11-17 Average Pb level, exp. I 11-91
11-18 Average PbB levels, exp. II 11-91
11-19 Lead in blood (mean values and range) in volunteers 11-93
11-20 Cube root regression of blood lead on first flush water lead 11-96
11-21 Relation of blood lead (adult female) to first flush water lead
i n combi ned estates 11-98
11-22 Cumulative distribution of lead levels in dwelling units 11-117
11-23 Correlation of children's blood lead levels with fractions of surfaces
within a dwell ing having lead concentrations £2 mg pb/cm2 11-119
11-24 Change in 206Pb/Zo7Pb ratios in petrol, airborne particulate
and blood from 1974 to 1981 11-123
11-25 Direct and indirect contributions of lead in gasoline to blood
lead in Italian men 11-126
11-26 Geometric mean blood lead levels of New York City children (aged 25-36
months) by ethnic group, and ambient air lead concentration vs.
quarterly sampling period, 1970-1976 11-131
11-27 Geometric mean blood lead levels of New York City children (ages 25-36
months) by ethnic group, and estimated amount of lead present in
gasoline sold in New York, New Jersey, and Connecticut vs.
quarterly sampling period, 1970-1976 11-132
1x
TCPBA/K 8/8/83
-------
PRELIMINARY DRAFT
LIST OF FIGURES (continued).
Figure Page
11-28 Geometric mean blood levels for blacks and Hispanics in the 25-to-36-
month age group and rooftop quarterly averages for ambient cityvride
lead levels 11-134
11-29 Time dependence of blood lead and gas lead for blacks, ages 24 to 35
months, in New York 11-135
11-30 Arithmetic mean air lead levels by traffic volume, Dallas, 1976 11-154
11-31 Blood lead concentration and traffic density by sex and age, Dallas, 1976 11-155
11-32 Geometric mean blood lead levels by race and age for younger children in
the NHANES II study, and the Kellogg/Silver Valley and the New York
childhood screening studies 11-159
11B-1 Residual sum of squares for nonlinear regression models for Azar data
(N=149) 11-170
11B-2 Hypothetical relationship between blood lead and air lead by inhalation
and non-inhalation 11-172
TCPBA/K X 8/8/83
-------
PRELIMINARY DRAFT
LIST OF TABLES
Table Page
10-1 Deposition of lead in the human respiratory tract 10-3
10-2 Regional distribution of lead in humans and animals 10-17
10-3 Comparative excretion and retention rates in adults and infants 10-25
10-4 Effect of nutritional factors on lead uptake in animals 10-35
11-1 Studies of past exposures to lead 11-9
11-2 NHANES II blood lead levels of persons 6 months-74 years, with weighted
arithmetic mean, standard error of the mean, weighted geometric mean,
median, and percent distribution, by race and age, United States,
1976-80 11-12
11-3 NHANES II blood lead levels of males 6 months-74 years, with weighted
arithmetic mean, standard error of the mean, weighted geometric mean,
median, and percent distribution, by race and age, United States,
1976-80 11-13
11-4 NHANES II blood lead levels of females 6 months-74 years, with weighted
arithmetic mean, standard error of the mean, weighted geometric mean,
median, and percent distribution, by race and age, United States,
1976-80 11-14
11-5 Weighted geometric mean blood lead levels from NHANES II survey by
degree of urbanization of place of residence in the U.S. by age
and race, United States 1976-80 11-17
11-6 Annual geometric mean blood lead levels from the New York blood lead
screening studies. Annual geometric means are calculated from
quarterly geometric means estimated by the method of
Hasselblad et al. (1980) 11-18
11-7 Characteristics of childhood lead poisoning screening data 11-21
11-8 Distribution of blood lead levels for 13 to 48 month old blacks
by season and year for New York screening data 11-21
11-9 Summary of unweighted blood lead levels in whites not living in an
SMSA with family income greater than $6,000 11-28
11-10 Summary of fits to NHANES II blood lead levels of whites not
living in an SMSA, income greater than $6,000, for five
different two parameter distributions 11-29
11-11 Estimated mean square errors resulting from analysis of variance on
various subpopulations of the NHANES II data using unweighted data 11-31
11-12 Multiple regression models for blood lead of children in
New Haven, Connecticut, September 1974 - February 1977 11-33
11-13 Griffin experiments - subjects exposed to air lead both years 11-43
11-14 Gri f f i n experiments - controls used both years 11-44
11-15 Griffin experiment - subjects exposed to air lead one year only 11-45
11-16 Inhalation slope estimates 11-47
11-17 Mean residence time in blood 11-47
11-18 Air lead concentrations (ng/ma) for two subjects in the
Rabinowitz studies 11-48
11-19 Estimates of Inhalation slope for Rabinowitz studies 11-49
11-20 Linear slope for blood lead vs. air lead at low air lead
exposures i n Kehoe's subjects 11-53
11-21 Geometric mean air and blood lead levels (pg/100 g) for five city-
occupation groups 11-56
11-22 Geometric mean blood lead levels by area compared with estimated
air-lead levels for 1- to 9-year-old children living near Idaho
smelter 11-61
xi
TCPBA/K 8/8/83
-------
PRELIMINARY DRAFT
LIST OF TABLES (continued).
Table page
11-23 Geometric mean blood lead levels by age and area for subjects
living near the Idaho smelter 11-61
11-24 Age specific regression coefficients for the analysis of log-blood-
lead levels in the Idaho smelter study 11-62
11-25 Estimated coefficients and standard errors for the Idaho
smelter study 11-63
11-26 Air, dustfall and blood lead concentrations in Omaha, NE, study,
1970-1977 11-66
11-27 Mean airborne and blood lead levels recorded during five distinct
surveys (1974 to 1978) for study populations of 11-year old
children living less than 1 km or 2.5 km from a lead smelter,
or living in a rural or urban area 11-69
11-28 Geometric mean air and blood lead values for 11 study populations 11-71
11-29 Mean air and blood lead values for five zones in Tokyo study 11-71
11-30 Blood lead-air lead slopes for several population studies as
calculated by Snee 11-73
11-31 A selection of recent analyses on occupational 8-hour exposures
to high air lead levels 11-74
11-32 Cross-sectional observational study with measured individual air
lead exposure 11-75
11-33 Cross-sectional observational studies on children with estimated
ai r exposures 11-76
11-34 Longitudinal experimental studies with measured individual
air lead exposures 11-77
11-35 Blood lead levels and lead intake values for infants
in the study of Ryu et al 11-82
11-36 Influence of level of lead in water on blood lead level in
blood and placenta 11-84
11-37 Blood lead and kettle water lead concentrations for adult
women living in Ayr 11-85
11-38 Relationship of blood lead (ug/dl) and water lead (ug/1) in 910
men aged 40-59 from 24 British towns 11-88
11-39 Dose response analysis for blood leads in the Kehoe study as
analyzed by Gross 11-90
11-40 Blood lead levels of 771 persons in relation to lead content of
drinking water, Boston, Mass 11-99
11-41 Studies relating blood lead levels (ug/dl) to dietary intakes (ug/day) 11-100
11-42 Studies relating blood lead levels (ug/dl) and experimental
dietary intakes 11-101
11-43 Studies relating blood lead levels (ug/dl) to
first-flush water lead 11-102
11-44 Studies relating blood lead levels (ug/dl) to running water
lead (ug/1) 11-104
11-45 Mean blood and soil lead concentrations in English study 11-108
11-46 Lead concentration of surface soil and children's blood
by residential area of trail, British Columbia 11-110
11-47 Analysis of relationship between soil lead and blood lead in children 11-113
11-48 Estimates of the contribution of soil lead to blood lead 11-114
11-49 Estimates to the contribution of housedust to blood lead 1n children 11-115
11-50 Results of screening and housing inspection in childhood lead
poisoning control project by fiscal year 11-120
xi i
"PBA/K 8/8/83
-------
PRELIMINARY DRAFT
LIST OF TABLES (continued).
Table Page
11-51 Estimated contribution of leaded gasoline to blood lead by inhalation
and non-inhalation pathways 11-124
11-52 Assumed air lead concentration for model 11-125
11-53 Regression model for blood lead attributable to gasoline 11-127
11-54 Rate of change of 266Pb/2°4Pb and 206Pb/267Pb in air and blood, and
percentage of airborne lead in blood of subjects 1, 3, 5, 6 and 9 11-128
11-55 Calculated blood lead uptake from air lead using Manton isotope study 11-129
11-56 Mean air lead concentrations during the various blood sampling periods
at the measurement sites described in the text (ug/m3) 11-136
11-57 Mean blood lead levels in selected Yugoslavian populations, by
estimated weekly time-weighted air lead exposure 11-140
11-58 Environmental parameters and methods: Arnhem lead study, 1978 11-144
11-59 Geometric mean blood lead levels for children based on reported
occupation of father, history of pica, and distance of residence
from smelter 11-146
11-60 Sources of lead 11-157
11-61 Summary of pooled geometric standard deviations and estimated
analyti c errors 11-160
11-62 Summary of blood inhalation slopes, (B)ug/dl per ug/m3 11-161
11-63 Estimated contribution of leaded gasoline to blood lead by
inhalation and non-inhalation pathways 11-165
xiii
TCPBA/K 8/8/83
-------
PRELIMINARY DRAFT
LIST OF ABBREVIATIONS
AAS
Ach
ACTH
ADCC
ADP/0 ratio
AIDS
AIHA
All
ALA
ALA-D
ALA-S
ALA-U
APDC
APHA
ASTM
ASV
ATP
B-cells
Ba
BAL
BAP
BSA
BUN
BW
C.V.
CaBP
CaEDTA
CBD
Cd
CDC
CEC
CEH
CFR
CMP
CNS
CO
COHb
CP-U
cBah
D.F.
DA
DCMU
DDP
DNA
DTH
EEC
EEC
EMC
EP
EPA
Atomic absorption spectrometry
Acetylcholine
Adrenocoticotrophic hormone
Antibody-dependent cell-mediated cytotoxicity
Adenosine diphosphate/oxygen ratio
Acquired immune deficiency syndrome
American Industrial Hygiene Association
Angiotensin II
Aminolevulinic acid
Aminolevulinic acid dehydrase
Aminolevulinic acid synthetase
Aminolevulinic acid in urine
Ammonium pyrrolidine-dithiocarbamate
American Public Health Association
Amercian Society for Testing and Materials
Anodic stripping voltammetry
Adenosine triphosphate
Bone marrow-derived lymphocytes
Barium
British anti-Lewisite (AKA dimercaprol)
benzo(a)pyrene
Bovine serum albumin
Blood urea nitrogen
Body weight
Coefficient of variation
Calcium binding protein
Calcium ethylenediaminetetraacetate
Central business district
Cadmium
Centers for Disease Control
Cation exchange capacity
Center for Environmental Health
reference method
Cytidine monophosphate
Central nervous system
Carbon monoxide
Carboxyhemoglobi n
Urinary coproporphyrin
plasma clearance of p-aminohippuric acid
Copper
Degrees of freedom
Dopami ne
[3-(3,4-dichlorophenyl)-l,l-dimethylurea
Differential pulse polarography
Deoxyribonucleic acid
Delayed-type hypersensitivity
European Economic Community
Electroencephalogram
Encephaloroyocardi ti s
Erythrocyte protoporphyrin
U.S. Environmental Protection Agency
TCPBA/D
xiv
8/8/83
-------
PRELIMINARY DRAFT
LIST OF ABBREVIATIONS (continued).
FA Fulvic acid
FDA Food and Drug Administration
Fe Iron
FEP Free erythrocyte protoporphyrin
FY Fiscal year
G.M. Grand mean
G-6-PD Glucose-6-phosphate dehydrogenase
GABA Gamma-aminobutyric acid
GALT Gut-associated lymphoid tissue
GC Gas chromatography
GFR Glomerular filtration rate
HA Humic acid
Hg Mercury
hi-vol High-volume air sampler
HPLC High-performance liquid chromatography
i.ro. Intramuscular (method of injection)
i.p. Intraperitoneally (method of infection)
i.v. Intravenously (method of injection)
IAA Indol-3-ylacetic acid
IARC International Agency for Research on Cancer
ICD International classification of diseases
ICP Inductively coupled plasma
IDMS Isotope dilution mass spectrometry
IF Interferon
ILE Isotopic Lead Experiment (Italy)
IRPC International Radiological Protection Commission
K Potassium
LAI Leaf area index
LDH-X Lactate dehydrogenase isoenzyme x
LCcn Lethyl concentration (50 percent)
LD?Q Lethal dose (50 percent)
LH Luteinizing hormone
LIPO Laboratory Improvement Program Office
In National logarithm
LPS Lipopolysaccharide
LRT Long range transport
mRNA Messenger ribonucleic acid
ME Mercaptoethanol
MEPP Miniature end-plate potential
MES Maximal electroshock seizure
MeV Mega-electron volts
MLC Mixed lymphocyte culture
MMD Mass median diameter
MMED Mass median equivalent diameter
Mn Manganese
MNO Motor neuron disease
MSV Moloney sarcoma virus
MTD Maximum tolerated dose
n Number of subjects
N/A Not Available
xv 8/8/83
TCPBA/D
-------
PRELIMINARY DRAFT
LIST OF ABBREVIATIONS
NA
NAAQS
NADB
NAMS
NAS
NASN
NBS
NE
NFAN
NFR-82
NHANES II
Ni
OSHA
P
P
PAH
Pb
PBA
Pb(Ac)?
PbB *
PbBrCl
PBG
PFC
pH
PHA
PH2
PIXE
PMN
PND
PNS
ppm
PRA
PRS
PWM
Py-5-N
RBC
RBF
RCR
redox
RES
RLV
RNA
S-HT
SA-7
son
S.D.
SOS
S.E.M.
SES
SGOT
Not Applicable
National ambient air quality standards
National Aerometric Data Bank
National Air Monitoring Station
National Academy of Sciences
National Air Surveillance Network
National Bureau of Standards
Norepinephrine
National Filter Analysis Network
Nutrition Foundation Report of 1982
National Health Assessment and Nutritional Evaluation Survey II
Nickel
Occupational Safety and Health Administration
Potassium
Significance symbol
Para-aminohippuric acid
Lead
Air lead
Lead acetate
concentration of lead in blood
Lead (II) bromochloride
Porphobilinogen
Plaque-forming cells
Measure of acidity
Phytohemagglutinin
Polyacrylamide-hydrous-zirconia
Proton-induced X-ray emissions
Polymorphonuclear leukocytes
Post-natal day
Peripheral nervous system
Parts per mil lion
Plasma renin activity
Plasma renin substrate
Pokeweed mitogen
Pyrimide-5'-nucleotidase
Red blood cell; erythrocyte
Renal blood flow
Respiratory control ratios/rates
Oxidation-reduction potential
Reticuloendothelial system
Rauscher leukemia virus
Ribonucleic acid
Serotonin
Simian adenovirus
Standard cubic meter
Standard deviation
Sodium dodecyl sulfate
Standard error of the mean
Socioeconomic status
Serum glutamic oxaloacetic transaminase
TCPBA/D
xvi
8/8/83
-------
PRELIMINARY DRAFT
LIST OF ABBREVIATIONS (continued).
slg
SLAMS
SMR
Sr
SRBC
SRMs
STEL
SW voltage
T-cells
t-tests
TBL
TEA
TEL
TIBC
TML
TMLC
TSH
TSP
U.K.
UMP
USPHS
VA
WHO
XRF
X^
Zn
ZPP
Surface immunoglobulin
State and local air monitoring stations
Standardized mortality ratio
Strontium
Sheep red blood cells
Standard reference materials
Short-term exposure limit
Slow-wave voltage
Thymus-derived lymphocytes
Tests of significance
Tri-n-butyl lead
Tetraethyl-ammoni urn
Tetraethyl1ead
Total iron binding capacity
Tetramethyllead
Tetramethyllead chloride
Thyroid-stimulating hormone
Total suspended particulate
United Kingdom
Uridine monophosphate
U.S. Public Health Service
Veterans Administration
Deposition velocity
Visual evoked response
World Health Organization
X-Ray fluorescence
Chi squared
Zinc
Erythrocyte zinc protoporphyrin
MEASUREMENT ABBREVIATIONS
dl
ft
g
g/gal
g/ha-mo
km/hr
1/min
mg/km
ug/m3
mm
umol
ng/cm2
run
nM
sec
deciliter
feet
gram
gram/gallon
gram/hectare•month
kilometer/hour
liter/minute
mi 11i gram/kilometer
microgram/cubic meter
millimeter
micrometer
nanograms/square centimeter
namometer
nanomole
second
TCPBA/D
xvli
8/8/83
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
Chapter 9: Quantitative Evaluation of Lead and Biochemical Indices of Lead
Exposure in Physiological Media
Principal Author
Dr. Paul Mushak
Department of Pathology
School of Medicine
University of North Carolina
Chapel Hill, NC 27514
The following persons reviewed this chapter at EPA's request. The evaluations
and conclusions contained herein, however, are not necessarily those of the
reviewers.
Dr. Carol Angle
Department of Pediatrics
University of Nebraska
College of Medicine
Omaha, NE 68105
Dr. Lee Annest
Division of Health Examin. Statistics
National Center for Health Statistics
3700 East-West Highway
Hyattsville, MD 20782
Dr. Donald Barltrop
Department of Child Health
Westminister Children's Hospital
London SW1P 2NS
England
Dr. Irv Billick
Gas Research Institute
8600 West Bryn Mawr Avenue
Chicago, IL 60631
Dr. Joe Boone
Clinical Chemistry and
Toxicology Section
Centers for Disease Control
Atlanta, GA 30333
Dr. Robert Bornschein
University of Cincinnati
Kettering Laboratory
Cincinnati, OH 45267
Dr. A. C. Chamberlain
Environmental and Medical
Sciences Division
Atomic Energy Research
Establishment
Harwell 0X11
England
Dr. Neil Chernoff
Division of Developmental Biology
MD-67
U.S. Environmental Protection
Agency
Research Triangle Park, NC 27711
Dr. Julian Chisolm
Baltimore City Hospital
4940 Eastern Avenue
Baltimore, MD 21224
Mr. Jerry Cole
International Lead-Zinc Research
Organization
292 Madison Avenue
New York, NY 10017
Dr. Max Costa
Department of Pharmacology
University of Texas Medical
School
Houston, TX 77025
Dr. Anita Curran
Commissioner of Health
Westchester County
White Plains, NY 10607
xvi ii
-------
Dr. Jack Dean
Immunobiology Program and
Immunotoxicology/Cell Biology program
CUT
P.O. Box 12137
Research Triangle Park, NC 27709
Dr. H. T. Delves
Chemical Pathology and Human
Metabolism
Southampton General Hospital
Southampton S09 4XY
England
Dr. Fred deSerres
Assoc. Director for Genetics
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Dr. Robert Dixon
Laboratory of Reproductive and
Developmental Toxicology
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Dr. Claire Ernhart
Department of Psychiatry
Cleveland Metropolitan General Hospital
Cleveland, OH 44109
Dr. Sergio Fachetti
Section Head - Isotope Analysis
Chemistry Division
Joint Research Center
121020 Ispra
Varese, Italy
Dr. Virgil Ferm
Department of Anatomy and Cytology
Dartmouth Medical School
Hanover, NH 03755
Dr. Alf Fischbein
Environmental Sciences Laboratory
Mt. Sinai School of Medicine
New York, NY 10029
Dr. Jack Fowle
Reproductive Effects Assessment Group
U.S. Environmental Protection Agency
RD-689
Washington, DC 20460
Dr. Bruce Fowler
Laboratory of Pharmacology
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Dr. Warren Galke
Department of Biostatistics
and Epidemiology
School of Allied Health
East Carolina University
Greenville, NC 27834
Mr. Eric Goldstein
Natural Resources Defense
Council, Inc.
122 E. 42nd Street
New York, NY 10168
Dr. Harvey Gonick
1033 Gayley Avenue
Suite 116
Los Angeles, CA 90024
Dr. Robert Goyer
Deputy Director
NIEHS
P.O. Box 12233
Dr. Stanley Gross
Hazard Evaluation Division
Toxicology Branch
U.S. Environmental Protection
Agency
Washington, DC 20460
Dr. Paul Hammond
University of Cincinnati
Kettering Laboratory
Cincinnati, OH 45267
Dr. Ronald D. Hood
Department of Biology
The University of Alabama
University, AL 35486
Dr. V. Houk
Centers for Disease Control
1600 Clifton Road, NE
Atlanta, GA 30333
xix
-------
Dr. Loren D. Koller
School of Veterinary Medicine
University of Idaho
Moscow, ID 83843
Dr. Kristal Kostial
Institute for Medical Research
and Occupational Health
Yu-4100 Zagreb
Yugoslavia
Dr. Lawrence Kupper
Department of Biostatistics
UNC School of Public Health
Chapel Hill, NC 27514
Dr. Phillip Landrigan
Division of Surveillance,
Hazard Evaluation and Field Studies
Taft Laboratories - NIOSH
Cincinnati, OH 45226
Dr. David Lawrence
Microbiology and Immunology Dept.
Albany Medical College of Union
University
Albany, NY 12208
Dr. Jane Lin-Fu
Office of Maternal and Child Health
Department of Health and Human Services
Rockville, MD 20857
Dr. Don Lynam
Air Conservation
Ethyl Corporation
451 Florida Boulevard
Baton Rouge, LA 70801
Dr. Kathryn Mahaffey
Division of Nutrition
Food and Drug Administration
1090 Tusculum Avenue
Cincinnati, OH 45226
Dr. Ed McCabe
Department of Pediatrics
University of Wisconsin
Madison, WI 53706
Dr. Chuck Nauman
Exposure Assessment Group
U.S. Environmental Protection
Agency
Washington, DC 20460
Dr. Herbert L. Needleman
Children's Hospital of Pittsburgh
Pittsburgh, PA 15213
Dr. H. Mitchell Perry
V.A. Medical Center
St. Louis, MO 63131
Dr. Jack Pierrard
E.I. duPont de Nemours and
Company, Inc.
Petroleum Laboratory
Wilmington, DE 19898
Dr. Sergio Piomelli
Columbia University Medical School
Division of Pediatric Hematology
and Oncology
New York, NY 10032
Dr. Magnus Piscator
Department of Environmental Hygiene
The Karolinska Institute 104 01
Stockholm
Sweden
Dr. Robert Putnam
International Lead-Zinc
Research Organization
292 Madison Avenue
New York, NY 10017
Dr. Michael Rabinowitz
Children's Hospital Medical
Center
300 Longwood Avenue
Boston, MA 02115
xx
-------
Dr. Harry Roels
Unite de Toxicologie
Industrie!le et Medicale
Universite de Louvain
Brussels, Belgium
Dr. John Rosen
Division of Pediatric Metabolism
Albert Einstein College of Medicine
Montefiore Hospital and Medical Center
111 East 210 Street
Bronx, NY 10467
Dr. Michael Rutter
Department of Psychology
Institute of Psychiatry
DeCrespigny Park
London SE5 SAL
England
Dr. Stephen R. Schroeder
Division for Disorders
of Development and Learning
Biological Sciences Research Center
University of North Carolina
Chapel Hill, NC 27514
Dr. Anna-Maria Seppalainen
Institutes of Occupational Health
Tyoterveyslaitos
Haartmaninkatu 1
00290 Helsinki 29
Finland
Dr. Ellen Silbergeld
Environmental Defense Fund
1525 18th Street, NW
Washington, DC 20036
Dr Ron Snee
E.I. duPont Nemours and
Company, Inc.
Engineering Department L3167
Wilmington, DE 19898
Dr. Gary Ter Haar
Toxicology and Industrial
Hygiene
Ethyl Corporation
451 Florida Boulevard
Baton Rouge, LA 70801
Mr. Ian von Lindern
Department of Chemical Engineering
University of Idaho
Moscow, Idaho 83843
Dr. Richard P. Wedeen
V.A. Medical Center
Tremont Avenue
East Orange, MJ 07019
xx1
-------
Chapter 10: Metabolism of Lead
Principal Author
Dr. Paul Mushak
Department of Pathology
School of Medicine
University of North Carolina
Chapel Hill, NC 27514
Contributing Author
Dr. Michael Rabinowitz
Children's Hospital Medical Center
300 Longwood Avenue
Boston, MA 02115
The following persons reviewed this chapter at EPA's request. The evaluations
and conclusions contained herein, however, are not necessarily those of the
reviewers.
Dr. Carol Angle
Department of Pediatrics
University of Nebraska
College of Medicine
Omaha, NE 68105
Dr. Lee Annest
Division of Health Exanrin. Statistics
National Center for Health Statistics
3700 East-West Highway
Hyattsville, MO 20782
Dr. Donald Barltrop
Department of Child Health
Westminister Children's Hospital
London SW1P 2NS
England
Dr. Irv Billick
Gas Research Institute
8600 West Bryn Mawr Avenue
Chicago, IL 60631
Or. Joe Boone
Clinical Chemistry and
Toxicology Section
Centers for Disease Control
Atlanta, GA 30333
Dr. Robert Bornschein
University of Cincinnati
Kettering Laboratory
Cincinnati, OH 45267
Dr. A. C. Chamberlain
Environmental and Medical
Sciences Division
Atomic Energy Research
Establishment
Harwell 0X11
England
Dr. Neil Chernoff
Division of Developmental Biology
MD-67
U.S. Environmental Protection
Agency
Research Triangle Park, NC 27711
Or. Julian Chisolm
Baltimore City Hospital
4940 Eastern Avenue
Baltimore, MD 21224
Mr. Jerry Cole
International Lead-Zinc Research
Organization
292 Madison Avenue
New York, NY 10017
xxii
-------
Dr. Max Costa
Department of Pharmacology
University of Texas Medical School
Houston, TX 77025
Dr. Anita Outran
Commissioner of Health
Westchester County
White Plains, NY 10607
Dr. Jack Dean
Immunobiology Program and
Immunotoxicology/Cell Biology program
CUT
P.O. Box 12137
Research Triangle Park, NC 27709
Dr. H.T. Delves
Chemical Pathology and Human Metabolism
Southampton General Hospital
Southampton S09 4XY
England
Dr. Fred deSerres
Assoc. Director for Genetics
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Dr. Robert Dixon
Laboratory of Reproductive and
Developmental Toxicology
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Dr. Claire Ernhart
Department of Psychiatry
Cleveland Metropolitan General Hospital
Cleveland, OH 44109
Dr. Sergio Fachetti
Section Head - Isotope Analysis
Chemistry Division
Joint Research Center
121020 Ispra
Varese, Italy
Dr. Virgil Perm
Department of Anatomy and Cytology
Dartmouth Medical School
Hanover, NH 03755
xxiii
Dr. Alf Fischbein
Environmental Sciences Laboratory
Mt. Sinai School of Medicine
New York, NY 10029
Dr. Dr. Jack Fowle
Reproductive Effects Assessment
Group
U.S. Environmental Protection
Agency
RD-689
Washington, DC 20460
Dr. Bruce Fowler
Laboratory of Pharmacology
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Dr. Warren Galke
Department of Biostatisties
and Epidemiology
School of Allied Health
East Carolina University
Greenville, NC 27834
Mr. Eric Goldstein
Natural Resources Defense
Council, Inc.
122 E. 42nd Street
New York, NY 10168
Dr. Harvey Gonick
1033 Gayley Avenue
Suite 116
Los Angeles, CA 90024
Dr. Robert Goyer
Deputy Director
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Or. Stanley Gross
Hazard Evaluation Division
Toxicology Branch
U.S. Environmental Protection
Agency
Washington, DC 20460
Dr. Paul Hammond
University of Cincinnati
Kettering Laboratory
Cincinnati, OH 45267
-------
Or. Ronald 0. Hood
Department of Biology
The University of Alabama
University, AL 35486
Dr. V. Houk
Centers for Disease Control
1600 Clifton Road, NE
Atlanta, GA 30333
Dr. Loren D. Koller
School of Veterinary Medicine
University of Idaho
Moscow, ID 83843
Dr. Kristal Kostial
Institute for Medical Research
and Occupational Health
Yu-4100 Zagreb
Yugoslavia
Dr. Lawrence Kupper
Department of Biostatisties
UNC School of Public Health
Chapel Hill, NC 27514
Dr. Phillip Landrigan
Division of Surveillance,
Hazard Evaluation and Field Studies
Taft Laboratories - NIOSH
Cincinnati, OH 45226
Dr. David Lawrence
Microbiology and Immunology Dept.
Albany Medical College of Union
University
Albany, NY 12208
Dr. Jane Lin-Fu
Office of Maternal and Child Health
Department of Health and Human Services
Rockville, MD 20857
Dr. Don Lynam
Air Conservation
Ethyl Corporation
451 Florida Boulevard
Baton Rouge, LA 70801
Dr. Kathryn Mahaffey
Division of Nutrition
Food and Drug Administration
1090 Tusculum Avenue
Cincinnati, OH 45226
xxiv
Dr. Ed McCabe
Department of Pediatrics
University of Wisconsin
Madison, WI 53706
Dr. Chuck Nauman
Exposure Assessment Group
U.S. Environmental Protection Agency
Washington, DC 20460
Dr. Herbert L. Neddleman
Children's Hospital of Pittsburgh
Pittsburgh, PA 15213
Dr. H. Mitchell Perry
V.A. Medical Center
St. Louis, MO 63131
Dr. Jack Pierrard
E.I. duPont de Nemours and
Company, Inc.
Petroleum Laboratory
Wilmington, DE 19898
Dr. Sergio Piomelli
Columbia University Medical School
Division of Pediatric Hematology
and Oncology
New York, NY 10032
Dr. Magnus Piscator
Department of Environmental Hygiene
The Karolinska Institute 104 01
Stockholm
Sweden
Dr. Robert Putnam
International Lead-Zinc
Research Organization
292 Madison Avenue
New York, NY 10017
Dr. Harry Roels
Unite de Toxicologie
Industrielle et Medicale
Universite de Louvain
Brussels, Belgium
Dr. John Rosen
Division of Pediatric Metabolism
Albert Einstein College of Medicine
Montefiore Hospital and Medical Center
111 East 210 Street
Bronx, NY 10467
-------
Dr. Michael Rutter
Department of Psychology
Institute of Psychiatry
DeCrespigny Park
London SE5 SAL
England
Dr. Stephen R. Schroeder
Division for Disorders
of Development and Learning
Biological Sciences Research Center
University of North Carolina
Chapel Hill, NC 27514
Dr. Anna-Maria Seppalainen
Institutes of Occupational Health
Tyoterveyslaitos
Haartmaninkatu 1
00290 Helsinki 29
Finland
Dr. Ellen Silbergeld
Environmental Defense Fund
1525 18th Street, NW
Washington, DC 20036
Dr. Ron Snee
E.I. duPont Nemours and
Company, Inc.
Engineering Department L3167
Wilmington, DE 19898
Dr. Gary Ter Haar
Toxicology and Industrial
Hygiene
Ethyl Corporation
451 Florida Boulevard
Mr. Ian von Lindern
Department of Chemical
Engineering
University of Idaho
Moscow, ID 83843
Dr. Richard P. Wedeen
V.A. Medical Center
Tremont Avenue
East Orange, NO 07019
xxv
-------
Chapter 11: Assessment of Lead Exposures and Absorption in Human Populations
Principal Authors
Dr. Warren Galke
Department of Biostatisties and Epidemiology
School of Allied Health
East Carolina University
Greenville, NC 27834
Dr. Alan Marcus
Department of Mathematics
Washington State University
Pullman, Washington 99164-2930
Contributing Author:
Dr. Dennis Kotchmar
Environmental Criteria and Assessment Office
MD-52
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
Dr. Vic Hasselblad
Biometry Division
MD-55
U.S. Environmental Protection
Agency
Research Triangle Park, NC 27711
The following persons reviewed this chapter at EPA's request. The evaluations
and conclusions contained herein, however are not necessarily those of the
reviewers.
Dr. Carol Angle
Department of Pediatrics
University of Nebraska
College of Medicine
Omaha, NE 68105
Dr. Lee Annest
Division of Health Examin. Statistics
National Center for Health Statistics
3700 East-West Highway
Hyattsville, MD 20782
Dr. Donald Barltrop
Department of Child Health
Westminister Children's Hospital
London SW1P 2NS
England
Dr. Irv Billick
Gas Research Institute
8600 West Bryn Mawr Avenue
Chicago, IL 60631
Dr. Joe Boone
Clinical Chemistry and
Toxicology Section
Centers for Disease Control
Atlanta, GA 30333
Dr. Robert Bornschein
University of Cincinnati
Kettering Laboratory
Cincinnati, OH 45267
Dr. A. C. Chamberlain
Environmental and Medical
Sciences Division
Atomic Energy Research
Establishment
Harwell 0X11
England
Dr. Neil Chernoff
Division of Developmental Biology
MD-67
U.S. Environmental Protection
Agnecy
Research Triangle Park, NC 27711
xxvi
-------
Dr. Julian Chisolm
Baltimore City Hospital
4940 Eastern Avenue
Baltimore, MD 21224
Mr. Jerry Cole
International Lead-Zinc Research Organization
292 Madison Avenue
New York, NY 10017
Dr. Max Costa
Department of Pharmacology
University of Texas Medical School
Houston, TX 77025
Dr. Anita Curran
Commissioner of Health
Westchester County
White Plains, NY 10607
Dr. Jack Dean
Immunobiology Program and
Immunotoxicology/Cell Biology Program
CUT
P.O. Box 12137
Research Triangle Park, NC 27709
Dr. Fred deSerres
Assoc. Director for Genetics
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Dr. Robert Dixon
Laboratory of Reproductive and
Developmental Toxicology
NIEHS
P.O. Box 12233
Research Tn angle Park, NC 27709
Dr. Claire Ernhart
Department of Psychiatry
Cleveland Metropolitan General Hospital
Cleveland, OH 44109
Dr. Sergio Fachetti
Section Head - Isotope Analysis
Chemistry Division
Joint Research Center
121020 Ispra
Varese, Italy
Dr. Virgil Ferm
Department of Anatomy and Cytology
Dartmouth Medical School
Hanover, NH 03755
Dr. Alf Fischbein
Environmental Sciences Laboratory
Mt. Sinai School of Medicine
New York, NY 10029
Dr. Jack Fowle
Reproductive Effects Assessment
Group
U.S. Environmental Protection
Agency
RD-689
Washington, DC 20460
Dr. Bruce Fowler
Laboratory of Pharmocology
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Mr. Eric Goldstein
Natural Resources Defense
Council, Inc.
School of Allied Health
122 E. 42nd Street
New York, NY 10168
Dr. Harvey Gonick
1033 Gayley Avenue
Suite 116
Los Angeles, CA 90024
Dr. Robert Goyer
Deputy Director
NIEHS
P.O. Box 12233
Research Triangle Park, NC 27709
Dr. Stanley Gross
Hazard Evaluation Division
Toxicology Branch
U.S. Environmental Protection
Dr. Paul Hammond
University of Cincinnati
Kettering Laboratory
3223 Eden Avenue
Cincinnati, OH 45267
xxvil
-------
Dr. Ronald D. Hood
Department of Biology
The University of Alabama
University, AL 35486
Dr. V. Houk
Centers for Disease Control
1600 Clifton Road, NE
Atlanta, GA 30333
Dr. Loren Koller
School of Veterinary Medicine
University of Idaho
Moscow, ID 83843
Dr. Kristal Kostial
Institute for Medical Research
and Occupational Health
Yu-4100 Zagreb
Yugoslavia
Dr. Lawrence Kupper
Department of Biostatisties
UNC School of Public Health
Chapel Hill, NC 27514
Dr. Phillip Landrigan
Division of Surveillance,
Hazard Evaluation and Field Studies
Taft Laboratories - NIOSH
Cincinnati, OH 45226
Dr. David Lawrence
Microbiology and Immunology Dept.
Albany Medical College of Union
University
Albany, NY 12208
Dr. Jane Lin-Fu
Office of Maternal and Child Health
Department of Health and Human Services
Rockville, MD 20857
Dr. Don Lynam
Air Conservation
Ethyl Corporation
451 Florida Boulevard
Baton Rouge, LA 70801
Dr. Kathryn Mahaffey
Division of Nutrition
Food and Drug Administration
1090 Tusculum Avenue
Cincinnati, OH 45226
Dr. Ed McCabe
Department of Pediatrics
University of Wisconsin
Madison, WI 53706
Dr. Paul Mushak
Department of Pathology
UNC School of Medicine
Chapel Hill, NC 27514
Dr. Chuck Nauman
Exposure Assessment Group
U.S. Environmental Protection
Agency
Washington, DC 20460
Dr. Herbert L. Needleman
Children's Hospital of Pittsburgh
Pittsburgh, PA 15213
Dr. H. Mitchell Perry
V.A. Medical Center
St. Louis, MO 63131
Dr. Jack Pierrard
E.I. duPoint de Nemours and
Company, Inc.
Petroleum Laboratory
Wilmington, DE 19898
Dr. Sergio Piomelli
Columbia University Medical School
Division of Pediatric Hematology
and Oncology
New York, NY 10032
Dr. Magnus Piscator
Department of Environmental Hygiene
The Karolinska Institute 104 01
Stockholm
Sweden
xxvlii
-------
Dr. Robert Putnam
International Lead-Zinc
Research Organization
292 Madison Avenue
New York, NY 10017
Dr. Michael Rabinowitz
Children's Hospital Medical Center
300 Longwood Avenue
Boston, MA 02115
Dr. Harry Roels
Unite de Toxicologie
Industrielle et Medicale
Universite de Louvain
Brussels, Belgium
Dr. John Rosen
Division of Pediatric Metabolism
Albert Einstein College of Medicine
Montefiore Hospital and Medical Center
111 East 210 Street
Bronx, NY 10467
Dr. Stephen R. Schroeder
Division for Disorders
of Development and Learning
Biological Sciences Research Center
University of North Carolina
Chapel Hill, NC 27514
Dr. Anna-Maria Seppalainen
Institutes of Occupational Health
Tyoterveysla i tos
Haartmaninkatu 1
00290 Helsinki 29
Finland
Dr. Ellen Silbergeld
Environmental Defense Fund
1525 18th Street, NW
Washington, DC 20036
Dr. Ron Snee
E.I. duPont Nemours and
Company, Inc.
Engineering Department L3267
Wilmington, DE 19898
Dr. Gary Ter Haar
Toxicology and Industrial
Hygiene
Ethyl Corporation
451 Florida Boulevard
Baton Rouge, LA 70801
Mr. Ivon von Lindern
Department of Chemical Engineering
University of Idaho
Moscow, ID 83843
Dr. Richard P. Weeden
V.A. Medical Center
Tremont Avenue
East Orange, NO 07019
xxix
-------
PRELIMINARY DRAFT
9. QUANTITATIVE EVALUATION OF LEAD AND BIOCHEMICAL INDICES
OF LEAD EXPOSURE IN PHYSIOLOGICAL MEDIA
9.1 INTRODUCTION
In order to completely understand a given agent's effects on an organism, e.g., dose-
effect relationships, a quantitative evaluation of the substance in some indicator medium and
knowledge of the physiological parameters associated with exposure is vital. This said, two
questions follow:
1) What are the most accurate, precise, and efficient ways to
carry out such measurements?
2) In the case of lead (lead itself or biological indicators),
which measurement methods in which media are most appropri-
ate for each particular exposure?
Under the rubric of "analysis" are a number of discrete steps, all of which are important
contributors to the quality of the final result: (1) collection of samples and transmission
to the laboratory; (2) laboratory manipulation of samples, physically and chemically, before
analysis by instruments; (3) instrumental analysis and quantitative measurement; and (4)
establishment of relevant criteria for accuracy and precision, namely, internal and external
quality assurance checks. Each of these steps is discussed in this chapter.
It is clear that the definition of "satisfactory analytical method" for lead has been
changing over the years in ways paralleling (1) the evolution of more sophisticated instrumen-
tation and procedures, (2) a greater awareness of such factors as background contamination and
loss of element from samples, and (3) development of new statistical methods to analyze data.
For example, current methods of lead analysis, such as anodic stripping voltammetry, back-
ground-corrected atomic absorption spectrometry, and isotope dilution mass spectrometry (par-
ticularly the latter), are more sensitive and specific than the older classical approaches.
Increasing use of the newer methods would tend to result in lower lead values being reported
for a given sample. Whether this trend in analytical improvement can be isolated from such
other variables as temporal changes in exposure is another matter.
Since lead is ubiquitously distributed as a contaminant, the constraints (i.e., ultra-
clean, ultra-trace analysis) placed upon a laboratory attempting analysis of geochemical
samples of pristine origin, or of extremely low lead levels in biological samples such as
plasma, are quite severe. Very few laboratories can credibly claim such capability. Ideally,
similar standards of quality should be adhered to across the rest of the analytical spectrum.
With many clinical, epidemiological, and experimental studies, however, this may be unrealis-
tic, given practical limitations and objectives of the studies. Laboratory performance is but
23PB12/C 9-1 7/1/83
-------
PRELIMINARY DRAFT
one part of the quality equation; the problems of sampling are equally important but less sub-
ject to tight control. The necessity of rapidly obtaining a blood sample in cases of suspec-
ted lead poisoning, or of collecting hundreds or thousands of blood samples in urban popu-
lations, limits the number of sampling safeguards to those that can be realistically achieved.
Sampling in this context will always be accompanied by a certain amount of analytical
"suspicion." Furthermore, a certain amount of biological lead analysis data is employed for
comparative purposes, as in experimental studies concerned with the relative increase in tis-
sue burden of lead associated with increases in doses or severity of effects. In addition
any major compromise of an analytical protocol may be statistically discernible. Thus, anal-
ysis of biological media for lead must be done under protocols that minimize the risk of in-
accuracy. Specific accuracy and precision characteristics of a method in a particular report
should be noted to permit some judgment on the part of the reader about the influence of
methodology on the reported results.
The choice of measurement method (see Question 2) and medium for analysis is dictated
both by the type of information desired and by technical or logistical considerations. As
noted elsewhere in this document, whole blood lead reflects recent or continuing exposure,
whereas lead in mineralized tissue, such as deciduous teeth, reflects an exposure period of
months and years. While urine lead values are not particularly good correlates of lead ex-
posure under steady-state conditions in populations at large, such measurements may be of con-
siderable clinical value. In acquisition of blood samples, the choice of venipuncture or
finger puncture will be governed by such factors as cost and feasibility, contamination risk,
the biological quality of the sample, etc. The use of biological indicators that strongly
correlate with lead burden may be more desirable since they provide evidence of actual re-
sponse and, together with blood lead data, provide a less risky diagnostic tool for assessment
of lead exposure.
9.2 DETERMINATIONS OF LEAD IN BIOLOGICAL MEDIA
9.2.1 Sampling and Sample Handling Procedures for Lead in Biological Media
Lead analysis in biological media requires careful collection and handling of samples for
two special reasons: (1) lead occurs at trace levels in most indicators of subject exposure
even under conditions of high lead exposure, and (2) such samples roust be obtained against a
backdrop of pervasive contamination, the full extent of which may still be unrecognized by
many laboratories.
The reports of Speecke et al. (1976), Patterson and Settle (1976), Murphy (1976), Berman
(1976), and Settle and Patterson (1980) review detailed aspects of the problems of sampling
and subsequent sample handling in the laboratory. It is clear from these discussions that the
23PB12/C 9-2 7/1/83
-------
PRELIMINARY DRAFT
normal precautions taken in the course of sample acquisition (detailed below for clinical and
epidemiological studies) should not be taken as absolute, but rather as what is practical and
feasible. Furthermore, it may also be the case that the inherent sensitivity or accuracy of a
given methodology or instrumentation is less of a determining factor in the overall analysis
than is quality of sample collection and handling.
9.2.1.1 Blood Sampling. Samples for blood lead determination may be collected by venipunc-
ture (venous blood) or finger tip puncture (capillary blood). Collection of capillary vs.
venous blood is normally decided by a number of factors, including the feasibility of obtain-
ing samples during screening of many subjects and the difficulty of securing subject compli-
ance, particularly in the case of children and their parents. Furthermore, capillary blood
may be collected as discrete quantities in small-volume capillary tubes or as spots on filter
paper disks. With capillary tubes, obtaining good mixing with anticoagulant to avoid clotting
is important, as is the problem of lead contamination of the tube. The use of filter paper
requires the selection of paper with uniform composition, low lead content, and uniform blood
dispersal.characteristics.
Whether venous or capillary blood is collected, much care must be exercised in cleaning
the site before puncture as well as in selecting lead-free receiving containers. Cooke et al.
(1974) employed vigorous scrubbing with a low-lead soap solution and deionized water rinsing,
while Marcus et al. (1975) carried out preliminary cleaning with an ethanolic citric acid
solution followed by 70 percent ethanol rinsing. The vigor in cleaning the puncture site is
probably as important as any particular choice of cleaning agent. Marcus et al. (1977) noted
that in one procedure for puncture site preparation, where the site is covered with wet paper
towels, contamination will occur if the paper towels are made from recycled paper, owing to
significant lead retention in recycled paper.
In theory, capillary and venous blood lead levels should be virtually identical, although
the available literature indicates that some differences, which mainly reflect problems of
sampling, do arise in the case of capillary blood. A given amount of contaminant has a
greater impact on a 100 ul fingerstick sample than on a 5 ml sample of venous blood. Finger
coating techniques may reduce some of the contamination problem (Mitchell et al., 1974). An
additional problem is the presence of lead in the anticoagulants used to coat capillary tubes.
Also, lower values of capillary vs. venous blood lead may reflect "dilution" of the sample by
extracellular fluid owing to excessive compression of the puncture site. When Joselow and
Bogden (1972) compared a method using finger puncture and spotting onto filter paper with a
procedure using venous blood and Hessel's procedure (1968) for flame atomic absorption spec-
trometry, they obtained a correlation coefficient of r = 0.9 (range, 20-46 pg/dl). Similarly,
Cooke et al. (1974) found an r value of 0.8 (no range given), while Mitchell et al. (1974)
23PB12/C 9-3 7/1/83
-------
PRELIMINARY DRAFT
obtained a value of 0.92 (10-92 ug/dl). Mahaffey et al. (1979) found that capillary blood
levels in a comparison test were approximately 20 percent higher than corresponding venous
blood levels in the same subjects, presumably reflecting sample contamination. Similar eleva-
tions have been described by DeSilva and Donnan (1980). Carter (1978) has found that blood
samples with lower hemoglobin levels may spread onto filter paper differently from normal
hemoglobin samples, requiring correction in quantification to obtain values that are reliable.
This complication should be kept in mind when considering children, who are frequently prone
to iron-deficiency anemia.
The relative freedom of the blood container from interior surface lead and the amount of
lead in the anticoagulant used are important considerations in venous sampling. For studies
focused on "normal" ranges, such tubes may add some lead to blood and still meet certification
requirements. The "low-lead" heparinized blood tubes commercially available (blue stopper
Vacutainer, Becton-Dickinson) were found to contribute less than 0.2 |jg/d1 to whole blood
samples (Rabinowitz and Needleman, 1982). Nackowski et al. (1977) surveyed a large variety of
commercially available blood tubes for lead and other metal contamination. Lead uptake by
blood over time from the various tubes was minimal with the "low-lead" Vacutainer tubes and
with all but four of the other tube types. In the large survey of Mahaffey et al. (1979),
5-ml Monoject (Sherwood) or 7-ml lavender-top Vacutainer (Becton-Dickinson) tubes were found
satisfactory. However, when more precision is needed, tubes are best recleaned in the labor-
atory and lead-free anticoagulant added (although this would be less convenient for sampling
efficiency than the commercial tubes). In addition, blank levels for every batch of samples
should be verified.
9.2.1.2 Urine Sampling. Urine samples require collection in lead-free containers and caps as
well as the addition of a low-lead bacteriocide if samples are to be stored for any period of
time. While not always feasible, 24-hour samples should be obtained, as such collection would
level out any effect of variation in excretion over time. If spot sampling is done, lead
levels should be expressed per unit creatinine. For 24-hour collections, corrections must be
made for urine density.
9.2.1.3 Hair Sampling. The usefulness of hair lead analysis depends on the manner of samp-
ling. Hair samples should be removed from subjects by some consistent method, either by a
predetermined length measured from the skin or by using the entire hair. Hair should be
placed in air-tight containers for shipment or storage. For segmental analysis, the entire
hair length is required.
9.2.1.4 Mineralized Tissue. An important consideration in deciduous tooth collection Is
consistency in the type of teeth collected from various subjects. Fosse and Justesen (1978)
reported no difference 1n lead content between molars and incisors, and Chatman and Wilson
(1975) reported comparable whole tooth levels for cuspids, incisors, and molars. On the other
hand, Mackie et al. (1977) and Lockeretz (1975) noted levels varying with tooth type, with a
23PB12/C 9-4 7/1/83
-------
PRELIMINARY DRAFT
statistically significant difference (Mackie et al., 1977) between second molar (lowest
levels) and incisors (highest levels). The fact that the former two studies found rather low
overall lead levels across groups, while Mackie et al. (1977) reported higher values, suggests
that dentition differences in lead content may be magnified at relatively higher levels of ex-
posure. Delves et al. (1982), comparing pairs of central incisors or pairs of central and
lateral incisors from the same child, found that lead levels may even vary within a specific
type of tooth. These data suggest the desirability of acquiring two teeth per subject to get
an average lead value.
Teeth containing fillings or extensive decay are best eliminated from analysis. Mackie
et al. (1977) discarded decayed teeth if the extent of decay exceeded approximately 30 per-
cent.
9.2.1.5 Sample Handling in the Laboratory. With blood samples, there is the potential prob-
lem of the effect of storage on the lead content. It is clear that dilute aqueous solutions
of lead will surrender a sizable portion of the lead content to the container surface, whether
glass or plastic (Issaq and Zielinski, 1974; Linger and Green, 1977); whether there is a com-
parable effect, or the extent of such an effect, with blood is not clear. Linger and Green
(1977) claim that lead loss from blood to containers parallels that seen with aqueous solu-
tions, but their data do not support this assertion. Moore and Meredith (1977) used isotopic
lead spiking (203Pb) with and without carrier in various containers at differing temperatures
to monitor lead stability in blood over time. The only material loss occurred with soda glass
at room temperature after 16 days. Nackowski et al. (1977) found that "low-lead" blood tubes,
while quite satisfactory in terms of sample contamination, began to show transfer of lead to
the container wall after four days. Meranger et al. (1981) studied movement of lead, spiked
to various levels, to containers of various composition as a function of temperature and time.
In all cases, reported lead loss to containers was significant. However, there are problems
with the above reports. Spiked samples probably are not incorporated Into the same biochemi-
cal environment as lead inserted in vivo. The Nackowski et al. (1977) study did not indicate
whether the blood samples were kept frozen or refrigerated between testing intervals.
Mitchell et al. (1972) found that the effect of blood storage depends on the method of anal-
ysis, with lower recoveries of lead from aged blood being seen using the Hessel (1968) method.
Lerner (1975) collected blood samples (35 originally) from a single subject into lead-
free tubes and, after freezing, forwarded them in blind fashion to a certified testing labor-
atory over a period of 9 months. Four samples were lost, while one was rejected as being
grossly contaminated (4 standard deviations from mean). Of the remaining 30 samples, the mean
was 18.3 pg/dl with a standard deviation (S.D.) of 3.9. The analytical method had a precision
of ±3.5 |jg Pb/dl (1 = S.D.) at normal levels of lead, suggesting that the overall stability of
the samples in terms of lead content, was good. Boone et al. (1979), reported that samples
23PB12/C 9-5 7/1/83
-------
PRELIMINARY DRAFT
frozen for periods of less than a year showed no effect of storage, while Piscator (1982)
noted no change in low levels (<10 (jQ/dl) when samples were stored at -20°C for 6 months.
Based on the above data, it appears that blood samples to be stored for any period of time
should be frozen rather than refrigerated, with care taken to prevent breaking of the tube
during freezing. Teeth and hair samples, when stored in containers to minimize contamination
are indefinitely stable.
The actual site of analysis should be as lead-free as possible. Given the uncommon
availability of an "ultra-clean" facility such as that described by Patterson and Settle
(1976), the next desirable level of laboratory cleanliness is the "Class 100" facility, in
which there are fewer than 100 airborne particles >0.5 urn. These facilities employ high ef-
ficiency particulate air filtering and laminar air flow (with movement away from sample
handling areas). Totally inert surfaces in the working area and an antechamber for removing
contaminated clothes, appliance cleaning, etc. are other necessary features.
All plastic and glass ware coming into contact with samples should be rigorously cleaned
and stored away from dust contact; materials such as ashing vessels should permit minimal lead
leaching. In this regard, Teflon and quartz ware is more desirable than other plastics or
borosilicate glass (Patterson and Settle, 1976).
Reagents, particularly for chemical degradation of biological samples, should be both
certified and periodically tested for retention of quality. Several commercial grades of re-
agents are available, although precise work may require doubly purified materials from the
National Bureau of Standards. These reagents should be stored with a minimum of surface con-
tamination around the top of the containers.
For a more detailed discussion of appropriate laboratory practices, the reader may con-
sult LaFleur (1976).
9.2.2 Methods of Lead Analysis
Detailed technical discussion of the array of instruments available to measure lead in
blood and other media is outside the scope of this Chapter (see Chapter 4). This discussion
is structured more appropriately to those aspects of methodology dealing with relative sensi-
tivity, specificity, accuracy and precision. While there is increasing acceptance of interna-
tional standardized units (SI units) for expressing lead levels in various media, units famil-
iar to clinicians and epidemiologists will be used here. (To convert ug Pb/dl blood to SI
units (umoles/liter), multiply by 0.048.)
Many reports over the years have purported to offer satisfactory analysis of lead in bio-
logical media, but in fact have shown rather meager adherence to criteria for accuracy and
precision or have shown a lack of demonstrable utility across a wide spectrum of analytical
applications. Therefore, discussion in this section is confined to "definitive" and reference
23PB12/C 9-6 7/1/83
-------
PRELIMINARY DRAFT
methods for lead analysis, except for a brief treatment of the traditional but now widely sup-
planted colorimetric method.
Using the definition of Cali and Reed (1976), a definitive method is one in which all
major or significant parameters are related by solid evidence to the absolute mass of the ele-
ment with a high degree of confidence. A reference method, by contrast, is one of demonstra-
ted accuracy, validated by a definitive method and arrived at by consensus through performance
testing by a number of different laboratories. In the case of lead in biological media, the
definitive method is isotope-dilution mass spectrometry (IDMS). IDMS accuracy comes from the
fact that all manipulations are on a weight basis involving simple procedures. The measure-
ments entail only ratios and not the absolute determinations of the isotopes involved, which
greatly reduces instrumental corrections or errors. Reproducible results to a precision of
one part in 104 or 10s are routine with specially designed instruments.
In terms of reference methods for lead in biological media, such a label cannot techni-
cally be attached to atomic absorption spectrometry in its various instrumentation/
methodology configurations or to the electrochemical technique, anodic stripping voltammetry.
However, these have been termed reference methods insofar as their precision and accuracy can
be verified or calibrated against IDMS.
Other methods that are recognized for trace metal analysis in general are not fully ap-
plicable to biological lead or have inherent shortcomings. X-ray fluorescence analysis lacks
the requisite sensitivity for media with low lead content and the associated sample prepara-
tion may present a high contamination risk. A notable exception may be X-ray fluorescence
analysis of teeth or bone iji situ as discussed below. Neutron activation analysis is the
method of choice with many elements, but is not technically feasible for lead analysis because
of the absence of long-lived isotopes.
9.2.2.1 Lead Analysis in Whole Blood. The first generally accepted technique for quantifying
lead in whole blood and other biological media was a colorimetric method that involved spec-
trophotometric measurement based on the binding of lead to a chromogenic agent to yield a
chromophoric complex. The complexing agent has typically been dithizone, 1,5-diphenylthio-
carbazone, yielding a lead complex that is spectrally measured at 510 nm.
Two variations of the spectrophotometric technique used when measuring low levels of lead
have been the USPHS (National Academy of Sciences, 1972) and APHA (American Public Health
Association, 1955) procedures. In both, venous blood or urine is wet ashed using concentrated
nitric acid of low lead content followed by adjustment of the ash with hydroxylamine and so-
dium citrate to a pH of 9-10. Cyanide ion is added and the solution extracted with dithizone
in chloroform. Back extraction removes the lead into dilute nitric acid; the acid layer is
treated with ammonia, then cyanide, and re-extracted with dithizone in chloroform. The
extracts are read in a spectrophotometer at 510 nm. Bismuth interference is handled (APHA
23PB12/C 9-7 7/1/83
-------
PRELIMINARY DRAFT
variation) by removal with dithizone at pH 3.4. According to Lerner (1975), the analytical
precision in the "normal" range is about ±3.5 ug Pb/dl (1 = S.D.), using 5 ml of sample.
The most accurate and precise method for lead measurement in blood is isotope dilution
mass spectrometry. As typified by the report of Machlan et al. (1976), whole blood samples
are accurately weighed, and a weighed aliquot of 206Pb-enriched isotope solution is added.
After sample decomposition with ultra-pure nitric and perchloric acids, samples are evapo-
rated, residues are taken up in dilute lead-free hydrochloric acid, and lead is isolated using
anion-exchange columns. Column eluates are evaporated with the above acids, and lead is
deposited onto high purity platinum wire from dilute perchloric acid. The 206Pb/208Pb ratio
is then determined by thermal ionization mass spectrometry. Samples without added isotope and
reagent blanks are also carried through the procedure. In terms of precision, the 95 percent
confidence level for lead samples overall is within 0.15 percent. Due to the expense incurred
by the requirements for operator expertise, the amount of time involved, and the high standard
of laboratory cleanliness, IDMS is mainly of practical value in the development of standard
reference materials and for the verification of other analytical methods.
Atomic absorption spectrometry (AAS) is widely used for lead measurements in whole blood,
with sample analysis involving analysis of venous blood with chemical degradation, analysis of
liquid samples with or without degradation, and samples applied to filter paper. It is thus
the most flexible for samples already collected or subject to manipulation.
By means of a flame or electrothermal excitation, ionic lead in some matrix is first vapor-
ized and then converted to the atomic state, followed by resonance absorption from either a
hollow cathode or electrodeless discharge lamp generating lead absorption lines at 217.0 and
283.3 nm. After monochrometer separation and photomultiplier enhancement of the differential
signal, it is measured electronically.
The earliest methods of atomic absorption spectrometric analysis involved the aspiration
into a flame of ashed samples of blood, usually subsequent to extraction into an organic sol-
vent to enhance sensitivity by preconcentration. Some methods did not involve digestion steps
prior to solvent extraction (Kopito et al., 1974). Of these various flame AAS methods, that
of Hessel's (1968) technique continues to be used with some frequency.
Currently, lead measurement in blood by AAS employs several different methods that permit
greater sensitivity, precision, and economy of sample and time. The flame method of Delves
(1970), called the "Delves cup" procedure, usually involves delivery of discrete small samples
(§100 jjl) of unmodified whole blood to nickel cups, with subsequent drying and peroxide decom-
position of organic content before positioning in the flame. The marked enhancement of sen-
sitivity over conventional flame aspiration is due to immediate, total consumption of the
sample and the generation of a localized population of atoms. In addition to discrete blood
volumes, blood-containing filter paper disks have been used (Joselow and Bogden, 1972; Cerni!
23PB12/C 9-8 7/1/83
-------
PRELIMINARY DRAFT
and Sayers, 1971; Piomelli et al., 1980). Several modifications of the Delves method include
that of Ediger and Coleman (1972), in which dried blood samples in the cups are pre-ignited to
destroy organic matter by placement near the flame in a precise, repeatable manner, and the
variation of Barthel et al. (1973), in which blood samples are mixed with dilute nitric acid
in the cups followed by drying in an oven at 200°C and charring at 450°C on a hot plate. A
number of laboratories eschew even these modifications and follow dispensing and drying with
direct placement of the cup into the flame (e.g., Mitchell et al., 1974). The Delves cup pro-
cedure may require correction for background spectral interference, which is usually achieved
by instrumentation equipped at a non-resonance absorption line. While the 217.0 nm line of
lead is less subject to such interference, precise work is best done with correction. This
method as applied to whole blood lead appears to have an operational sensitivity down to 1.0
ug Pb/dl, or somewhat below when competently employed, and a relative precision of approxi-
mately 5 percent in the range of levels encountered in the United States.
AAS methods using electrothermal (furnace) excitation in lieu of a flame can be approxi-
mately 10-fold more sensitive than the Delves procedure. A number of reports describing whole
blood lead analysis have appeared in the literature (Lawrence, 1982, 1983). Because of in-
creased sensitivity, the "flameless" AAS technique permits the use of small blood volumes
(1-5 pi) with samples undergoing drying and dry ashing jin situ. Physicochemical and spectral
interferences are inherently severe with this approach, requiring careful background cor-
rection. In one flameless AAS configuration, background correction exploits the Zeeman
effect, where correction is made at the specific absorption line of the element and not over a
band-pass region, as is the case with the deuterium arc. While control of background inter-
ference up to 1.5 molecular absorbence is claimed with the Zeeman system (Koizumi and Yasuda,
1976), it is technically preferable to employ charring before atomization. Hinderberger et
al. (1981) used dilute ammonium phosphate solution to minimize chemical interference in their
furnace AAS method.
Precision can be a problem in the flameless technique unless careful attention is paid to
the problem of sample diffusibility over and into the graphite matrix of the receiving recep-
tacle — tube, cup, or rod. With the use of diluted samples and larger applied volumes, the
relative precision of this method can approach that of the Delves technique (Delves, 1977).
In addition to the various atomic absorption spectral methods noted above, electro-
chemical techniques have been applied to blood lead analysis. Electrochemical methods, in
theory, differ from AAS methods in that the latter are "concentration" methods regardless of
sample volumes available, while electrochemical analysis involves bulk consumption of sample
and hence would have infinite sensitivity, given an infinite sample volume. This intrinsic
property is of little practical advantage given usual sample volume, instrumentation design,
and blank limits.
23PB12/C 9-9 7/1/83
-------
PRELIMINARY DRAFT
The most widely used electrochemical method for lead measurement in whole blood and other
biological media is anodic stripping voltammetry (ASV) which is also probably the most sensi-
tive, as it involves an electrochemical preconcentration (deposition) step in the analysis
(Matson and Roe, 1966; Matson et al., 1970). In this method, samples such as whole blood
(50-100 ul), are preferably but not commonly wet ashed and reconstituted in dilute acid or
made electro-available with metal exchange reagents. Using freshly prepared composite elec-
trodes of mercury film deposited on carbon, lead is plated out from the solution for a speci-
fic amount of time and at a selected negative voltage. The plated lead is then reoxidized in
the course of anodic sweeping, generating a current peak that may be recorded on a chart or
displayed on commercial instruments as units of concentration (ug/dl).
One alternative to the time and space demands of wet ashing blood samples is the use of
metal exchange reagents that displace lead from binding sites in blood by competitive binding
(Morell and Giridhar, 1976; Lee and Meranger, 1980). In one commercial preparation, this re-
agent consists of a solution of calcium, chromium, and mercuric ions. Use of the metal ex-
change reagent adds a chemical step that must be carefully controlled for full recovery of
lead from the sample.
The working detection limit of ASV for blood is comparable to that of the AAS fTameless
methods while the relative precision is best with prior sample degradation, approximately 5
percent, but less when the blood samples are run directly with the ion exchange reagents
(Morrell and Giridhar, 1976), particularly at the low end of "normal" blood lead values.
While AAS methods require attention to various spectral interferences to achieve satisfactory
performance, electrochemical methods such as ASV require consideration of such factors as the
effects of co-reducible metals and agents that complex lead and alter its reduction-oxidation
(redox) potential properties. Chelants used in therapy, particularly penicillamine, may in-
terfere, as does blood copper, which may be elevated in pregnancy and such disease states as
leukemia, lymphoma, and hyperthyroidism (Herman, 1981). At very low levels of lead in blood,
then, ASV may pose more problems than atomic absorption spectrometric techniques.
Correction of whole blood lead values for hematocrit, although carried out in the past,
is probably not appropriate and not commonly done at present. While the erythrocyte is the
carrier for virtually all lead, in blood, the saturation capacity of the red blood cell for
lead is so high that it can still carry lead even at highly toxic levels (Kochen and Greener,
1973). Kochen and Greener (1973) also showed that acute or chronic dosing at a given lead
level in rats with a wide range of hematocrits (induced by bleeding) gave similar blood lead
values. Rosen et al. (1974), based on studies of hematocrit, plasma, and whole blood lead in
children, noted hematocrit correction was not necessary, a view supported by Chisolm (1974).
9.2.2.2 Lead in Plasma. While virtually all of the lead present in whole blood is bound to
the erythrocyte (Robinson et al., 1958; Kochen and Greener, 1973), lead in plasma is trans-
ported to affected tissues. It is very important, therefore, that every precaution be taken
23PB12/C 9-10 7/1/83
-------
PRELIMINARY DRAFT
to use non-hemolyzed blood samples for plasma isolation. The very low levels of lead in
plasma require that more attention be paid to "ultra clean" methods.
Rosen et al. (1974) used fTameless atomic absorption spectrometry and microliter samples
of plasma to measure plasma lead, with background correction for the smoke signal generated
for the unmodified sample. Cavalleri et al. (1978) used a combination of solvent extraction
of modified plasma with preconcentrating and flameless atomic absorption. These authors noted
that the method used by Rosen et al. (1974) permitted less precision and accuracy than did
their technique, because a significantly smaller amount of lead was delivered to the furnace
accessory.
DeSilva (1981) used a technique similar to that of Cavalleri et al. (1978), but collected
samples in heparinized tubes, claiming that the use of EDTA as anticoagulant disturbs the
cell-plasma distribution of lead enough to yield erroneous data. Much more care was given in
this procedure to background contamination. In both cases, increasing levels of plasma lead
were measured with increasing whole blood lead, suggesting an equilibrium ratio in contradic-
tion to the data of Rosen et al. (1974), who found a fixed level of 2-3 ug Pb/dl plasma over a
wide range of blood lead. However, the actual levels of lead in plasma in the DeSilva (1981)
study were much lower than those reported by Cavalleri et al. (1978).
Using isotope-dilution mass spectrometry and sample collection/manipulation in an
"ultra-clean" facility, Everson and Patterson (1980) measured the plasma lead levels in two
subjects, a control and a lead-exposed worker. The control had a plasma lead level of 0.002
ug Pb/dl, several orders of magnitude lower than that seen with studies using less precise
analytical approaches. The lead-exposed worker had a plasma level of 0.2 ug Pb/dl. Several
other reports in the literature using isotope-dilution mass spectrometry noted somewhat higher
values of plasma lead (Manton and Cook, 1979; Rabinowitz et al., 1974), which Everson and
Patterson (1980) have ascribed to problems of laboratory contamination. Utilizing tracer lead
to minimize the impact of contamination results in a value of 0.15 ug/dl (Rabinowitz et al.,
1974).
With appropriate plasma lead methodology, reported lead levels are extremely low, the de-
gree varying with the methods used to measure such concentrations. While the data of Everson
and Patterson (1980) were obtained from only two subjects, it seems unlikely that using more
subjects would result in a plasma lead range extending upward to the levels seen with ordinary
methodology in ordinary laboratory surroundings. The above considerations are necessary when
discussing appropriate methodology for plasma analysis, and the Everson and Patterson (1980)
report indicates that some doubt surrounds results obtained with conventional methods. Al-
though not the primary focus of their study, the values obtained by Everson and Patterson
(1980) for whole blood lead, unlike the data for plasma, are within the ranges for unexposed
(11 M9 Pb/dl) and exposed (80 ug Pb/dl) subjects generally reported with other methods. This
23PB12/C 9-11 7/1/83
-------
PRELIMINARY DRAFT
would suggest that, for the most part, reported values do actually reflect HI vivo blood lead
levels rather than sampling problems or inaccurate methods.
9.2.2.3 Lead in Teeth. When carrying out analysis of shed deciduous or extracted permanent
teeth, some reports have used the whole tooth after surface cleaning to remove contaminating
lead (e.g., Moore et al., 1978; Fosse and Justesen, 1978; Mackie et al., 1977), while others
have measured lead in dentine (e.g., Shapiro et al., 1973; Needleman et al , 1979; Al-Naimi et
al., 1980). Several reports (Grandjean et al., 1978; Shapiro et al., 1973) have also de-
scribed the analysis of secondary (circumpulpal) dentine, that portion of the tooth found to
have the highest relative fraction of lead. Needleman et al. (1979) separated dentine by em-
bedding the tooth in wax, followed by thin central sagittal sectioning. The dentine was then
isolated from the sawed sections by careful chiseling.
The mineral and organic composition of teeth and their components requires the use of
thorough chemical decomposition techniques, including wet ashing and dry ashing steps, sample
pulverizing or grinding, etc. In the procedure of Steenhout and Pourtois (1981), teeth are
dry ashed at 450°C, powdered, and dry ashed again. The powder is then dissolved in nitric
acid. Fosse and JusteSen (1978) reduced tooth samples to a coarse powder by crushing in a
vise, followed by acid dissolution. Oehme and Lund (1978) crushed samples to a fine powder in
an agate mortar and dissolved the samples in nitric acid. Mackie et al. (1977) and Moore et
al. (1978) dissolved samples directly in concentrated acids. Chatman and Wilson (1975) and
Needleman et al. (1974) carried out wet ashing with nitric acid followed by dry ashing at
450°C. Oehme and Lund (1978) found that acid wet ashing of tooth samples yielded better re-
sults if carried out in a heated Teflon bomb at 200°C.
With regard to methods of measuring lead in teeth, atomic absorption spectrometry and
anodic stripping voltammetry have been employed most often. With the AAS methods, the high
mineral content of teeth tends to argue for isolating lead from this matrix before analysis.
In Needleman et al.'s (1974) and Chatman and Wilson's (1975) method, ashed residues in nitric
acid were treated with ammonium nitrate and ammonium hydroxide to a pH of 2.8, followed by
dilution and extraction with a methylisobutylketone solution of ammonium pyrrolidine-
carbodithioate. Analysis is by flame AAS using the 217.0 nm lead absorption line. A similar
procedure was employed by Fosse and Justesen (1978).
Anodic stripping voltammetry has been successfully used in tooth lead measurement
(Shapiro et al., 1973; Needleman et al., 1979; Oehme and Lund, 1978). As typified by the
method of Shapiro et al. (1973), samples of dentine-were dissolved in a small volume of low-
lead concentrated perchloric acid and diluted (5.0 ml) with lead-free sodium acetate solution.
With deoxygenation, samples were analyzed in a commercial ASV unit, using a plating time of 10
minutes at a plating potential of -1.05 V. Anodic sweeping was at a rate of 60 mV/sec with a
variable current of 100-500 uA.
23PB12/C 9-12 7/1/83
-------
PRELIMINARY DRAFT
Since lead content of teeth is higher than in most samples of biological media, the rela-
tive precision of analysis with appropriate accommodation of the matrix effect, such as the
use of matrix-matched standards, in the better studies indicates a value of approximately 5-7
percent.
All of the above methods involve shed or extracted teeth and consequently provide a ret-
rospective determination of lead exposure. In Bloch et al.'s (1976) procedure, tooth lead is
measured i_n situ using an X-ray fluorescence technique. A collimated beam of radiation from
57Co was allowed to irradiate the upper central incisor teeth of the subject. Using a rela-
tively safe 100-second irradiation time and measurement of K t and K 2 lead lines via a ger-
a a
manium diode and a pulse height analyzer for signal processing, lead levels of 15 ppm or
higher could be measured. Multiple measurement by this method would be very useful in pros-
pective studies because it would show the "on-going" rate of increase in body lead burden.
Furthermore, when combined with serial blood sampling, it would provide data for blood lead-
tooth lead relationships.
9.2.2.4 Lead in Hair. Hair constitutes a non-invasive sampling source with virtually no
problems with sample stability on extended storage. However, the advantages of accessibility
and stability are offset by the problem of assessing external contamination of the hair sur-
face by atmospheric fallout, hand dirt, lead in hair preparations, etc. Thus, such samples
are probably of less value overall than those from other media.
The various methods that have been employed for removal of external lead have been
reviewed (Chatt et al., 1980; Gibson, 1980; Chattopadhyay et al., 1977). Cleaning techniques
obviously should be vigorous enough to remove surface lead but not so vigorous as to remove
the endogenous fraction. To date, it remains to be demonstrated that any published cleaning
procedure is reliable enough to permit acceptance of reported levels of lead in hair. Such a
demonstration would have to use lead isotopic studies with both surface and endogenous
isotopic lead removal monitored as a function of a particular cleaning technique.
9.2.2.5 Lead in Urine. Analysis of lead in urine is complicated by its relatively low con-
centrations (lower than in blood in many cases) as well as by the complex mixture of mineral
elements present. Lead levels are higher, of course, in cases where lead mobilization or
therapy with chelants is in progress, but in these cases samples must be analyzed to account
for lead bound to chelants such as EDTA. This requires either sample ashing or the use of
standards containing the chelant. Although analytical methods have been published for the
direct analysis of lead in urine, samples are probably best wet ashed before analysis, using
the usual mixtures of nitric plus sulfuric and/or perchloric acids.
Both atomic absorption spectrometric and anodic stripping voltammetric methods have been
applied to urine lead analyses, the former employing either direct analysis of ashed residues
or a preliminary chelation-extraction step. With flame AAS, ashed urine samples must invari-
23PB12/C 9-13 7/1/83
-------
PRELIMINARY DRAFT
ably be extracted with a chelant such as ammonium pyrrolidinecarbodithioate in methylisobutyl-
ketone to achieve reasonably satisfactory results. Direct analysis, furthermore, creates me-
chanical problems with burner operation, due to the high mineral content of urine, and results
in considerable maintenance problems with equipment. The procedure of Lauwerys et al. (1975)
is typical of flame AAS methods with preliminary lead separation. Owing to the relatively
greater sensitivity of graphite furnace (flameless) AAS, this variation of the method has been
applied to urine analysis in scattered reports where it appears that adequate performance for
direct sample analysis requires steps to minimize matrix interference. A typical example of
one of the better direct analysis methods is that of Hodges and Skelding (1981). Urine
samples were mixed with iodine solution and heated, then diluted with a special reagent con-
taining ammonium molybdate, phosphoric acid, and ascorbic acid. Small aliquots (5 ul) were
delivered to the furnace accessory of an AAS unit containing a graphite tube pretreated with
ammonium molybdate. The relative standard deviation of the method is reported to be about 6
percent. In the method of Legotte et al. (1980), such tube treatment and sample modifications
were not employed and the average precision figure was 13 percent.
Compared with various atomic absorption spectrometric methods, anodic stripping voltam-
metry has been less frequently employed for urine lead analysis, and it would appear from
available electrochemical methods in general that such techniques applied to urine require
further development. Franke and de Zeeuw (1977) used differential pulse anodic stripping vol-
tammetry as a screening tool for lead and other elements in urine. Jagner et al. (1979) de-
scribed analysis of urine lead using potentiometric stripping. In their procedure the element
was pre-concentrated at a thin-film mercury electrode as in conventional ASV, Jaut deoxygenated
samples were reoxidized with either oxygen or mercuric ions after the circuitry was disconnec-
ted.
As noted in Section 9.1.1.2, spot sampling of lead in urine should be expressed per unit
creatinine, if it is not possible to obtain 24-hour collection.
9.2.2.6 Lead in Other Tissues. Bone samples of experimental animal or human autopsy origin
require preliminary cleaning procedures for removal of muscle and connective tissue, with care
being taken to minimize sample contamination. As is the case with teeth, samples must be che-
mically decomposed before analysis. Satisfactory instrumental methods for bone lead analysis
comprise a much smaller literature than is the case for other media.
Wittmers et al. (1981) have described the measurement of lead in dry-ashed (450°C) bone
samples using flameless atomic absorption spectrometry. Ashed samples were weighed and dis-
solved in dilute nitric acid containing lanthanum ion, the latter being used to suppress in-
terference from bone elements. Small volumes (20 ul) and high calcium content required that
atomization be done at 2400°C to avoid condensation of calcium within the furnace. Quantifi-
cation was by the method of additions. Relative precision was 6-8 percent at relatively high
lead content (60 ug/g ash) and 10-12 percent at levels of 14 (jg/g ash or less.
23PB12/C 9-14 7/1/83
-------
PRELIMINARY DRAFT
Ahlgren et al. (1980) described the application of X-ray fluorescence analysis to ui vivo
lead measurement in the human skeleton, using tibia and phalanges. In this technique, ir-
radiation is carried out with dual 57Co gamma ray source. The generated K l and K 2 lead
lines are detected with a lithium-drifted germanium detector. The detection limit is 20 parts
per mil 1 ion.
Soft organs differ from other biological media in the extent of anatomic heterogeneity as
well as lead distribution, e.g., brain vs. kidney. Hence, sample analysis involves either
discrete regional sampling or the homogenizing of an organ. The efficiency of the latter can
vary considerably, depending on the density of the homogenate, the efficiency of rupture of
the formed elements, and other factors. Glass-on-glass homogenizing is to be avoided because
lead is liberated from the glass matrix with abrasion.
Atomic absorption spectrometry, in its flame or fTameless variations, appears to be the
method of choice in many studies. In the procedure of Slavin et al. (1975), tissues were wet
ashed and the residues taken up in dilute acid and analyzed with the furnace accessory of an
AAS unit. A large number of reports representing slight variations of this basic technique
have appeared over the years (Lawrence, 1982, 1983). Flame procedures, being less sensitive
than the graphite furnace method, require more sample than may be available or are restricted
to measurement in tissues where levels are relatively high, e.g., kidney. In the method of
Farris et al. (1978), samples of brain, liver, lung, or spleen (as discrete segments) were
lyophilized and solubilized at room temperature with nitric acid. Following neutralization,
lead was extracted into methylisobutylketone with ammonium pyrrolidinecarbodithioate and
aspirated into the flame of an AAS unit. The reported relative precision was 8 percent.
9.2.3 Quality Assurance Procedures In Lead Analysis
Regardless of technical differences among the different methodologies for lead analysis,
one can define the quality of such techniques as being of: (1) poor accuracy and poor pre-
cision; (2) poor accuracy and good precision; or (3) good accuracy and good precision. In
terms of available information, the major focus in assessing quality has been on blood lead
determinations.
According to Boutwell (1976), the use of quality control testing for lead measurement
rests on four assumptions: (1) the validity of the specific procedure for lead in some matrix
has been established; (2) the stability of the factors making up the method has been both es-
tablished and manageable; (3) the validity of the calibration process and the calibrators with
respect to the media being analyzed has been established; and (4) surrogate quality control
materials of reliably determined analyte content can be provided. These assumptions, when
translated into practice, revolve around steps employed within the laboratory, using a battery
of "internal checks" and a further reliance on "external checks" such as a formal, well-
organized, multi-laboratory proficiency testing program.
23PB12/C 9-15 7/1/83
-------
PRELIMINARY DRAFT
Analytical quality protocols can be further divided into start-up and routine procedures,
the former entailing the establishment of detection limits, "within-run" and "between-run"
precision, recovery of analyte, etc. When a new method is adopted for some specific analyti-
cal advantage, the procedure is usually tested in the laboratory or outside the laboratory for
comparative performance. For example, Hicks et al. (1973) and Kubasik et al. (1972) reported
that flameless techniques for measuring lead in whole blood were found to have a satisfactory
correlation with results using conventional flame procedures. Matson et al. (1970) noted a
good agreement between anodic stripping voltammetry and both atomic absorption spectral and
dithizone colorimetric techniques. The problem with such comparisons is that the reference
method is assumed to be accurate for the particular level of lead in a given matrix. High
correlations obtained in this manner may simply indicate that two inaccurate methods are
simultaneously performing with the same level of precision.
Preferable approaches for assessing accuracy are the use of certified samples determined
by a definitive method, or a direct comparison of different techniques with a definitive pro-
cedure. For example, Eller and Hartz (1977) compared the precision and accuracy of five
available methods for measuring lead in blood: dithizone spectrometry, extraction and tanta-
lum boat AAS, extraction and flame aspiration AAS, direct aspiration AAS, and graphite furnace
AAS techniques. Porcine whole blood certified by the National Bureau of Standards (NBS) using
isotope-dilution mass spectrometry at 1.00 ug Pb/g (±0.023) was tested and all methods were
found to be equally accurate. The tantalum boat technique was found to be the least precise.
The obvious limitation of these data is that they relate to a high blood lead content, suit-
able for use in measuring the exposure of lead workers or in some other occupational context,
but less appropriate for clinical or epidemiological investigations.
Boone et al. (1979) compared the analytical performance of 113 laboratories using various
methods and 12 whole blood samples (blood from cows fed a lead salt) certified as to lead con-
tent using isotope-dilution mass spectrometry at the NBS. Lead content ranged from 13 to 102
ug Pb/dl, determined by anodic stripping voltammetry and five variations of AAS. The order of
agreement with NBS values, i.e., relative accuracy, was: extraction > ASV > tantalum strip >
graphite furnace > Delves cup > carbon rod. The AAS methods all tended to show bias, being
positive at values less than 40 ug Pb/dl and negative at levels greater than 50 ug Pb/dl. ASV
tended to show less of a positive bias problem, although it was not bias-free within either of
the blood lead ranges. In terms of relative precision, the ranking was: ASV > Delves cup >
tantalum strip > graphite furnace > extraction > carbon rod. The overall ranking in accuracy
and precision indicated: ASV > Delves cup > extraction > tantalum strip > graphite furnace >
carbon rod. As the authors cautioned, the above data should not be taken to indicate that any
established laboratory using one particular technique would not perform better than this-
rather, it should be used as a guide for newer facilities choosing among methods.
23PB12/C 9-16 7/1/83
-------
PRELIMINARY DRAFT
There are a number of necessary steps in quality assurance pertinent to the routine
measurement of lead that should be used in an ongoing program. With respect to internal
checks of routine performance, these include calibration and precision and accuracy testing.
With biological matrices, the use of matrix-matched standards is quite important, as is an
understanding of the range of linearity and variation of calibration curve slopes from day to
day. It is common practice to analyze a given sample in duplicate, further replication being
carried out if the first two determinations vary beyond a predetermined range. A second de-
sirable step is the analysis of samples collected in duplicate but analyzed "blind" to avoid
bias.
Monitoring of accuracy within the laboratory is limited to the availability of control
samples having a certified lead content in the same medium as the samples being analyzed.
Controls should be as physically close to the media being analyzed as possible. Standard re-
ference materials (SRMs), such as orchard leaves and lyophilized bovine liver, are of help in
some cases, but there is need for NBS-certified blood samples for the general laboratory com-
munity. There are commercially available whole blood samples, prepared and certified by the
marketing facility (TOX-EL, A.R. Smith Co., Los Angeles, CA; Kaulson Laboratories, Caldwell,
NJ; Behringwerke AG, Marburg, W. Germany; and Health Research Institute, Albany, NY). With
these samples, attention must be paid to the reliability of the methods used by reference
laboratories. The use of such materials, from whatever source, must minimize bias; for exam-
ple, the attention given control specimens should be the same as that given routine samples.
Finally, the most important form of quality assurance is the ongoing assessment of lab-
.oratory performance by proficiency testing programs using externally provided specimens for
analysis. Earlier interlaboratory surveys of lead measurement in blood and in urine indicated
that a number of laboratories had performed unsatisfactorily, even at high levels of lead
(Keppler et al., 1970; Donovan et al., 1971; Berlin et al., 1973), although there may have
been problems in the preparation and status of the blood samples during and after distribution
(World Health Organization, 1977). These earlier tests for proficiency indicated that: (1)
many laboratories were able to achieve a good degree of precision within their own facilities;
(2) the greater the number of samples routinely analyzed by a facility, the better the per-
formance; and (3) 30 percent of the laboratories routinely analyzing blood lead reported
values differing by more than 15 percent from the true level (Pierce et al., 1976).
In the more recent, but very limited, study of Paulev et al. (1978), five facilities par-
ticipated in a survey, using samples to which known amounts of lead were added. For lead in
both whole blood and urine, the interlaboratory coefficient of variation was reported to be
satisfactory, ranging from 12.3 to 17.2 percent for blood and urine samples. Aside from its
23PB12/C 9-17 7/1/83
-------
PRELIMINARY DRAFT
limitation of scope, this study used "spikeci" instead of jn vivo lead, so that extraction
techniques used in most of the laboratories surveyed would have given misleadingly better
results in terms of actual recovery.
Maher et al. (1979) described the outcome of a proficiency study involving up to 38 lab-
oratories that analyzed whole blood pooled from a large number of samples submitted for blood
lead testing. The Delves cup technique was the most heavily represented, followed by the
chelation-extraction plus flame AAS method and the graphite furnace AAS method. Anodic strip-
ping voltammetry was used by only approximately 10 percent of the laboratories, so that the
results basically portray AAS methods. All laboratories had about the same degree of ac-
curacy, with no evidence of consistent bias, while the interlaboratory coefficient of
variation was approximately 15 percent. A subset of this group, certified by the American
Industrial Hygiene Association (AIHA) for air lead, showed a corresponding precision figure of
approximately 7 percent. Over time, the subset of AIHA-certified laboratories remained about
the same in proficiency, while the other facilities showed continued improvement in both ac-
curacy and precision. This study indicates that program participation does help the per-
formance of a laboratory doing blood lead determinations.
The most comprehensive proficiency testing program is that carried out by the Centers for
Disease Control of the U.S. Public Health Service. This consists of two operationally and ad-
ministratively distinct subprograms, one conducted by the Center for Environmental Health
(CEH) and the other by the Licensure and Proficiency Testing Division, Laboratory Improvement
Program Office (LIPO). The CEH program is directed at facilities involved in lead poisoning
prevention and screening, while LIPO is concerned with laboratories seeking certification
under the Clinical Laboratories Improvement Act of 1967 as well as under regulations of the
Occupational Safety and Health Administration (OSHA). Both the CEH and LIPO protocols involve
the use of bovine whole blood certified as to content by reference laboratories (6 in the CEH
program, 20-23 in LIPO) with an ad hoc target range of ±6 ug Pb/dl for values of 40 ug Pb/dl
or less and ±15 percent for higher levels. Three samples are provided monthly from CEH, for
a total of 36 yearly, while LIPO participants receive 3 samples quarterly (12 samples yearly).
Use of a fixed range rather than a standard deviation has the advantage of allowing the moni-
toring of overall laboratory improvement.
For Fiscal Year (FY) 1981, 114 facilities were in the CEH program, 92 of them partici-
pating for the entire year. Of these, 57 percent each month reported all three samples within
the target range, and 85 percent on average reported two out of three samples correctly. Of
the facilities reporting throughout the year, 95 percent had a 50 percent or better perfor-
mance, i.e., 18 blood samples or better. If one compares these summary data for FY 1981 with
earlier annual reports, it would appear that there has been considerable improvement in the
23PB12/C 9-18 7/1/83
-------
PRELIMINARY DRAFT
number of laboratories achieving higher levels of proficiency. For the interval FY 1977-79,
there was a 20 percent increase in the number correctly analyzing more than 80 percent of all
samples and a 33 percent decrease in those reporting less than 50 percent correct. In the
last several years, FY 1979-81, overall performance appears to have more or less stabilized.
With the LIPO program for 1981 (Dudley, 1982), the overall laboratory performance
averaged across all quarters was 65 percent of the laboratories analyzing all samples cor-
rectly and approximately 80 percent performing well with two of three samples. Over the four
years of this program, an increasing ability to correctly analyze lead in blood appears to
have been demonstrated. Dudley's survey (1982) also indicates that reference laboratories in
the LIPO program are becoming more accurate relative to isotope-dilution mass spectrometry
values, i.e., bias over the blood lead range is contracting.
Current OSHA criteria for certification of laboratories measuring occupational blood lead
levels require that eight of nine samples be correctly analyzed in the previous quarter (U.S.
Occupational Safety and Health Administration, 1982). These criteria appear to reflect the
ability of a number of laboratories to perform at this level.
It should be noted that most proficiency programs, including the CEH and LIPO surveys,
are appropriately concerned with blood lead levels encountered in such cases as pediatric
screening for excessive exposure to lead or in occupational exposures. As a consequence,
there does appear to be an underrepresentation of lead values in the low end of the "normal"
range. In the CEH distribution for FY 1981, four samples (11 percent) were below 25 ug Pb/dl.
The relative performance of the 114 facilities with these samples indicates outcomes much
better than with the whole sample range.
9.3 DETERMINATION OF ERYTHROCYTE PORPHYRIN (FREE ERYTHROCYTE PROTOPORPHYRIN,
ZINC PROTOPORPHYRIN)
9.3.1 Methods of Erythrocyte Porphyrin Analysis
Lead exposure results in inhibition of the final step in heme biosynthesis, the insertion
of iron into protoporphyrin IX to form heme. This leads to an accumulation of the porphyrin,
with zinc (II) occupying the position normally filled by iron. Depending on the particular
method of analysis, zinc protoporphyrin (ZPP) itself or the metal-free form, free erythrocyte
protoporphyrin (FEP), is measured. PEP generated as a consequence of chemical manipulation
should be kept distinct from the metal-free form biochemically produced in the porphyria,
erythropoietic protoporphyria. The chemical or "wet" methods measure free erythrocyte
porphyrin or zinc protoporphyrin, depending upon the relative acidity of the extraction
medium. The hematof1uorometer in its commercially available form measures zinc proto-
porphyrin.
23PB12/C 9-19 7/1/83
-------
PRELIMINARY DRAFT
Porphyn'ns are labile due to photochemical decomposition; hence, samples must be pro-
tected from light during collection and handling and analyzed as soon as possible.
Hematocrits must also be obtained to adjust for anemic subjects.
In terms of methodological approaches for EP analysis, virtually all methods now in use
exploit the ability of porphyrins to undergo intense fluorescence when excited at the appro-
priate wavelength of light. Such fluorometric techniques can be further classified as wet
chemical micromethods or as micro methods using a recently developed instrument, the hemato-
fluorometer. The latter involves direct measurement in whole blood. Because the mammalian
erythrocyte contains all of the EP in whole blood, either packed cells or whole blood may be
used, although the latter is more expedient.
Due to the relatively high sensitivity of fluorometric measurement for FEP or ZPP,
laboratory methods for spectrofluorometric analysis require a relatively small sample of
blood; hence, microtechniques are currently the most popular in most laboratories. These in-
volve either liquid samples or blood collected on filter paper, the latter of use particularly
in field sampling.
As noted above, chemical methods for EP analysis measure either free erythrocyte proto-
porphyn'n, where zinc is chemically removed, or zinc protoporphyrin, where zinc is retained.
The procedures of Piomelli and Davidow (1972), Granick et al., (1972), and Chisholm and Brown
(1975) typify "free" EP methods, while those of Lamola et al. (1975), Joselow and Flores
(1977), and Chisholm and Brown (1979) involve measurement of zinc-EP.
In Piomelli and Davidow's (1972) micro procedure, small volumes of whole blood, analyzed
either directly or after collection on filter paper, were treated with a suspension of Celite
in saline followed by a 4:1 mixture of ethyl acetate to glacial acetic acid. After agitation
and centrifugation, the supernatant was extracted with 1.5N HC1. The acid layer was analyzed
fluorometrically using an excitation wavelength of 405 nm and measurement at 615 nm. Blood
collected on filter paper discs was first eluted with 0.2 ml H20. The filter paper method was
found to work just as well as liquid samples of whole blood. Protoporphyrin IX was employed
as a quantitative standard. Granick et al. (1972) use similar microprocedure, but it differs
1n the concentration of acid employed and the use of a ratio of maxima.
In Chisolm and Brown's (1975) variation, volumes of 20 pi of whole blood were treated
with ethyl acetate/acetic acid (3:1) and briefly mixed. The acid extraction step was done
with 3N HC1, followed by a further dilution step with more acid if the value was beyond the
range of the calibration curve. In this procedure, protoporphyrin IX was used as the working
standard, with coproporphyrin used to monitor the calibration of the fluorometer and any
variance with the protoporphyrin standard.
23PB12/C 9-20 7/1/83
-------
PRELIMINARY DRAFT
The above microfluorometric methods all involve double extraction. In the single-
extraction variation of Orfanos et al. (1977), liquid samples of whole blood (40 pi) or blood
on filter paper were treated with acidified ethanol, the mixtures agitated and centrifuged,
and the supernatants analyzed directly in fluorometer cuvettes. For blood samples on filter
paper, blood was first leached from the paper with saline by soaking for 60 minutes. Copro-
porphyrin was used as the quantitative standard. The correlation coefficient with the
Piomelli and Davidow (1972) procedure (see above) over the range 40-650 |jg EP/dl RBCs was
r = 0.98.
Lamola et al. (1975) analyzed the zinc protophyrin as such in their procedure. Small
volumes of blood (20 ul) were worked up in a detergent (dimethyl dodecylamine oxide) and
phosphate buffer solution, and fluorescence measured at 594 nm with excitation at 424 nm. In
the variation of Joselow and Flores (1977), 10 ul of whole blood was diluted 1000-fold, along
with protoporphyrin (Zn) standards, with the detergent-buffer solution. It should be noted
that it is virtually impossible to obtain the ZPP standard in pure form, and Chisolm and Brown
(1979) reported the use of protoporphyrin IX plus very pure zinc salt for such standards.
Regardless of the extraction methods used, some instrumental parameters are of impor-
tance, including the variation between cut-offs in secondary emission filters and variation
among photomultiplier tubes in the red region of the spectrum. Hanna et al. (1976) compared
four micromethods for EP analysis: double extraction with ethyl acetate/acetic acid and HC1
(Piomelli and Davidow, 1972), single extraction with either ethanol or acetone (Chisolm et
al., 1974), and direct solubilization with detergent (Lamola et al., 1975). Of these, the
ethyl acetate and ethanol procedures were satisfactory; complete extraction occurred only with
the ethyl acetate/acetic acid method. In the method of Chisholm et al. (1974), it appears that
the choice of acid and its concentration is more significant than the choice of organic
solvent.
The levels of precision with these wet micromethods appears to differ with the specifics
of analysis. Piomelli (1973) reported a coefficient of variation (C.V.) of 5 percent, com-
pared to Herber's (1980) observation of 2-4 percent for the methods per se and 6-11 percent
total C.V., which included precision of samples, standards, and day-to-day variation. The
Lamola et al. (1975) method for ZPP measurement was found to have a C.V. of 10 percent (same
day, presumably), whereas Herber (1980) reported a day-to-day C.V. of 9.3-44.6 percent.
Herber (1980) also found that the wet chemical micro method of Piomelli (1973) had a detection
limit of 20 ug EP/dl whole blood, while that of Lamola et al. (1975) was sensitive to 50 ug
EP/dl whole blood.
23PB12/C 9-21 7/1/83
-------
PRELIMINARY DRAFT
The recent development of direct instrumental measurement of ZPP with the hematofluoro-
meter has added a dimension to the use of EP measurement for field screening the lead exposure
of large groups of subjects. As originally developed by Bell Laboratories (Blumberg et al.,
1977) and now produced commercially, the apparatus employs front-face optics, in which exci-
tation of the fluorophore is at an acute angle to the sample surface, with emitted light
emerging from the same surface and thus being detected. Routine calibration requires a stable
fluorescing material with spectra comparable to ZPP; the triphenylmethane dye Rhodamine B is
used for this purpose. Absolute calibration requires adjusting the microprocessor-controlled
readout system to read the known concentration of ZPP in reference blood samples, the latter
calibration being performed as frequently as possible.
Hematofluorometers are designed for the measurement of EP in samples containing oxyhemo-
globin, i.e., capillary blood. Venous blood, therefore, must first be oxygenated, usually by
moderate shaking for approximately 10 minutes (Blumberg et al., 1977; Grandjean and Lintrup,
1978). A second problem with hematofluorometer use, in contrast to wet chemical methods, is
interference by bilirubin (Karacic et al., 1980; Grandjean and Lintrup, 1978); this would oc-
cur with relatively low levels of EP. At levels normally encountered in lead workers or sub-
jects with anemia or nonoccupational lead exposure, the degree of such interference is not
considered significant (Grandjean and Lintrup, 1978). Karacic et al. (1980) have found that
carboxyhemoglobin (COHb) may pose a potential problem, but its relevance to EP levels of sub-
jects exposed to lead has not been fully elucidated. Background fluorescence in cover glass
may be a problem and should be tested in advance. Finally, the accuracy of the hematofluoro-
meter appears to be affected by hemolyzed blood.
Competently employed, the hematofluorometer appears to be reasonably precise but its ac-
curacy may still be biased (see below). Blumberg et al. (1977) reported a C.V. of 3 percent
over the entire range of ZPP values measured when using a prototype apparatus. Karacic et al.
(1980) found the relative standard deviation to vary from 1 percent (0.92 mM ZPP/M Hb) to 5
percent (0.41 mM ZPP/M Hb) depending on concentration. Grandjean and Lintrup (1978) obtained
a day-to-day C.V. of 5 percent using blood samples refrigerated for up to 9 weeks. Herber
(1980) obtained a total C.V. of 4.1-11.5 percent.
A number of investigators have compared EP measured by the hematofluorometer with the
laboratory or wet chemical techniques, ranging from a single, intralaboratory comparison to
interlaboratory performance testing. The latter included the EP proficiency testing program
of the Centers for Disease Control. Working with prototype instrumentation, Blumberg et al.
(1977) obtained correlation coefficients of r = 0.98 (range: 50-800 ug EP/dl RBCs) and 0.99
(range: up to 1000 ug EP/dl RBCs) for comparisons with the Granick and Piomelli methods,
respectively. Grandjean and Lintrup (1978), Castoldi et al. (1979) and Karacic" et al. (1980)
have achieved equally good correlation results.
23PB12/C 9-22 7/1/83
-------
PRELIMINARY DRAFT
Several reports (Culbreth et al., 1979; Scoble et al., 1981; Smith et al., 1980) have
described the application of high-performance liquid chromatography (HPLC) to the analysis of
either free or zinc protoporphyrin in whole blood. In one of the studies (Scoble et al.,
1981), the protoporphyrins as well as coproporphyrin and mesoporphyrin IX were reported to be
determined on-line fluorometrically in less than 6 minutes using 0.1 ml of blood sample. The
HPLC approach remains to be tested in interlaboratory proficiency programs.
9.3.2 Inter!aboratory Testing of Accuracy and Precision in EP Measurement
In a relatively early attempt to assess interlaboratory proficiency in EP measurement,
Jackson (1978) reported results of a survey of 65 facilities that analyzed 10 whole blood
samples by direct measurement with the hematofluorometer or by one of the wet chemical
methods. In this survey, the instrumental methods had a low bias compared to the extraction
techniques but tended to show better interlaboratory correlation.
At present, CDC's ongoing EP proficiency testing program constitutes the most comprehen-
sive assessment of laboratory performance (U.S. Centers for Disease Control, 1981). Every
month, three samples of whole blood prepared at the University of Wisconsin Laboratory of
Hygiene are forwarded to participants. Reference means are determined by a group of reference
laboratories with a target range of ±15 percent across the whole range of EP values. For
Fiscal Year 1981, of the 198 laboratories participating, 139 facilities were involved for the
entire year. Three of the 36 samples in the year were not included. Of the 139 year-long
participants, 93.5 percent had better than half of the samples within the target range, 84.2
percent performed satisfactorily with 70 percent or more of the samples within range, and 50.4
percent of all laboratories had 90 percent or more of the samples yielding the correct re-
sults. The participants as a whole showed greater proficiency than in the previous year. Of
the various methods currently used, the hematofluorometer direct measurement technique was
most heavily represented. For example, the January 1982 survey of the three major techniques
154 participants used the h$matofluorometer, 30 used the Piomelli method, and 7 used the
Chisolm/Brown method.
The recent survey of Balamut et al. (1982) raises the troublesome observation that the
use of commercially available hematofluorometers may yield satisfactory proficiency results
but still be inaccurate when compared to the wet chemical method using freshly-drawn whole
blood. Two hematofluorometers in wide use performed well in proficiency testing but showed an
approximately 30 percent negative bias with clinical samples analyzed by both instrument and
chemical microtechniques. This bias leads to false negatives when used in screening. It ap-
pears that periodic testing of split samples by both fluorometer and chemical means is neces-
sary to monitor, and correct for, instrument negative bias. The basis of the bias is much
more than can be explained by the difference between FEP and ZZP.
23PB12/C 9-23 7/1/83
-------
PRELIMINARY DRAFT
9.4 MEASUREMENT OF URINARY COPROPORPHYRIN
The elevation of urinary coproporphyrin (CP-U) with lead intoxication served as a useful
indicator of such intoxication in children and lead workers for many years. Although analysis
of CP-U has declined considerably in recent times with the development of other testing
methods, such as measurement of erythrocyte protoporphyrin, it still possesses the advantage
of showing active intoxication (Piomelli and Graziano, 1980).
The standard method of CP-U determination is the fluorometric procedure described by
Schwartz et al. (1951). Urine samples are treated with acetate buffer and aqueous iodine, the
latter converting coproporphyrinogen to CP. The porphyrin is partitioned into ethyl acetate
and back-extracted (4 X) with 1.5N HC1. Coproporphyrin is employed as the quantitative stan-
dard. Working curves are linear below 5 ug CP/1 urine.
In the absorption spectrometric technique of Haeger-Aronsen (1960), iodine is also used
to convert coproporphyrinogen to CP. The extractant is ethyl ether, from which the CP is re-
moved with 0.1N HC1. Absorption is read at three wavelengths, 380, 430, and the Soret maximum
at 402 nm; and quantification is carried out using an equation involving the three wave
lengths.
9.5 MEASUREMENT OF DELTA-AMINOLEVULINIC ACID DEHYDRASE ACTIVITY
Oelta-aminolevulinic acid dehydrase (5-aminolevulinate hydrolase; porphobilinogen
synthetase; E.C. 4.2.1.24; ALA-D) is an allosteric sulfhydryl enzyme that mediates the con-
version of two units of 6-aminolevulinic acid to porphobilinogen, a precursor in the heme bio-
synthetic pathway to the porphyrins. Lead's inhibition of the activity of this enzyme is the
enzymological basis of ALA-D1s diagnostic utility in assessing lead exposure using erythro-
cytes.
A number of sampling precautions are necessary when measuring this enzyme's activity.
ALA-D activity is modified by the presence of zinc as well as by lead. Consequently, blood
collection tubes that have high background zinc content, mainly in the rubber stoppers, must
be avoided completely or care taken to avoid stopper contact with blood. Nackowski et al.
(1977) observed that the presence of zinc in blood collection tubes is a pervasive problem,
and it appears that plastic-cup tubes are the only practical means to avoid it. To guard
against zinc in the tube itself, it would appear prudent to determine the extent of zinc
Teachability by blood and to use one tube lot, if possible. Heparin is the anticoagulant of
choice, as the lead binding agent, EDTA, or other chelants would affect the lead-enzyme inter-
action. The relative stability of the enzyme in blood makes rapid determinations of activity
necessary, preferably as soon after collection as possible. Even with refrigeration, analysis
of activity should be done within 24 hours (Berlin and Schaller, 1974). Furthermore, porpho-
bilinogen is light-labile, which requires that the assay be done under restricted light.
23PB12/C 9-24 7/1/83
-------
PRELIMINARY DRAFT
Various procedures for ALA-D activity measurement are chemically based on measurement of
porphobilinogen generated from the substrate, 6-ALA porphobilinogen is condensed with p-di-
methylaminobenzaldehyde (Ehrlich's reagent) to yield a chromophore measured at 553 nm in a
spectrophotometer. In the European Standardized Method for ALA-D activity measurement (Berlin
and Schaller, 1974), developed with the collaboration of nine laboratories for use with blood
samples having relatively low lead content, triplicate blood samples (0.2 ml) are hemolyzed,
along with a blood blank, with water for 10 minutes at 37°C. Samples are then mixed with
6-ALA solution followed by a 60-minute incubation. The enzyme reaction is terminated by ad-
dition of a solution of mercury (II) in trichloroacetic acid, followed by centrifugation and
filtration. Filtrates are mixed with modified Ehrlich's reagent (p-dimethylaminobenzalehyde
in trichloroacetic/perchloric acid mixture) and allowed to react for 5 minutes, followed by
chromophore measurement in a spectrophotometer at 555 nm. Activity is quantified in terms of
pM 6-ALA/min-l erythrocytes. It should be noted that the amount of phosphate for Solution A
in Berlin & Schaller's report should be 1.78 g, not the 1.38 g stated. In a micro scale
variation, Granick et al. (1973) used only 5 ul of blood and terminated the assay by tri-
chloroacetic acid.
In comparing various reports concerning the relationship between lead exposure and ALA-D
inhibition, attention should be paid to the units of activity measurement employed with the
different techniques. Berlin and Schaller's (1974) procedure expresses activity as uM
ALA/min/1 cells, while Tomokuni's (1974) method expresses activity as uM porphobilinogen/hr/ml
cells. Similarly, when comparing the Bonsignore et al. (1965) procedure to that of Berlin and
Schaller (1974), a conversion factor of 3.8 is necessary when converting from Bonsignore to
European Standard Method units (Trevisan et al., 1981).
Several factors have been shown to affect ALA-D activity. Rather than measuring enzyme
activity in blood once, Granick et al. (1973) measured activity before and after treatment
with dithiothreitol, an agent that reactivates the enzyme by complexing lead. The ratio of
activated to unactivated enzymes vs. blood lead levels accommodates inherent differences in
enzyme activity among individuals due to genetic factors and other reasons. Other agents for
such activation include zinc (Finelli et al., 1975) and zinc plus glutathione (Mitchell et
al., 1977). In the Mitchell et al. (1977) study, non-physiological levels of zinc were used.
Wigfield and Farant (1979) found that enzyme activity is related to assay pH; thus, reduced
activity from such a pH-activity relationship could be misinterpreted as lead inhibition.
These researchers find that pH shifts away from optimal, in terms of activity, as blood lead
content increases and the incubation step proceeds.
23PB12/C 9-25 7/1/83
-------
PRELIMINARY DRAFT
9.6 MEASUREMENT OF DELTA-AMINOLEVULINIC ACID IN URINE AND OTHER MEDIA
Delta-aminolevulinic acid (6-ALA) levels increase with elevated lead exposure, due to the
inhibitory effect of lead on the activity of ALA dehydrase and/or the increase of ALA synthe-
tase activity by feedback derepression. The result is that this intermediate in heme bio-
synthesis rises in the body and eventually results in increased urinary excretion. The meas-
urement of this metabolite in urine provides an indication of the level of lead exposure.
The ALA content of urine samples is stable for approximately 2 weeks or more if urine
samples are acidified with tartaric or acetic acid and kept refrigerated. Values of ALA-U are
adjusted for urine density, if concentration is expressed in mg/1 or is measured per gram
creatinine. As noted in the case of urinary lead measurement, 24-hour collection is more de-
sirable than spot sampling.
Five manual and one automated procedure for urinary ALA measurement are most widely used.
Mauzerall and Granick (1956) and Davis and Andelman (1967) described the most involved proce-
dures, requiring the initial chromatographic separation of ALA. The approach of Grabecki et
al. (1967) omitted chromatographic isolation, whereas the automated variation of Lauwerys et
al. (1972) omitted prechromatography but included the use of an internal standard. Tomokuni
and Ogata (1972) omitted, chromatography but employed solvent extraction to isolate the pyr-
role intermediate.
Mauzerall and Granick (1956) condensed ALA with a p-dicarbonyl compound, acetylacetone,
at pH 4.6 to yield a pyrrole intermediate (Knorr condensation reaction), which was further re-
acted with p-dimethylaminobenzaldehyde in perchloric/acetic acid. The samples were then read
in a spectrophotometer at 553 nm 15 minutes after mixing. In this method, there is separation
of both porphobillnogen and ALA from urine by means of a dual column configuration of cation
and anion exchange resins. The latter retains the porphobilinogen and the former separates
ALA from urea. The detection limit is 3 umoles/1 urine. In the modification of this method
by Davis and Andelman (1967), disposable cation/anion resin cartridges were used, in a
sequential configuration, to expedite chromatographic separation and increase sample analysis
rate. Commercial (Bio-Rad) disposable columns based on this design are now available and
appear satisfactory.
In these two approaches (Mauzerall and Granick, 1956; Davis and Andelman, 1967), the pro-
blem of interference due to aminoacetone, a metabolite occurring in urine, is not taken into
account. However, Marver et al. (1966) used Dowex-1 in a chromatographic step subsequent to
the condensation reaction,to form the pyrrole. This separates the ALA derivative from that of
the aminoacetone. Similarly, Schlenker et al. (1964) used an IRC column to retain amino-
ace tone.
23PB12/C g-26 7/1/83
-------
PRELIMINARY DRAFT
Tomokuni and Ogata (1972) condensed ALA with ethylacetoacetate and extracted the re-
sulting pyrrole with ethyl acetate. The extract was then treated with Ehrlich's reagent and
the resulting chromophore measured spectrophotometrically. Lauwerys et al. (1972) developed
an automated ALA analysis method for lead worker screening, in which ALA was added in known
amount as an internal standard and the pre-chromatography avoided. They reported a high cor-
relation (r = 0.98, no range available) with the procedure of Mauzerall and Granick (1956).
Roels et al. (1974) compared the relative proficiency of four methods — those of
Mauzerall and Granick (1956), Davis and Andelman (1967), the Lauwerys et al. (1972) automated
version, and the Grabecki et al. (1967) method, which omits chromatographic separation and is
normally used with occupational screening. The chromatographic methods gave identical results
over the range of 0-60 mg ALA/1 urine, while the automated method showed a positive bias at <6
ng/1. The Grabecki et al. (1967) technique was the least satisfactory of the procedures com-
pared. Roels et al. (1974) also noted that commercial ion-exchange columns resulted in low
variability (<10 percent).
Della-Fiorentina et al. (1979) combined the Tomokuni and Ogata (1972) extraction method
with a correction equation for urine density. Up to 25 mg ALA/1, the C.V. was ^4 percent
along with a good correlation (r = 0.937) with the Davis and Andelman (1967) technique. While
there is a time saving in avoiding prechromatography, it is necessary to prepare a curve re-
lating urine density to a correction factor for quantitative measurement.
Although ALA analysis is normally done with urine as the indicator medium, Haeger-Aronsen
(1960) reported a similar colorimetric method for blood and MacGee et al. (1977) described a
gas-liquid chromatographic method for ALA in plasma as well as urine. Levels of ALA in plasma
are much lower than those in urine. In the latter method, ALA was isolated from plasma, re-
acted with acetyl-acetone, and partitioned into a solvent (trimethylphenylhydroxide), which
also served for pyrolytic methylation in the injection port of the gas-liquid chromatograph,
the methylated pyrrole being more amenable to chromatographic isolation than the more polar
precursor. t For quantification, an internal standard, 6-amino-5-oxohexanoic acid, was used.
The sample requirement is 3 ml plasma. Measured levels ranged from 6.3 to 73.5 ng ALA/ml
plasma, and yielded values that were approximately 10-fold lower than the colorimetric techni-
ques (O1Flaherty et al., 1980).
9.7 MEASUREMENT OF PYRIMIDINE-5'-NUCLEOTIDASE ACTIVITY
Erythrocyte pyrimidine-5'-nucleotidase (5'-ribonucleotide phosphohydrolase, E.C. 3.1.3.5,
Py5N) catalyzes the hydrolytic dephosphorylation of the pyrlmidlne nucleotides uridine mono-
phosphate (UMP) and cytidinemonophosphate (CMP) to uridine and cytidine (Paglia and Valentine,
1975). Enzyme Inhibition by lead in humans and animals results in Incomplete degradation of
23PB12/C 9-27 7/1/83
-------
PRELIMINARY DRAFT
reticulocyte RNA fragments, accumulation of the nucleotides, and increased cell hemolysis
(Paglia et al., 1975; Paglia and Valentine, 1975; Angle and Mclntire, 1978; George and Duncan,
1982).
There are two methods for measurement of Py5N activity. One is quite laborious in terms
of time and manipulation, while the other is shorter but requires the use of radioisotopes and
radiometric measurement. In Paglia and Valentine's (1975) method, heparinized venous blood
was filtered through cotton or a commercial cellulose preparation to separate erythrocytes
from platelets and leukocytes. Cells were given multiple saline washings, packed lightly, and
subjected to freeze hemolysis. The hemolysates were dialyzed against a saline-Tris buffer
containing MgCl2 and EDTA to remove nucleotides and other phosphates. The assay system con-
sists of dialyzed hemolysate, MgCl2, Tris buffer at pH 8.0, and either UMP or CMP; incubation
is for 2 hours at 37°C. Activity is terminated by treatment with 20 percent trichloroacetic
acid, followed by centrifugation. The supernatant inorganic phosphate, P^, is measured by the
classic method of Fiske and Subbarow (1925), the phosphontolybdic acid complex being measured
spectrophotometrically at 660 nm. A unit of enzyme activity is expressed as umol P^hr/g
hemoglobin. Hemolysates appear to be stable (90 percent) with refrigeration at 4°C for up to
6 days, provided that mercaptoethanol is added at the time of assay. Like the other method,
activity measurement requires the determination of hemoglobin.
In the simpler approach of Torrance et al. (1977), which can be feasibly applied to much
larger numbers of samples, erythrocytes were separated from leukocytes and platelets with a
1:1 mixture of microcrystalline and alphacellulose, followed by saline washing and hemolysis
with a solution of mercaptoethanol and EDTA. Hemolysates were incubated with a medium con-
taining purified 14C-CMP and MgCl2 for 30 minutes at 37°C. The reaction was terminated by
sequential addition of barium hydroxide and zinc sulfate solution. Proteins and unreacted
nucleotide were precipitated, leaving the labeled cytidine in the supernatant. Aliquots were
measured for 14C activity in a liquid scintillation counter. Enzyme activity was expressed as
nM CMP/min/g hemoglobin. The blank activity was determined for each sample by carrying out
the precipitation step as soon as the hemolysate was mixed with the labeled CMP, i.e., t = 0.
This procedure shows a good correlation (r = 0.94; range: 135-189 enzyme units) with the
method of Paglis and Valentine (1975). The two methods express units of enzyme activity dif-
ferently, so that one must know which method is used when comparing enzyme activity.
23PB12/C 9-28 7/1/83
-------
PRELIMINARY DRAFT
9.8 SUMMARY
The sine qua non of a complete understanding of a toxic agent's effects on an organism,
e.g., dose-effect relationships, is quantitative measurement of either that agent in some bio-
logical medium or a physiological parameter associated with exposure to the agent. Quantita-
tive analysis involves a number of discrete steps, all of which contribute to the overall
reliability of the final analytical result: sample collection and shipment, laboratory
handling, instrumental analysis, and criteria for internal and external quality control.
From a historical perspective, it is clear that the definition of "satisfactory analyt-
ical method" for lead has been steadily changing as new and more sophisticated equipment
becomes available and understanding of the hazards of pervasive contamination along the
analytical course increases. The best example of this is the use of the definitive method for
lead analysis, isotope-dilution mass spectrometry in tandem with "ultra-clean" facilities and
sampling methods, to demonstrate conclusively not only the true extent of anthropogenic input
of lead to the environment over the years but also the relative limitations of most of the
methods for lead measurement used today.
9.8.1 Determinations of Lead in Biological Media
The low levels of lead in biological media, even in the face of excessive exposure, and
the fact that sampling of such media must be done against a backdrop of pervasive lead contam-
ination necessitates that samples be carefully collected and handled. Blood lead sampling is
best done by venous puncture and collection into low-lead tubes after careful cleaning of the
puncture site. The use of finger puncture as an alternative method of sampling should be
avoided, if feasible, given the risk of contamination associated with the practice in indus-
trialized areas. While collection of blood onto filter paper enjoyed some popularity in the
past, paper deposition of blood requires special correction for hematrocrit/hemoglobin level.
Urine sample collection requires the use of lead-free containers as well as addition of a
bacteriocide. If feasible, 24-hour sampling is preferred to spot collection. Deciduous teeth
vary in lead content both within and across type of dentition. Thus a specific tooth type
should be uniformly obtained for all study subjects and, if possible, more than a single
sample should be obtained from each subject.
Measurements of Lead in Blood. Many reports over the years have purported to offer
satisfactory analysis of lead in blood and other biological media, often with severe inherent
limitations on accuracy and precision, meager adherence to criteria for accuracy and pre-
cision, and a limited utility across a spectrum of analytical applications. Therefore, it is
only useful to discuss "definitive" and, comparatively speaking, "reference" methods presently
used.
23PB12/C 9-29 7/1/83
-------
PRELIMINARY DRAFT
In the case of lead in biological media, the definitive method is isotope-dilution mass
spectrometry (IDMS). The accuracy and unique precision of IDMS arise from the fact that all
manipulations are on a weight basis involving simple procedures, and measurements entail only
lead isotope ratios and not the absolute determinations of the isotopes involved, greatly re-
ducing instrumental corrections and errors. Reproducible results to a precision of one part
in 104-10S are routine with appropriately designed and competently operated instrumentation.
Although this methodology is still not recognized in many laboratories, it was the first
breakthrough, in tandem with "ultra-clean" procedures end facilities, to definitive methods
for indexing the progressive increase in lead contamination of the environment over the
centuries. Given the expense, required level of operator expertise, and time and effort
involved for measurements by IDMS, this methodology mainly serves for analyses that either
require extreme accuracy and precision, e.g., geochronometry, or for the establishment of
analytical reference material for general testing purposes or the validation of other
methodologies.
While the term "reference method" for lead in biological media cannot be rigorously ap-
plied to any procedures in popular use, the technique of atomic absorption spectrometry in its
various configurations or the electrochemical method, anodic stripping voltammetry, come
closest to meriting the designation. Other methods that are generally applied in metal anal-
yses are either limited in sensitivity or are not feasible for use on theoretical grounds for
lead analysis.
Atomic absorption spectrometry (AAS) as applied to analysis of whole blood generally In-
volves flame or flameless micromethods. One macromethod, the Hessel procedure, still enjoys
some popularity. Flame microanalysis, the Delves cup procedure, applied to blood lead appears
to have an operational sensitivity of about 10 ug Pb/dl blood and a relative precision of
approximately 5 percent in the range of blood lead seen in populations in industrialized
areas. The flameless, or electrothermal, method of AAS enhances sensitivity about 10-fold,
but precision can be more problematical because of chemical and spectral interferences.
The most widely used and sensitive electrochemical method for lead in blood is anodic
stripping voltammetry (ASV). For most accurate results, chemical wet ashing of samples must
be carried out, although this .process is time-consuming and requires the use of lead-free
reagents. The use of metal exchange reagents has been employed in lieu of the ashing step to
liberate lead from binding sites, although this substitution is associated with less
precision. For the ashing method, relative precision is approximately 5 percent. In terms of
accuracy and sensitivity, it appears that there are problems at low levels, e.g., 5 ug/dl or
below, particularly if samples contain elevated cooper levels.
23PB12/C 9-30 7/1/83
-------
PRELIMINARY DRAFT
Lead in Plasma. Since lead in whole blood is virtually all confined to the erythrocyte,
plasma levels are quite low and it appears that extreme care must be employed to reliably
measure plasma levels. The best method for such measurement is IDMS, in tandem with ultra-
clean facility use. Atomic absorption spectrometry is satisfactory for comparative analyses
across a range of relatively high whole blood values.
Lead in Teeth. Lead measurement in teeth has involved either whole tooth sampling or
analysis of specific regions, such as primary or circumpulpal dentine. In either case, sam-
ples must be solublized after careful surface cleaning to remove contamination; solubilization
is usually accompanied by either wet ashing directly or ashing subsequent to a dry ashing
step.
Atomic absorption spectrometry and anodic stripping have been employed more frequently
for such determinations than any other method. With AAS, the high mineral content of teeth
argues for preliminary isolation of lead via chelation-extraction. The relative precision of
analysis for within-run measurement is around 5-7 percent, with the main determinant of vari-
ance in regional assay being the initial isolation step. One change from the usual methods
for such measurement is the jn sjtu measurement of lead by X-ray fluorescence spectrometry in
children. Lead measured in this fashion allows observation of on-going lead accumulation,
rather than waiting for exfoliation.
Lead in Hair. Hair as an exposure indicator for lead offers the advantages of being non-
s
invasive and a medium of indefinite stability. However, there is still the crucial problem of
external surface contamination, which is such that it is still not possible to state that any
cleaning protocol reliably differentiates between external and internally deposited lead.
Studies that demonstrate a correlation between increasing hair lead and increasing sever-
ity of a measured effect probably support arguments for hair being an external indicator of
exposure. It is probably also the case, then, that such measurement, using cleaning protocols
that have not been independently validated, will overstate the relative accumulation of "in-
ternal" hair lead in terms of some endpoint and will also underestimate the relative sensiti-
vity of changes in internal lead content with exposure. One consequence of this would be, for
example, an apparent threshold for a given effect in terms of hair lead which is significantly
above the actual threshold. Because of these concerns, hair is best used with the simultan-
eous measurement of blood lead.
Lead in Urine. Analysis of lead in urine is complicated by the relatively low levels of
the element in this medium as well as the complex mixture of mineral elements present. Urine
lead levels are most useful and also somewhat easier to determine in cases of chelation mobil-
ization or chelation therapy, where levels are high enough to permit good precision and dilu-
tion of matrix interference.
23PB12/C 9-31 7/1/83
-------
PRELIMINARY DRAFT
Samples are probably best analyzed by prior chemical wet ashing, using the usual mixture
of acids. Both anodic stripping voltammetry and atomic absorption spectrometry have been
applied to urine analysis, with the latter more routinely used and usually with a chelation/
extraction step.
Lead in Other Tissues. Bone samples require cleaning procedures for removal of muscle
and connective tissue and chemical solubilization prior to analysis. Methods of analysis are
comparatively limited and it appears that flameless atomic absorption spectrometry is the
technique of choice.
Lead measurements in bone, jm vivo, have been reported with lead workers, using X-ray
fluorescence analysis and a radioisotopic source for excitation. One problem with this
approach with moderate lead exposure is the detection limit, approximately 20 ppm. Soft organ
analysis poses a problem in terms of heterogeneity in lead distribution within an organ (e.g.
brain and kidney. In such cases, regional sampling or homogenization must be carried out.
Both flame and flameless atomic absorption spectrometry appear to be satisfactory for soft
tissue analysis and are the most widely used.
Quality Assurance Procedures in Lead Analyses. In terms of available information, the
major focus in establishing quality control protocols for lead has involved whole blood meas-
urements. Translated into practice, quality control revolves around steps employed within the
laboratory, using a variety of internal checks, and the further reliance on external checks,
such as a formal continuing multi-laboratory proficiency testing program.
Within the laboratory, quality assurance protocols can be divided into start-up and rou-
tine procedures, the former involving establishment of detection limits, within-run and
between-run precision, analytical recovery, and comparison with some reference technique
within or outside the laboratory. The reference method is assumed to be accurate for the par-
ticular level of lead in some matrix at a particular point in time. Correlation with such a
method at a satisfactory level, however, may simply indicate that both methods are equally
inaccurate but performing with the same level of precision proficiency. More preferable is
the use of certified samples having lead at a level established by the definitive method.
For blood lead, the Centers for Disease Control periodically survey overall accuracy and
precision of methods used by reporting laboratories. In terms of overall accuracy and preci-
sion, one such survey found that anodic stripping voltammetry as well as the Delves cup and
extraction variations of atomic absorption spectrometry performed better than other proce-
dures. These results do not mean that a given laboratory cannot perform better with a partic-
ular technique; rather, such data are of assistance for new facilities choosing among methods.
Of particular value to laboratories carrying out blood lead analysis are the external
quality assurance programs at both the state and federal levels. The most comprehensive
23PB12/C 9-32 7/1/83
-------
PRELIMINARY DRAFT
proficiency testing program is that carried out by the Centers for Disease Control, USPHS.
This program actually consists of two subprograms, one directed at facilities involved in lead
poisoning prevention and screening (Center for Environmental Health) and the other concerned
with laboratories seeking certification under the Clinical Laboratories Improvement Act of
1967 as well as under regulations of the Occupational Safety and Health Administration's
(OSHA) Laboratory Improvement Program Office. Overall, the proficiency testing programs have
served their purpose well, judging from the relative overall improvements in reporting
laboratories over the years of the programs' existence. In this regard, OSHA criteria for
laboratory certification require 8 of 9 samples be correctly analyzed for the previous
quarter. This level of required proficiency reflects the ability of a number of laboratories
to actually perform at this level.
9.8.2 Determination of Erythrocyte Porphyrin (Free Erythrocyte Protoporphyrin, Zinc
Protoporphyrin)
With lead exposure, there is an accumulation of erythrocyte protoporphyrin IX, owing to
impaired placement of divalent iron to form heme. Divalent zinc occupies the place of the na-
tive iron. Depending upon the method of analysis, either metal-free erythrocyte porphyrin or
zinc protoporphyrin (ZPP) is measured, the former arising from loss of zinc in the chemical
manipulation. Virtually all methods now in use for EP analysis exploit the ability of the
porphyrin to undergo intense fluorescence when excited by ultraviolet light. Such fluoro-
metric methods can be further classified as wet chemical micromethods or direct measuring
fluorometry using the hematofluorometer. Owing to the high sensitivity of such measurement,
relatively small blood samples are required, with liquid samples or blood collected on filter
paper.
The most common laboratory or wet chemical procedures now in use represent variations of
several common chemical procedures: 1) treatment of blood samples with a mixture of ethyl
acetate/acetic acid followed by a repartitioning into an inorganic acid medium, or 2) solu-
bilization of a blood sample directly into a detergent/buffer solution at a high dilution.
Quantification has been done using protoporphyrin, coproporphyrin, or zinc protoporphyrin IX
plus pure zinc ion. The levels of precision for these laboratory techniques vary somewhat
with the specifics of analysis. The Piomelli method has a coefficient of variation of 5
percent, while the direct ZPP method using buffered detergent solution is higher and more
variable.
The recent development of the hematofluorometer has made it possible to carry out EP
measurements in high numbers, thereby making population screening feasible. Absolute calibra-
tion is necessary and requires periodic adjustment of the system using known concentrations of
23PB12/C 9-33 7/1/83
-------
PRELIMINARY DRAFT
EP in reference blood samples. Since these units are designed for oxygenated blood, i.e.,
capillary blood, use of venous blood requires an oxygenation step, usually a moderate shaking
for several minutes. Measurement of low or moderate levels of EP can be affected by interfer-
ence with bilirubin. Competently employed, the hematofluorometer appears to be reasonably
precise, showing a total coefficient of variation of 4.11-11.5 percent. While the comparative
accuracy of the unit has been reported to be good relative to the reference wet chemical tech-
nique, a very recent study has shown that commercial units carry with them a significant nega-
tive bias, which may lead to false negatives in subjects having only moderate EP elevation.
Such a bias in accuracy has been difficult to detect in existing EP proficiency testing
programs. It appears that, by comparision to wet methods, the hematofluorometer should be
restricted to field use rather than becoming a substitute in the laboratory for chemical meas-
urement, and field use should involve periodic split-sample comparison testing with the wet
method.
9.8.3 Measurement of Urinary Coproporphyrin
Although EP measurement has largely supplanted the use of urinary coproporphyrin analysis
(CP-U) to monitor excessive lead exposure in humans, this measurement is still of value in
that it reflects active intoxication. The standard analysis is a fluorometric technique
whereby urine samples are treated with buffer, and an oxidant (iodine) is added to generate CP
from its precursor. The CP-U is then partitioned into ethyl acetate and re-extracted with
dilute hydrochloric acid. The working curve is linear below 5 |jg CP/dl urine.
9.8.4 Measurement of Delta-Aminolevulinic Acid Dehydrase Activity
Inhibition of the activity of the erythrocyte enzyme, delta-aminolevulinic acid dehydra-
tase (ALA-D), by lead is the basis for using such activity in screening for excessive lead
exposure. A number of sampling and sample handling precautions attend such analysis. Since
zinc (II) ion will offset the degree of activity inhibition by lead, blood collecting tubes
must have extremely low zinc content. This essentially rules out the use of rubber-stoppered
blood tubes. Enzyme stability is such that the activity measurement is best carried out
within 24 hours of blood collection. Porphobilinogen, the product of enzyme action, is light-
labile and requires the assay be done in restricted light. Various procedures for ALA-D meas-
urement are based on measurement of the level of the chromophoric pyrrole (approximately
555 nm) formed by condensation of the porphobilinogen with p-dimethylaminobenzaldehyde.
In the European Standardized Method for ALA-D activity determination, blood samples are
hemolyzed with water, ALA solution added, followed by incubation at 37°C, and the reaction
terminated by a solution of mercury (II) in trichloroacetic acid. Filtrates are treated with
23PB12/C 9-34 7/1/83
-------
PRELIMINARY DRAFT
modified Ehrlich's reagent (p-dimethyl aminobenzaldehyde) in trichloroacetic/perchloroacetic
acid mixture. Activity is quantified in terms of micromoles ALA/min/liter erythrocytes.
One variation in the above procedure is the initial use of a thiol agent, such as dithio-
threotol, to reactivate the enzyme, giving a measure of the full native activity of the
enzyme. The ratio of activated/unactivated activity vs. blood lead levels accomodates genetic
differences between individuals.
9.8.5 Measurement of Delta-Aminolevulinic Acid in Urine and Other Media
Levels of delta-aminolevulinic acid (6-ALA) in urine and plasma increase with elevated
lead exposure. Thus, measurement of this metabolite, generally in urine, provides an index of
the level of lead exposure. ALA content of urine samples (ALA-U) is stable for about two
weeks or more with sample acidification and refrigeration. Levels of ALA-U are adjusted for
urine density or expressed per unit creatinine. If feasible, 24-hour collection is more
desirable than spot sampling.
Virtually all the various procedures for ALA-U measurement employ preliminary isolation
of ALA from the balance of urine constituents. In one method, further separation of ALA from
the metabolite aminoacetone is done. Aminoacetone can interfere with colorimetric measure-
ment. ALA is recovered, condensed with a beta-dicarbonyl compound, e.g., acetyl acetone, to
yield a pyrrole intermediate. This intermediate is then reacted with p-dimethyl amino-
benzaldehyde in perchloric/acetic acid, followed by colorimetric reading at 553 nm. In one
variation of the basic methodology, ALA is condensed with ethyl acetoacetate directly and the
resulting pyrrole extracted with ethyl acetate. Ehrlich's reagent is then added as in other
procedures and the resulting chromophore measured spectrophotometrically.
Measurement of ALA in plasma is much more difficult than in urine, since plasma ALA is at
nanogram/milliter levels. In one gas-liquid chromatographic procedure, ALA is isolated from
plasma, reacted with acetyl acetone and partitioned into a solvent that also serves for
pyrolytic methylation of the involatile pyrrole in the injector port of the chromatograph,
making the derivative more volatile. For quantification, an interval standard, 6-amino-5-
oxohexanoic acid, is used. While the method is more involved, it is more specific than the
older colorimetric technique.
9.8.6 Measurement of Pyrimidine-5'-Nuc1eotidase Activity
Erythrocyte pyrimidine-S'-nucleotidase (Py5N) activity is inhibited with lead exposure.
Presently two different methods are used for assaying the activity of this enzyme. The older
method is quite laborious in time and effort, whereas the more recent approach is shorter but
uses radioisotopes and radiometric measurement.
23PB12/C 9-35 7/1/83
-------
PRELIMINARY DRAFT
In the older method, heparinized venous blood is filtered through cellulose to separate
erythrocytes from platelets and leukocytes. Cells are then freeze-fractured and the hemo-
lysates dialyzed to remove nucleotides and other phosphates. This dialysate is then incubated
in the presence of a nucleoside monophosphate and cofactors, the enzyme reaction being termi-
nated by treatment with trichloroacetic acid. The inorganic phosphate isolated from added
substrate is measured colorimetrically as the phosphomolybdic acid complex.
In the radiometric assay, hemolysates obtained as before are incubated with pure 14C-CMP.
By addition of a barium hydroxide/zinc sulfate solution, proteins and unreacted nucleotide are
precipitated, leaving labeled cytidine in the supernatant. Aliquots are measured for 14C ac-
tivity in a liquid scintillation counter. This method shows a good correlation with the ear-
lier technique.
23PB12/C 9-36 7/1/83
-------
PRELIMINARY DRAFT
9.9 REFERENCES
Ahlgren, L.; Haeger-Aronsen, B. ; Mattson, S.; Schutz, A. (1980) In vivo determination of lead
in the skeleton after occupational exposure to lead. Br. J. Ind. Med. 37: 109-113.
Al-Naimi, T.; Edmonds, M. I.; Fremlin, J. H. (1980) The distribution of lead in human teeth,
using charged particle activation analysis. Phys. Med. Biol. 25: 719-726.
American Public Health Association, Committee on Clinical Procedures. (1955) Methods for
determining lead in air and biological materials. New York, NY: American Public Health
Association.
Angle, C. R.; Mclntire, M. S. (1978) Low level lead and inhibition of erythrocyte pyrimidine
nucleotidase. Environ. Res. 17: 296-302.
Balamut, R.; Doran, D.; Giridhar, G.; Mitchell, D.; Soule, S. (1982) Systematic error between
erythrocyte protoporphyrin in proficiency test samples and patients' samples as measured
with two hematofluorometers. Clin. Chem. (Winston Salem N.C.) 28: 2421-2422.
Barthel, W. F.; Smrek, A. L. ; Angel, G. P.; Liddle, J. A.; Landrigan, P. J.; Gehlbach, S. H.;
Chisolra, J. J. (1973) Modified Delves cup atomic absorption determination of lead in
blood. J. Assoc. Off. Anal. Chem. 56: 1252-1256.
Berlin, A.; Schaller, K. H. (1974) European standardized method for the determination of
6-aminolevulinic acid dehydratase activity in blood. Z. Klin. Chem. Klin. Biochem. 12:
389-390.
Berlin, A.; Del Castilho, P.; Smeets, J. (1973) European intercomparison programmes. In:
Barth, D.; Berlin, A.; Engel, R.; Recht, P.; Smeets, J. , eds. Environmental health
aspects of lead: proceedings, international symposium; October 1972; Amsterdam, The
Netherlands. Luxembourg: Commission of the European Communities, Centre for Information
and Documentation; pp. 1033-1046.
Berman, E. (1976) The challenge of getting the lead out. In: LaFleur, P. D., ed. Accuracy in
trace analyses: sampling, sample handling, analysis - volume 2. Proceedings of the 7th
materials research symposium; October 1974; Gaithersburg, MD. Washington, DC: U.S.
Department of Commerce, National Bureau of Standards; NBS special publication no. 422;
pp. 715-719.
Berman, E. (1981) Heavy metals. Lab. Med. 12: 677-684.
Bloch, P.; Garavaglia, G.; Mitchell, G.; Shapiro, I. M. (1976) Measurement of lead content of
children's teeth jm situ by X-ray fluorescence. Phys. Med. Biol. 20: 56-63.
Blumberg, W. E.; Eisinger, J.; Lamola, A. A.; Zuckerman D. M. (1977) Zinc protoporphyrin level
in blood determined by a portable hematofluorometer: a screening device for lead
poisoning. J. Lab. Clin. Med, 89: 712-723.
Bonsignore, D.; Calissano, P.; Cartasegna, C. (1965) Un semplice metodo per la determinazione
della 6-amino-levul1nico-deidratasi nel sangue: comportamento dell'enzima nelV
intossicazione saturnine. [A simple method for determining 6-aminolevulinic dehydratase
in the blood: behavior of the enzyme in lead poisoning.] Med. Lav. 56: 199-205.
SRD13REF/E 9-37 7/1/83
-------
PRELIMINARY DRAFT
Boone, J.; Hearn, T.; Lewis, S. (1979) A comparison of interlaboratory results for blood lead
with results from a definitive- method. Clin. Chem. (Winston Salem N.C.) 25: 389-393.
Boutwell, J. H. (1976) Accuracy and quality control in trace element analyses. In: LaFleur
P. D., ed. Accuracy in trace analysis: sampling, sample handling, analysis - volume 1*
Proceedings of the 7th materials research symposium; October 1974; Gaithersburg, MD*
Washington, DC: U.S. Department of Commerce, National Bureau of Standards; NBS special
publication no. 422; pp. 35-40.
Cali, S. P.; Reed, W. P. (1976) The role of the National Bureau of Standards reference
materials in accurate trace analysis. In: LaFleur, P. D., ed. Accuracy in trace analysis:
sampling, sample handling, analysis - volume 1. Proceedings of the 7th materials research
symposium; October 1974; Gaithesburg, MD. Washington, DC: U.S. Department of Commerce
National Bureau of Standards; NBS special publication no. 422; pp. 41-63.
Carter, G. F. (1978) The paper punched disc technique for lead in blood samples with abnormal
haemoglobin values. Br. J. Ind. Med. 35: 235-240.
Castoldi, M. R. ; Odone, P.; Buratti, M.; Alessio, L. (1979) Determination of erythrocyte zinc
protoporphyrin: methodological problems. In: International conference: management and
control of heavy metals in the environment; September; London, United Kingdom. Edinburgh
United Kingdom: CEP Consultants, Ltd.; pp. 113-117. '
Cavalleri, A.; Minoia, C.; Pozzoli, L.; Baruffini, A. (1978) Determination of plasma lead
levels in normal subjects and in lead-exposed workers. Br. J. Ind. Med. 35: 21-26.
Cernik, A. A.; Sayers, M. P. H. (1971) Determination of lead in capillary blood using a paper
punched disc atomic absorption technique: applications to the supervision of lead
workers. Br. J. Ind. Med. 28: 392-398.
Chatman, T.; Wilson, D. J. (1975) Lead levels in human deciduous teeth in Tennessee. Environ
Lett. 8: 173-183.
Chatt, A.; Secord, C. A.; Tiefenbach, B.; Jervis, R. E. (1980) Scalp hair as a monitor of com-
munity exposure to environmental pollutants. In: Brown, A. C.; Crounse, R. C., eds. Hair,
trace elements and human illness. New York, NY: Praeger; pp. 46-73.
Chattopadhyay, A.; Roberts, T. M.; Jervis, R. E. (1977) Scalp hair as a monitor of community
exposure to lead. Arch. Environ. Health 32: 226-236.
Chisolm, J. J., Jr. (1974) Lead in red blood cells and plasma. J. Pediatr. (St. Louis) 84-
163-164.
Chisolm, J. J., Jr.; Brown, D. H. (1975) Micro-scale photofluormetric determination of "free
erythrocyte porphyrin" (protoporphyrin IX). Clin. Chem. (Winston Salem, NC) 21-
1669-1682.
Chisolm, J. J., Jr-; Brown, D. H. (1979) Micromethod for zinc protoporphyrin in erythrocytes:
including new data on the absorptivity of zinc protoporphyrin and new observations in
neonates and sickle cell disease. Biochem. Med. 22: 214-237.
Chisolm, J. J., Jr-; Hastings, C. W.; Cheung, 0. K. K. (1974) Microphotofluorometric assay for
protoporphyrin in acidified acetone extracts of whole blood. Biochem. Med. 9: 113-135.
SR013REF/E 9-38 7/1/83
-------
PRELIMINARY DRAFT
Cooke, R. E.; Glynn, K. L.; Ullman, W. W.; Lurie, N.; Lepow, M. (1974) Comparative study of a
micro-scale test for lead in blood, for use in mass screening programs. Clin. Chem.
(Winston Salem, N.C.) 20: 582-585.
Culbreth, P.; Walter, G.; Carter, R.; Burtis, C. (1979) Separation of protoporphyrins and
related compounds by reversed-phase liquid chromatography. Clin. Chem. (Winston Salem
N.C.) 25: 605-610.
Davis, J. R.; Andelman, S. L. (1967) Urinary delta-aminolevulinic acid (ALA) levels in lead
poisoning. Arch. Environ. Health 15: 53-59.
Delia Fiorentina, H.; Grogna, M.; Dewiest, F. (1979) Simplified determination of urinary
6-aminolevulinic acid in a wide range of concentrations. Clin. Chem. (Winston Salem N.C.)
25: 581-583.
Delves, H. T. (1970) A micro-sampling method for the rapid determination of lead in blood by
atomic absorption spectrophotometry. Analyst (London) 95: 431-438.
Delves, H. T. (1977) Analytical techniques for blood-lead measurements. J. Anal. Toxicol. 1:
261-264.
Delves, H. T.; Clayton, B. E.; Carmichael, A.; Bubear, M.; Smith M. (1982) An appraisal of the
analytical significance of tooth-lead measurements as possible indices of environmental
exposure of children to lead. Ann. Clin. Biochem. 19: 329-337.
DeSilva, P. E. (1981) Determination of lead in plasma and studies on its relationship to lead
in erythrocytes. Br. J. Ind. Med. 38: 209-217.
DeSilva, P. E.; Donnan, M. B. (1980) Blood lead levels in Victorian children. Med. J. Aust.
2: 315-318.
Donovan, D. T.; Vought, V. M.; Rakow, A. B. (1971) Laboratories which conduct lead analysis on
biologic specimens. Arch. Environ. Health 23: 111-113.
Dudley, D. M. T. (1982) Critique: blood lead analyses 1981. Atlanta, GA: U.S. Centers for
Disease Control.
Ediger, R. D.; Coleman, R. L. (1972) A modified Delves cup atomic absorption procedure for the
determination of lead in blood. At. Absorp. Newsl. 11: 33-36.
Eller, P. M.; Hartz, J. C. (1977) A study of methods for the determination of lead and cad-
mium. Am. Ind. Hyg. Assoc. J. 38: 116-124.
Everson, J.; Patterson, C. C. (1980) "Ultra-clean" isotope dilution/mass spectrometric anal-
yses for lead in human blood plasma Indicate that most reported values are artificially
high. Clin. Chem. (Winston Salem N.C.) 26: 1603-1607.
FarHs, F. F.; PokHs, A.; Griesmann, G. E. (1978) Atomic absorption spectroscopic determina-
tion of lead extracted from acid-solubilized tissues. J. Assoc. Off. Anal. Chem. 61:
660-663.
FinelH, V. N.; Klauder, D. S.; Karaffa, M. A.; Petering, H. G. (1975) Interaction of zinc and
lead on 6-am1nolevul1nate dehydratase. Biochem. Biophys. Res. Commun. 65: 303-311.
SRD13REF/E 9-39 7/1/83
-------
PRELIMINARY DRAFT
Fiske, C. H.; Subbarow, Y. (1925) The colorimetric determination of phosphorous. J. Biol
Chem. 66: 375-400.
Fosse, G.; Justesen, N. P. B. (1978) Lead in deciduous teeth of Norwegian children. Arch
Environ. Health 33: 166-175.
Franke, J. P.; de Zeeuw, R. A. (1977) Toxic metal analysis by differential pulse anodic
stripping voltammetry in clinical and forensic toxicology. J. Anal. Toxicol. 1: 291-295.
George, J. W.; Duncan, J. R. (1982) Pyrimidine-specific 5' nucleotidase activity in bovine
erythrocytes: effect of phlebotomy and lead poisoning. Am. J. Vet. Res. 43: 17-20.
Gibson, R. S. (1980) Hair as a biopsy material for the assessment of trace element status in
infancy: a review. J. Human Nutr. 34: 405-416.
Grabecki, J.; Haduch, T.; Urbanowicz, H. (1967) Die einfachen Bestimmungsmethoden der
d-aminolavulinsaure im Harn. [Simple methods for the determination of 6-aminolevulinic
acids in the urine.] Int. Arch. Gewerbepathol. Gewerbehyg. 23: 226-240.
Grandjean, P.; Lintrup, J. (1978) Erythrocyte-Zn-protoporphyrin as an indicator of lead
exposure. Scand. J. Clin. Lab. Invest. 38: 669-675.
Grandjean, P.; Nielsen, 0. V.; Shapiro, I. M. (1978) Lead retention in ancient Nubian and
contemporary populations. J. Environ. Pathol. Toxicol. 2: 781-787.
Granick, S.; Sassa, S.; Granick, J. L.; Levere, R. D.; Kappas, A. (1972) Assays for por-
phyrins, 6-aminolevulinic-acid dehydratase, and porphyrinogen synthetase in microliter
samples of whole blood: applications to metabolic defects involving the heme pathway
Proc. Natl. Acad. Sci. U.S.A. 69: 2381-2385.
Granick, J. L.; Sassa, S.; Granick, S.; Levere, R. D.; Kappas, A. (1973) Studies in lead
poisoning. II: Correlation between the ratio of activated and inactivated
6-aminolevulinic acid dehydratase of whole blood and the blood lead level. Biochem. Med
8: 149-159.
Haeger-Aronsen, B. (1960) Studies on urinary excretion of 6-aminolaevulic acid and other haem
precursors in lead workers and lead-intoxicated rabbits. Scand. J. Clin. Lab. Invest. 12
(Suppl. 47) 1-128.
Hanna, T. L.; Dietzler, D. N.; Smith, C. H.; Gupta, S.; Zarkowsky, H. S. (1976) Erythrocyte
porphyrin analysis in the detection of lead poisoning in children: evaluation of four
micromethods. Clin. Chem. (Winston Salem N.C.) 22: 161-168.
Herber R. F. M. (1980) Estimation of blood lead values from blood porphyrin and urinary
5-aminolevulinic acid levels in workers. Int. Arch. Occup. Environ. Health 45: 169-179.
Hessel, D. W. (1968) A simple and rapid quantitative determination of lead in blood. At
Absorp. Newsl. 7: 55-56.
Hicks, J. M.; Gutierrez, A. N.; Worthy, B. E. (1973) Evaluation of the Delves micro system for
blood lead analysis. Clin. Chem. (Winston Salem N.C.) 19: 322-325.
Hinderberger, E. J.; Kaiser, M. L.; Koirtyohann, S. R. (1981) Furnace atomic absorption
analysis of biological samples using the L'vov platform and matrix modification. At
Spectr. 2: 1-7.
SRD09REF/B 9-40 7/1/83
-------
PRELIMINARY DRAFT
Hodges, D. J. ; Skelding, D. (1981) Determination of lead in urine by atomic-absorption
spectroscopy with electrothermal atomisation. Analyst (London) 106: 299-304.
Issaq, H. J.; Zielinski, W. L., Jr. (1974) Loss of lead from aqueous solutions during storage.
Anal. Chem. 46: 1328-1329.
Jackson, K. W. (1978) Interlaboratory comparison of results of erythrocyte protoporphyrin
analysis. Clin. Chem. (Winston Salem N.C.) 24: 2135-2138.
Jagner, D. ; Danielsson, L. G. ; Aren, K. (1979) Potentiometric stripping analysis for lead in
urine. Anal. Chim. Acta 106: 15-21.
Joselow, M. M. ; Bogden, J. D. (1972) A simplified micro method for collection and determina-
tion of lead in blood using a paper disk-in-Delves cup technique. At. Absorp. Newsl. 11:
99-101.
Joselow, M. M. ; Flores, J. (1977) Application of the zinc protoporphyrin (ZP) test as a
monitor of occupational exposure to lead. Am. Ind. Hyg. Assoc. J. 38: 63-66.
Karacic, V. ; Prpic"-Majic D. ; Telisman, S. (1980) The relationship between zinc protoporphyrin
(ZPP) and "free" erythrocyte protoporphyrin (FEP) in lead-exposed individuals. Int. Arch.
Occup. Environ. Health 47: 165-177.
Keppler, J. F.; Maxfield, M. E. ; Moss, W. D.; Tietjen, G.; Linch, A. L. (1970) Interlaboratory
evaluation of the reliability of blood lead analyses. J. Am. Ind. Hyg. Assoc. 31:
412-429.
Kochen, J. A.; Greener, Y. (1973) Levels of lead in blood and hematocrit: implications for the
evaluation of the newborn and anemic patient. Pediatr. Res. 7: 937-944.
Koizumi, H.; Yasuda, K. (1976) Determination of lead, cadmium, and zinc using the Zeeman
effect in atomic absorption spectrometry. Anal. Chem. 48: 1178-1182.
Kopito, L. E. ; Davis, M. A.; Schwachman, H. (1974) Sources of error in determining lead in
blood by atomic absorption spectrophotometry. Clin. Chem. (Winston Salem N.C.) 20:
205-211.
Kubasik, N. P.; Volosin, M. T.; Murry, M. H. (1972) Carbon rod atomizer applied to measurement
of lead in whole blood by atomic absorption spectrophotometry. Clin. Chem. (Winston Salem
N.C.) 18: 410-412.
LaFleur, P. D. , ed. (1976) Accuracy in trace analysis: sampling, sample handling, analysis:
vols. I and II. Proceedings of the 7th materials research symposium; October 1974;
Gaithersburg, MD. Washington, DC: U.S. Department of Commerce, National Bureau of
Standards; NBS special publication no. 422. Available from: U.S. GPO, Washington, DC; SN
003-003-01458-3.
Lamola, A-A.; Joselow, M.; Yamane, T. (1975) Zinc protoporphyrin (ZPP): a simple, sensitive,
fluorometric screening test for lead poisoning. Clin. Chem. (Winston Salem N.C.) 21:
93-97.
Lauwerys, R.; Delbroeck, R.; Vens, M. D. (1972) Automated analysis of delta-aminolaevulinic
acid in urine. Clin. Chim. Acta 40: 443-447.
SRD13REF/E 9-41 7/1/83
-------
PRELIMINARY DRAFT
Lauwerys, R. ; Buchet, J-p.; Roels, H.; Berlin, A.; Smeets, J. (1975) Intercomparison program
of lead, mercury and cadmium analysis in blood, urine, and aqueous solutions. Clin. Chem
(Winston Salem, N.C.) 21: 551-557.
Lawrence, D. M. (1982) An Atomic Spectroscopy bibliography for January - June 1982 At
Spectr. 3: 95-115.
Lawrence, D. M. (1983) An Atomic Spectroscopy bibliography for July - December 1982. At
Spectr. 4: 10-33.
Legotte, P. A.; Rosa, W. C.; Sutton, D. C. (1980) Determination of cadmium and lead in urine
and other biological samples by graphite-furnace atomic-absorption spectrometry. Talanta
27: 39-44.
Lerner, S. (1975) Blood lead analysis--precision and stability. J. Occup. Med. 17: 153-154.
Lockeretz, W. (1975) Lead content of deciduous teeth of children in different environments
Arch. Environ. Health 30: 583-587.
MacGee, J. ; Roda, S. M. B.; Elias, S. V.; Lington, E. A.; Tabor, M. W.; Hammond, P. B. (1977)
Determination of 6-aminolevulinic acid in blood plasma and urine by gas-liquid chromato-
graphy. Biochem. Med. 17: 31-44.
Machlan, L. A.; Gramlich, J. W.; Murphy, T. J.; Barnes, I. L. (1976) The accurate determina-
tion of lead in biological and environmental samples by isotope dilution mass spectro-
metry. In: LaFleuer, P. D., ed. Accuracy in trace analysis: sampling, sampling handling,
analysis - volume 2. Proceedings of the 7th materials research symposium; October 1974J
Gaithersburg, MD. Washington, DC: U.S. Department of Commerce, National Bureau of
Standards; NBS special publication no. 422; pp. 929-935
Mackie, A. C.; Stephens, R.; Towns.end, A.; Waldron, H. A. (1977) Tooth lead levels in
Birmingham children. Arch. Environ. Health 32: 178-185.
Mahaffey, K. R.; Annest, J. L.; Barbano, H. E.; Murphy, R. S. (1979) Preliminary analysis of
blood lead concentrations for children and adults: HANES II, 1976-1978. In: Hemphill, D.
D., ed. Trace substances in environmental health-XIII: [proceedings of University of
Missouri's 13th annual conference on trace substances in environmental health]; June;
Columbia, MO. Columbia, MO: University of Missouri-Columbia; pp. 37-51.
Maher, C. C.; Roettgers, D. M.; Conlon, H. J. (1979) Interlaboratory comparison of blood lead
determinations. Am. Ind. Hyg. Assoc. J. 40: 230-237.
Manton, W. I.; Cook, J. D. (1979) Lead content of cerebrospinal fluid and other tissue in
amyotrophic lateral sclerosis (ALS). Neurology 29: 611-612.
Marcus, M.; Hollander, M.; Lucas, R. E.; Pfeiffer, N. C. (1975) Micro-scale blood lead deter-
minations in screening: evaluation of factors affecting results. Clin. Chem. (Winston
Salem N.C.) 21: 533-536.
Marcus, S. M. ; Joselow, M. M. ; Kemp, F. ; Ziering, R. ; Milhalovic, D.; Anderson, L. (1977)
Warning: spurious elevations of blood lead in micro puncture techniques [Letter]. J
Pediatr. (St. Louis) 91: 164.
SR013REF/E 9-42 7/1/83
-------
PRELIMINARY DRAFT
Marver, H. S.; Tschudy, D. P.; Perlroth, M. G.; Collins, A.; Hunter, G., Jr. (1966) The deter-
mination of aminoketones in biological fluids. Anal. Biochem. 14: 53-60.
Matson, W. R.; Roe, D. K. (1966) Trace metal analysis of natural media by anodic stripping
voltammetry. Anal. Instrum. 4: 19-22.
Matson, W. R.; Griffin, R. M.; Schreiber, G. B. (1970) Rapid sub-nanogram simultaneous
analysis of Zn, Cd, Pb, Cu, Bi and TI. In: Hemphill, D. D. , ed. Trace substances in en-
vironmental health - IV: [proceedings of University of Missouri's 4th annual conference
on trace substances in environmental health]; June; Columbia, MO. Columbia, MO:
University of Missouri-Columbia; pp. 396-406.
Mauzerall, D.; Gram'ck, S. (1956) The occurrence and determination of d-aminolevulinic acid
and porphobilinogen in urine. J. Biol. Chem. 219: 435-446.
Meranger, J. C.; Hollebone, B. R.; Blanchette, G. A.; (1981) The effects of storage times,
temperatures and container types on the accuracy of atomic absorption determinations of
Cd, Cu, Hg, Pb and Zn in whole heparinized blood. J. Anal. Toxicol. 5: 33-41.
Mitchell, D. G.; Ryan, F. J.; Aldous, K. M. (1972) The precise determination of lead in blood
by solvent extraction - atomic absorption spectrometry. At. Absorp. Newsl. 11: 120-121.
Mitchell, D. G.; Aldous, K. M.; Ryan, F. J. (1974) Mass screening for lead poisoning:
capillary blood sampling and automated Delves-cup atomic-absorption analysis. N.Y. State
J. Med. 74: 1599-1603.
Mitchell, R. A.; Drake, J. E.; Wittlin, L. A.; Rejent, T. A. (1977) Erythrocyte por-
phobilinogen synthase (delta-aminolaevulinate dehydratase) activity: a reliable and
quantitative indicator of lead exposure in humans. Clin. Chem. (Winston Salem N.C.) 23:
105-111.
Moore, M. R.; Meredith, P. A. (1977) The storage of samples for blood and water lead analysis.
Clin. Chim. Acta 75: 167-170.
Moore, M. R.; Campbell, B. C.; Meredith, P. A.; Beattie, A. D.; Goldberg, A.; Campbell, D.
(1978) The association between lead concentrations in teeth and domestic water lead
concentrations. Clin. Chim. Acta 87: 77-83.
Morrell, G.; Giridhar, G. (1976) Rapid micromethod for blood lead analysis by anodic stripping
voltammetry. Clin. Chem. (Winston Salem N.C.) 22: 221-223.
Murphy, T. J. (1976) The role of the analytical blank in accurate trace analysis. In: LaFleur,
P. D., ed. Accuracy in trace analysis: sampling, sample handling, analysis - volume 1.
Proceedings of the 7th materials research symposium; October 1974; Gaithersburg, MD.
Washington, DC: U.S. Department of Commerce, National Bureau of Standards; NBS special
publication no. 422; pp. 509-539.
Nackowski, S. B.; Putnam, R. D.; Robbins, D. A.; Varner, M. 0.; White, L. D.; Nelson, K. W.
(1977) Trace metal contamination of evacuated blood collection tubes. Am. Ind. Hyg.
Assoc. J. 38: 503-508.
National Academy of Sciences. (1972) Lead: airborne lead in perspective. Washington, DC:
National Academy of Sciences. (Biologic effects of atmospheric pollutants.)
SRD13REF/E 9-43 7/1/83
-------
PRELIMINARY DRAFT
Needleman, H. L. ; Davidson, I.; Sewell, E. M. ; Shapiro, I. M. (1974) Subclinical lead exposure
in Philadelphia school children: identification by dentine lead analysis. N. Engl j
Med. 290: 245-248. ' '
Needleman, H. L. ; Gunnoe, C. ; Leviton, A.; Reed, R.; Peresie, H.; Maher, C.; Barrett, P.
(1979) Deficits in psychological and classroom performance of children with elevated
dentine lead levels. N. Engl. J. Med. 300: 689-695.
Oehme, M. ; Lund, W. (1978) The determination of copper, lead, cadmium and zinc in human teeth
by anodic stripping voltametry. Anal. Chim. Acta 100: 389-398.
0'Flaherty, E. J. ; Hammond, P. B. ; Lerner, S. I.; Hanenson, I. B. ; Roda, S. M. B. (1980) The
renal handling of 6-aminolevulinic acid in the rat and in the human. Toxicol Appl
Pharmacol. 55: 423-432. '
Orfanos, A. P.; Murphey, W. H.; Guthrie, R. (1977) A simple fluorontetric assay of proto-
porphyrin in erythrocytes (EPP) as a screening test for lead poisoning. J. Lab. Clin
Med. 89: 659-665.
Paglia, D. E.; Valentine, W. N. (1975) Characteristics of a pyrimidine-specific
5'-nucleotidase in human erythrocytes. J. Biol. Chem. 250: 7973-7979.
Paglia, D. E.; Valentine, W. N.; Dahlgren, J. G. (1975) Effects of low-level lead exposure on
pyrimidine 5'-nucleotidase and other erythrocyte enzymes: possible role of pyrimidine
5'-nucleotidase in the pathogenesis of lead-induced anemia. J. Clin. Invest. 56-
1164-1169.
Patterson, C. C. (1980) An alternative perspective - lead pollution in the human environment:
origin, extent and significance. In: National Academy of Sciences, Committee on Lead in
the Human Environment. Lead in the human environment. Washington, DC: National Academy of
Sciences; pp. 265-350.
Patterson, C. C.; Settle, D. M. (1976) The reduction of orders of magnitude errors in lead
analyses of biological materials and natural waters by evaluating and controlling the
extent and sources of industrial lead contamination introduced during sample collecting
handling, and analyses. In: LaFleur, P. D., ed. Accuracy in trace analysis: sampling*
sample handling, and analysis - volume 1. Proceedings of the 7th materials research
symposium; October 1974; Gaithersburg, MD. Washington, DC: U.S. Department of Commerce
National Bureau of Standards; NBS special publication no. 422; pp. 321-352. '
Paulev, P-E.; Solgaard, P.; Tjell, J. C. (1978) Interlaboratory comparison of lead and cadmium
in blood, urine and aqueous solutions. Clin. Chem. (Winston Salem N.C.) 24: 1797-1800.
Pierce, J. 0.; Koirtyoharm, S. R.; Clevenger, T. E.; Lichte, F. E. (1976) The determination of
lead in blood: a review and critique of the state of the art. New York, NY: International
Lead Zinc Research Organization, Inc.
Piomelli, S. (1973) A micro method for free erythrocyte porphyrins: the FEP test. J. Lab
Clin. Med. 81: 932-940.
Piomelli, S.; Davidow, B. (1972) Free erythrocyte protoporphrin concentration: a promising
screening test for lead poisoning. Pediatr. Res. 6: 366.
SRD13REF/E 9-44 7/1/83
-------
PRELIMINARY DRAFT
Piomelli, S.; Graziano, J. (1980) Laboratory diagnosis of lead poisoning. Pediatr. Clin. North
Am. 27: 843-853.
Piomelli, S.; Corash, L.; Corash, M. B.; Seaman, C. ; Mushak, P.; Glover, B.; Padgett, R."
(1980) Blood lead concentrations in a remote Himalayan population. Science (Washington
D.C.) 210: 1135-1137.
Piscator, M. (1982) The importance of quality control for estimating dose-effect and dose-
response relationships. In: Schramel, P.; Bratter, P., eds. Trace element analytical
chemistry in medicine and biology. New York, NY: deGruyter and Co.; PAGES. (IN PRESS)
Rabinowitz, M. B.; Needleman, H. L. (1982) Temporal trends in the lead concentrations of
umbilical cord blood. Science (Washington D.C.) 216: 1429-1432.
Rabinowitz, M.; Wetherill, G. W.; Kopple, J. D. (1974) Studies of human lead metabolism by
using stable isotope tracers. Environ. Health Perspect. 7: 145-152.
Robinson, M. J.; Karpenski, F. E.; Brieger, H. (1958) The concentration of lead in plasma,
whole blood and erythroctyes of infants and children. Pediatrics 21: 793-796.
Roels, H.; Lauwerys, R.; Buchet, J-P.; Berlin, A.; Smeets, J. (1974) Comparison of four
methods for determination of 6-aminolevulinic acid in urine and evaluation of clinical
factors. Clin. Chem. (Winston Salem N.C.) 20: 753-760.
Rosen, J. F.; Zarate-Salvador, C. ; Trinidad, E. E. (1974) Plasma lead levels in normal and
lead-intoxicated children. J. Pediatr. (St. Louis) 84: 45-48.
Schlenker, F. S.; Taylor, N. A.; Kiehn, B. P. (1964) The chromatographic separation, deter-
mination, and daily excretion of urinary porphobilinogen, ami no acetone, and 6-amino-
levulinic acid. Am. J. Clin. Pathol. 42: 349-354.
Schwartz, S. ; Zieve, L.; Watson, C. J. (1951) An improved method for the determination of
urinary coproporphyrin and an evaluation of factors influencing the analysis. J. Lab.
Clin. Med. 37: 843-859.
Scoble, H. A.; McKeag, M.; Brown, P. R.; Kavarnos, G. J. (1981) The rapid determination of
erythrocyte porphyrins using reversed-phase high performance liquid chromatography. Clin.
Chim. Acta 113: 253-265.
Settle, D. M.; Patterson, C. C. (1980) Lead in albacore: guide to lead pollution in Americans.
Science (Washington D.C.) 207: 1167-1176.
Shapiro, I. M.; Dobkin, B.; Tuncay, 0. C.; Needleman, H. L. (1973) Lead levels in dentine and
circumpulpal dentine of deciduous teeth of normal and lead poisoned children. Clin. Chim.
Acta 46: 119-123.
Slavin, S.; Peterson, G. E.; Lindahl, P. C. (1975) Determination of heavy metals in meats by
atomic absorption spectrometry. At. Absorp. Newsl. 14: 57-59.
Smith, R. M.; Doran, D.; Mazur, M.; Bush, B. (1980) High performance liquid chromatographic
determination of protoporphyrin and zinc protoporphyrin in blood. J. Chromatogr. 181:
319-327.
SRD13REF/E 9-45 7/1/83
-------
PRELIMINARY DRAFT
Speecke, A.; Hosts, J. ; Versieck, J. (1976) Sampling of biological materials. In: LaFleur,
P. D., ed. Accuracy in trace analysis: sampling, sample handling, analysis - volume 1.'
Proceedings of the 7th materials research symposium; October 1974; Gaithersburg, MD.
Washington, DC: U.S. Department of Commerce, National Bureau of Standards; NBS special
publication no. 422; pp. 299-310.
Steenhout, A.; Pourtois, M. (1981) Lead accumulation in teeth as a function of age with dif-
ferent exposures. Br. J. Ind. Med. 38: 297-303.
Tomokuni, K. (1974) 6-aminolevulinic acid dehydratase test for lead exposure. Arch. Environ.
Health 29: 274-281.
Tomokuni, K.; Ogata, M. (1972) Simple method for determination of urinary 6-aminolevulinic
acid as an index of lead exposure. Clin. Chem. (Winston Salem N.C.) 18: 1534-1536.
Torrance, J.; West, C.; Beutler, E. (1977) A simple rapid radiometric assay for pyrimidine-
5'-nucleotidase. J. Lab. Clin. Med. 90: 563-568.
Trevisan, A.; Buzzo, A,; Scarpa, F. M. (1981) Studio comparative delle metodiche di deter-
minazione dell'attivita araino levulinico deidratasi eritrocitaria. [Comparative study of
methods for measurements of erythrocyte aminolevulinic acid dehydratase activity.] Med.
Lav. 72: 113-117.
U.S. Centers for Disease Control. (1981) Erythrocyte protoporphyrin proficiency testing: 1981
data summary. Atlanta, GA: U.S. Department of Health and Human Services, Centers for
Disease Control.
U.S. Occupational Safety and Health Administration. (1982) OSHA criteria for laboratory pro-
ficiency in blood lead analysis. Arch. Environ. Health 37: 58-60.
Unger, B. C.; Green. V. A. (1977) Blood lead analysis—lead loss to storage containers. Clin.
foxicol.' 11: 237-243.
Wigfield D C • Farant, J-P. (1979) Factors influencing the pH-activity relationship of
6-aminolevul'in1c acid dehydratase from human blood and their relevance to blood lead
assay. J. Anal. Toxicol. 3: 161-168.
Wittmers L E Jr.; Alich, A.; Aufderheide, A. C. (1981) Lead in bone. 1: Direct analysis
for' lead in milligram quantities of bone ash by graphite furnace atomic absorption
spectroscopy. Am. J. Clin. Pathol. 75: 80-85.
World Health Organization/United Nations Environmental Programme. (1977) Lead. Geneva,
Switzerland: World Health Organization. (Environmental health criteria 3.)
Zielhuis R. L. (1975) Dose-response relationships for inorganic lead. I: Biochemical and
haematological responses. Int. Arch. Occup. Environ. Health 35: 1-18.
SRD13REF/E 9-46 7/1/83
-------
PRELIMINARY DRAFT
10. METABOLISM OF LEAD
10.1 INTRODUCTION
The absorption, distribution, retention, and excretion of lead in humans and animals as
well as the various factors that mediate the extent of toxicokinetic processes are discussed
in this chapter. While inorganic lead is the form of the element that has been most heavily
studied, organolead compounds are also emitted into the environment and, as they are quite
toxic, they are also included in the discussion. Since the preparation of the 1977 Air
Quality Criteria Document for Lead (U.S. Environmental Protection Agency, 1977), a number of
reports have appeared that have proved particularly helpful in both quantifying the various
processes to be discussed in this chapter and assessing the interactive impact of factors such
as nutritional status in determining internal exposure risk.
10.2 LEAD ABSORPTION IN HUMANS AND ANIMALS
The amounts of lead entering the bloodstream from various routes of absorption are deter-
mined not only by the levels of the element in the particular media, but also by the various
physical and chemical parameters that characterize lead. Furthermore, specific host factors,
such as age and nutritional status, are important, as is interindividual variability. Addi-
tionally, in order to assess absorption rates, it is necessary to know whether or not the sub-
ject is in "equilibrium" with respect to a given level of lead exposure.
10.2.1 Respiratory Absorption of Lead
The movement of lead from ambient air to the bloodstream is a two-part process: a frac-
tion of air lead is deposited in the respiratory tract and, of this deposited amount, some
fraction is subsequently absorbed directly into the bloodstream or otherwise cleared from the
respiratory tract. At present, enough data exist to make some quantitative statements about
both of these components of respiratory absorption of lead.
The 1977 Air Quality Criteria Document for Lead described the model of the International
Radiological Protection Commission (IRPC) for the deposition and removal of lead from the
lungs and the upper respiratory tract (International Radiological Protection Commission,
1966). Briefly, the model predicts that 35 percent of lead inhaled from ambient air is depos-
ited in the airways, with most of this going to the lung. The IRPC model predicts a total de-
position of 40-50 percent for particles with an aerodynamic diameter of 0.5 urn and indicates
that the absorption rate would vary, depending on the solubility of the particular form.
NEW10A/A 10-1 7/1/83
-------
PRELIMINARY DRAFT
10.2.1.1 Human Studies. Table 10-1 tabulates the various studies of human subjects that pro-
vide data on the deposition of inorganic lead in the respiratory tract. Studies of this type
have involved diverse methodology to characterize the inhaled particles in terms of both size
(and size ranges) and fractional distribution. The use of radioisotopic or stable lead iso-
topes to directly or indirectly measure lead deposition and uptake into the bloodstream has
been particularly helpful in quantifying these processes.
From the studies of Kehoe (1961a,b,c) and their update by Gross (1981) as well as data
from Chamberlain et al. (1978), Morrow et al. (1980), and Nozaki (1966), it appears that the
respiratory deposition of airborne lead as encountered in the general population is approx-
imately 30-50 percent, depending on particle size and ventilation rates. Ventilation rate is
particularly important with submicron particles, where Brownian diffusion governs deposition,
since a slower breathing rate enhances the frequency of collisions of particles with the alve-
olar wall.
Figure 10-1 reproduces a composite figure of Chamberlain et al. (1978) that compares
data, both calculated and experimentally measured, on the relationship of percentage deposi-
tion to particle size. With increasing particle size, deposition rate decreases to a minimum
over the range where Brownian diffusion predominates, followed by an increase in deposition
with size (>0.5 urn MMAD) as impaction and sedimentation become the main deposition factors.
In contrast to the ambient air or chamber data tabulated in Table 10-1, higher deposition
rates in some occupational settings are associated with relatively large particles. However,
much of this deposition will be in the upper respiratory tract, with eventual movement to the
gastrointestinal tract by ciliary action and swallowing. Mehani et al. (1966) measured depo-
sition rates in battery workers and workers in marine scrap yards and observed total depositor
rates of 28-70 percent. Chamberlain and Heard (1981) calculated an absorption rate for parti-
cle sizes encountered in workplace air of appproximately 47 percent.
Systemic absorption of lead from the lower respiratory tract occurs directly, while much
of the absorption from the upper tract involves swallowing and some uptake in the gut. From
the radioactive isotope data of Chamberlain et al. (1978) and Morrow et al. (1980), and the
stable isotope studies of Rabinowitz et al. (1977), it can be concluded that lead deposited in
the lower respiratory tract is quantitatively absorbed.
Chamberlain et al. (1978) used 203Pb-labeled lead in engine exhaust, lead oxide, or lead
nitrate aerosols in experiments where human subjects inhaled the lead from a chamber through a
mouthpiece or in wind tunnel aerosols. By 14 days, approximately 90 percent of the label was
removed from the lung. Lead movement into the bloodstream could not be described by a simple
exponential function; 20 percent was absorbed within 1 hour and 70 percent within 10 hours.
NEW10A/A 10-2 7/1/83
-------
TABLE 10-1. DEPOSITION OF LEAD IN THE HUMAN RESPIRATORY TRACT
Form
Particle
size
Exposure
Percent
deposition
Reference
Pb203 aerosols
from engine.
exhaust
0.05 urn median
count diameter
in 38 studies;
5 subjects
exposed to average
of 0.9 urn
Lead "fumes" 0.05-1.0 pm mean
made in indue- diameter
tion furnace
203Pb-labeled
Pb203 aerosol
o
I
CO
Ambient air
lead near
motorway and
other urban
areas in U.K.
203Pb-labeled
Pb(OH)2 or
PbCl2 aero-
sols
Lead in work-
place air;
battery
factory and
shipbreaking
operations
Mean densities
of 0.02, 0.04,
0.09 urn
Mainly 0.1 urn
Both forms at
0.25 urn MMAD
Not determined;
defined as fumes,
fine dust, or
coarse dust
Chamber studies; 10, 20,
or 150 ug/m3; 3 hr on
alternate days;
12 subjects
Mouthpiece/aerosol chamber;
10 mg/m3; adult subjects
Mouthpiece/aerosol chamber;
adult subjects
2-10 ug/m3; adult subjects
50 liters air; 0.2 uCi/
liter; adult subjects
3 adult groups:
23 ug/m3 - controls
86 ug/m3 - battery workers
180 ug/m3 - scrap yard
30-70% (mean: 48%)
for mainly
0.05 urn particles
42% 0.05 urn;
63% 1.0 urn
80% 0.02 urn;
45% 0.04 urn;
30% 0.09 urn
60%, fresh exhaust;
50% other urban
area
23%, chloride;
26%, hydroxide
47%, battery workers;
39%, shipyard and
controls
Kehoe, 1961a,b,c;
Gross, 1981
Nozaki, 1966
Chamberlain et al.,
1978
Chamberlain et al.,
1978
Morrow et al. , 1980
Mehani, 1966
-------
PRELIMINARY DRAFT
z
3
O
tu
O
oc
UJ
0.
80
70
60
5 5°
O
Q.
UJ
° «o
30
20 -
10 -
0.01
1 ~
(T) MJPb DATA (VT = 1000 cm1)
(T) HEYDER 1975 (VT = 1000 cmj)
(3) MITCHELL 1977 (VT = 1000 cm3)
(T) JAMES 1978 (VT = 1000 cm1) CALCULATED
IT) JAMES 1978 (VT = 500 cm1) CALCULATED
YU 1977
-------
PRELIMINARY DRAFT
Rabinowitz et al. (1977) administered 204Pb tracer to young adult volunteers and were
able to determine by isotope tracer as well as balance data that 14 |jg of lead was absorbed by
these subjects daily at ambient air lead levels of 1-2 ug/m3. Assuming a daily ventilation
rate of 20 m3 a deposition rate of 50 percent of ambient air (Chamberlain et al., 1978), and a
mean air lead level of 1.5 ug/m3 (2.0 ug/m3 outside the study unit, 1.0 ug/m3 inside, as de-
termined by the authors), then 15 |jg lead was available for absorption. Hence, better than 90
percent of deposited lead was absorbed daily.
Morrow et al. (1980) followed the systemic uptake of 203Pb-labeled lead in 17 adult sub-
jects using either lead chloride or lead hydroxide aerosols with an average size of 0.25
(±0.1) um MMAD. Half of the deposited fraction of either aerosol was absorbed in 14 hours or
less. The radiolabel data described above are consistent with the results of Hursh and Mercer
(1970), who studied the systemic uptake of 212Pb on a carrier aerosol.
Given the apparent invariance of absorption rate for deposited lead in the above studies
as a function of chemical form of the element (Chamberlain et al., 1978; Morrow et al., 1980),
it seems that inhaled lead lodging deep in the respiratory tract is absorbed equally, regard-
less of form. Supporting evidence for total human systemic uptake of lead comes from autopsy
tissue analysis for lead content. Barry (1975) found that lead was not accumulated in the
lungs of lead workers. This may also be seen in the data of Gross et al. (1975) for non-occu-
pational ly exposed subjects.
All of the available data for lead deposition and uptake from the respiratory tract in
humans have been obtained with adults, and quantitative comparisons with the same exposures in
children are not possible. Although children 2 years of age weigh one-sixth as much as an
adult, they inhale 40 percent as much air lead as adults (Barltrop, 1972). James (1978) has
also taken into account differences in airway dimensions in adults vs. children, and has es-
timated that, often controlling for weight, the 10-year-old child has a deposition rate 1.6-
to 2.7-fold higher than the adult.
10.2.1.2 Animal Studies. Experimental animal data for quantitative assessment of lead de-
position and absorption for the lung and upper respiratory tract are limited. The available
information does, however, support the finding that respired lead is extensively and rapidly
absorbed.
Morgan and Holmes (1978) exposed adult rats, by nose-only technique, to a 203Pb-labeled
engine exhaust aerosol generated in the same manner as by Chamberlain et al. (1978) over a
period of 8 days. Exposure was at a level of 21.9-23.6 nCi label/liter chamber air. Ad-
justing for deposition on the animal pelt, 20-25 percent of the label was deposited in the
lungs. Deposited lead was extensively taken up in blood: 50 percent within 1 hour and, 98
percent within 7 days. The absorption rate kinetic profile was similar to that reported for
humans (Chamberlain et al., 1978).
NEW10A/A 10-5 7/1/83
-------
PRELIMINARY DRAFT
Boudene et al. (1977) exposed rats to 210Pb-labeled aerosols at a level of 1 |jg label/m3
and 10 MO/1"3. tne majority of the particles being 0.1-0.5 urn in size. At 1 hour, 30 percent
of the label had left the lung; by 48 hours 90 percent was gone.
Bianco et al. (1974) used 212Pb aerosol (^0.2 urn) inhaled briefly by dogs and found a
clearance half-time from the lung of approximately 14 hours. Greenhalgh et al. (1979) found
that direct instillation of 203Pb~labeled lead nitrate solution into the lungs of rats led to
an uptake of approximately 42 percent within 30 minutes, compared with an uptake rate of 15
percent within 15 minutes in the rabbit. These instillation data are consistent with the
report of Pott and Brockhaus (1971), who noted that intratracheal instillation of lead in
solution (as bromide) or suspension (as oxide) serially over 8 days resulted in systemic lead
levels in tissues indistinguishable from injected lead. Rendall et al. (1975) found that the
movement of lead into blood of baboons inhaling a lead oxide (Pb304) was more rapid and
resulted in higher levels when coarse (1.6 urn mean diameter) rather than fine (0.8 urn mean
diameter) particles were used. This suggests that considerable fractions of both size parti-
cles were eventually lodged in the gut, where absorption of lead tends to be higher in baboons
than in other animal species (Pounds et al., 1978). In addition, the larger particles appear
to move more rapidly to the gut.
10.2.2 Gastrointestinal Absorption of Lead
Gastrointestinal absorption of lead mainly involves uptake from food and beverages as
well as lead deposited in the upper respiratory tract and eventually swallowed. It also in-
cludes ingestion of non-food material, primarily in children via normal mouthing activity and
pica. Two issues of concern with lead uptake from the gut are the comparative rates of such
absorption in developing vs. adult organisms, including humans, and how the bioavailability of
lead affects such uptake.
10.2.2.1 Human Studies. Based on the long-term metabolic studies with adult volunteers,
Kehoe (1961a,b,c) estimated that approximately 10 percent of dietary lead is absorbed from the
gut of humans. According to Gross (1981), there can be considerable variation of various
balance parameters among subjects. These studies did not take into account the contribution
of biliary clearance of lead into the gut, which would have affected measurements for both
absorption and total excretion. Chamberlain et al. (1978) also determined that the level of
endogenous fecal lead is approximately 50 percent of urinary lead values. Chamberlain et al.
(1978) have estimated that 15 percent of dietary lead is absorbed, if the amount of endogenous
fecal lead is taken into account.
NEW10A/A 10-6 7/1/83
-------
PRELIMINARY DRAFT
Following the Kehoe studies, a number of reports determined gastrointestinal (GI) absorp-
tion using both stable and radioisotopic labeling of dietary lead. Generally, these reports
support the observation that in the adult human there is limited absorption of lead when taken
with food. Harrison et al. (1969) determined a mean absorption rate of 14 percent for three
adult subjects ingesting 203Pb-labeled lead in diet, a figure in accord with the results of
Hursh and Suomela (1968). Chamberlain et al. (1978) studied the absorption of 203Pb in two
forms (as the chloride and as the sulfide) taken with food. The corresponding absorption
rates were 6 percent (sulfide) and 7 percent (chloride), taking into account endogenous fecal
excretion. Using adult subjects who ingested the stable isotope 204Pb in their diet,
Rabinowitz et al. (1974) reported an average gut absorption of 7.7 percent. In a later study,
Rabinowitz et al. (1980) measured an absorption rate of 10.3 percent.
A number of recent studies indicate that lead ingested under fasting conditions is absor-
bed to a much greater extent than when it is taken with or incorporated into food. For exam-
ple, Blake (1976) measured a mean absorption rate of 21 percent when 11 adult subjects in-
gested 203Pb-labeled lead chloride several hours after breakfast. Chamberlain et al. (1978)
found that lead uptake in six subjects fed 203Pb as the chloride was 45 percent after a fast-
ing period, compared to 6 percent with food. Heard and Chamberlain (1982) obtained a rate of
63.3 percent using a similar procedure with eight subjects. Rabinowitz et al. (1980) reported
an absorption rate of 35 percent in five subjects when 204Pb was ingested after 16 hours of
fasting. To the extent that lead in beverages is ingested between meals, these isotope
studies support the observations of Barltrop (1975) and Garber and Wei (1974) that beverage
•lead is absorbed to a greater extent than is lead in food.
The relationship of lead bioavailability in the human gut to the chemical/biochemical
form of lead can be determined from available data, although interpretation is complicated by
the relatively small amounts given and the presence of various components of food already pre-
sent in the gut. Harrison et al. (1969) found no difference in lead absorption from the human
gut when lead isotope was given either as the chloride or incorporated into alginate. Cham-
berlain et al. (1978) found that labeled lead as the chloride or sulfide was absorbed to the
same extent when given with food, while the sulfide form was absorbed at a rate of 12 percent
compared with 45 percent for the chloride when given under fasting conditions. Rabinowitz et
al. (1980) obtained similar absorption rates for the chloride, sulfide, or cysteine complex
forms when administered with food or under fasting conditions. Heard and Chamberlain (1982)
found no difference in absorption rate when isotopic lead (203Pb) was given with unlabeled
liver and kidney or when the label was first incorporated into these organs.
Three studies have focused on the question of differences in gastrointestinal absorption
rates between adults and children. Alexander et al. (1973) carried out 11 balance studies
with 8 children, aged 3 months to 8 years. Intake averaged 10.6 (jg Pb/kg body weight/day
NEW10A/A 10-7 7/1/83
-------
PRELIMINARY DRAFT
(range 5-17); the mean absorption rate determined from metabolic balance studies was 53 per-
cent. Ziegler et al. (1978) carried out a total of 89 metabolic balance studies with 12 nor-
mal infants aged 2 weeks to 2 years. Diets were closely controlled and lead content was
measured. Two discrete studies were carried out and in the first, 51 balance studies using 9
children furnished a mean absorption rate of 42.7 percent. In the second study, 6 children
were involved in 38 balance studies involving dietary lead intake at 3 levels. For all daily
intakes of 5 ug Pb/kg/day or higher, the mean absorption rate was 42 percent. At low levels
of lead intake data were variable, with some children apparently in negative balance, probably
due to the difficulty in controlling low lead intake.
In contrast to these studies, Barltrop and Strehlow (1978) found that with children hos-
pitalized as orthopedic or "social" admissions, the results were highly variable. A total of
104 balance studies were carried out in 29 children ranging in age from 3 weeks to 14 years.
Fifteen of the subjects were in net negative balance, with an average dietary absorption of
-40 percent and, when weighted by number of balance studies, -16 percent.
It is difficult to closely compare these data with those of Ziegler et al. (1978). Sub-
jects were inpatients, represented a much greater age range, and were not classified in terms
of mineral nutrition or weight change status. As an urban pediatric group, the children in
this study may have had higher prior lead exposure so that the "washout" phenomenon (Kehoe,
1961a,b,c; Gross, 1981) may have contributed to the highly variable results. The calculated
mean daily lead intake in the Barltrop and Strehlow group (6.5 ug/kg/day) was lower than those
for all but one study group described by Ziegler et al. (1978). In the latter study it ap-
pears that data for absorption became more variable as the daily lead intake was lowered.
Finally, in those children classified as orthopedic admissions, it is not clear that skeletal
trauma was without effect on lead equilibrium between bone and other body compartments.
As typified by the results of the NHANES II survey (Mahaffey et al., 1979), children at
2-3 years of age show a small peak in blood lead during childhood. The question arises
whether this peak indicates an intrinsic biological factor, such as increased absorption or
retention when compared with older children, or whether this age group is exposed to lead in
some special way. Several studies are relevant to the question. Zielhuis et al. (1978) re-
ported data for blood lead levels in 48 hospitalized Dutch children ranging in age from 2
months to 6 years. Children up to 3 years old had a mean blood lead level of 11.9 ug/dl vs. a
level of 15.5 in children aged 4-6 years. A significant positive relationship between child
age and blood lead was calculated (r = 0.44, p <0.05). In the Danish survey by Nygaard et al.
(1977), a subset of 126 children representing various geographical areas and age groups
yielded the following blood lead values by mean age group: children (N = 8) with a mean age of
1.8 years had a mean blood lead of 4.3 ug/dl; those with a mean age of 3.7-3.9 had values
ranging from 5.6 to 8.3 ug/dl children 4.6-4.8 years of age had a range of 9.2 to 10 ug/dl.
NEW10A/A 10-8 7/1/83
-------
PRELIMINARY DRAFT
These authors note that the youngest group was kept at a nursery whereas the older kindergar-
ten children had more interaction with the outside environment. Sartor and Rondia (1981) sur-
veyed two population groups in Belgium, one of which consisted of groups of children aged 1-4,
5-8, and 9-14 years. Children under the age of 1 had a mean blood lead of 10.7 ug/dl; the 1-4
and 5-8 age groups were comparable, 13.9 and 13.7 pg/dl respectively; and those 9-14 years old
had a blood lead of 17.2 ug/dl. In this study, all of the children were hospital patients.
While these European studies suggest that any significant restriction of children in terms of
environmental interaction, e.g., hospitalization or nurseries, is associated with an apparen-
tly different age-blood lead relationship than the U.S. NHANES II subjects, it remains to be
demonstrated that European children in the 2-3 year age group show a similar peak. The issue
merits further study.
The normal mouthing activity of young children, as well as the actual ingestion of non-
food items, i.e., pica, is a major concern in pediatric lead exposure, particularly in urban
areas with deteriorating housing stock and high automobile density and in non-urban areas
contiguous to lead production facilities. The magnitude of such potential exposures is dis-
cussed in Chapter 7, while an integrated assessment of impact on human intake appears in
Chapter 13. Such intake is intensified for children with pica and would include paint, dust
and dirt.
Drill et al. (1979), using data from Day et al. (1975) and Lepow et al. (1974), have at-
tempted to quantify the daily intake of soil/dust in young children from such mouthing acti-
vities as thumb sucking and finger licking. A total of 100 mg/day was obtained for children
2-3 years old, with the amount of lead in this ingested quantity varying considerably from
site to site. In the report, a gastrointestinal absorption rate of 30 percent was taken for
lead in soil and dust. Of relevance to this estimate of absorption rate in children are the
animal data discussed in the next section, which show that lead of variable chemical form in
soil or dust is as available for absorption as food lead. The j_n vitro studies relating lead
solubility in street dusts with acidity clearly demonstrate that the acidity of the human
stomach is adequate to extensively solubilize lead assimilated from soil and dust. To the
extent that ingestion of such material by children occurs other than at mealtime, the fasting
factor in enhancing lead absorption from the human gastrointestinal tract (vides supra) must
also be considered. Hence, a factor of 30 percent for lead absorption from dusts and soils is
not an unreasonable value.
Paint chip ingestion by children with pica has been estimated in the NAS report on lead
poisoning in children to be considerable (National Academy of Sciences, 1976). In the case of
paint chips, Drill et al. (1979) estimated an absorption rate as high as 17 percent. This
value may be compared with the animal data in Section 10.2.2.2 which indicate that lead in old
paint films can undergo significant absorption in animals.
NEW10A/A 10-9 7/1/83
-------
PRELIMINARY DRAFT
10.2.2.2 Animal Studies. Lead absorption via the gut of various adult experimental animal
species appears to resemble that for the adult human, on the order of 1-15 percent in most
cases. Kostial and Kello (1979), Kostial et al. (1978), and Kostial et al. (1971) reported a
value of 1 percent or less in adult rats maintained on commercial rat chow. These studies
were carried out using radioisotopic tracers. Similarly, Barltrop and Meek (1975) reported an
absorption rate of 4 percent in control diets, while Aungst et al. (1981) found the value to
range from 0.9 to 6.9 percent, depending on the level of lead given in the diet. In these rat
studies, lead was given with food. Quarterman and Morrison (1978) administered 203Pb label in
small amounts of food to adult rats and found an uptake rate of approximately 2 percent at 4
months of age. Pounds et al. (1978) obtained a value of 26.4 percent with four adult Rhesus
monkeys given 210Pb by gastric intubation. The higher rate, relative to the rat, may reflect
various states of fasting at time of intubation or differences in dietary composition (vide
infra), two factors that affect rates of absorption.
As seen above with human subjects, fasting appears to enhance the rate of lead uptake in
experimental animals. Garber and Wei (1974) found that fasting markedly enhanced gut uptake
of lead in rats. Forbes and Reina (1972) found that lead dosing by gastric intubation of rats
yielded an absorption rate of 16 percent, which is higher than other data for the rat. It is
likely that intubation was done when there was little food in the gut. The data of Pounds et
al. (1978), as described above, may also suggest a problem with giving lead by gastric intuba-
tion or with water as opposed to mixing it with food.
The bioavailability of lead in the gastrointestinal tract of experimental animals has
been the subject of a number of reports. The designs of these studies differed in accordance
with how "bioavailability" is defined by different investigators. In some cases, the dietary
matrix was kept constant, or nearly so, while the chemical or physical form of the lead was
varied. By contrast, other data described the effect of changes in bioavailability as the
basic diet matrix was changed. The latter case is complicated by the simultaneous operation
of lead-nutrient interactive relationships, which are described in Section 10.5.2 within this
chapter.
Allcroft (1950) observed comparable effects when calves were fed lead in the form of the
phosphate, oxide, or basic carbonate (PbC03-Pb(OH)2), or incorporated into wet or dry paint.
By contrast, lead sulfide in the form of finely ground galena ore was less toxic. Criteria
for relative effect included kidney and blood lead levels and survival rate over time.
In the rat, Barltrop and Meek (1975) carried out a comparative absorption study using
lead in the form of the acetate as the reference substance. The carbonate and thallate were
absorbed to the greatest extent, while absorption of the sulfide, chromate, napthenate, and
octoate was 44-67 percent of the reference agent. Gage and Litchfield (1968, 1969) found
that lead napthenate and chromate can undergo considerable absorption from the rat gut when
NEW10A/A 10-10 7/1/83
-------
PRELIMINARY DRAFT
incorporated into dried paint films, although less than when given with other vehicles. Ku et
al. (1978) found that lead in the form of the acetate or as a phospholipid complex was equally
absorbed from the GI tract of both adult and young rats at a level of 300 ppm. Uptake was
assessed by weight change, tissue levels of lead, and urinary aminolevulinic acid levels.
In a study relevant to the problem of lead bioavailability in soils and dusts, particu-
larly in exposed children, Dacre and Ter Haar (1977) compared the effects of lead as acetate
with lead contained in roadside soil and in house paint soil, at a level of approximately 50
ppm, in commercial rat chow. Uptake of lead was indexed by weight change, tissue lead con-
tent, and inhibition of ALA-D activity. There was no significant difference in any of these
parameters across the three groups, suggesting that neither the geochemical matrix in the
soils or the various chemical forms—basic carbonate in paint soil, and the oxide, carbonate,
and basic carbonate in roadside soil — affect lead uptake.
These data are consistent with the behavior of lead in dusts upon acid extraction as re-
ported by Day et al. (1979), Harrison (1979), and Duggan and Williams (1977). In the Day et
al. study, street dust samples from England and New Zealand were extracted with hydrochloric
acid over the pH range of 0-5. At an acidity that may be equalled by gastric secretions,
i.e., pH of 1, approximately 90 percent of the dust lead was solubilized. Harrison (1979)
noted that at this same acidity, up to 77 percent of Lancaster, England, street dust lead was
soluble, while an average 60 percent solubility was seen in London dust samples (Duggan and
Williams, 1977). Because gastric solubilization must occur for lead in these media to be ab-
sorbed, the above data are useful in determining relative risk.
Kostial and Kello (1979) compared the absorption of 203Pb from the gut of rats maintained
on commercial rat chow vs. rats fed such "human" diets as baby foods, porcine liver, bread,
and cow's milk. Absorption in the latter cases varied from 3 to 20 percent, compared with
<1.0 percent with rat chow. This range of uptake for the non-chow diet compares closely with
that reported for human subjects (vide supra). Similarly, Jugo et al. (1975a) observed that
rats maintained on fruit diets had an absorption rate of 18-20 percent. It would appear,
then, that the generally observed lower absorption of lead in the adult rat vs. the adult
human is less reflective of a species difference than of a dietary difference.
Barltrop and Meek (1979) studied the relationship of particle size of lead in two
forms--as the metal or as lead octoate or chromate in powdered paint films—to the amount of
gut absorption in the rat and found that there was an inverse relationship between uptake and
particle size for both forms.
A number of studies have documented that the developing animal absorbs a relatively
greater fraction of ingested lead than does the adult, thus supporting those studies that have
shown this age dependency in humans. For example, the adult rat absorbs approximately 1 per-
cent lead or less when contained in diet vs. a corresponding value 40-50 times greater in the
NEW10A/A 10-11 7/1/83
-------
PRELIMINARY DRAFT
rat pup (Kostial et al., 1971, 1978;. Forbes and Reina, 1972). In the rat, this difference
persists through weaning (Forbes and Reina, 1972), at which point uptake resembles that of
adults. Part of this difference can be ascribed to the nature of the diet (mother's milk vs.
regular diet), although it should be noted that the extent of absorption enhancement with milk
vs. rat chow in the adult rat (Kello and Kostial, 1973) falls short of what is seen in the
neonate. An undeveloped, less selective intestinal barrier may also exist in the rat neonate.
In non-human primates, Munro et al. (1975) observed that infant monkeys absorbed 65-85 percent
via the gut vs. 4 percent in adults. Similarly, Pounds et al. (1978) noted that juvenile
Rhesus monkeys absorbed approximately 50 percent more lead than adults.
The question of the relationship of level of lead intake through the GI tract and rate of
lead absorption was addressed by Aungst et al. (1981), who exposed adult and suckling rats to
doses of lead by intubation over the range 1-100 mg Pb/kg or by variable concentrations in
drinking water. With both age groups and both forms of oral exposure, lead absorption as a
percentage of dose decreased, suggesting a saturation phenomenon for lead transport across the
gut wall.
10.2.3 Percutaneous Absorption of Lead
Absorption of inorganic lead compounds through the skin appears to be considerably less
significant than the respiratory and gastrointestinal routes of uptake. This is in contrast
to the .observations for lead alkyls and other organic derivatives (U.S. Environmental
Protection Agency, 1977). Uptake of alkyl lead through the skin is discussed in Section 10.7.
Rastogi and Clausen (1976) found that cutaneous or subcutaneous administration of lead
napthenate in rat skin was associated with higher tissue levels and more severe toxic effects
than was the case for lead acetate. Laug and Kunze (1948) applied lead as the acetate, ortho-
arsenate, oleate, and ethyl lead to rat skin and determined that the greatest levels of kidney
lead were associated with the alkyl contact.
Moore et al. (1980) studied the percutaneous absorption of 203Pb-labeled lead acetate in
cosmetic preparations using eight adult volunteers. Applied in wet or dry forms, absorption
was indexed by blood, urine, and whole body counting. Absorption rates ranged from 0 to 0.3
percent, with the highest values obtained when the application sites were scratched. These
researchers estimated that the normal use of such preparations would result in an absorption
of approximately 0.06 percent.
10.2.4 Transplacental Transfer of Lead
Lead uptake by the human and animal fetus occurs readily, based on such indices as fetal
tissue lead measurements and, in the human, cord blood lead levels. Barltrop (1969) and
Horiuchi et al. (1959) demonstrated by fetal tissue analysis that placental transfer in the
NEW10A/A 10-12 7/1/83
-------
PRELIMINARY DRAFT
human occurs by the 12th week of gestation, with increasing fetal lead uptake throughout deve-
lopment. Highest levels occur in bone, kidney, and liver, followed by blood, brain, and
heart. Cord blood contains significant amounts of lead, generally correlating with maternal
blood values and being slightly but significantly lower than mothers' in concentration
(Scanlon, 1971; Harris and Holley, 1972; Gershanik et al., 1974; Buchet et al., 1978;
Alexander and Delves, 1981; Rabinowitz and Needleman, 1982).
A cross-sectional study of maternal blood lead carried out by Alexander and Delves (1981)
showed that a significant decrease in maternal blood lead occurs throughout pregnancy, a de-
crease greater than the dilution efffect of the concurrent increase in plasma volume. Hence,
during pregnancy there is either an increasing deposition of lead in placental or fetal tissue
or an increased loss of body lead via other routes. Increasing absorption by the fetus during
gestation, as demonstrated by Barltrop (1969), suggests that the former explanation is a
likely one. Hunter (1978) found that summer-born children showed a trend to higher blood lead
than those born in the spring, suggesting increased fetal uptake in the summer due to in-
creases in circulating maternal lead. This observation was confirmed in the report of
Rabinowitz and Needleman (1982). Ryu et al. (1978) and Singh et al. (1978) both reported that
infants born to women having a history of lead exposure had significantly elevated blood lead
values at
10.3 DISTRIBUTION OF LEAD IN HUMANS AND ANIMALS
A quantitative understanding of the sequence of changes in levels of lead in various body
pools and tissues is essential in interpreting measured levels of lead with respect to past
exposure as well as present and future risks of toxicity. This section discusses the dis-
tribution kinetics of lead in various portions of the body—blood, soft tissues, calcified
tissues, and the "chelatable" or toxicologically active body burden—as a function of such
parameters as exposure history and age.
A given quantity of lead taken up from the GI tract or the respiratory tract into the
bloodstream is initially distributed according to the rate of delivery by blood to the various
organs and systems. Lead is then redistributed to organs and systems in proportion to their
respective affinities for the element. With consistent exposure for an extended period, a
near steady-state of intercompartmental distribution is achieved.
Fluctuations in the near steady-state will occur whenever short-term lead exposures are
guperimposed on a long-term uptake pattern. Furthermore, the steady-state description is im-
perfect because on a very short (hourly) time scale, intake is not constant. Lead intake with
meals and changes in ambient air lead—outside to inside and vice versa—will cause quick
changes in exposure levels which may be viewed as short-term alterations in the small, labile
NEW10A/A 10-13 7/1/83
-------
PRELIMINARY DRAFT
lead pool. Metabolic stress could remobilize and redistribute body stores, although documen-
tation of the extent to which this happens is very limited (Chisolm and Harrison, 1956).
10.3.1 Lead in Blood
Viewed from different time scales, lead in whole blood may be seen as residing in several
distinct, interconnected pools. More than 99 percent of blood lead is associated with the
erythrocytes (DeSilva, 1981; Everson and Patterson, 1980; Manton and Cook, 1979) under typical
conditions, but it is the very small fraction of lead transported in plasma and extracellular
fluid that provides lead to the various body organs (Baloh, 1974).
Most of the erythrocyte lead is bound within the cell, although toxicity of the element
to the erythrocyte (Raghavan et al., 1981) is mainly associated with membrane lead content.
Within erythrocytes from non-exposed subjects, lead is primarily bound to hemoglobin, in par-
ticular HbA2, which binds approximately 50 percent of cell lead although it comprises only 1-2
percent of total hemoglobin (Bruenger et al., 1973). A further 5 percent is bound to a
10,000-dalton molecular weight fraction, about 20 percent to a much heavier molecule, and
about 25 percent is considered "free" or bound to lower weight molecules (Ong and Lee, 1980a;
Raghavan and Gonick, 1977). Raghavan et al. (1980) have observed that, among workers exposed
to lead, those who develop signs of toxicity at relatively low blood lead levels seem to have
a diminished binding of intracellular lead with the 10,000-dalton fraction, suggesting an im-
paired biosynthesis of a protective species. According to Ong and Lee (1980b), fetal hemo-
globin has a higher affinity for lead than adult hemoglobin. Whole blood lead in daily equi-
librium with other compartments was found to have a mean life of 35 days (25-day half-life)
and a total content of 1.9 mg, based on studies with a small number of subjects (Rabinowitz et
al., 1976). Chamberlain et al. (1978) established a similar half-time for 203Pb in blood when
volunteers were given the label by ingestion, inhalation, or injection. The inhaled lead
studies in adults, described by Griffin et al. (1975), permit calculation of half-times of 28
and 26 days for inhalation of 10.4 and 3.1 ug Pb/m3 respectively.
Alterations in blood lead levels in response to abrupt changes in exposure apparently oc-
cur over somewhat different periods, depending on whether the direction of change is greater
or smaller. With increased lead intake, blood lead achieves a new value in approximately 60
days (Griffin et al., 1975; Tola et al., 1973), while a decrease may involve a longer period
of time, depending on the magnitude of the past higher exposure (O'Flaherty et al., 1982;
Rabinowitz et al. 1977; Gross, 1981). With age, there appears to be a modest increase in
blood lead, Awad et al. (1981) reporting an increase of 1 ug for each 14 years of age. In the
latter case, particularly with occupational exposure, it appears that the time for re-estab-
lishing near steady-state is more dependent upon the extent of lead resorption from bone and
the total quantity deposited, extending the "washout" interval.
NEW10A/A 10-14 7/1/83
-------
PRELIMINARY DRAFT
Lead levels in newborn children are similar to but somewhat lower than those of their
mothers: 8.3 vs. 10.4 pg/dl (Buchet et al., 1978) and 11.0 vs. 12.4 (jg/dl (Alexander and
Delves, 1981). Alexander and Delves (1981) also reported that maternal blood lead levels de-
crease throughout pregnancy, such decreases being greater than the expected dilution via the
concurrent increase in plasma volume. - These data are consistent with increasing fetal uptake
during gestation (Barltrop, 1969). Increased tissue retention may also be a factor.
Levels of lead in blood are sex-related, adult women invariably showing lower levels than
adult males (e.g., Mahaffey et al., 1979). Of interest in this regard is the study of Stuik
(1974) showing lower blood lead response in women than in men for an equivalent level of lead
intake.
The small but biologically significant lead pool in blood plasma has proven technically
difficult to measure reliable values have become available only recently, and (see Chapter 9).
Chamberlain et al. (1978) found that injected 203Pb was removed from plasma (and, by infer-
ence, extracellular fluid) with a half-life of less than 1 hour. These data support the ob-
servation of DeSilva (1981) that lead is rapidly cleared from plasma. Ong and Lee (1980a), in
their in vitro studies, found that 203PB is virtually all bound to albumin and that only trace
amounts are bound to high weight globulins. It is not possible to state which binding form
constitutes an "active" fraction for movement to tissues.
Although Rosen et al. (1974) reported that plasma lead was invariant across a range of
whole blood levels, the findings of Everson and Patterson (1980), DeSilva (1981), and
Cavalleri et al. (1978) indicate that there is an equilibrium between red cell and plasma,
such that levels in plasma rise with levels in whole blood. This is consistent with the data
of Clarkson and Kench (1958) who found that lead in the red cell is relatively labile to ex-
change and a logical prerequisite for a dose-effect relationship in various organs. Ong and
Lee (I980c), furthermore, found that plasma calcium is capable of displacing RBC membrane
lead, suggesting that plasma calcium is a factor in the cell-plasma lead equilibrium.
10.3.2 Lead Levels in Tissues
Of necessity, various relationships of tissue lead to exposure and toxicity in humans
generally must be obtained from autopsy samples, although in some studies biopsy data have
been described. There is, then, the inherent question of how such samples adequately repre-
sent lead behavior in the living population, particularly in cases where death was preceded by
prolonged illness or disease states. Also, victims of fatal accidents are not well character-
ized as to exposure status, and are usually described as having no "known" lead exposures.
Finally* these studies are necessarily cross-sectional in design, and in the case of body
accumulation of lead it is assumed that different age groups have been similarly exposed.
Some important aspects of the available data include the distribution of lead between soft and
NEW10A/A 10-15 7/1/83
-------
PRELIMINARY DRAFT
calcifying tissue, the effect of age and development on lead content of soft and mineral tis-
sue, and the relationship between total and "active" lead burdens in the body.
10.3.2.1 Soft Tissues. In humans after age 20 most soft tissues do not show age-related
changes in lead levels, in contrast to the case with bone (Barry and Mossman, 1970; Barry
1975, 1981; Schroeder and Tipton, 1968; Butt et al., 1964). Kidney cortex also shows in-
creases in lead with age that may be associated with formation of lead nuclear inclusion
bodies (Indraprasit et al., 1974). Based on these rates of accumulation, the total body bur-
den may be divided into pools that behave differently: the largest and kinetically slowest
pool is the skeleton, which accumulates lead with age; and the much more labile lead pool is
in soft tissue.
Soft tissue levels generally stabilize in early adult life and show a turnover rate
similar to blood, sufficient to prevent accumulation except in the renal cortex, which may be
reflecting formation of lead-containing nuclear inclusion bodies (Cramer et al., 1974-
Indraprasit et al., 1974). The data of Gross et al. (1975) and Barry (1975) indicate that
aortic levels appear to rise with age, although this may reflect entrapment of lead in athero-
sclerotic deposits. Biliary and pancreatic secretions, while presumably reflecting some of
the organ levels, have tracer lead concentrations distinct from either blood or bone pools
(Rabinowitz et al., 1973).
For levels of lead in soft tissue, the reports of Barry (1975, 1981), Gross et al. (1975)
and Horiuchi et al. (1959) indicate that soft tissue lead content generally is below 0.5 ug/g
wet weight, with higher values for aorta and kidney cortex. The higher values in aorta may or
may not reflect lead in plaque deposits, while higher kidney levels may be associated with the
presence of lead-accumulating tubular cell nuclear inclusions. The relatively constant lead
concentration in lung tissue across age groups suggests no accumulation of respired lead and
is consistent with data for deposition and absorption (see Section 10.2.2). Brain tissue was
generally under 0.2 ppm wet weight and appeared to show no change with increasing age. Since
these data were collected by cross-sectional study, age-related changes in the low levels of
lead in brain would have been difficult to discern. Barry (1975) found that tissues in a
small group of samples from subjects with known or suspected occupational exposure showed
higher lead levels in aorta, liver, brain, skin, pancreas, and prostate.
Levels of lead in whole brain are less illuminating to the issue of sensitivity of cer-
tain regions within the organ to toxic effects of lead than is regional analysis. The distri-
bution of lead across brain regions has been reported from various laboratories and the
relevant data for humans and animals are set forth in Table 10-2. The data of Grandjean
(1978) and Niklowitz and Mandybur (1975) for human adults, and those of Okazaki et al. (1963)
for autopsy samples from young children who died of lead poisoning, are consistent in showing
that lead is selectively accumulated in the hippocampus. The correlation of lead level with
NEW10A/A 10-16 7/1/83
-------
TABLE 10-2. REGIONAL DISTRIBUTION OF LEAD IN HUMANS AND ANIMALS
Species
Exposure status
Relative distribution
Reference
Humans
Adult Males
Children
Child, 2 yrs. old
Adults
Animals
Adult rats
Adult rats
Unexposed
Fatal lead poisoning
Fatal lead poisoning
3 subjects unexposed;
1 subject with lead
poisoning as child
Unexposed
Unexposed
Hippocampus = amygdala > medulla
oblongata > half brain > optic
tract £ corpus callosum. Pb
correlated with K.
Hippocampus > frontal cortex »
occipital white matter, pons
Cortical gray matter > basal
gangli > cortical white matter
Hippocampus > cerebellum s temporal
lobes > frontal cortex in 3
unexposed subjects; temporal
lobes > frontal cortex >
hippocampus > cerebellum > in
case with prior exposure
Hippocampus
brain
> amygdala » whole
Hippocampus had 50 percent of
brain lead with a 4:1 ratio
of hippocampus.-whole brain
Grandjean, 1978
Okazaki et al., 1963
Klein et al., 1970
Niklowitz and
Mandybur, 1975
-f.
O
Danscher et al., 1975
Fjerdingstad et al. ,
1974
-------
TABLE 10-2 (continued)
Species
Exposure status
Relative distribution
Reference
O
I
00
Neonatal rats
Young dogs
Controls and
daily i.p. injection,
5.0 or 7.5 mg/kg
Controls and dietary
exposure, 100 ppm;
12 weeks of exposure
In both treated and control
animals: cerebellum > cerebral
cortex > brainstem + hippocampus
Klein and Koch, 1981
Controls: cerebellum = medulla >
caudate > occipital gray > frontal
gray
Exposed: occipital gray > frontal
gray = caudate > occipital
white = thalamus > medulla > cerebellum
Stowe et al. , 1973
•XI
-<
O
-------
PRELIMINARY DRAFT
potassium level suggests that uptake of lead is greater in cellulated areas. The involvement
of the cerebellum in lead encephalopathy in children (see Section 12.4) and in adult intoxica-
tion from occupational exposure indicates that the sensitivity of various brain regions to
lead as well as their relative uptake characteristics are factors in lead neuropathology.
In adult rats, selective uptake of lead is shown by the hippocampus (Fjerdingstad et al.,
1974; Danscher et al. , 1975) and the amygdala (Danscher et al., 1975). By contrast, lead-
exposed neonate rats show greatest uptake of lead into cerebellum, followed by cerebral cor-
tex, then brainstem plus hippocampus. Hence, there is a developmental difference in lead dis-
tribution in the rat with or without increased lead exposure (Klein and Koch, 1981).
In studies of young dogs, unexposed animals showed highest levels in the cerebellum,
while lead exposure was associated with selective uptake into gray matter; cerebellar levels
were relatively low. Unlike the young rat, then, the distribution of lead in brain regions of
dogs appears to be dose-dependent (Stowe et al., 1973).
Barry (1975, 1981) compared lead levels in soft tissues of children vs. adults. Tissue
lead of infants under 1 year old was generally lower than in older children, while children
aged 1-16 years had values that were comparable to adult women. In the Barry (1981) study,
the absolute concentration of lead in brain cortex or the ratios of brain cortex to blood lead
levels did not appear to be different in infants or older children compared to adults. Such
direct comparisons do not account for relative tissue mass changes with age, but this factor
is comparatively less with soft tissue than with the skeletal system (see Section 10.4).
Subcellular distribution of lead in soft tissue is not uniform, with high amounts of lead
•being sequestered in the mitochondria and nucleus. Cramer et al. (1974) studied renal biopsy
tissue in lead workers having exposures of variable duration and observed lead-binding nuclear
inclusion bodies in renal proximal tubules of subjects having short exposure, with all showing
mitochondrial changes. A considerable body of animal data (see Section 10.3.5) documents the
selective uptake of lead into these organelles. Pounds and Wright (1982) describe these
organellar pools in kinetic terms as having half-lives of comparatively short duration in cul-
tured rat hepatocytes, while McLachlin et al. (1980) found that rat kidney epithelial cells
form lead-sequestering nuclear inclusions within 24 hours.
10.3.2.2 Mineralizing Tissue. Biopsy and autopsy data have shown that lead becomes localized
and accumulates in human calcified tissues, i.e., bones and teeth. The accumulation begins
with fetal development (Barltrop, 1969; Horiuchi et al. , 1959).
Total lead content in bone may exceed 200 mg in men aged 60 to 70 years, but in women the
accumulation is somewhat lower. Various investigators (Barry, 1975; Horiguchi and Utsonomiya,
1973; Schroeder and Tipton, 1968; Horiuchi et al., 1959) have documented that approximately 95
percent of total body lead is lodged in bone. These reports not only establish the affinity
of bone for lead, but also provide evidence that lead increases in bone until 50-60 years, the
NEW10A/A 10-19 7/1/83
-------
PRELIMINARY DRAFT
later fall-off reflecting some combination of diet and mineral metabolism changes. Tracer
data show accumulation in both trabecular and compact bone (Rabinowitz et al., 1976).
In adults, bone lead is the most inert pool as well as the largest, and accumulation can
serve to maintain elevated blood lead levels years after past, particularly occupational, ex-
posure has ended. This accounts for the observation that duration of exposure correlates with
the rate of reduction of blood lead after termination of exposure (O'Flaherty et al., 1982).
The proportion of body lead lodged in bone is reported to be lower in children than in adults,
although concentrations of lead in bone increase more rapidly than in soft tissue during
childhood (Barry, 1975, 1981). In 23 children, bone lead was 9 mg, or 73 percent of total
body burden vs. 94 percent in adults. Expression of lead in bone in terms of concentration
across age groups, however, does not accommodate the "dilution" factor, which is quite large
for the skeletal system in children (see Section 10.4).
The isotope kinetic data of Rabinowitz et al. (1976) and Holtzman (1978) indicate biolo-
gical half-times of lead in bone on the order of several decades, although it appears that
there are two bone compartments, one of which is a repository for relatively labile lead
(Rabinowitz et al., 1977).
Tooth lead levels also increase with age at a rate proportional to exposure (Steenhout
and Pourtois, 1981), and are also roughly proportional to blood lead levels in man (Winneke et
al., 1981) and experimental animals (Kaplan et al., 1980). Dentine lead is perhaps the most
responsive component of teeth to lead exposure since it is laid down from the time of eruption
until the tooth is shed. Needleman. and Shapiro (1974) have documented the utility of dentine
lead as an indicator of the degree of subject exposure. Fremlin and Edmonds (1980), using
alpha particle excitation and micro-autoradiography, have shown dentine zones of lead enrich-
ment related to abrupt changes in exposure. The rate of lead deposition in teeth appears to
vary with the type of tooth, being highest in the central incisors and lowest in the molars, a
difference that must be taken into account when using tooth lead data for exposure assessment
particularly for low levels of lead exposure (Mackie et al., 1977; Delves et al., 1982).
10.3.3 Chelatable Lead
Mobile lead in organs and-systems is potentially more "active" toxicologically in terms
of being available to sites of action. Hence, the presence of diffusible, mobilizable, or ex-
changeable lead may be a more significant predictor of imminent toxicity or recent exposure
than total body or whole blood burdens. In reality, however, these would be quite difficult
assays.
In this regard, "chelatable" urinary lead has been shown to provide an index of this
mobile portion of total body burden. Chelation challenge is now viewed as the most useful
probe of undue body burden in children and adults (U.S. Centers for Disease Control, 1978;
NEW10A/A 10-20 7/1/83
-------
PRELIMINARY DRAFT
World Health Organization, 1977; Chisolm and Barltrop, 1979; Chisolm et al., 1976; Saenger et
al., 1982; Hansen et al., 1981), based mainly on the relationship of chelatable lead to in-
dices of heme biosynthesis impairment. In general, the amount of plumburesis associated with
chelant challenge is related to the dose and the schedule of administration.
A quantitative description of inputs to the fraction of body lead that is chelatable from
various body compartments is difficult to fully define, but it very likely includes a sizable,
fairly mobile compartment within bone as well as soft tissues this assertion is based on: 1)
the fact that the amount of lead mobilized by chelation is age dependent in non-exposed adults
(Araki, 1973; Araki and Ushio, 1982) while blood and soft tissue lead levels are not (Barry,
1975)> indicating a lead pool labile to chelation but kinetically distinct from soft tissue;
2) the studies of chelatable lead in animals (Hammond, 1971, 1973) suggesting removal of some
bone lead fraction and the response of explanted fetal rat bone lead to chelants (Rosen and
Markowitz, 1980); 3) the tracer modeling estimates of Rabinowitz et al. (1977) which suggest a
mobile bone compartment; and 4) the complex, non-linear relationship of lead intake by air,
food, and water (see Chapter 11) to blood lead, as well as the exponential relationship of
chelatable lead to blood lead (Chisolm et al., 1976).
The logarithmic relationship of chelatable lead to blood lead in children (Chisolm et
a!,f 1976) is consistent with the studies of Saenger et al. (1982), who reported that levels
of mobilizable lead in "asymptomatic" children with moderate elevations in blood lead were
quite similar in many cases to those values obtained in children with signs of overt toxicity.
Hansen et al. (1981) reported that lead workers challenged with CaNa2EDTA showed 24-hour urine
lead levels that in many cases exceeded the accepted limit levels even though blood lead was
only moderately elevated in many of those workers. The action level corresponded, on the re-
gression curve, to a blood value of 35 ug/dl.
Several reports provide insight into the behavior of labile lead pools in children
treated with chelating agents over varying periods of time. Treatment regimens using
CaNaoEDTA or CaNa2EDTA + BAL for up to 5 days have been invariably associated with "rebound"
in blood lead, ascribed to a redistribution of lead among mobile lead compartments (Chisolm
and Barltrop, 1979). Marcus (1982) reported that 41 children given oral D-penicillamine for 3
months showed a significant drop in blood lead by 2 weeks (mean initial value of 53.2 ug/dl)
then a slight rise that was within measurement error with a peak at 4 weeks, and a fall at 6
eks followed by no further change at a blood lead of 36 ug/dl. Hence, there was a near
ady~state a^ an eie-Vated level for 10 of the 12 weeks with continued treatment. This ob-
rvation may indicate that re-exposure was occurring, with oral penicillamine and ingested
lead leading to increased lead uptake, as seen by Jugo et al. (1975a). However, Marcus
tates that an effort was made to limit further lead intake as much as possible. From these
oorts, it appears that a re-equilibration does occur, varying in characteristics with type
NEW10A/A 10-21 7/1/83
-------
PRELIMINARY DRAFT
and duration of chelation. The rebound seen in short-term treatment with CaNa2EDTA or
CaNa2EDTA + BAL, although attributed to soft tissue, could well include a shift of lead from a
larger mobile bone compartment to soft tissues and blood. The apparent steady state between
the blood lead pool and other compartments that is achieved in the face of plumburesis, in-
duced by D-penicillamine (Marcus, 1982), suggests a rather sizable labile body pool which, in
quantitative terms, would appear to exceed that of soft tissue alone.
10.3.4 Mathematical Descriptions of Physiological Lead Kinetics
In order to account for observed kinetic data and make predictive statements, a variety
of mathematical models have been suggested, including those describing "steady state" condi-
tions. Tracer experiments have suggested compartmental models of lead turnover based on a
central blood pool (Holtzman, 1978; Rabinowitz et al., 1976; Batschelet et al. , 1979). These
experiments have hypothesized well-mixed, interconnected pools and have utilized coupled dif-
ferential equations with linear exponential solutions to predict blood and tissue lead ex-
change rates. Were lead to be retained in these pools in accordance with a power-law distri-
bution of residence times, rather than being uniform, a semi-Markov model would be more appro-
priate (Marcus, 1979).
Lead pools with more rapid turnover than whole blood (on the order of minutes) have been
detected within isolated cells (Pounds and Wright, 1982). Evidence of an extracellular lead
pool in humans exists in observations of lead plasma (DeSilva, 1981) and urine (Rabinowitz et
al., 1974) after oral lead exposure, as well as from 203Pb studies using injection, ingestion,
and inhalation exposure routes (Chamberlain and Heard, 1981). No single model has been deve-
loped to utilize what has been learned about lead behavior in these highly labile pools
existing around and within permanent and concentrated sites.
Extant steady-state models are also deficient, not only because they are based on small
numbers of subjects but also because there may be a dose dependency for some of the interpool
transfer coefficients. In this case, a non-linear dose-indicator response model would be more
appropriate when considering changes in blood lead levels. For example, the relationship
between blood lead and air lead (Hammond et al., 1981) as well as that for diet (United
Kingdom Central Directorate on Environmental Pollution, 1982) and tap drinking water (Sherlock
et al., 1982) are all non-linear in mathematical form. In addition, alterations in
nutritional status or the onset of metabolic stresses can complicate steady-state relation-
ships.
The above discussions of both the non-linear relationship of intake to the blood lead
pool and the non-linear relationship of chelatable, or toxicologically active, lead to blood
levels logically indicate that intake at elevated levels can add substantially to this
chelatable pool and be substantially unrecognized in blood lead measurements.
NEW10A/A 10-22 7/1/83
-------
PRELIMINARY DRAFT
10.3.5 Animal Studies
The relevant questions to be asked of animal data are those that cannot be readily or
fully satisfied in human subjects: (1) What is the effect of exposure level on distribution
within the body at specific time points? (2) What is the relationship of age or developmental
stage on the distribution of lead in organs and systems, particularly the nervous system?
(3) What are the relationships of physiological stress and nutritional status to the redistri-
bution kinetics? (4) Can the relationship of chelatable lead to such indicator lead pools as
blood be defined better?
Administration of a single dose of lead to rats produces high initial lead concentrations
in soft tissues, which then fall rapidly as the result of excretion and transfer to bone
(Hammond, 1971), while the distribution of lead appears to be independent of the dose.
Castellino and Aloj (1964) reported that single dose exposure of rats to lead was associated
with a fairly constant ratio of red cell to plasma, a rapid distribution to tissues and rela-
tively higher uptake in liver, kidney, and particularly bone. Lead loss from organs and tis-
sues follow first-order kinetics except for bone. The data of Morgan et al. (1977),
Castellino and Aloj (1964), and Keller and Doherty (1980a) document that the skeletal system
in rats and mice is the kinetically rate-limiting step in whole-body lead clearance.
Subcellular distribution studies involving either tissue fractionation after rn vivo lead
exposure or in vitro data document that lead is preferentially sequestered in the nucleus
(Castellino and Aloj, 1964; Goyer et al. , 1970) and mitochondrial fractions (Castellino and
Aloj, 1964; Barltrop et al., 1974) of cells from lead-exposed animals. Lead enrichment in the
mitochondrion is consistent with the high sensitivity of this organelle to the toxic effects
of lead.
The neonatal animal seems to retain proportionately higher levels of tissue lead compared
with the adult (Goldstein et al., 1974; Momcilovic and Kostial, 1974; Mykkanen et al., 1979;
Klein and Koch, 1981) and shows slow decay of brain lead levels while other tissue levels sig-
nificantly decrease over time. This appears to be the result of enhanced entry by lead due to
a poorly developed brain barrier system in the developing animals, as well as enhanced body
retention in the young animals. The effects of such changes as metabolic stress and nutri-
tional status have been noted in the literature. Keller and Doherty (1980b) have documented
that tissue redistribution of lead, specifically bone lead mobilization, occurs in lactating
female mice, both lead and calcium transfer occurring from mother to pups. Changes in lead
movement from body compartments, particularly bone, with changes in nutrition are described in
Section 10.5.
In studies with rats that are relevant both to the issue of chelatable lead vs. lead in-
dicators in humans and to the relative lability of lead in the young vs. the adult, Jugo et
al. (I975b) and Jugo (1980) studied the chelatability of lead in neonate vs. adult rats and
NEW10A/A 10-23 7/1/83
-------
PRELIMINARY DRAFT
its lability in the erythrocyte. Challenging young rats with metal chelants yielded propor-
tionately lower levels of urinary lead than in the adult, a finding that has been ascribed to
tighter binding of lead in the young animal (Jugo et al., 1975b). In a related observation,
the chelatable fraction of lead bound to erythrocytes of young animals given 203Pb was approx-
imately 3-fold greater than in the adult rat (Jugo, 1980), although the fraction of dose in
the cells was higher in the suckling rat. The difference in the suckling rat erythrocyte re-
garding the binding of lead and relative content compared with the adult may be compared with
the Ong and Lee's (1980b) observation that human fetal hemoglobin binds lead more avidly than
does mature hemoglobin.
10.4 LEAD EXCRETION AND RETENTION IN HUMANS AND ANIMALS
Dietary lead in humans and animals that is not absorbed passes through the gastro-
instestinal tract and is eliminated with feces, as is that deposited fraction of air lead that
is swallowed and not absorbed. Lead absorbed into the blood stream and not retained is excre-
ted through the renal and gastrointestinal tracts, the latter by biliary clearance. The
amounts appearing in urine and feces appear to be a function of such factors as species, age,
and differences in dosing.
10.4.1 Human Studies
Booker et al. (1969) found that 212Pb injected into two adult volunteers led to initial
appearance of the label first in urine (4.4 percent of dose in 24 hours), then in both urine
and feces in approximately equal amounts. By use of the stable isotope 204Pb, Rabinowitz et
al. (1973) reported that urinary and fecal excretion of the label amounted to 38 and 8 ug/day
in adult subjects, accounting for 76 and 16 percent, respectively, of the measured recovery.
Fecal excretion was thus approximately twice that of all the remaining modes of excretion:
hair, sweat, and nails (8 percent).
Perhaps the most detailed study of lead excretion in adult humans was done by Chamberlain
et al. (1978), who used 203Pb administered by injection, inhalation and ingestion. Following
injection or oral intake, the amounts in urine (Pb-U) and feces (Pb-Fe, endogenous fecal lead)
were compared for the two administration routes. Endogenous fecal lead was 50 percent of that
in urine, or a 2:1 ratio of urinary/fecal lead, after allowing for increased transit time of
fecal lead through the GI tract.
Based on the metabolic balance and isotope excretion data of Kehoe (1961a,b,c),
Rabinowitz et al. (1976), and Chamberlain et al. (1978), as well as some recalculations of the
Kehoe and Rabinowitz data by Chamberlain et al. (1978), it appears that short-term lead excre-
tion amounts to 50-60 percent of the absorbed fraction, the balance moving primarily to bone
NEW10A/A 10-24 7/1/83
-------
PRELIMINARY DRAFT
TABLE 10-3. COMPARATIVE EXCRETION AND RETENTION
RATES3 IN ADULTS AND INFANTS
Dietary intake (ng/kg)
Fraction absorbed6
Diet lead absorbed (|jg/kg)
Air lead absorbed (ug/kg)
Total absorbed lead (yg/kg
Daily urinary Pb (|jg/kg)
Ratio: urinary/absorbed Pb
Endogenous fecal Pb
Total excreted Pb
Ratio: total excreted/
absorbed Pb
Fraction of intake retained
Children
10.76
0.46 (0.55)f
4.95 (5.92)
0.20
5.15 (6.12)
1.00
0.19 (0.16)
0.5 (1.56)h
1.50 (2.56)
0.29 (0.42)
0.34 (0.33)
Adult
group A
3.63
0.159
0.54
0.21
0.75
0.47
0.62
0.241
0.71
0.92
0.01
Adult .
group B
3.86
0.159
0.58
0.11
0.68
0.34
0.50
0.171
0.51
0.75
0.04
DZiegler et al., 1978.
JjRabinowitz et al., 1977.
Thompson, 1971, and estimates of Chamberlain et al., 1978.
^Corrected for endogenous fecal Pb; Pb-Fe = 0.5 x Pb-U.
'Corrected for endogenous fecal Pb at extrapolated value from
Ziegler et al., 1978.
^Corrected for Pb-Fe.
.Extrapolated value for endogenous fecal Pb of 1.56.
'For a ratio of 0.5, Pb-Fe/Pb-U.
HEW10A/A
10-25
7/1/83
-------
PRELIMINARY DRAFT
with some subsequent fraction, (approximately half) of this stored amount eventually being
excreted. The rapidly excreted fraction was determined by Chamberlain et al. (1978) to have
an excretion half-time of about 19 days. This is consistent with the estimates of Rabinowitz
et al. (1976), who expressed clearance in terms of mean-times. Mean-times are multiplied by
In 2 (0.693) to arrive at half-times. The similarity of blood 203Pb half-times with that of
body excretion noted by Chamberlain et al. (1978) indicates a steady rate of clearance from
the body.
The age dependency of lead excretion rates in humans, has not been well studied, for all
of the above lead excretion data involved only adults. Table 10-3 combines available data
from adults and infants for purposes of comparison. Intake, urine, fecal, and endogenous
fecal lead data from two studies involving adults and one. report with infants are used. For
consistency in the adult data, 70 kg is used as an average adult weight, and a Pb-Fe/Pb-U
value of 0.5 used. Lead intake, absorption, and excretion are expressed as ug Pb/kg/day. For
the Ziegler et al. (1978) data with infants, endogenous fecal lead excretion is calculated
using the adult ratio as well as the extrapolated value of 1.5 (jg Pb/kg/day. The respiratory
intake value for the infants is an upper value (0.2 ug Pb/m3), since Ziegler et al. found air
lead to be <0.2 ug/m3. In comparison with the two representative adult groups, infants appear
to have a lower total excretion rate, although the excretion of endogenous fecal lead may be
higher than for adults.
Lead is accumulated in the human body with age, mainly in bone, up to approximately 60
years of age, when a decrease occurs with changes in intake as well as in bone mineral
metabolism. Total accumulation by 60 years of age ranges up to approximately 200 mg (see
review by Barry, 1978), although occupational exposure can raise this figure several-fold
(Barry, 1975). Holtzman (1978) has reviewed the available literature on studies of lead
retention in bone. In normally exposed humans a biological half-time of approximately 17
years has been calculated, while data for uranium miners yield a range of 1320-7000 days (4-19
years). Chamberlain et al. (1978) have estimated life-time averaged daily retention at 9.5 pg
using data of Barry (1975). Within shorter time frames, however, retention can vary con-
siderably due to such factors as disruption of the individual's equilibrium with lead intake
at a given level of exposure, the differences between children and adults, and, in elderly
subjects, the presence of osteoporosis (Gross and Pfitzer, 1974).
Lead labeling experiments, such as those of Chamberlain et al. (1978), indicate a short-
term or initial retention of approximately 40-50 percent of the fraction absorbed, much of
which is by bone. It is difficult to determine how much lead resorption from bone will even-
tually occur using labeled lead, given the extremely small fraction of labeled to unlabeled
lead (i.e.,. label dilution) that would exist. Based on the estimates of Kehoe (1961a,b,c),
the Gross (1981) evaluation of the Kehoe studies, the Rabinowitz et al. (1976) study, the
NEW10A/A 10-26 7/1/83
-------
PRELIMINARY DRAFT
Chamberlain et al. (1978) assessments of the aforementioned reports, and the data of Thompson
(1971). approximately 25 percent of the lead absorbed daily undergoes long-term bone storage.
The above estimates relate either to adults or to long-term retention over most of an
individual's lifetime. Studies with children and developing animals (see Section 10.4.2)
indicate lead retention in childhood can be higher than in adults. By means of metabolic
balance studies, Ziegler et al. (1978) obtained a retention figure (as percentage of total
intake) of 31.5 percent for infants, while of Alexander et al. (1973) provided an estimate of
18 percent. Corrected retention data for both total and absorbed intake for the pediatric
subjects of Ziegler et al. (1978) are shown in Table 10.3, using the two values for endogenous
fecal excretion as noted. Barltrop and Strehlow (1978) calculated a net negative lead reten-
tion in their subjects, but problems in comparing this _report with the others were noted
above. Given the increased retention of lead in children relative to adults, as well as the
greater rate of lead intake on a body weight basis, increased uptake in soft tissues and/or
bone is indicated.
Barry (1975, 1981) measured the lead content of soft and mineral tissue in a small group
of autopsy samples from children 16 years of age and under, and noted that average soft tissue
values were comparable to those in female adults, while mean bone lead values were lower than
in adults. This suggests that bone in children has less retention capacity for lead than
adults. It should be noted, however, that "dilution" of bone lead will occur because of the
significant growth rate of the skeletal system through childhood. Trotter and Hi/on (1974)
studied changes in skeletal mass, density, and mineral content as a function of age, and noted
•that skeletal mass increases exponentially in children until the early teens, increases less
up to the early 20s, levels off in adulthood, and then slowly decreases. From infancy to the
late teens, bone mass increases up to 40-fold. Barry (1975) noted an approximate doubling in
bone lead concentration over this interval, indicating that total skeletal lead had actually
increased 80-fold, and obtained a mean total bone lead content up to 16 years of approximately
8 mg, compared with a value of approximately 18 mg estimated from both the bone concentrations
in his study at different ages and the bone growth data of Trotter and Hixon (1974). In a
later study (Barry, 1981), autopsy samples from infants and children between 1 and 9 years old
showed an approximate 3.5-fold increase in mean bone concentrations across the three bone
tvpes studied, compared with a skeletal mass increase from 0-6 mos. to 3-13 years old of
areater than 10-fold, for an estimated increase in total lead of approximately 35-fold. Five
reports (see Barry, 1981) noted age vs. tissue lead relationships indicating that overall bone
lead levels in infants and children were less than in adults, whereas while 4 reports observed
comparable levels in children and adults.
If one estimates total daily retention of lead in the infants studied by Ziegler et al.
(1978), using a mean body weight of approximately 10 kg and the corrected retention rate in
Table 10.3, one obtains a total daily retention of approximately 40 ug Pb. By contrast, the
NEW10A/A 10-27 7/1/83
-------
PRELIMINARY DRAFT
total reported or estimated skeletal .lead accumulated between 2 and 14 years is 8-18 mg (vide
supra), which averages out to a daily long-term retention of 2.0-4.5 ug/day or 6-13 percent of
total retention. It may be the case that lead retention is highest in infants up to about 2
years of age (the subjects of the Ziegler et al. study), then decreases in older children.
The mean retention in the Alexander et al. (1973) study was 18 percent, about half that seen
by Ziegler et al. (1978). This difference is possibly due to the greater age range in the
former study.
"Normal" blood lead levels in children either parallel adult males or are approximately
30 percent greater than adult females (Chamberlain et al., 1978), indicating (1) that the soft
tissue lead pool in very young children is not greatly elevated and thus, (2) that there is a
huge labile lead pool in bone which is still kinetically quite distinct from soft tissue lead
or (3) that in young children, blood lead is a much less reliable indicator of greatly ele-
vated soft tissue or labile bone lead than is the case with adults. Barry (1981) found that
soft tissue lead levels were comparable in infants ^1 year old and children 1-5 and 6-9 years
old.
Given the implications of the above discussion, that retention of lead in the young child
is higher than in adults and possibly older children, while at the same time their skeletal
system is less effective for long-term lead sequestration, the very young child is at greatly
elevated risk to a toxicologically "active" lead burden. For a more detailed discussion, see
Chapter 13.
10.4.2 Animal Studies
In rats and other experimental animals, both urinary and fecal excretion appear to be
important routes of lead removal from the organism; the relative partitioning between the two
modes is species and dose dependent. Morgan et al. (1977), injected 203Pb into adult rats and
noted that lead initially appeared in urine, followed by equivalent elimination by both
routes; by 5 days, lead was proportionately higher in feces. Castellino and Aloj (1964),
using 210Pb, observed that fecal excretion was approximately twice that of urine (35.7 vs.
15.9 percent) by 14 days. In the report of Klaassen and Shoeman (1974), relative excretion by
the two routes was seen to be -dose-dependent up to 1.0 mg/kg, being much higher by biliary
clearance into the gut. At 3.0 mg/kg, approximately 90 percent of the excreted amount was
detected in feces. The relatively higher proportion appearing in feces in the studies of
Castellino and Aloj (1964) and Klaassen and Shoeman (1974), compared with the results of
Morgan et al. (1977), is possibly due to the use of carrier dosing, since Morgan et al. (1977)
used carrier-free injections. Hence, it appears that increasing dose does favor biliary excre-
tion, as noted by Klaassen and Shoeman (1974).
NEW10A/A 10-28 7/1/83
-------
PRELIMINARY DRAFT
With regard to species differences, Klaassen and Shoeman (1974) found that the amount of
biliary clearance in dogs was about 2 percent of that in rats, while rabbits showed 50 percent
of the rate of the rat at equivalent dosing. These data for the dog are in contrast to the
results of Lloyd et al. (1975), who observed 75 percent of the excreted lead eliminated
through biliary clearance. It should be noted that the latter researchers used carrier-free
label while the other investigators used injections with carrier at 3.0 mg Pb/kg levels. In
mice, Keller and Doherty (1980a) observed that the cumulative excretion rate of 210Pb in urine
was 25-50 percent of that in feces. In nonhuman primates, Cohen (1970) observed that baboons
excreted lead at the rate of 40 percent in feces and 60 percent in urine. Pounds et al.
(1978) noted that the Rhesus monkey lost 30 percent of lead by renal excretion and 70 percent
in feces. This may also be reflecting a carrier dosing difference.
The extent of total lead excretion in experimental animals given labeled lead orally or
parenterally varies, in part due to the time frames for post-exposure observation. In the
adult rat, Morgan et al. (1977) found that 62 percent of injected 203Pb was excreted by 6
days. By 8 days, 66 percent of injected 203Pb was eliminated in the adult rats studied by
Momcilovic and Kostial (1974), while the 210Pb excretion data of Castellino and Aloj (1964)
for the adult rat showed 52 percent excreted by 14 days. Similar data were obtained by
Klaassen and Shoeman (1974). Lloyd et al. (1975) found that dogs excreted 52 percent of
injected lead label by 21 days, 83 percent by 1 year, and 87 percent by 2 years. In adult
mice (Keller and Doherty, 1980a), 62 percent of injected lead label was eliminated by 50 days.
In the nonhuman primate, Pounds et al. (1978) measured approximately 18 percent excretion in
adult Rhesus monkeys by 4 days.
Kinetic studies of lead elimination in experimental animals indicate that excretion is
described by two or more components. From the elimination data of Momcilovic and Kostial
(1974)> Morgan et al. (1977) estimated that in the rat the excretion curve obeys a two-compo-
nent exponential expression with half-times of 21 and 280 hours. In dogs, Lloyd et al. (1975)
found that excretion could be described by three components, i.e., a sum of exponentials with
half-times of 12 days, 184 days, and 4951 days. Keller and Doherty (1980a) reported that the
half-time of whole-body clearance of injected 203Pb consisted of an initial rapid and a much
slower terminal component, the latter having a half-time of 110 days in the adult mouse.
The excretion rate dependency on dose level has been investigated in several studies.
Although Castellino and Aloj (1964) saw no difference in total excretion rate when label was
Injected with 7 or 100 M9 of carrier, Klaassen and Shoeman (1974) did observe that the excre-
tion rate by biliary tract was dose dependent at 0.1, 1.0, and 3.0 mg Pb/kg (urine values were
not provided for obtaining estimates of total excretion). Momcilovic and Kostial (1974) saw
Increased rate of excretion into urine over the added carrier range of 0.1 to 2.0 ug Pb with
no change in fecal excretion. In the report of Aungst et al. (1981) there was no change in
NEW10A/A 10-29 7/1/83
-------
PRELIMINARY DRAFT
excretion rate in the rat over the injected lead dosing range of 1.0 to 15.0 mg/kg. It thus
appears that rat urinary excretion rates are dose-dependent over a narrow range less than <7
ug, while elimination of lead through biliary clearance is dose-dependent up to an exposure
level of 3 mg Pb/kg.
Lead movement from lactating animals to their offspring via milk constitutes both a route
of excretion for the mother and a route of exposure to lead for the young. Investigations
directed at this phenomenon have examined both prior-plus-ongoing maternal lead exposure
during lactation and the effects of immediate prior treatment. Keller and Doherty (1980b)
exposed two groups of female rats to 210Pb-labeled lead: one group for 105 days before mat-
ing; the second before and during gestation and nursing. During lactation, there was an over-
all loss of lead from the bodies of the lactating females compared with controls while the
femur ash weights were inversely related to level of lead excretion, indicating that such
enhancement is related to bone mineral metabolism. Lead transfer via milk was approximately
3 percent of maternal body burden, increasing with continued lead exposure during lactation.
Lorenzo et al. (1977) found that blood lead in nursing rabbits given injected lead peaks
rather rapidly (within 1 hour), while milk lead shows a continuous increase for about 8 days,
at which point its concentration of lead is 8-fold higher than blood. This indicates that
lead transfer to milk can occur against a concentration gradient in blood. Momcilovic (1978)
and Kostial and Momcilovic (1974) observed that transfer of 203Pb in the late stage of lacta-
tion occurs readily in the rat, with higher overall excretion of lead in nursing vs. control
females. Furthermore, it appeared that the rate of lead movement to milk was dose-dependent
over the added lead carrier range of 0.2-2.0 ug Pb.
The comparative retention of lead in developing vs. adult animals has been investigated
in several studies using rats, mice, and nonhuman primates. Momcilovic and Kostial (1974)
compared the kinetics of lead distribution in suckling vs. adult rats after injection of
203Pb. Over an 8-day interval, 85 percent of the label was retained in the suckling rat,
compared with 34 percent in the adult. Keller and Doherty (1980a) compared the levels of
210Pb in 10-day-old mice and adults, noting from the clearance half-times (vide supra) that
lead retention was greater in the suckling animals than in the adults. In both adult and
young mice, the rate of long-term retention was governed by the rate of release of lead from
bone, indicating that in the mouse, skeletal lead retention in the young is greater than in
the adult. With infant and adult monkeys orally exposed to 210Pb, Pounds et al. (1978)
observed that at 23 days the corresponding amounts of initial dose retained were 92.7 and 81.7
percent, respectively.
The studies of Rader et al. (1981; 1982) are of particular interest as they not only
demonstrate that young experimental animals continue to show greater retention of lead in
tissue when exposure occurs after weaning, but also that such retention occurs in terms of
NEW10A/A 10-30 7/1/83
-------
PRELIMINARY DRAFT
either uniform exposure (Rader et al., 1981) or uniform dosing (Rader et al., 1982) when com-
pared with adult animals. With uniform exposure, 30-day-old rats given lead in drinking water
showed significantly higher lead levels in blood and higher percentages of dose retained in
brain, femur, and kidney, as well as higher indices (ALA-U, EP) of hematopoietic impairment
when compared with adult animals. As a percentage of dose retained, tissues in the young ani-
mals were approximately 2-3-fold higher. In part, the difference is due to a higher ingestion
rate of lead. However, in the uniform dosing study where this was not the case, an increased
retention of lead still prevailed, the amount of lead in brain being approximately 50 percent
higher in young vs. adult animals. Comparison of values in terms of percent retained is more
meaningful for such assessments, because the factor of changes in organ mass (see above) is
taken into account. Delayed excretion in the young animal may reflect an immature excretory
system or a tighter binding of lead in various body compartments.
10.5 INTERACTIONS OF LEAD WITH ESSENTIAL METALS AND OTHER FACTORS
Deleterious agents, particularly toxic metals such as lead, do not express their toxico-
Icinetic or toxicological behavior in a physiological vacuum, but rather are affected by inter-
actions of the agent with a variety of biochemical factors such as nutrients. Growing recog-
nition of this phenomenon and its implications for lead toxicity in humans have prompted a
number of studies, many of them recent, that address both the scope and mechanistic nature of
such interactive behavior.
10.5.1 Human Studies
In humans, the interactive behavior of lead and various nutritional factors is appropri-
ately viewed as being particularly significant for children, since this age group is not only
particularly sensitive to lead's effects, but also represents the time of greatest flux in
relative nutrient status. Such interactions occur against a backdrop of rather widespread
deficiencies in a number of nutritional components in children. While such deficiencies are
inore pronounced in lower income groups, they exist in all socioeconomic strata. Mahaffey and
Michaelson (1980) have summarized the three nutritional status surveys carried out in the
United States for infants and young children: the Preschool Nutrition Survey, the Ten State
Nutrition Survey, and the National Health Assessment and Nutritional Evaluation Survey (NHANES
I) The most recent body of data of this type is the NHANES II study (Mahaffey et al., 1979),
Ithough the dietary information from it has yet is to be reported. In the older surveys,
iron deficiency was the most common nutritional deficit in children under 2 years of age,
rticularly cnii
-------
PRELIMINARY DRAFT
calcium. Owen and Lippmann (1977) reviewed the regional surveys of low-income groups within
Hispanic, white, and black populations. In these groups, iron deficiency was a common
finding, while low intakes of calcium and vitamins A and C were observed regularly. Hambidge
(1977) concluded that zinc intake in low-income groups is generally inadequate, relative to
recommended daily allowances.
Available data from a number of reports document the association of lead absorption with
suboptimal nutritional status. Mahaffey et al. (1976) summarized their studies showing that
children with blood lead greater than 40 ug/dl had significantly (p <0.01) lower intake of
phosphorus and calcium compared with a control group, while iron intake in the two groups was
comparable. This study involved children 1-4 years old from an inner-city, low-income popula-
tion, with close matching for all parameters fexcept the blood lead level. Sorrell et al.
(1977), in their nutritional assessment of 1- to 4-year-old children with a range of blood
lead levels, observed that blood lead content was inversely correlated with calcium intake,
while children with blood lead levels >60 ug/dl had significantly (p <0.001) lower intakes of
calcium and vitamin D.
Rosen et al. (1981) found that children with elevated blood lead (33-120 ug/dl) had sig-
nificantly lower serum concentrations of the vitamin D metabolite 1,25-(OH)2D (p <0.001) com-
pared with age-matched controls, and showed a negative correlation of serum 1,25-(OH)2D with
lead over the range of blood leads measured. These observations and animal data (Barton et
al., 1978a, see Section 10.5.2) may suggest an increasingly adverse interactive cycle of
1,25-(OH)2D, lead, and calcium in which lead reduces biosynthesis of the vitamin D metabolite.
This then leads to reduced induction of calcium binding protein (CaBP), less absorption of
calcium from the gut, and greater uptake of lead, thus increasing uptake of lead and further
reducing metabolite levels. Barton et al. (1978a) isolated two mucosal proteins in rat intes-
tine, one of which bound mainly lead and was not vitamin D-stimulated; the second bound mainly
calcium and was under vitamin control. The authors suggested direct site binding competition
between lead and calcium in these proteins. Hunter (1978) investigated the possible inter-
active role of seasonal vitamin D biosynthesis in adults and children; it is a common obser-
vation that lead poisoning occurs more often in summer than in other seasons (see Hunter,
1977, for review). In children, seasonality accounts for 16 percent of explained variance
of blood lead in black children, 12 percent in Hispanics, and 4 percent in whites. More
recently, it has been documented that there is no seasonal variation in circulating levels of
1,25-(OH)2D the metabolite that affects the rate of lead absorption from the GI tract (Chesney
et al., 1981). These results suggest that seasonality is related to changes in exposure.
Johnson and Tenuta (1979) determined that calcium intake was negatively correlated
(r = -0.327, p <0.05) with blood lead in 43 children aged 1-6 years. The high lead group also
consumed less zinc than children with lower blood levels. Yip et al. (1981) found that 43
NEW10A/A 10-32 7/1/83
-------
PRELIMINARY DRAFT
children with elevated blood lead (>30 ug/dl) and EP (>35 ug/dl) had an increased prevalence
of iron deficiency as these two parameters increased. Children classed as CDC Ib and II had a
79 percent deficiency rate, while those in Class III were all iron-deficient. Chisolm (1981)
demonstrated an inverse relationship between "chelatable" iron and chelatable body lead levels
as indexed by urinary ALA levels in 66 children with elevated blood lead. Watson et al.
(1980) reported that adult subjects who were iron-deficient (determined from serum ferritin
measurement) showed a lead absorption rate 2-3 times greater than subjects who were iron
replete. In a group of 13 children, Markowitz and Rosen (1981) reported that the mean serum
zinc levels in children with plumbism were significantly below the values seen in normal chil-
dren. Chelation therapy reduced the mean level even further. Chisolm (1981) reported that
there was an inverse relationship between ALA-U and the amount of "chelatable11 or systemically
active zinc in 66 children challenged with EDTA and having blood lead levels ranging from 45
to 60 ug Pb/dl. These two studies suggest that zinc status is probably as important an inter-
active modifier of lead toxicity as is either calcium or iron.
The role of nutrients in lead absorption has been reported in several metabolic balance
studies for both adults and children. Ziegler et al. (1978), in their investigations of lead
absorption and retention in infants, observed that lead retention was inversely correlated
with calcium intake, expressed either as intake percentage (r = -0.284, .p <0.01) or on a
weight basis (r = -0.279, p <0.01). Of interest is the fact that the range of calcium intake
measured was within the range considered adequate for infants and toddlers by the National
Research Council (National Academy of Sciences, National Research Council, 1974). These data
also support the premise that severe deficiency need not be present for an interactive rela-
tionship to occur. Using adults, Heard and Chamberlain (1982) monitored the uptake of 203Pb
from the gut in eight subjects as a function of the amounts of dietary calcium and phosphorus.
Without supplementation with either of these minerals in fasting subjects, the label absorp-
tion rate was approximately 60 percent, compared with 10 percent with 200 mg calcium plus
140 mg phosphorus, the amounts present in an average meal. Calcium alone reduced uptake by a
factor of 1.3 and phosphorus alone by 1.2; both together yielded a reduction factor of 6.
This work suggests that insoluble calcium phosphate is formed and co-precipitates any lead
resent. This interpretation is supported by animal data (see Section 10.5.2).
jO.5.2 Animal Studies
Reports of lead-nutrient interactions in experimental animals have generally described
uch relationships in terms of a single nutrient, using relative absorption or tissue reten-
tion 1" the an"ima^ to index the effect. Most of the recent data are concerned with the impact
f dietary levels of calcium, iron, phosphorus, and vitamin D. Furthermore, some investigat-
have attempted to elucidate the site(s) of interaction as well as the mechanism(s)
MEW1OA/A 10-33 7/1/83
-------
PRELIMINARY DRAFT
governing the interactions. Lead's interactions involve the effect of the nutrient on lead
uptake, as well as lead's effect on nutrients; the focus of this discussion is on the former.
These interaction studies are tabulated in Table 10-4.
10.5.2.1 Interactions of Lead with Calcium. The early report of Sobel et al. (1940) noted
that variation of dietary calcium and other nutrients affected the uptake of lead by bone and
blood in animals. Subsequent studies by Mahaffey-Six and Goyer (1970) in the rat demonstrated
that a considerable reduction in dietary calcium was necessary from (0.7 percent to 0.1 per-
cent), at which level blood lead was increased 4-fold, kidney lead content was elevated 23-
fold, and relative toxicity (Mahaffey et al., 1973) was increased. The changes in calcium
necessary to alter lead's effects in the rat appear to be greater than those seen by Ziegler
et al. (1978) in young children, indicating species differences in terms of sensitivity to
basic dietary differences, as well as to levels of all interactive nutrients. These observa-
tions in the rat have been confirmed by Kostial et al. (1971), Quarterman and Morrison (1975),
Barltrop and Khoo (1975), and Barton et al. (1978a). The inverse relationship between dietary
calcium and lead uptake has also been noted in the pig (Hsu et al., 1975), horse (Willoughby
et al., 1972), lamb (Morrison et al., 1977), and domestic fowl (Berg et al., 1980).
The mechanism(s) governing lead's interaction with calcium operate at both the gut wall
and within body compartments. Barton et al. (1978a), using everted duodenal sac preparations
in the rat, reported that: (1) interactions at the gut wall require the presence of intubated
calcium to affect lead label absorption - (pre-existing calcium deficiency in the animal and
no added calcium have no effect on lead transport); (2) animals having calcium deficiency show
increased retention of lead rather than absorption (confirmed by Quarterman et al., 1973); and
(3) lead transport may be mediated by two mucosal proteins, one of which has high molecular
weight, a high proportion of bound lead, and is affected in extent of lead binding with
changes in lead uptake. The second protein binds mainly calcium and is vitamin D-dependent.
Smith et al. (1978) found that lead is taken up at a different site in the duodenum of
rats than is calcium but absorption does occur at the site of phosphate uptake, suggesting a
complex interaction of phosphorus, calcium, and lead. This is consistent with the data of
Barltrop and Khoo (1975) for rats and the data of Heard and Chamberlain (1982) for humans,
thus showing that the combined action of the two mineral nutrients is greater than the sum of
either's effects.
Mykkanen and Wassermann (1981) observed that lead uptake in the intestine of the chick
occurs in 2 phases: a rapid uptake (within 5 minutes) followed by a rate-limiting slow trans-
fer of lead into blood. Conrad and Barton (1978) have observed a similar process in the rat.
NEW10A/A 10-34 7/1/83
-------
TABLE 10-4. EFFECT OF NUTRITIONAL FACTORS ON LEAD UPTAKE IN ANIMALS
Factor
Species
Index of effect
Interactive effect
Reference
o
i
CO
Calcium
Calcium
Calcium
Calcium
Calcium
Iron
Iron
Iron
Rat Lead in tissues and
effect severity at
low levels of dietary
calcium
Pig Lead in tissues at
low levels of
dietary calcium
Horse Lead in tissues at
low levels of
dietary calcium
Lamb Lead in tissue at
low levels of
dietary calcium
Rat Lead retention
Rat Tissue levels and
relative toxicity
of lead
Rat Lead absorption in
everted duodenal
sac preparation
Mouse Lead retention
Low dietary calcium (0.1%)
increases lead absorption
and severity of effects
Increased absorption of
lead with low dietary
calcium
Increased absorption of
lead with low dietary
calcium
Increased absorption of
lead with low dietary
calcium
Retention increased in
calcium deficiency
Iron deficiency increases
lead absorption and
toxicity
Reduction in intubated
iron increases lead
absorption; increased
levels decrease lead uptake
Iron deficiency has no
effect on lead
retention
Mahaffey-Six and Goyer,
1970; Mahaffey et al.,
1973
Hsu et al., 1975
Willoughby et al., 1972
Morrison et al. , 1977
Barton et al., 1978a
Mahaffey-Six and Goyer,
1972
Barton et al. , 1978b
Hamilton, 1978
O
;u
3>
-------
TABLE 10-4. (continued)'
Factor
Species
Index of effect
Interactive effect
Reference
Iron
Phosphorus
Phosphorus
o Phosphorus
Vitamin D
Vitamin D
Lipid
Protein
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
lr\ utero or milk
transfer of lead in
pregnant or lactating
rats
Lead uptake in tissues
Lead retention
Lead retention
Lead absorption
using everted sac
techniques
Lead absorption
using everted sac
techniques
Lead absorption
Lead uptake by tissues
Iron deficiency increases
both ui utero and milk
transfer of lead to
sucklings
Reduced P increased
203Pb uptake 2.7-fold
Low dietary P enhances
lead retention; no
effect on lead resorption
in bone
Low dietary P enhances
both lead retention *and
deposition in bone
Increasing vitamin D
increases intubated
lead abosrption
Both low and excess
levels of vitamin D
increase lead uptake
by affecting motility
Increases in lipid (corn
oil) content up to
40 percent enhances lead
absorption
Both low and high protein
in diet increase lead
absorption
Cerklewski, 1980
Barltrop and Khoo, 1975
Quarterman and Morrison,
1975
Barton and Conrad, 1981
Smith et al., 1978
Barton et al., 1980
Barltrop and Khoo, 1975
Barltrop and Khoo, 1975
O
TO
f-
-------
TABLE 10-4. (continued)
Factor
Species Index of effect
Interactive effect
Reference
CO
Protein
Protein
Zinc/Copper
Rat
Rat
Milk components Rat
Milk components Rat
Rat
Zinc/Copper *• Rat
Zinc/Copper
Rat
Body lead retention
Tissue levels of
lead
Lead absorption
Lead absorption
Lead absorption
Lead transer j^n
utero and in milk
during lactation
Lead absorption
Low dietary protein either
reduces or does not affect
retention in various
tissues
Casein in diet increases
lead uptake compared to
soybean meal
Lactose-hydrolyzed milk
does not increase lead
absorption, but ordinary
milk does
Lactose in diet enhances
lead absorption compared
to glucose
Low zinc in diets
increases lead absorption
Low-zinc diet of mother
increases lead transfer
in utero and in maternal
JnTlk
Low copper in diet
increases lead absorption
Quarter-man et al., 1978b
Anders et al., 1982
Bell and Spickett, 1981
Bushnell and DeLuca, 1981
Cerklewski and Forbes,
1976; El-Gazzar et al. ,
1978
Cerklewski, 1979
Klauder et al. , 1973;
Klauder and Petering, 1975
-o
TO
f>
73
70
f>
-------
PRELIMINARY DRAFT
Hence, there is either a saturation process occurring, i.e., carrier-mediated transport, or
simply lead precipitation in the lumen. In the former case, calcium interacts to saturate the
carrier proteins as isolated by Barton et al. (1978a) or may precipitate lead in the lumen by
initial formation of calcium phosphate.
Quarterman et al. (1978a) observed that calcium supplementation of the diet above normal
also resulted in increased body retention of lead in the rat. Because both deficiency (Barton
et al., 1978a) and excess in calcium intake enhance retention, two sites of influence on
retention are suggested. Goyer (1978) has suggested that body retention of lead in calcium
deficiency, i.e., reduced excretion rate, may be due to renal impairment, while Quarterman et
al. (1978a) suggest that excess calcium suppresses calcium resorption from bone, hence also
reducing lead release.
10.5.2.2 Interactions of Lead with Iron. Mahaffey-Six and Goyer (1972) reported that iron-
deficient rats had increased tissue levels of lead and manifested greater toxicity compared
with control animals. This uptake change was seen with but minor alterations in hematocrit,
indicating a primary change in lead absorption over the time of the study. Barton et al.
(1978b) found that dietary restriction of iron, using 210Pb and everted sac preparations in
the rat, led to enhanced absorption of iron; iron loading suppressed the extent of lead
uptake, using normal intake levels of iron. This suggests receptor binding competition at a
common site, consistent with the isolation by these workers of two iron-binding mucosa frac-
tions. While iron level of diet affects lead absorption, the effect of changes in lead con-
tent in the gut on iron absorption is not clear. Barton et al. (1978b) and Dobbins et al.
(1978) observed no effect of lead in the gut on iron absorption in the rat, while Flanagan et
al. (1979) reported that lead reduced iron absorption in mice.
In the mouse, Hamilton (1978) found that body retention of 203Pb was unaffected by iron
deficiency, using intraperitoneal administration of the label, while gastric intubation did
lead to increased retention. Animals with adequate iron showed no changes in lead retention
at intubation levels of 0.01 to 10 nM. Cerklewski (1980) observed that lead transfer both in
utero and in milk to nursing rats was enhanced when dams were maintained from gestation
through lactation on low iron diets compared with controls.
10.5.2.3 Lead Interactions with Phosphate. The early studies of Shelling (1932), Grant et
al. (1938), and Sobel et al." (1940) documented that dietary phosphate influenced the extent of
lead toxicity and tissue retention of lead in animals, with low levels enhancing those para-
meters while excess intake retarded the effects. More recently, Barltrop and Khoo (1975)
reported that reduced phosphate increased the uptake of 203Pb approximately 2.7-fold compared
with controls. Quarterman and Morrison (1975) found that low dietary phosphate enhanced lead
retention in rats but had no effect on skeletal lead mobilization nor was injected lead
label affected by such restriction. In a related study, Quarterman et al. (1978a) found that
NEW10A/A 10-38 7/1/83
-------
PRELIMINARY DRAFT
doubling of the nutrient over normal levels resulted in lowering of lead absorption by appro-
ximately half. Barton and Conrad (1981) found that reduced dietary phosphorus increased the
retention of labeled lead and deposition in bone, in contrast to the results of Quarterman and
Morrison (1975). Increasing the intraluminal level of phosphorus reduced lead absorption,
possibly by increasing intraluminal precipitation of lead as the mixed lead/calcium phosphate.
Smith et al. (1978) reported that lead uptake occurs at the same site as phosphate, suggesting
that lead absorption may be more related to phosphate than calcium transport.
10.5.2.4 interactions of Lead with Vitamin D. Several studies had earlier indicated that a
positive relationship might exist between dietary vitamin D and lead uptake, resulting in
either greater manifestations of lead toxicity or a greater extent of lead uptake (Sobel et
al., 1938, 1940). Using the everted sac technique and testing with 210Pb, Smith et al. (1978)
observed that increasing levels of intubated vitamin D in the rat resulted in increased
absorption of the label, with uptake occurring at the distal end of the rat duodenum, the site
of phosphorus uptake and greatest stimulation by the vitamin. Barton et al. (1980) used 210Pb
to monitor lead absorption in the rat under conditions of normal, deficient, and excess
amounts of dietary vitamin D. Lead absorption is increased with either low or excess vitamin
0. This apparently occurs because of increased retention time of fecal mass containing the
lead due to alteration of intestinal motility rather than because of direct enhancement of
mucosal uptake rate. Hart and Smith (1981) reported that vitamin D repletion of diet enhanced
lead absorption (210Pb) in the rat, while also enhancing femur and kidney lead uptake when the
label was given by injection.
10.5.2.5 Interactions of Lead with Lipids. Barltrop and Khoo (1975) observed that varying
the lipid (corn oil) content of rat diet from 5 up to 40 percent resulted in an increase of
lead in blood 13.6-fold higher compared with the normal level. Concomitant increases were
observed in lead levels in kidney, femur, and carcass. Reduction of dietary lipid below the
5 percent control figure was without effect on lead absorption rate. As an extension of this
carlier work, Barltrop (1982) has noted that the chemical composition of the lipid is a signi-
ficant factor in affecting lead absorption. Study of triglycerides of saturated and unsatura-
ted fatty acids showed that polyunsaturated, trilinolein increased lead absorption by 80 per-
cent in rats, when given as 5 or 10 percent loadings in diet, compared with monounsaturated
triolein or any of the saturates in the series tricaproin to tristearin.
10 5,2.6 Lead Interaction with Protein. Quarterman et al. (1978b) have drawn attention to
of the inherent difficulties of measuring lead-protein interactions, i.e., the effect of
rotein on both growth and the toxicokinetic parameters of lead. Der et al. (1974) found that
reduction of dietary protein, from 20 to 4 percent, led to increased uptake of lead in rat
tissues, but the approximately 6-fold reduction'in body weight over the interval of the study
makes it difficult to draw any firm conclusions. Barltrop and Khoo (1975) found that lead
NEW10A/A 10-39 7/1/83
-------
PRELIMINARY DRAFT
(203Pb) uptake by rat tissue could be enhanced with either suboptimal or excess levels of pro-
tein in diet. Quarterman et al. (1978b) reported that retention of labeled lead in rats main-
tained on a synthetic diet containing approximately 7 percent protein was either unaffected or
reduced compared with controls, depending on tissues taken for study.
It appears that not only levels of protein but also the type of protein affects tissue
levels of lead. Anders et al. (1982) found that rats maintained on either of two synthetic
diets varying only as to having casein or soybean meal as the protein source showed signifi-
cantly higher lead levels in the casein group.
10.5.2.7 Interactions of Lead with Milk Components. For many years, milk was recommended
prophyTactically for lead poisoning among lead workers (Stephens and Waldron, 1975). More
recent data, however, suggest that milk may actually enhance lead uptake. Kello and Kostial
(1973) found that rats maintained on milk diets absorbed a greater amount of 203Pb than those
having access to commercial rat chow. This was ascribed to relatively lower levels of certain
nutrients in milk compared with the rat chow. These observations were confirmed by Bell and
Spickett (1981), who also observed that lactose-hydrolyzed milk was less effective than the
ordinary form in promoting lead absorption, suggesting that lactose may be the enhancing prin-
ciple. Bushnell and DeLuca (1981) demonstrated that lactose significantly increased lead
(210Pb) absorption and tissue retention by weanling rats by comparing diets identical in all
respects except for carbohydrate source. These results provide one rationale for why nursing
mammals tend to absorb greater quantities of lead than adults; lactose is the major carbohy-
drate source in suckling rats and is known to enhance the uptake of many essential metals.
10.5.2.8 Lead Interactions with Zinc and Copper. The studies of Cerklewski and Forbes (1976)
and El-Gazzar et al. (1978) documented that zinc-deficient diets promote lead absorption in
the rat, while repletion with zinc reduces lead uptake. The interaction continues within the
body, particularly with respect to ALA-D activity (see Chapter 11). In a study of zinc-lead
interactions in female rats during gestation and lactation, Cerklewski (1979) observed that
zinc-deficient diets resulted in more transfer of lead through milk to the pups as well as
reduced litter body weights.
Klauder et al. (1973) reported that low dietary copper enhanced lead absorption in rats
fed a high lead diet (5000 ppm). These observations were confirmed by Klauder and Petering
(1975) at a level of 500 ppm lead in diet. These researchers subsequently observed that
reduced copper enhanced the hematological effects of lead (Klauder and Petering, 1977), and
that both copper and iron deficiencies must be corrected to restore hemoglobin levels to
normal.
NEW10A/A 10-40 7/1/83
-------
PRELIMINARY DRAFT
10.6 INTERRELATIONSHIPS OF LEAD EXPOSURE, EXPOSURE INDICATORS, AND TISSUE LEAD BURDENS
Information presented so far in this chapter sets forth the quantitative and qualitative
aspects of lead toxicokirtetics, including the compartmental modeling of lead distribution HI
vivo, and leads up to the critical issue of the various interrelationships of lead toxico-
kinetics to lead exposure, toxicant levels in indicators of such exposure, and exposure-target
tissue burdens of lead.
Chapter 11 (Sections 11.4, 11.5, 11.6) discusses the various experimental and epidemiolo-
gical studies relating the relative impact of various routes of lead exposure on blood lead
levels in human subjects, including the description of mathematical models for such relation-
ships. In these sections, the basic question is: what is the mathematical relationship of
lead in air, food, water, etc. to lead in blood? This question is descriptive and does not
address the biological basis of the observed relationships. Nor does it consider the impli-
cations for adverse health risk in the sequence of exposure leading from external lead to lead
in some physiological indicator to lead in target tissues.
For purposes of discussion, this section separately considers 1) the temporal character-
istics of physiological indicators of lead exposure, 2) the biological aspects of the rela-
tionship of external exposure to internal indicators of exposure, and 3) internal indicator-
tissue lead relationships, including both steady-state lead exposure and abrupt changes in
lead exposure. The relationship of internal indicators of body lead, such as blood lead, to
biological indicators such as EP or urinary ALA is discussed in Chapter 13, since any compara-
tive assessment of the latter should follow the chapter on biological effects, Chapter 12.
10,B.I Temporal Characteristics of Internal Indicators of Lead Exposure
The biological half-time for blood lead or the non-retained fraction of body lead is
relatively short (see Sections 10.3 and 10.4); thus, a given blood or urine lead value
reflects rather recent exposure. In cases where lead exposure can be reliably assumed to have
occurred at a given level, a blood lead value is more useful than in cases where some inter-
mittent, high level of exposure may have occurred. The former most often occurs with occupa-
tional exposure, while the latter is of particular relevance to young children.
Accessible mineralizing tissue, such as shed teeth, extend the time frame for assessing
lead exposure from weeks or several months to years (Section 10.3), since teeth accumulate
lead up to the time of shedding or extraction. Levels of lead in teeth increase with age in
roportion to exposure (Steenhout and Pourtois, 1981). Furthermore, tooth levels are propor-
tional to blood lead levels in humans (Shapiro et al., 1978) and animals (Kaplan et al.,
1980). The technique of Fremlin and Edmonds (1980), employing micro-autoradiography of
'rradiated teeth, permits the identification of dentine zones high in lead content, thus
allowing the disclosure of past periods of abrupt increases in lead intake.
HEW10A/A
10-41 7/1/83
-------
PRELIMINARY DRAFT
While levels of lead in shed teeth are more valuable than blood lead in assessing expo-
sure at more remote time points, such information is retrospective in nature and would not be
of use in monitoring current exposure. In this case, serial blood lead measurements must be
employed. With the development of methodology for j_n situ measurement of tooth lead in chil-
dren (described in Chapter 9), serial ui situ tooth analysis in tandem with serial blood lead
determining would provide comparative data for determination of both time-concordant blood/
tooth relationships as well as which measure is the better indicator of ongoing exposure.
Given the limitations of an indicator such as blood lead in reflecting lead uptake in target
organs, as discussed below, it may well be the case that the rate of accumulation of lead in
teeth, measured j_n situ, is a better index of ongoing tissue lead uptake than is blood lead.
This aspect merits further study, especially as Shapiro et al. (1978) were able to demonstrate
the feasibility of using i_n situ tooth lead analysis in a large group of children screened for
lead exposure.
10.6.2 Biological Aspects of External Exposure-Internal Indicator Relationships
Information provided in Chapter 11 as well as the critique of Hammond et al. (1981) indi-
cate that the relationship of levels of lead in air, food, and water to lead in blood is
curvilinear, with the result that as "baseline" blood lead rises, i.e., as one moves up the
curve, the relative change in the dependent variable, blood lead, per unit change of lead in
some intake medium (such as air) becomes smaller. Conversely, as one proceeds down the curve
with reduction in "baseline" lead, the corresponding change in blood lead becomes larger. One
assumption in this "single medium" approach is that the baseline is not integrally related to
the level of lead in the particular medium being studied. This assumption is not necessarily
appropriate in the case of air vs. food lead, nor, in the case of young children, air lead vs.
total oral intake of the element.
Hammond et al. (1981) have noted that the shape of the blood lead curves seen in human
subjects is similar to that discernible in certain experimental animal studies with dogs,
rats, and rabbits (Azar et al., 1973; Prpic-Majic et al., 1973). Also, Kimmel et al. (1980)
exposed adult female rats to lead at four levels in drinking water for 6-7 weeks and reported
values of blood lead that showed curvilinear relationship to the dose levels. Over the dosing
range of 5 to 250 ppm in water, the blood lead range was 8.5 to 31 ug/dl. In a related study
(Grant et al. , 1980) rats were exposed to lead ui utero, through weaning, and up to 9 months
of age at the dosing range used in the Kimmel et al. study the weanlings, 0.5 to 250 ppm in
the dams' drinking water until weaning of pups; then the same levels in the weanlings' drink-
ing water) showed a blood lead range of 5 to 67 ug/dl. It may be assumed in all of the above
studies that lead in the various dosing groups was near or at equilibrium within the various
body compartments.
NEW10A/A 10-42 7/1/83
-------
PRELIMINARY DRAFT
The biological basis of the curvilinear relationship of blood lead to lead intake does
not appear to be due to reduced absorption or enhanced excretion of the element with changes
in exposure level. In other words, a decrease in the ratio of blood lead to medium lead as
blood lead increases cannot be taken to indicate reduced uptake rate of lead into target tis-
sues. In the study of Prpic-Majic et al. (1973), dosing was by injection so that the GI
absorption rate of lead was not a factor. Azar et al. (1973) reported values for urinary lead
across the dosing groups that indicated the excretion rate for the 10, 50, 100, and 500 ppm
dietary lead groups was fairly constant. As suggested by Hammond et al. (1981), the shape of
the blood lead curves in the context of external exposure is probably related to the tissue
distribution of lead. Other supporting evidence is the relationship of blood lead to chela-
table lead and that of tissue burden to dosing level as discussed below.
10.6.3 Internal Indicator-Tissue Lead Relationships
In living human subjects it is not possible to directly determine tissue burdens of lead
(or relate these levels to adverse effects associated with target tissue) as a function of
lead intake. Instead, measurement of lead in an accessible indicator such as blood, along
with determination of some biological indicator of impairment, e.g., ALA-U or EP, is used.
Evidence continues to accumulate in both the clinical and experimental animal literature
that the use of blood lead as an indicator has limitations in reflecting both the amounts of
lead in target tissues and the temporal changes in tissue lead with changes in exposure. Per-
haps the best example of the problem is the relationship of blood lead to chelatable lead (see
Section 10.3.3). Presently, measurement of the plumburesis associated with challenge by a
single dose of a chelating agent such as CaNa2EDTA is considered the best measure of the mo-
bile, potentially toxic, fraction of body lead in children and adults (Chisolm et al., 1976;
U.S. Centers for Disease Control, 1978; Chisolm and Barltrop, 1979; Hansen et al., 1981).
Chisolm et al. (1976) have documented that the relationship of blood lead to chelatable
lead is curvilinear, such that a given incremental increase in blood lead is associated with
an increasingly larger increment of mobilizable lead. The problems associated with this cur-
vilinear relationship in exposure assessment are typified by the recent reports of Saenger et
al (1982) concerning children and Hansen et al. (1981) concerning on adult lead workers. In
the former study, it was noted that significant percentages of children having mild to moder-
ate lead exposure, as discernible by blood lead and EP measurements, were found to have uri-
nary outputs of lead upon challenge with CaNa2EDTA qualifing them for chelation therapy under
CDC guidelines. In adult workers, Hansen et al. (1981) observed that a sizable fraction of
subjects with only modest elevations in blood lead excreted lead upon CaNa2EDTA challenge sig-
nificantly exceeding the upper end of normal. This occurred at blood lead levels of 35 ug/dl
and above.
NEW10A/A 10-43 7/1/83
-------
PRELIMINARY DRAFT
The biological basis for the non-linearity of the relationship between blood lead and
chelatable lead, appears in a major part, to be the existence of a sizeble pool of lead in
bone that is labile to chelation. Evidence pointing to this was summarized in Section 10.3.3.
The question of how long any lead in this compartment of bone remains labile to chelation has
been addressed by several investigators in studies of both children and adults. The question
is relevant to the issue of the utility of EDTA challenge in assessing evidence for past lead
exposure.
Chisolm et al. (1976) found that a group of adolescent subjects (N = 55; 12-22 yrs old),
who had a clinical history of lead poisoning as young children and whose mean blood lead was
22.1 ug/dl at the time of study, yielded chelatable lead values that placed them on the same
regression curve as a second group of young children with current elevations of blood lead.
The results with the adolescent subjects did not provide evidence that they might have had a
past history of lead poisoning. According to the authors, this suggests that chelatable lead
at the time of excessive exposure was not retained in a pool that remained labile to chelation
years later, but underwent subsequent excretion or transfer to the inert compartment of bone.
One problem with drawing conclusions from this study is that all of the adolescents apparently
had one or more courses of chelation therapy and were removed to housing where re-exposure
would be minimal as part of their clinical management after lead poisoning was diagnosed. One
must assume that chelation therapy removed a significant portion of the mobile lead burden and
placement in lead-free housing reduced the extent of any further exposure. The obvious
question is how would this group of adolescents compare with subjects who had excessive
chronic lead exposure as young children but who did not require or receive chelation therapy?
Former lead workers challenged with CaNa2EDTA show chelatable lead values that are sig-
nificantly above normal years after workplace exposure ceases (e.g., Alessio et al., 1975.
Prerovska and Teisinger, 1970). In the case of former lead workers, blood lead also remains
elevated, suggesting that the mobile lead pool in bone remains in equilibrium with blood.
The closer correspondence of chelatable lead with actual tissue lead burdens, compared to
blood lead, is also reflected in a better correlation of this parameter with such biological
indicators of impairment as EP. Saenger et al. (1982), in the study noted above, found that
the only significant correlation with erythrocyte protoporphyrin was obtained with the pM
Pb/mM EDTA ratio. Similarly, Alessio et al. (1976) found that EP in former lead workers was
more significantly correlated with chelatable lead than with blood lead.
Consideration of both the intake vs. blood lead and the blood lead vs. chelatable lead
curves leads to the prediction that the level of lead exposure per se is more closely related
to tissue lead burden than is blood lead; this appears to be the case in experimental animals
Azar et al. (1973) and Grant et al. (1980) reported that levels of lead in brain, kidney, and
femur followed more of a direct proportionality with the level of dosing than with blood lead
NEW10A/A 10-44 7/1/83
-------
PRELIMINARY DRAFT
Finally, there is the question of how adequately an internal indicator such as blood lead
reflects changes in tissue burden when exposure changes abruptly. In the study of Bjb'rklund
et al. (1981). lead levels in both blood and brain were monitored over a 6-week period in rats
exposed to lead through their drinking water. Blood lead rose rapidly by day 1, during which
time brain lead content was only slightly elevated. After day 1, the rate of increase in
blood lead began to taper off while brain lead began to rise in a near-linear fashion up to
the end of the experiment. From day 7 to 21, blood lead increased from approximately 45 to 55
pg/dl, while brain lead increased approximately 2-fold.
Abrupt reduction in exposure similarly appears to be associated with a more rapid
response in blood than in soft tissues, particularly brain. Goldstein and Diamond (1974)
reported that termination of intravenous administration of lead to 30-day-old rats resulted in
a 7-fold drop of lead in blood by day 7. At the same time, there was no significant decrease
In brain lead. A similar difference in brain vs. blood response was reported by Momcliovic
and Kostial (1974).
In all of the above studies, it may be seen that blood lead was of limited value in
reflecting changes in the brain, which is, for children, the significant target organ for lead
exposure. With abrupt increases in exposure level, the problem concerns a much more rapid
approach to steady-state in blood than in brain. Conversely, the biological half-time for
lead clearance from blood in the young rats of both the Goldstein and Diamond (1974) and
Momcilovic and Kostial (1974) studies was much less than it appeared to be for lead movement
from brain.
Despite the limitations in indexing tissue burden and exposure changes, blood lead
remains the one measure that can reliably demonstrate the relationship of various effects.
10.7 METABOLISM OF LEAD ALKYLS
The lower alkyl lead compounds used as gasoline additives, tetraethyl lead (TEL) and
tetramethyl lead (TML), are much more toxic, i.e., neurotoxic, on an equivalent dose basis
than inorganic lead. These agents are emitted in auto exhaust and their rate of environmental
degradation depends on such factors as sunlight, temperature, and ozone levels. There is also
some concern that organolead compounds may result from biomethylation in the environment (see
Chapter 6). Finally, there appears to be a problem with the practice among children of snif-
fing leaded gasoline. The available information dealing with metabolism of lead alkyls is
derived mainly from experimental animal studies, workers exposed to the agents and cases of
lead alkyl poisoning.
NEW10A/A 10-45 7/1/83
-------
PRELIMINARY DRAFT
10.7.1 Absorption of Lead Alkyls In Humans and Animals
The respiratory intake and absorption of TEL and TML in the vapor state was investigated
by Heard et al. (1979), who used human volunteers inhaling 203Pb-labeled TEL and TML. Initial
lung deposition rates were 37 and 51 percent for TEL and TML, respectively. Of these amounts
40 percent of TEL was lost by exhalation within 48 hours, while the corresponding figure for
TML was 20 percent. The remaining fraction was absorbed. The effect of gasoline vapor on
these parameters was not investigated. In this study Mortensen (1942) reported that adult
rats inhaling TEL labeled with 203Pb (0.07-7.00 mg TEL/1) absorbed 16-23 percent of the frac-
tion reaching the alveoli. Gasoline vapor had no effect on the absorption rates.
Respiratory absorption of organolead bound to particulate matter has not been specif-
ically studied as such. According to Harrison and Laxen (1978), TEL or TML does not adher to
particulate matter to any significant extent, but the toxicologically equivalent trialkyl
derivatives, formed from photolytic dissociation or ozonolysis in the atmosphere, may do so
10.7.1.1 Gastrointestinal Absorption. Information on the rate of absorption of lead alkyls
through the gastrointestinal tract is not available in the literature. Given the level of
gastric acidity (pH 1.0) in humans, one would expect TML and TEL to be rapidly converted to
the corresponding trialkyl forms, which are comparatively more stable (Bade and Huber, 1970).
Given the similarity of the chemical and biochemical behavior of trialkyl leads to their Group
IV analogs, the trialkyltins, the report of Barnes and Stoner (1958) that triethyltin is
quantitatively absorbed from the GI tract indicates that triethyl and trimethyllead would be
extensively absorbed via this route.
'10.7.1.2 Percutaneous Absorption of Lead Alkyls. In contrast to inorganic lead salts, both
TEL and TML are rapidly and extensively absorbed through the skin in rabbits and rats (Kehoe
and Thamann, 1931; Laug and Kunze, 1948), and lethal effects can be rapidly induced in these
animals by merely exposing the skin. Laug and Kunze (1948) observed that systemic uptake of
TEL was still 6.5 percent even though most of the TEL was seen to have evaporated from the
skin surface. The rate of passage of TML was somewhat slower than that of TEL in the study of
Davis et al. (1963); absorption of either agent was retarded somewhat when applied in gaso-
line.
10.7.2 Biotransformation and Tissue Distribution of Lead Alkyls
In order to have an understanding of the |n vivo fate of lead alkyls, it is useful to
first discuss the biotransformation processes of lead alkyls known to occur in mammalian
systems. Tetraethyl and tetramethyl lead both undergo oxidative dealkylation in mammals to
the triethyl or trimethyl metabolites, which are now accepted as the actual toxic forms of
these alkyls.
NEW10A/A 10-46 7/1/83
-------
PRELIMINARY DRAFT
Studies of the biochemical mechanisms for these transformations, as noted by Kimmel et
al. (1977). indicate a dealkylation mediated by a P-450 dependent mono-oxygenase system in
liver microsomes, with intermediate hydroxylation. In addition to rats (Cremer, 1959; Stevens
et al., I960; Bolanowska, 1968), mice (Hayakawa, 1972), and rabbits (Bolanowska and
Garczyriski, 1968) this transformation also occurs in humans accidentally poisoned with TEL
(Bolanowska et al., 1967) or workers chronically exposed to TEL (Adamiak-Ziemka and
Bolanowska, 1970).
The rate of hepatic oxidative de-ethylation of TEL in mammals appears to be rather rapid;
Cremer (1959) reported a maximum conversion rate of approximately 200 (jg TEL/g rat liver/hour.
In comparison with TEL, TML may undergo transformation at either a slower rate (in rats) or
more rapidly (in mice), according to Cremer and Calloway (1961) and Hayakawa (1972).
Other transformation steps involve conversion of triethyl lead to diethyl form, the pro-
cess appearing to be species-dependent. Bolanowska (1968) did not report the formation of
diethyl lead in rats, while significant amounts of it are present in the urine of rabbits
(Arai et al., 1981) and humans (Chiesura, 1970). Inorganic lead is formed in various species
treated with tetraethyl lead, which may arise from degradation of the diethyl lead metabolite
or some other direct process (Bolanowska, 1968). The latter process appears to occur in rats,
as little or no diethyllead is found, whereas significant amounts of inorganic lead are
present. Formation of inorganic lead with lead alkyl exposure may account for the hematolo-
gical effects seen in humans chronically exposed to the lead alkyls (see Section 12.3),
including children who inhale leaded gasoline vapor.
Partitioning of triethyl or trimethyl lead, the corresponding active metabolites of TEL
and TML, between the erythrocyte and plasma appears to be species-dependent. Byington et al.
(1980) studied the partitioning of triethyl lead between cells and plasma i_n vitro using
washed human and rat erythrocytes and found that human cells had a very low affinity for the
allcyl lead while rat cells bound the alkyl lead in the globin moiety at a ratio of three mole-
cules per Hb tetramer. Similarly, it was found that injected triethyl lead was associated
with whole blood levels approximately 10-fold greater than in rat plasma. The available
literature on TEL poisoning in humans concurs, as significant plasma values of lead have been
routinely reported (Boeckx et al., 1977; Golding and Stewart, 1982). These data indicate that
the rat is a poor model to use in studying the adverse effects of lead alkyl s in human sub-
jects.
The biological half-time in blood for the lead alkyls depends on whether clearance of the
tetraalkyl or trialkyl forms is being observed. Heard et al. (1979) found that 203Pb-labeled
TML and TEL inhaled by human volunteers was rapidly cleared from blood (by 10 hours), followed
by a reappearance of lead. The fraction of lead in plasma initially was quite high, approxi-
mately 0.7, suggesting tetra/trialkyl lead; but the subsequent rise in blood lead showed all
NEW10A/A 10-47 7/1/83
-------
PRELIMINARY DRAFT
of it essentially present in the cell, which would indicate inorganic or possibly diethyl
lead. Triethyl lead in rabbits was more rapidly cleared from the blood of rabbits (3-5 days)
than was the trimethyl form (15 days) when administered as such (Hayakawa, 1972).
Tissue distribution of lead in both humans and animals exposed to TEL and TML primarily
involves the trialkyl metabolites. Levels are highest in liver, followed by kidney, then
brain (Bolanowska et al., 1967; Grandjean and Nielsen, 1979). Nielsen et al. (1978) observed
that measurable amounts of trialkyl lead were present in samples of brain tissue from subjects
with no known occupational exposure.
The available studies on tissue retention of triethyl or trimethyl lead provide variable
findings. Bolanowska (1968) noted that tissue levels of triethyl lead in rats were almost
constant for 16 days after a single injection of TEL. Hayakawa (1972) found that the half-
time of triethyl lead in brain was 7-8 days for rats; the half-time for trimethyl lead was
much longer. In humans, Yamamura et al. (1975) reported two tissue compartments for triethyl
lead having half-times of 35 and 100 days (Yamamura et al., 1975).
10.7.3 Excretion of Lead Alkyls
Excretion of lead through the renal tract is the main route of elimination in various
species exposed to lead alkyls (Grandjean and Nielsen, 1979). The chemical forms of lead in
urine suggest that the differing amounts of the various forms are species-dependent. Arai et
al. (1981) found that rabbits given TEL parenterally excreted lead primarily in the form of
diethyl lead (69 percent) and inorganic lead (27 percent), triethyl lead accounting only for 4
percent. In rats, Bolanowska and Garczynski (1968) found that levels of triethyl lead were
somewhat higher in urine than was the case for rabbits. In humans, Chiesura (1970) found that
trialkyl lead never was greater than 9 percent of total lead content in workers with heavy TEL
exposure. Adamiak-Ziemka and Bolanowska (1970) reported similar data; the fraction of tri-
ethyl lead in the urine was approximately 10 percent of total lead.
The urinary rates of lead excretion in human subjects with known levels of TEL exposure
were also reported by Adamiak-Ziemka and Bolanowska (1970). In workers involved with the
blending and testing of leaded gasoline, workplace air levels of lead (as TEL) ranged from
0.037 to 0.289 mg Pb/m3 and the corresponding urine levels ranged from 14 to 49 ug Pb/1, Of
which approximately 10 percent was triethyl lead.
NEW10A/A 10-48 7/1/83
-------
PRELIMINARY DRAFT
10.8 SUMMARY
Toxicokinetic parameters of lead absorption, distribution, retention, and excretion con-
necting external environmental lead exposure to various adverse effects are discussed in this
section. Also considered are various influences on these parameters, e.g., nutritional
status, age, and stage of development.
A number of specific issues in lead metabolism by animals and humans merit special focus
and these include:
1. How does the developing organism from gestation to maturity differ from the adult in
toxicokinetic response to lead intake?
2. What do these differences in lead metabolism portend for relative risk for adverse
effects?
3. What are the factors that significantly change the toxicokinetic parameters in ways
relevant to assessing health risk?
4. How do the various interrelationships among body compartments for lead translate to
assessment of internal exposure and changes in internal exposure?
10.8.1 Lead Absorption in Humans and Animals
The amounts of lead entering the bloodstream via various routes of absorption are influ-
enced not only by the levels of the element in a given medium but also by various physical and
chemical parameters and specific host factors, such as age and nutritional status.
10.8.1-1 Respiratory Absorption of Lead. The movement of lead from ambient air to the blood-
stream is a two-part process: deposition of some fraction of inhaled air lead in the deeper
part of the respiratory tract and absorption of the deposited fraction. For adult humans, the
deposition rate of particulate airborne lead as likely encountered by the general population
Is around 30-50 percent, with these rates being modified by such factors as particle size and
ventilation rates. It also appears that essentially all of the lead deposited in the lower
respiratory tract is absorbed, so that the overall absorption rate is governed by the deposi-
tion rate, i.e., approximately 30-50 percent. Autopsy results showing no lead accumulation in
the lung indicate quantitative absorption of deposited lead.
All of the available data for lead uptake via the respiratory tract in humans have been
obtained with adults. Respiratory uptake of lead in children, while not fully quantifiable,
appears to be comparatively greater on a body weight basis, compared to adults. A second fac-
tor influencing the relative deposition rate in children has to do with airway dimensions.
One report has estimated that the 10-year-old child has a deposition rate 1.6- to 2.7-fold
higher than the adult on a weight basis.
It appears that the chemical form of the lead compound inhaled is not a major determinant
of the extent of alveolar absorption of lead. While experimental animal data for quantitative
assessment of lead deposition and absorption for the lung and upper respiratory tract are
NEW10B/A 10-49 7/1/83
-------
PRELIMINARY DRAFT
limited, available information from the rat, rabbit, dog, and nonhuman primate support the
findings that respired lead in humans is extensively and rapidly absorbed.
10.8.1.2 Gastrointestinal Absorption of Lead. Gastrointestinal absorption of lead mainly
involves lead uptake from food and beverages as well as lead deposited in the upper respira-
tory tract which is eventually swallowed. It also includes ingestion of non-food material,
primarily in children via normal mouthing activity and pica. Two issues of concern with lead
uptake from the gut are the comparative rates of such absorption in developing vs. adult
organisms, including humans, and how the relative bioavailability of lead affects such uptake.
By use of metabolic balance and isotopic (radioisotope or stable isotope) studies, var-
ious laboratories have provided estimates of lead absorption in the human adult on the order
of 10-15 percent. This rate can be significantly increased under fasting conditions to 45
percent, compared to lead ingested with food. The latter figure also suggests that beverage
lead is absorbed to a greater degree since much beverage ingestion occurs between meals.
The relationship of the chemical/biochemical form of lead in the gut to absorption rate
has been studied, although interpretation is complicated by the relatively small amounts given
and the presence of various components in food already present in the gut. In general, how-
ever, chemical forms of lead or their incorporation into biological matrices seems to have a
minimal impact on lead absorption in the human gut. Several studies have focused on the ques-
tion of differences in gastrointestinal absorption rates for lead between children and adults.
It would appear that such rates for children are considerably higher than for adults: 10-15
percent for adults vs. approximately 50 percent for children. Available data for the absorp-
tion of lead from non-food items such as dust and dirt on hands are limited, but one study has
estimated a figure of 30 percent. For paint chips, a value of about 17 percent has been esti-
mated.
Experimental animal studies show that, like humans, the adult absorbs much less lead from
the gut than the developing animal. Adult rats maintained on ordinary rat chow absorb 1 per-
cent or less of the dietary lead. Various animal species studies make it clear that the new-
born absorbs a much greater amount of lead than the adult, supporting studies showing this age
dependency in humans. Compared to an absorption rate of about 1 percent in adult rats, the
rat pup has a rate 40-50 times greater. Part, but not most, of the difference can be ascribed
to a difference in dietary composition. In nonhuman primates, infant monkeys absorb 65-85
percent of lead from the gut, compared to 4 percent for the adults.
The bioavailability of lead in the gastrointestinal (GI) tract as a factor in its absorp-
tion has been the focus of a number of experimental studies. These data show that: 1) iead
in a number of forms is absorbed about equally, except for the sulfide; 2) lead in dirt and
dust and as different chemical forms is absorbed at about the same rate as pure lead salts
NEW10B/A 10-50 7/1/83
-------
PRELIMINARY DRAFT
added to diet; 3) lead in paint chips undergoes significant uptake from the gut; and 4) in
some cases, physical size of particulate lead can affect the rate of GI absorption.
10.8.1-3 Percutaneous Absorption of Lead. Absorption of inorganic lead compounds through the
skin is of much less significance than through the respiratory and gastrointestinal routes.
This is in contrast to the case with lead alkyls (See Section 1.10.6). One recent study using
human volunteers and 203Pb-labeled lead acetate showed that under normal conditions, absorp-
tion approaches 0.06 percent.
10.8.1.4 Transplacental Transfer of Lead. Lead uptake by the human and animal fetus readily
occurs, such transfer going on by the 12th week of gestation in humans, with increasing fetal
uptake throughout development. Cord blood contains significant amounts of lead, correlating
with but somewhat lower than maternal blood lead levels. Evidence for such transfer, besides
lead content of cord blood, includes fetal tissue analyses and reduction in maternal blood
lead during pregnancy. There also appears to be a seasonal effect on the fetus, summer-born
children showing a trend to higher blood lead levels than those born in the spring.
10.8.2 Distribution of Lead in Humans and Animals
In this subsection, the distributional characteristics of lead in various portions of the
body-blood, soft tissue, calcified tissue, and the "chelatable" or potentially toxic body
burden—are discussed as a function of such variables as exposure history and age.
10.8.2.1 Lead in Blood. More than 99 percent of blood lead is associated with the erythro-
cyte in humans under steady-state conditions, but it is the very small fraction transported in
plasma and extracellular fluid that provides lead to the various body organs. Most (~50 per-
cent) of erythrocyte lead is bound within the cell, primarily associated with hemoglobin (par-
ticularly HbA2), with approximately 5 percent bound to a 10,000-dalton fraction, 20 percent to
a heavier molecule, and 25 percent to lower weight species.
Whole blood lead in daily equilibrium with other compartments in adult humans appears to
have a biological half-time of 25-28 days and comprises about 1.9 mg in total lead content.
Human blood lead responds rather quickly to abrupt changes in exposure. With increased lead
intake, blood lead achieves a new value in approximately 40-60 days, while a decrease in expo-
sure may be associated with variable new blood values, depending upon the exposure history.
This dependence presumably reflects lead resorption from bone. With age, furthermore, there
appears to be little change in blood lead during adulthood. Levels of lead in blood of child-
ren tend to show a peaking trend at 2-3 years of age, probably due to mouthing activity, fol-
lowed by a decline. In older children and adults, levels of lead are sex-related, females
showing lower levels than men even at comparable levels of exposure.
NEW10B/A 10-51 7/1/83
-------
PRELIMINARY DRAFT
In plasma, lead is virtually all bound to albumin and only trace amounts to high weight
globulins. It is not possible to state which binding form constitutes an "active" fraction
for movement to tissues. The most recent studies of the erythrocyte-plasma relationship in
humans indicate that there is an equilibrium between these blood compartments, such that
levels in plasma rise with levels in whole blood.
10.8.2.2 Lead Levels in Tissues. Of necessity, various relationships of tissue lead to expo-
sure and toxicity in humans must generally be obtained from autopsy samples. Limitations on
such data include questions of how samples represent lead behavior in the living population,
particularly with reference to prolonged illness and disease states. The adequate characteri-
zation of exposure for victims of fatal accidents is a problem, as is the fact that such
studies are cross-sectional in nature, with different age groups assumed to have had similar
exposure in the past.
10.8.2.2.1 Soft tissues. After age 20, most soft tissues in humans do not show age-related
changes; in contrast to bone. Kidney cortex shows increase in lead with age which may be
associated with formation of nuclear inclusion bodies. Absence of lead accumulation in most
soft tissues is due to a turnover rate for lead which is similar to that in blood.
Based on several autopsy studies, it appears that soft tissue lead content for individ-
uals not occupationally exposed is generally below 0.5 ug/g wet weight, with higher values for
aorta and kidney cortex. Brain tissue lead level is generally below 0.2 ppm wet weight with
no change with increasing age, although the cross-sectional nature of these data would make
changes in low brain lead levels difficult to discern. Autopsy data for both children and
adults indicate that lead is selectively accumulated in the hippocampus, a finding that Is
also consistent with the reginal distribution in experimental animals.
Comparisons of lead levels in soft tissue autopsy samples from children with results from
adults indicate that such values are lower in infants than in older children, while children
aged 1-16 years had levels comparable to adult women. In one study, lead content of brain
regions did not materially differ for infants and older children compared to adults. Compli-
cating these data somewhat are changes in tissue mass with age, although such changes are less
than for the skeletal system.
Subcellular distribution of lead in soft tissue is not uniform, with high amounts of lead
being sequestered in the mitochondria and nucleus. Nuclear accumulation is consistent with
the existence of lead-containing nuclear inclusions in various species and a large body of
data demonstrating the sensitivity of mitochondria to injury by lead.
10.8.2.2.2 Mineralizing tissue. Lead becomes localized and accumulates in human calcified
tissues, i.e., bones and teeth. This accumulation in humans begins with fetal development and,
continues to approximately 60 years of age. The extent of lead accumulation in bone ranges up
NEW10B/A 10-52 7/1/83
-------
PRELIMINARY DRAFT
to 200 mg In men ages 60-70 years, while in women lower values have been measured. Based upon
various studies, approximately 95 percent of total body lead is lodged in the bones of human
adults, with uptake distributed over trabecular and compact bone. In the human adult, bone
lead is both the most inert and largest body pool, and accumulation can serve to maintain ele-
vated blood lead levels years after exposure, particularly occupational exposure, has ended.
Compared to the human adult, 73 percent of body lead is lodged in the bones of children,
which is consistent with other information that the skeletal system of children is more meta-
bolically active than in the adult. While the increase in bone lead across childhood is mod-
est, about 2-fold if expressed as concentration, the total accumulation rate is actually 80-
fold, taking into account a 40-fold increase in skeletal mass. To the extent that some sig-
nificant fraction of total bone lead in children and adults is relatively labile, it is more
appropriate in terms of health risk for the whole organism to consider the total accumulation
rather than just changes in concentration.
The traditional view that the skeletal system was a "total" sink for body lead (and by
implication a biological safety feature to permit significant exposure in industrialized popu-
lations) never did accord with even older information on bone physiology, e.g., bone remodel-
ling, and is now giving way to the view that there are at least several bone compartments for
lead, with different mobility profiles. It would appear, then, that "bone lead" may be more
of an insidious source of long-term internal exposure than a sink for the element. This
aspect of the issue is summarized more fully in the next section. Available information from
studies of such subjects as uranium miners and human volunteers ingesting stable isotopes
Indicates that there is a relatively inert bone compartment for lead, having a half-time of
several decades, and a rather labile compartment which permits an equilibrium between bone and
tissue lead.
Tooth lead also increases with age at a rate proportional to exposure and roughly propor-
tional to blood lead in humans and experimental animals. Dentine lead is perhaps the most
responsive component of teeth to lead exposure since it is laid down from the time of eruption
until shedding. It is this characteristic which underlies the utility of dentine lead levels
In assessing long-term exposure.
10.8.2.2.3 Chelatable lead. Mobile lead in organs and systems is potentially more active
toxicologically in terms of being available to biological sites of action. Hence, this frac-
tion of total body lead burden is a more significant predictor of imminent toxicity. In
reality, direct measurement of such a fraction in human subjects would not be possible. In
this regard, "chelatable" lead, measured as the extent of plumburesis in response to admini-
stration of a chelating agent, is not viewed as the most useful probe of undue body burden in
children and adults.
NEW10B/A 10-53 7/1/83
-------
PRELIMINARY DRAFT
A quantitative description of the inputs to the body lead fraction that is chelant-
mobilizable is difficult to fully define, but it most likely includes a labile lead compart-
ment within bone as well as in soft tissues. Support for this view includes: 1) the age
dependency of chelatable lead, but not lead in blood or soft tissues; 2) evidence of removal
of bone lead in chelation studies with experimental animals; 3) jm vitro studies of lead
mobilization in bone organ explants under closely defined conditions; 4) tracer modelling
estimates in human subjects; and 5) the complex nonlinear relationship of blood lead and lead
intake through various media. Data for children and adults showing a logarithmic relationship
of chelatable lead to blood lead and the phenomenon of "rebound" in blood lead elevation after
chelation therapy regimens (without obvious external re-exposure) offer further support.
10.8.2.2.4 Animal studies. Animal studies have been of help in sorting out some of the rela-
tionships of lead exposure to ui vivo distribution of the element, particularly the impact of
skeletal lead on whole body retention. In rats, lead administration results in an initial
increase in soft tissues, followed by loss from soft tissue via excretion and transfer to
bone. Lead distribution appears to be relatively independent of dose. Other studies have
shown that lead loss from organs follows first-order kinetics except for bone, and the skele-
tal system in rats and mice is the kinetically rate-limiting step in whole-body lead clear-
ance.
The neonatal animal seems to retain proportionally higher levels of tissue lead compared
to the adult and manifests slow decay of brain lead levels while showing a significant decline
over time in other tissues. This appears to be the result of enhanced lead entry to the brain
because of a poorly developed brain barrier system as well as enhanced body retention of lead
by young animals.
The effects of such changes as metabolic stress and nutritional status on body redistri-
bution of lead have been noted. Lactating mice, for example, are known to demonstrate tissue
redistribution of lead, specifically bone lead resorption with subsequent transfer of both
lead and calcium from mother to pups.
10.8.3 Lead Excretion and Retention in Humans and Animals
10.8.3.1 Human Studies. Dietary lead in humans and animals that is not absorbed passes
through the gastrointestinal tract and is eliminated with feces, as is the fraction of air
lead that is swallowed and not absorbed. Lead entering the bloodstream and not retained is
excreted through the renal and GI tracts, the latter via biliary clearance. The amounts
excreted through these routes are a function of such factors as species, age, and exposure
characteristics.
NEW10B/A 10-54 7/1/83
-------
PRELIMINARY DRAFT
Based upon the human metabolic balance data and isotope excretion findings of various
investigators, it appears that short-term lead excretion in adult humans amounts to 50-60 per-
cent of the absorbed fraction, with the balance moving primarily to bone and some fraction
(approximately half) of this stored amount eventually being excreted. This overall retention
figure of 25 percent necessarily assumes that isotope clearance reflects that for body lead in
all compartments. The rapidly excreted fraction has a biological half-time of 20-25 days,
similar to that for lead removal from blood. This similarity indicates a steady rate of lead
clearance from the body. In terms of partitioning of excreted lead between urine and bile,
one study indicates that the biliary clearance is about 50 percent that of renal clearance.
Lead is accumulated in the human body with age, mainly in bone, up to around 60 years of
age, when a decrease occurs with changes in intake as well as in bone mineral metabolism. As
noted earlier, the total amount of lead in long-term retention can approach 200 mg, and even
much higher in the case of occupational exposure. This corresponds to a lifetime average
retention rate of 9-10 ug Pg/day. Within shorter time frames, however, retention will vary
considerably due to such factors as development, disruption in the individuals' equilibrium
with lead intake, and the onset of such states as osteoporosis.
The age dependency of lead retention/excretion in humans has not been well studied, but
most of the available information indicates that children, particularly infants, retain a sig-
nificantly higher amount of lead. While autopsy data indicate that pediatric subjects at iso-
lated points in time actually have a lower fraction of body lead lodged in bone, a full under-
standing of longer-term retention over childhood must consider the exponential growth rate oc-
curring in a child's skeletal system over the time period for which bone lead concentrations
have been gathered. This parameter itself represents a 40-fold mass increase. This signifi-
cant skeletal growth rate has an impact on an obvious question: if children take in more lead
on a body weight basis than adults, absorb and retain more lead than adults, and show only
modest elevations in blood lead compared to adults in the face of a more active skeletal sys-
tem where does the lead go? A second factor is the assumption that blood lead in children
relates to body lead burden in the same quantitative fashion as in adults, an assumption that
remains to be adequately proven.
10.8.3-2 Animal Studies. In rats and other experimental animals, both urinary and fecal
excretion appear to be important routes of lead removal from the organism; the relative parti-
tioning between the two modes is species- and dose-dependent. With regard to species differ-
ences, biliary clearance of lead in the dog is but 2 percent of that for the rat, while such
excretion in the rabbit is 50 percent that of the rat.
Lead movement from laboratory animals to their offspring via milk constituents is a route
of excretion for the mother as well as an exposure route for the young. Comparative studies
NEW10B/A 10-55 7/1/83
-------
PRELIMINARY DRAFT
of lead retention in developing vs. adult animals, e.g., rats, mice, and non-human primates
make it clear that retention is significantly greater in the young animal. These observations
support those studies showing greater lead retention in children. Some recent data indicate
that a differential retention of lead in young rats persists into the post-weaning period
calculated as either uniform dosing or uniform exposure.
10.8.4 Interactions of Lead with Essential Metals and Other Factors
Toxic elements such as lead are affected in their toxicokinetic or toxicological behavior
by interactions with a variety of biochemical factors such as nutrients.
10.8.4.1 Human Studies. In humans the interactive behavior of lead and various nutritional
factors is expressed most significantly in young children, with such interactions occurring
against a backdrop of rather widespread deficiencies in a number of nutritional components
Various surveys have indicated that deficiency in iron, calcium, zinc, and vitamins are wide-
spread among the pediatric population, particularly the poor. A number of reports have docu-
mented the association of lead absorption with suboptimal nutritional states for iron and cal-
cium, reduced intake being associated with increased lead absorption.
10.8.4.2 Animal Studies. Reports of lead-nutrient interactions in experimental animals have
generally described such relationships for a single nutrient, using relative absorption or
tissue retention in the animal to index the effect. Most of the recent data are for calcium
iron, phosphorus, and vitamin D. Many studies have established that diminished dietary calci-
um is associated with increased blood and soft tissue lead content in such diverse species as
the rat, pig, horse, sheep, and domestic fowl. The increased body burden of lead arises from
both increased GI absorption and increased retention, indicating that the lead-calcium inter-
action operates at both the gut wall and within body compartments. Lead appears to traverse
the gut via both passive and active transfer, involves transport proteins normally operating
for calcium transport, and is taken up at the site of phosphorus, not calcium, absorption.
Iron deficiency is associated with an increase in lead of tissues and increased toxicity
an effect which is expressed at the level of lead uptake by the gut wall. Iji vitro studies
indicate an interaction through receptor binding competition at a common site. This probably
involves iron-binding proteins. Similarly, dietary phosphate deficiency enhances the extent
of lead retention and toxicity via increased uptake of lead at the gut wall, both lead and
phosphate being absorbed at the same site in the small intestine. Results of various studies
of the resorption of phosphate along with lead as one further mechanism of elevation of tissue
lead have not been conclusive. Since calcium plus phosphate retards lead absorption to a
greater degree than simply the sums of the interactions, it has been postulated that an insol-
uble complex of all these elements may be the basis of this retardation.
NEW10B/A 10-56 7/1/83
-------
PRELIMINARY DRAFT
Unlike the inverse relationship existing for calcium, iron, and phosphate vs. lead
uptake, vitamin D levels appear to be directly related to the rate of lead absorption from the
GI tract, since the vitamin stimulates the same region of the duodenum where lead is absorbed.
A number of other nutrient factors are known to have an interactive relationship with lead:
1. Increases in dietary lipids increase the extent of lead absorption, with the extent
of the increase being highest with polyunsaturates and lowest with saturated fats,
e.g., tristearin.
2. The interactive relationship of lead and dietary protein is not clearcut, and either
suboptimal or excess protein intake will increase lead absorption.
3. Certain milk components, particularly lactose, will greatly enhance lead absorption
in the nursing animal.
4. Zinc deficiency promotes lead absorption as does reduced dietary copper.
10.8.5 Interrelationships of Lead Exposure with Exposure Indicators and Tissue Lead Burdens
There are three issues involving lead toxicokinetics which evolve toward a full connec-
tion between lead exposure and its adverse effects: 1) the temporal characteristics of inter-
nal indices of lead exposure; 2) the biological aspects of the relationship of lead in vari-
ous media to various indicators in internal exposure; and 3) the relationship of various
internal indicators of exposure to target tissue lead burdens.
10.8.5.1 Temporal Characteristics of Internal Indicators of Lead Exposure. The biological
half-time for newly absorbed lead in blood appears to be of the order of weeks or several
months, so that this medium reflects relatively recent exposure. If recent exposure is fairly
representative of exposure over a considerable period of time, e.g., exposure of lead workers,
then blood lead is more useful than for cases where exposure is intermittent or different
across time, as in the case of lead exposure of children. Accessible mineralized tissue, such
as shed teeth, extend the time frame back to years of exposure, since teeth accumulate lead
with age and as a function of the extent of exposure. Such measurements are, however, retro-
spective in nature, in that identification of excessive exposure occurs after the fact and
thus limits the possibility of timely medical intervention, exposure abatement, or regulatory
policy concerned with ongoing control strategies.
Perhaps the most practical solution to the dilemma posed by both tooth and blood lead
analyse5 is in situ measurement of lead in teeth or bone during the time when active accumu-
lation occurs, e.g., 2-3-year-old children. Available data using X-ray fluorescence analysis
do suggest that such approaches are feasible and can be reconciled with such issues as accept-
able radiation hazard risk to subjects.
-Q 3,5.2 Biological Aspects of External Exposure-Internal Indicator Relationships. It is
clear from a reading of the literature that the relationship of lead in relevant media for
human exposure to blood lead is curvilinear when viewed over a relatively broad range of blood
NEW10B/A
10-57 7/1/83
-------
PRELIMINARY DRAFT
lead values. This implies that the unit change in blood lead per unit intake of lead in some
medium varies across this range of exposure, with comparatively smaller blood lead changes as
internal exposure increases.
Given our present knowledge, such a relationship cannot be taken to mean that body uptake
of lead is proportionately lower at higher exposure, for it may simply mean that blood lead
becomes an increasingly unreliable measure of target tissue lead burden with increasing expo-
sure. While the basis of the curvilinear relationship remains to be identified, available
animal data suggest that it does not reflect exposure-dependent absorption or excretion rates.
10.8.5.3 Internal Indicator-Tissue Lead Relationships. In living human subjects, it is not
possible to directly determine tissue lead burdens or how these relate to adverse effects in
target tissues; some accessible indicator, e.g., lead in a medium such as blood or a biochem-
ical surrogate of lead such as EP, must be employed. While blood lead still remains the only
practical measure of excessive lead exposure and health risk, evidence continues to accumulate
that such an index has limitations in either reflecting tissue lead burdens or changes in such
tissues with changes in exposure.
At present, the measurement of plumburesis associated with challenge by a single dose of
a lead chelating agent such as CaNa2EDTA is considered the best indicator of the mobile,
potentially toxic fraction of body lead. Chelatable lead is logarithmically related to blood
lead, such that incremental increase in blood lead is associated with an increasingly larger
increment of mobilizable lead. The problems associated with this logarithmic relationship may
be seen in studies of children and lead workers in whom moderate elevation in blood lead can
•disguise levels of mobile body lead. This reduces the margin of protection against severe
intoxication. The biological basis of the logarithmic chelatable lead-blood lead relationship
rests, in large measure, with the existence of a sizable bone lead compartment that is mobile
enough to undergo chelation removal and, hence, potentially mobile enough to move into target
tissues.
Studies of the relative mobility of chelatable lead over time indicate that, in former
lead workers, removal from exposure leads to a protracted washing out of lead (from bone
resorption of lead) to blood and tissues, with preservation of a bone burden amenable to sub-
sequent chelation. Studies with children are inconclusive, since the one investigation
directed to th'is end employed pediatric subjects who all underwent chelation therapy during
periods of severe lead poisoning. Animal studies demonstrate that changes in blood lead with
increasing exposure do not agree with tissue uptake in a time-concordant fasion, nor does
decrease in blood lead with reduced exposure signal a similar decrease in target tissue, par-
ticularly in the brain of the developing organism.
NEW10B/A 10-58 7/1/83
-------
PRELIMINARY DRAFT
jO.8.6 Metabolism of Lead AlkyIs
The lower alkyl lead components used as gasoline additives, tetraethyl lead (TEL) and
tetramethyl lead (TML), may themselves poise a toxic risk to humans. In particular, there is
among children a problem of sniffing leaded gasoline.
10.8.6.1 Absorption of Lead Alkyls in Humans and Animals. Human volunteers inhaling labeled
TEL and TML show lung deposition rates for the lead alkyls of 37 and 51 percent, respectively,
values which are similar to those for particulate inorganic lead. Significant portions of
these deposited amounts were eventually absorbed. Respiratory absorption of organolead bound
to particulate matter has not been specifically studied as such.
While specific data for the GI absorption of lead alkyls in humans and animals are not
available, their close similarity to organotin compounds, which are quantitatively absorbed,
would argue for extensive GI absorption. In contrast to inorganic lead salts, the lower lead
alkyls are extensively absorbed through the skin and animal data show lethal effects with per-
cutaneous uptake as the sole route of exposure.
10.8.6.2 Biotransformation and Tissue Distribution of Lead Alkyls. The lower lead alkyls TEL
and TML undergo monodealkylation in the liver of mammalian species via the P-450-dependent
mono-oxygenase enzyme system. Such transformation is very rapid. Further transformation
involves conversion to the dialkyl and inorganic lead forms, the latter accounting for the
effects on heme biosynthesis and erythropoiesis observed in alkyl lead intoxication. Alykl
lead is rapidly cleared from blood, shows a higher partitioning into plasma than inorganic
lead with triethyl lead clearance being more rapid than the methyl analog.
Tissue distribution of alkyl lead in humans and animals primarily involves the trialkyl
metabolites. Levels are highest in liver, followed by kidney, then brain. Of interest is the
fact that there are detectable amounts of trialkyl lead from autopsy samples of human brain
even in the absence of occupational exposure. In humans, there appear to be two tissue com-
partments for triethyl lead, having half-times of 35 and 100 days.
10.8.6.3 Excretion of Lead Alkyls. With alkyl lead exposure, excretion of lead through the
renal tract is the main route of elimination. The chemical forms being excreted appear to be
species-dependent. In humans, trialkyl lead in workers chronically exposed to alkyl lead is a
minor component of urine lead, approximately 9 percent.
NEW10B/A 10-59 7/1/83
-------
PRELIMINARY DRAFT
10.9 REFERENCES
Adamiak-Ziemba, J. ; Bolanowska, W. (1970) Proba oceny ekspozycji zawodowej na czteroetylek
o/owiu na podstawie jego stezen w powietrzu i stezen trojetylku oJTowiu w moczu. [Occupa-
tional exposure to tetraethyl lead as estimated by tetraethyl lead concentration in air
and triethyl lead concentration in urine.] Med. Pr. 21: 172-179.
Alessio, L. ; Bertazzi, P. A.; Monelli, 0.; Toffoletto, F. (1976a) Free erythrocyte proto-
porphyrin as an indicator of the biological effect of lead in adult males. Ill: Behavior
of free erythrocyte protoporphyrin in workers with past lead exposure. Int. Arch. Occup
Environ. Health 37: 77-86.
Alexander, F. W. ; Delves, H. T. (1981) Blood lead levels during pregnancy. Int. Arch. Occup
Environ. Health 48: 35-39.
Alexander, f. W.; Delves, H. T.; Clayton, B. E. (1973) The uptake and excretion by children of
lead and other contaminants. In: Barth, D.; Berlin, A.; Engel, R.; Recht, P.; Smeets, J.
eds. Environmental health aspects of lead: proceedings, international symposium; October
1972; Amsterdam, The Netherlands. Luxembourg: Commission of the European Communities
Centre for Information and Documentation; pp. 319-330. '
Allcroft, R. (1950) Lead as a nutritional hazard to farm livestock. IV: Distribution of lead
in the tissues of bovines after ingestion of various lead compounds. J. Comp. Pathol 60-
190-208.
Anders, E.; Bagnell, C. R., Jr.; Krigman, M. R.; Mushak, P. (1982) Influence of dietary pro-
tein composition on lead absorption in rats. Bull. Environ. Contain. Toxicol. 28: 61-67.
Araki, S. (1973) On the behaviour of active deposit of lead (Teisinger) in the Japanese free
from occupational exposure to lead. Ind. Health 11: 203-224.
Araki, S. ; Ushio, K. (1982) Assessment of the body burden of chelatable lead: a model and its
application to lead workers. Br. J. Ind. Med. 39: 157-160.
Arai, F. ; Yamamura, Y.; Yoshida, M. (1981) Excretion of triethyl lead, diethyl lead and in-
organic lead after injection of tetraethyl lead in rabbits. Sangyo Igaku 23: 496-504.
Aungst, B. J.; Dolce, J. A.; 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.
Awad, L.; Huel, G.; Lazar, P.; Boudene, C. (1981) Facteurs de variation interindividuelle de
la plombe'mie. [Factors of interindividual variations of blood lead levels.] Rev.
Epidemiol. Sante Publique 29: 113-124.
Azar, A.; Trochimowicz, H. J.; Maxfield, M. E. (1973) Review of lead studies in animals
carried out at Haskell Laboratory: two year feedings study and response to hemmorhage
study. In: Barth, D.; Berlin, A.; Engel, R.; Recht, P.; Smeets, J., eds. Environmental
health aspects of lead: proceedings, international symposium; October 1972; Amsterdam,
The Netherlands. Luxembourg: Commission of the European Communities, Centre for Informa-
tion and Documentation; pp. 199-210.
SR010REF/A 10-60 7/1/83
-------
PRELIMINARY DRAFT
Bade, V. ; Huber, F. (1970) Reaktionen von Organometallverbindungen. IV: Uber den Einfluss der
'organischen Liganden und das Zentralatoms auf die Geschwindigkeit der Acidolyse von
Tetraorganoplumbanen und Tetraorganostannanen. [Reactions of organometallic compounds.
IV: The influence of the organic ligand and the central atom on the speed of acidolysis
of organolead compounds and analogous tin compounds.] J. Organometal. Chem. 24: 387-397.
pa1oh, R- W. (1974) Laboratory diagnosis of increased lead absorption. Arch. Environ. Health
28: 198-208.
Barltrop, 0. (1969) Transfer of lead to the human foetus. In: Barltrop, 0.; Burland, W. L.,
eds. Mineral metabolism in pediatrics. Philadelphia, PA: Davis Co.; pp. 135-151.
Barltrop, D. (1972) Children and environmental lead. In: Hepple, P., ed. Lead in the environ-
ment: proceedings of a conference; London, United Kingdom. London, United Kingdom:
Institute of Petroleum; pp. 52-60.
Barltrop, 0. (1975) Assessment of the health hazard of various lead compounds. Atlanta, GA:
U.S. Department of Health, Education and Welfare, Centers for Disease Control.
Barltrop, D. (1982) Nutritional and maturational factors modifying the absorption of inorganic
lead from the gastrointestinal tract. In: Ames, B. ; Infante, P.; Reitz, R. , eds. Banbury
report 11: proceedings of a meeting; November 1981; Lloyd Harbor, NY. Cold Spring Harbor,
NY: Cold Spring Harbor Laboratory; pp. 35-41.
Barltrop, D.; Khoo, H. E. (1975) The influence of nutritional factors on lead absorption.
Postgrad. Med. J. 51: 795-800.
Barltrop, 0.; Meek, F. (1975) Absorption of different lead compounds. Postgrad. Med. J. 51:
805-809.
Barltrop, D. ; Meek, F. (1979) Effect of particle size on lead absorption from the gut. Arch.
Environ. Health 34: 280-285.
Barltrop, D.; Strehlow, C. D. (1978) The absorption of lead by children. In: Kirchgessner, M.,
ed. Trace element metabolism in man and animals - 3: proceedings of the 3rd international
symposium; July 1977; Freising, West Germany. Freising-Weihenstephan, West Germany:
Technische Universitat Munchen; pp. 332-334.
Barltrop, D. ; Strehlow, C. D. ; Thorton, I.; Webb, J. S. (1974) Significance of high soil lead
concentrations for childhood lead burdens. Environ. Health Perspect. 7: 75-82.
Barnes J M.; Stoner, H. B. (1958) Toxic properties of some dialkyl and trialkyl tin salts.
Br. J. Ind. Med. 15: 15-22.
aa»v>\s P S. I. (1975) A comparison of concentrations of lead in human tissues. Br. J. Ind.
83 yMed.' 32: 119-139.
Barry P- $. *• (1978) Distribution and storage of lead in human tissues. In: Nriagu, J. 0.,
gd. The biogeochemistry of lead in the environment. Amsterdam, The Netherlands: Elsevier/
North Holland Biomedical Press; pp. 97-150. (Topics in environmental health: v. IB.)
Barry, P- s- *• (19B1) Concentrations of lead in the tissues of children. Br. J. Ind. Med. 38:
61-71.
SRD10REF/A
10-61 7/1/83
-------
PRELIMINARY DRAFT
Barry, P. S. I. ; Mossman, D. B. (1970) Lead concentrations in human tissues. Br. J. Ind. Med
27: 339-351.
Barton, J. C. ; Conrad, M. E. (1981) Effect of phosphate on the absorption and retention of
lead in the rat. Am. J. Clin. Nutr. 34: 2192-2198.
Barton, J. C.; Conrad, M. E.; Harrison, L.; Nuby, S. (1978a) Effects of calcium on the absorp-
tion and retention of lead. J. Lab. Clin. Med. 91: 366-376.
Barton, J. C. ; Conrad, M. E. ; Nuby, S. ; Harrison, L. (1978b) Effects of iron on the absorp-
tion and retention of lead. J. Lab. Clin. Med. 92: 536-547.
Barton, J. C. ; Conrad, M. E. ; Harrison, L. ; Nuby, S. (1980) Effects of vitamin D on the ab-
sorption and retention of lead. Am. J. Physiol. 238: G124-G130.
Batschelet, E.; Brand, L.; Steiner, A. (1979) On the kinetics of lead in the human body J
Math. Biol. 8: 15-23.
Bell, R. R.; Spickett, J. T. (1981) The influence of milk in the diet on the toxicity of
orally ingested lead in rats. Food Cosmet. Toxicol. 19: 429-436.
Berg, L R.; Nordstrom, J. 0.; Ousterhout, L. E. (1980) The prevention of chick growth depres-
sion due to dietary lead by increased dietary calcium and phosphorous levels. Poult. Sci
59: 1860-1863.
Bianco, A.; Gibb, F. R. ; Morrow, P. E. (1974) Inhalation study of a submicron size lead-212
aerosol. In: Snyder, W. S. , ed. Proceedings of the third international congress of the
International Radiation Protection Association; September 1973; Washington, DC. Oak
Ridge, TN: U.S. Atomic Energy Commission; pp. 1214-1219. Available from: NTIS, Spring-
field, VA; CONF-730907-P2.
Bjorklund, H. ; Lind, B. ; Piscator, M. ; Hoffer, B. ; Olson, L. (1981) Lead, zinc, and copper
levels in intraocular brain tissue grafts, brain, and blood of lead-exposed rats.
Toxicol. Appl. Pharmacol. 60: 424-430.
Blake, K. C. H. (1976) Absorption of 203Pb from gastrointestinal tract of man. Environ. Res
11: 1-4.
Boeckx, R. L. ; Posti, B. ; Coodin, F. J. (1977) Gasoline sniffing and tetraethyl lead poisoning
in children. Pediatrics 60: 140-145.
Bolanowska, W. (1968) Distribution and excretion of triethyllead in rats. Br. J. Ind. Med. 25:
203-208.
Bolanowska, W.; Garczyriski, H. (1968) Metabolizm czteroetylku olowiu u krdlikow. [Metabolism
of tetraethyl lead in rabbits.] Med. Prac. 19: 235-243.
Bolanowska, W.; Piotrowski, J. ; Garczynski, H. (1967) Triethyllead in the biological material
in cases of acute tetraethyllead poisoning. Arch. Toxicol. 22: 278-282.
Booker, D. V.; Chamberlain, A. C. ; Newton, D. ; Stott, A. N. B. (1969) Uptake of radioactive
lead following inhalation and injection. Br. J. Radiol. 42: 457-466.
SRD10REF/A 10-62 7/1/83
-------
PRELIMINARY DRAFT
Boudene, C.; Malet, D.; Masse, R. (1977) Fate of Pb inhaled by rats. Toxicol. Appl. Pharmacol.
41: 271-276.
Bruenger, F. W. ; Stevens, W. ; Stover, B. J. (1973) The association of 210Pb with constituents
of erythrocytes. Health Phys. 25: 37-42.
Buchet, J-P. ; Roels, H. ; Hubermont, G.; Lauwerys, R. (1978) Placental transfer of lead,
mercury, cadmium, and carbon monoxide in women. II: Influence of some epidemiological
factors on the frequency distributions of the biological indices in maternal and
umbilical cord blood. Environ. Res. 15: 494-503.
Bushnell, P. J.; DeLuca, H. F. (1981) Lactose facilitates the intestinal absorption of lead
in weanling rats. Science (Washington D.C.) 221: 61-63.
Butt, E. M. ; Nusbaum, R. E. ; Gilmore, T. C.; Didio, S. L.; Mariaro, Sister. (1964) Trace metal
levels in human serum and blood. In: Proceedings of the sixth annual air pollution
medical research conference; January 1963; San Francisco, CA. Arch. Environ. Health 8:
52-57.
Byington, K. H.; Yates, D. A.; Mullins, W. A. (1980) Binding of triethyllead chloride by hemo-
globin. Toxicol. Appl. Pharmacol. 52: 379-385.
Castellino, N.; Aloj, S. (1964) Kinetics of the distribution and excretion of lead in the rat.
Br. J. Ind. Med. 21: 308-314.
Cavalleri, A.; Minoia, C. ; Pozzoli, L. ; Baruffini, A. (1978) Determination of plasma lead
levels in normal subjects and in lead-exposed workers. Br. J. Ind. Med. 35: 21-26.
Cerklewski, F. L. (1979) Influence of dietary zinc on lead toxicity during gestation and
lactation in the female rat. J. Nutr. 109: 1703-1709.
Cerklewski, F. L. (1980) Reduction in neonatal lead exposure by supplemental dietary iron
during gestation and lactation in the rat. J. Nutr. 110: 1453-1457.
Terklewski, F. L.; Forbes, R. M. (1976) Influence of dietary zinc on lead toxicity in the rat.
J. Nutr. 106: 689-696.
Chamberlain, A. C. ; Heard, M. J. (1981) Lead tracers and lead balances. In: Lynam, D. R.;
piantainida, L. G. ; Cole, J. F. , eds. Environmental lead. New York, NY: Academic Press;
pp. 175-198.
Chamberlain, A. C.; Heard, M. J.; Little P.; Newton, D.; Wells, A. C.; Wiffen, R. D. (1978)
Investigations into lead from motor vehicles. Harwell, United Kingdom: United Kingdom
Atomic Energy Authority; report no. AERE-R9198.
Chesney, R- W.; Rosen, J. F.; Hamstra, A. J. ; Smith, C.; Mahaffey, K. ; DeLuca, H. F. (1981)
Absence of seasonal variation in serum concentrations of 1,25-dihydroxyvitamin D despite
a rise in 25-hydroxyvitamin D in summer. J. Clin. Endocrinol. Metab. 53: 139-142.
Chiesura, P. (1970) Escrezione urinaria di cataboliti del piombotetraetile nell'uomo.
[Urinary excretion of tetraethyl lead catabolites in man.] Med. Lav. 61: 437-441.
SRD10REF/A
10-63 7/1/83
-------
PRELIMINARY DRAFT
Chisolm, J. J. , Jr. (1931) Dose-effect relationships for lead in young children: evidence in
children for interactions among lead, zinc and iron. In: Lynam, D. R.; Piantanida, L. G-
Cole, J. R., eds. Environmental lead. New York, NY: Academic Press; pp. 1-7. ' '
Chisolm, J. J., Jr.; Barltrop, D. (1979) Recognition and management of children with increased
lead absorption. Arch. Dis. Child. 54: 249-262.
Chisolm, J. J. , Jr.; Harrison, H. E. (1956) The exposure of children to lead. Pediatrics 18-
943-958.
Chisolm, J. J., Jr.; Mellits, E. D.; Barrett, M. B. (1976) Interrelationships among blood lead
concentration, quantitative daily ALA-U and urinary lead output following calcium EDTA.
In: Nordberg, G. F.; ed. Proceedings of third meeting of the subcommittee on the toxi-
cology of metals under the Permanent Commission and International Association on Occupa-
tional Health; November 1974; Tokyo, Japan. Amsterdam, The Netherlands: Elsevier
Publishing Co.; pp. 416-433.
Clarkson, T. W. ; Kench, J. E. (1958) Uptake of lead by human erythrocytes j_n vitro. Biochem
J. 69: 432-439.
Cohen, N. (1970) The retention and distribution of lead-210 in the adult baboon. Ph.D. thesis.
New York University.
Conrad, M. E.; Barton, J. C. (1978) Factors affecting the absorption and excretion of lead in
the rat. Gastroenterology 74: 731-740.
Cramer, K. ; Goyer, R. A.; Jagenburg, R. ; Wilson, M. H. (1974) Renal ultrastructure, renal
function, and parameters of lead toxicity in workers with different periods of lead ex-
posure. Br. J. Ind. Med. 31: 113-127.
Cremer, J. E. (1959) Biochemical studies on the toxicity of tetraethyl lead and other organo-
lead compounds. Br. J. Ind. Med. 16: 191-199.
Cremer, J. E. ; Callaway, S. (1961) Further studies on the toxicity of some tetra and trialkyl
lead compounds. Br. J. Ind. Med. 18: 277-282.
Dacre, J. C. ; Ter Haar, G. L. (1977) Lead levels in tissues from rats fed soils containing
lead. Arch. Environ. Contam. Toxicol. 6: 111-119.
Danscher, G.; Hall, E.; Fredens, K.; Fjerdingstad, E.; Fjerdingstad, E. J. (1975) Heavy metals
in the amygdala of the rat: zinc, lead and copper. Brain Res. 94: 167-172.
Davis, R. K. ; Norton, A. W. ; Larson, E. E.; Stemmer, K. L. (1963) Inhalation of tetramethyl-
lead and tetraethyllead: a comparison of the affects in rats and dogs. Arch. Environ
Health 6: 473-479.'
Day, J. P.; Hart, M.; Robinson, M. S. (1975) Lead in urban street dusts. Nature (London) 235-
343-345.
Day, J. P.; Fergusson, J. E. ; Chee, T. M. (1979) Solubility and potential toxicity of lead in
urban street dust. Bull. Environ. Contam. Toxicol. 23: 497-502.
SRD10REF/A 10-64 7/1/83
-------
PRELIMINARY DRAFT
Delves, H. T.; Clayton, B. E.; Carmichael, A.; Bubear, M.; Smith M. (1982) An appraisal of the
analytical significance of tooth-lead measurements as possible indices of environmental
exposure of children to lead. Ann. Clin. Biochem. 19: 329-337.
Der, R.; Fahim, Z.; Hilderbrand, D.; Fahim, M. (1974) Combined effect of lead and low protein
diet on growth, sexual development, and metabolism in female rats. Res. Commun. Chem.
Pathol. Pharmacol. 9: 723-738.
DeSilva, P. E. (1981) Determination of lead in plasma and studies on its relationship to lead
in erythrocytes. Br. J. Ind. Med. 38: 209-217.
Dobbins, A.; Johnson, D. R. ; Nathan, P. (1978) Effect of exposure to lead on maturation of
intestinal iron absorption of rats. J. Toxicol. Environ. Health 4: 541-550.
Drill, S.; Konz, J.; Mahar, H.; Morse, M. (1979) The environmental lead problem: an assessment
of lead in drinking water from a multi-media perspective. Washington, DC: U.S. Environ-
mental Protection Agency; EPA report no. EPA-57079-79-003. Available from: NTIS,
Springfield, VA; PB 296556.
Duggan, M. J.; Williams, S. (1977) Lead-in-dust in city streets. Sci. Total Environ. 7: 91-97.
El-Gazzar, R. M.; Finelli, V. N. ; Boiano, J.; Petering, H. G. (1978) Influence of dietary zinc
on lead toxicity in rats. Toxicol. Lett. 1: 227-234.
Everson, J. ; Patterson, C. C. (1980) "Ultra-clean" isotope dilution/mass spectrometric anal-
yses for lead in human blood plasma indicate that most reported values are artificially
high. Clin. Chem. (Winston Salem N.C.) 26: 1603-1607.
Fjerdingstad, E. J. ; Danscher, G.; Fjerdingstad, E, (1974) Hippocampus: selective concentra-
tion of lead in the normal rat brain. Brain Res. 80: 350-354.
Flanagan, P. R.; Hamilton, D. L.; Haist, J.; Velberg, L. S. (1979) Interrelationships between
iron and lead absorption in iron-deficient mice. Gastroenterology 77: 1074-1081.
Forbes, G. B.; Reina, J. C. (1972) Effect of age on gastrointestinal absorption (Fe, Sr, Pb)
in the rat. J. Nutr. 102: 647-652.
Fremlin, J. M.; Edmonds, M. I. (1980) The determination of lead in human teeth. Nucl. Instrum.
Methods 173: 211-215.
Gage, J. C.; Litchfield, M. H. (1968) The migration of lead from polymers in the rat gastro-
intestinal tract. Food Cosmet. Toxicol. 6: 329-338.
Gaae J- C.; Litchfield, M. H. (1969) The migration of lead from paint films in the rat
'gastro-intestinal tract. J. Oil Colour Chem. Assoc. 52: 236-243.
Garber, B. T.; Wei, E. (1974) Influence of dietary factors on the gastrointestinal absorption
of lead. Toxicol. App. Pharmacol. 27: 685-691.
Gershanik, J. J-; Brooks, G. G.; Little, J. A. (1974) Blood lead values in pregnant women and
their offspring. Am. J. Obstet. Gynecol. 119: 508-511.
SRD10REF/A 10-65 7/1/83
-------
PRELIMINARY DRAFT
Goldings, A. S.; Stewart, R. M. (1982) Organic lead encephalopathy: behavioral change and
movement disorder following gasoline inhalation. J. Clin. Psycho!. 43: 70-72.
Goldstein, G. W.; Diamond, I. (1974) Metabolic basis of lead encephalopathy. In: Plum, F.
ed. Brain dysfunction in metabolic disorders. Res. Publ. Assoc. Nerv. Ment. Dis* 53-
293-304.
Goldstein, G. W. ; Asbury, A. K. ; Diamond, I. (1974) Pathogenesis of lead encephalopathy: up-
take of lead and reaction of brain capillaries. Arch. Neurol. (Chicago) 31: 382-389.
Goyer, R. A. (1978) Calcium and lead interactions: some new insights. J. Lab. Clin. Med 91-
363-365.
Goyer, R. A.; May, P.; Cates, M. M. ; Krigman, M. R. (1970) Lead and protein content of
isolated intranuclear inclusion bodies from kidneys of lead-poisoned rats. Lab Invest
22: 245-251.
Grandjean, P. (1978) Regional distribution of lead in human brains. Toxicol. Lett. 2: 65.
Grandjean, P. ; Nielsen, T. (1979) Organolead compounds: environmental health aspects Residue
Rev. 97-148.
Grant, L. D.; Kimmel, C. A.; West, G. L.; Martinez-Vargas, C. M. ; Howard, J. L. (1980) Chronic
low-level lead toxicity in the rat. II: Effects on postnatal physical and behavioral
development. Toxicol. App. Pharmacol. 56: 42-58.
Grant, R. L. ; Calvery, H. 0.; Laug, E. P.; Morris, H. J. (1938) The influence of calcium and
phosphorus on the storage and toxicity of lead and arsenic. J. Pharmacol. Exp. Ther 64-
446-457.
Greenhalgh, J. R.; James, A. C. ; Smith, H. ; Hodgson, A. (1979) Absorption of lead ions from
the lung. Harwell, United Kingdom: National Radiological Protection Board; annual
research and development report no. NRPB/R&D 3; pp. 38-42.
Griffin, T. B. ; Coulston, F. ; Wills, H.; Russell, J. C.; Knelson, J. H. (1975) Clinical
studies of men continuously exposed to airborne particulate lead. In: Griffin, T. B. •
Knelson, J. H. , eds. Lead. New York, NY: Academic Press; pp. 221-240. (Coulston, F. •
Korte, F. , eds. Environmental quality and safety: supplement v. 2). '
Gross, S. B. (1981) Human oral and inhalation exposures to lead: summary of Kehoe balance ex-
periments. J. Toxicol. Environ. Health 8: 333-377.
Gross, S. B.; Pfitzer, E. A. (1974) Influence of pathological change on lead in human tissues.
In: Hemphill, D. D. , ed. Trace substances in environmental health-VIII: [proceedings of
University of Missouri's 8th annual conference on trace substances in environmental
health]; June; Columbia, MO. Columbia, MO: University of Missouri-Columbia; pp. 335-34Q.
Gross, S. B.; Pfitzer, E. A.; Yeager, D. W.; Kehoe, R. A. (1975) Lead in human tissues
Toxicol. Appl. Pharmacol. 32: 638-651.
Hambidge, K. M. (1977) The role of zinc and other trace metals in pediatric nutrition and
health. Pediatr. Clin. North Am. 24: 95-106.
SRD10REF/A 10-66 7/1/83
-------
PRELIMINARY DRAFT
Hamilton, D. L. (1978) Interrelationships of lead and iron retention in iron-deficient mice.
Hammond, P. B. (1971) The effects of chelating agents on the tissue distribution and excretion
of lead. Toxicol. Appl. Pharmacol. 18: 296-310.
Hammond, P. B. (1973) The effects of D-penicillamine on the tissue distribution and excretion
of lead. Toxicol. Appl. Pharmacol. 26: 241-246.
Hammond, P. B. ; O'Flaherty, E. J.; Gartside, P. S. (1981) The impact of air-lead on blood-lead
in man - a critique of the recent literature. Food Cosmet. Toxicol. 19: 631-638.
Hansen, J. P. B. ; Dossing, M. ; Paulev, P-E. (1981) Chelatable lead body burden (by calcium-
disodium EDTA) and blood lead concentration in man. J. Occup. Med. 23: 39-43.
Harris, P.; Holley, M. R. (1972) Lead levels in cord blood. Pediatrics 49: 606-608.
Harrison, G. E.; Carr, T. E. F.; Sutton, A.; Humphreys, E. R. (1969) Effect of alginate on the
absorption of lead in man. Nature (London) 224: 1115-1116.
Harrison, R. M. (1979) Toxic metals in street and household dusts. Sci. Total Environ. 11:
89-97.
Harrison, R. M. ; Laxen, D. P. H. (1978) Sink processes for tetraalkyllead compounds in the
atmosphere. Environ. Sci. Technol. 12: 1384-1392.
Hart, M. H. ; Smith, J. L. (1981) Effect of vitamin D and low dietary calcium on lead uptake
'and retention in rats. J. Nutr. Ill: 694-698.
Hayakawa, K. (1972) Microdetermination and dynamic aspects of u> vivo alkyl lead compounds.
II: .Studies on the dynamic aspects of alkyl lead compounds in vivo. Jpn. J. Hyg. 26:
526-535.
Heard, M. J.; Chamberlain, A. C. (1982) Effect of minerals and food on uptake of lead from the
gastrointestinal tract in humans. Hum. Toxicol. 1: 411-415.
Heard, M. 0. ; Wells, A. C.; Newton, D.; Chamberlain, A. C. (1979) Human uptake and metabolism
of tetra ethyl and tetra methyl lead vapour labelled with 203Pb. In: International
conference: management and control of heavy metals in the environment; September; London,
United Kingdom. Edinburgh, United Kingdom: CEP Consultants, Ltd.; pp. 103-108.
Holtzman, R. B. (1978) Application of radiolead to metabolic studies. In: J. 0. Nriagu, ed.
The biogeochemistry of lead in the environment. Part B: Biological effects. Amsterdam,
The Netherlands: Elsevier/North-Holland Biomedical Press; pp. 37-96.
Horiguchi, S.; Utsunomiya, T. (1973) An estimate of the body burden of lead in the healthy
Japanese population: an attempt to assume absorption and excretion of lead in the healthy
Japanese population, part 2. Osaka City Med. J. 19: 1-5.
Horiuchi, K.; Horiquchi, S.; Suekane, M. (1959) Studies on the industrial lead poisoning. 1:
Absorption, transportation, deposition and excretion of lead. 6: The lead contents in
organ-tissues of the normal Japanese. Osaka City Med. J. 5: 41-70.
SRD10REF/A 10-67 7/1/83
-------
PRELIMINARY DRAFT
Hsu, F. S. ; Krook, L. ; Pond, W. G.; Duncan, J. R. (1975) Interaction of dietary calcium with
toxic levels of lead and zinc in pigs. J. Nutr. 105: 112-118.
Hunter, J. M. (1977) The summer disease: an integrative model of the seasonality aspects of
childhood lead poisoning. Soc. Sci. Med. 11: 691-703.
Hunter, J. M. (1978) The summer disease: some field evidence on seasonality in childhood lead
poisoning. Soc. Sci. Med. 12: 85-94.
Hursh, J. B.; Mercer, T. T. (1970) Measurement of 212Pb loss rate from human lungs. J ADD!
Physiol. 28: 268-274. "
Hursh, J. B. \ Suomela, J. (1968) Absorption of 212Pb from the gastrointestinal tract of man
Acta Radiol. 7: 108-120.
Indraprasit, S.; Alexander, G. V.; Gonick, H. C. (1974) Tissue composition of major and trace
elements in uremia and hypertension. J. Chronic Dis. 27: 135-161.
International Radiological Protection Commission, Task Group on Lung Dynamics. (1966) Deposi-
tion and retention models for internal dosimetry of the human respiratory tract. Health
Phys. 12: 173-207.
James, A. C. (1978) Lung deposition of submicron aerosols calculated as a function of age and
breathing rate. Harwell, United Kingdom: National Radiological Protection Board; annual
research and development report; pp. 71-75.
Johnson, N. E.; Tenuta, K. (1979) Diets and lead blood levels of children who practice pica.
Environ. Res. 18: 369-376.
Jugo, S. (1980) Chelatable fraction of 203Pb in blood of young and adult rats. Environ Res
21: 336-342.
Jugo, S.; Maljkovic", T. ; Kostial, K. (1975a) Influence of chelating agents on the gastro-
intestinal absorption of lead. Toxicol. Appl. Pharmacol. 34: 259-263.
Jugo, S.; Maljkovid, T.; Kostial, K. (1975b) The effect of chelating agents on lead excretion
in rats in relation to age. Environ. Res. 10: 271-279.
Kaplan, M. L.; Peresie, H. J.; Jeffcoat, M. K. (1980) The lead content of blood and deciduous
teeth in lead-exposed beagle pups. In: Needleman, H. L., ed. Low level lead exposure: the
clinical implications of current research. New York, NY: Raven Press; pp. 221-230.
Kehoe, R. A. (1961a) The metabolism of lead in man in health and disease: the normal metab-
olism of lead. (The Harben lectures, 1960). J. R. Inst. Public Health Hyg. 24: 81-97.
Kehoe, R. A. (1961b) The metabolism of lead in man in health and disease: the metabolism of
lead under abnormal conditions. (The Harben lectures, 1960). J. R. Inst. Public Health
Hyg. 24: 129-143.
Kehoe, R. A. (1961c) The metabolism of lead in man in health and disease: present hygienic
problems relating to the absorption of lead. (The Harben lectures, 1960). J. R. Inst
Public Health Hyg. 24: 177-203.
Kehoe, R. A.; Thamann, F. (1931) The behavior of lead in the animal organism. II: Tetraethvl
lead. Am. J. Hyg. 13: 478-498.
SRD10REF/A 10-68 7/1/83
-------
PRELIMINARY DRAFT
Keller, C. A.; Doherty, R. A. (1980a) Distribution and excretion of lead in young and adult
female mice. Environ. Res. 21: 217-228.
Keller, C. A.; Doherty, R. A. (1980b) Bone lead mobilization in lactating mice and lead trans-
fer to suckling offspring. Toxicol. Appl. Pharmacol. 55: 220-228.
Keller, C. A.; Doherty, R. A. (1980c) Lead and calcium distributions in blood, plasma, and
milk of the lactating mouse. J. Lab. Clin. Med. 95: 81-99.
Kello, D.; Kostial, K. (1973) The effect of milk diet on lead metabolism in rats. Environ.
Res. 6: 355-360.
Kirnnel, C. A.; Grant, L. D.; Sloan, C. S.; Gladen, B. C. (1980) Chronic low-level lead toxic-
ity in the rat. Toxicol. Appl. Pharmacol. 56: 28-41.
Kimmel, E. C.; Fish, R. H.; Casida, J. E. (1977) Bioorganotin chemistry: metabolism of
organotin compounds in microsomal monooxygenase systems and in mammals. J. Agric. Food
Chem. 25: 1-9.
Klaassen, C. D.; Shoeman, D. W. (1974) Biliary excretion of lead in rats, rabbits, and dogs.
Toxicol. Appl. Pharmacol. 29: 434-446.
Klauder, D. S.; Petering, H. G. (1975) Protective value of dietary copper and iron against
some toxic effects of lead in rats. Environ. Health Perspect. 12: 77-80.
Klauder, D. S.; Petering, H. G. (1977) Anemia of lead intoxication: a role for copper. J.
Nutr. 107: 1779-1785.
Klauder, D. S.; Murthy, L; Petering, H. G. (1973) Effect of dietary intake of lead acetate on
copper metabolism in male rats. In: Hemphill, D. D., ed. Trace substances in environ-
mental health-VI: [proceedings of University of Missouri's 6th annual conference on trace
substances in environmental health]; June 1972; Columbia, MO. Columbia, MO: University of
Missouri-Columbia; pp. 131-136.
Klein, A. W.; Koch, T. R. (1981) Lead accumulations in brain, blood, and liver after low
dosing of neonatal rats. Arch. Toxicol. 47: 257-262.
Klein, M.; Namer, R.; Harpur, E.; Corbin, R. (1970) Earthenware containers as a source of
fatal lead poisoning. N. Engl. J. Med. 283: 669-672.
Kostial, K.; Kello, D. (1979) Bioavailability of lead in rats fed "human" diets. Bull.
Environ. Contam. Toxicol. 21: 312-314.
Kostial K.; Momcilovic", B. (1974) Transport of lead 203 and calcium 47 from mother to off-
spring. Arch. Environ. Health 29: 28-30.
K.; Simonovic, J.; Pisonid, M. (1971) Lead absorption from the intestine in newborn
rats. Nature (London) 233: 564.
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.
SRD10REF/A 10-69 7/1/83
-------
PRELIMINARY DRAFT
Ku, Y.; Alvarez, G. H.; Mahaffey, K. R. (1978) Comparative effects of feeding lead acetate and
phospholipid-bound lead on blood and tissue lead concentrations in young and adult rats.
Bull. Environ. Contam. Toxicol. 20: 561-567.
Laug, E. P.; Kunze, F. M. (1948) The penetration of lead through the skin. J. Ind. Hyq
Toxicol. 30: 256-259.
Lepow, M. L.; Bruckman, L.; Rubino, R. A.; Markowitz, S. ; Gillette, M.; Kapish, J. (1974) Role
of airborne lead in increased body burden of lead in Hartford children. Environ. Health
Perspect. 7: 99-102.
Lloyd, R. D. ; Mays, C. W. ; Atherton, D. R. ; Bruenger, F. W. (1975) 210Pb studies in beagles
Health Phys. 28: 575-583.
Lorenzo, A. V.; Gewirtz, M. ; Maher, C. ; Davidowski, L. I. (1977) The equilibration of lead
between blood and milk of lactating rabbits. Life Sci. 21: 1679-1683.
Mackie, A. C. ; Stephens, R. ; Townsend, A.; Waldron, H. A. (1977) Tooth lead levels in
Birmingham children. Arch. Environ. Health 32: 178-185.
Mahaffey-Six, K.; Goyer, R. A. (1970) Experimental enhancement of lead toxicity by low dietarv
calcium. J. Lab. Clin. Med. 76: 933-942. y
Mahaffey-Six, K. ; Goyer, R. A. (1972) The influence of iron deficiency on tissue content and
toxicity of ingested lead in the rat. J. Lab. Clin. Med. 79: 129-136.
Mahaffey, K. R. ; Michaelson, I. A. (1980) The interaction between lead and nutrition. In-
Needleman, H. L. , ed. Low level lead exposure: the clinical implications of current
research. New York, NY: Raven Press; pp. 159-200.
Mahaffey, K. R. ; Goyer, R. ; Haseman, J. K. (1973) Dose-response to lead ingestion in rats fed
low dietary calcium. J. Lab. Clin. Med. 82: 92-100.
Mahaffey, K. R. ; Annest, J. L. ; Barbano, H. E. ; Murphy, R. S. (1979) Preliminary analysis of
blood lead concentrations for children and adults: HANES II, 1976-1978. In: Hemphill, D
D. , ed. Trace substances in environmental health-XIU: [proceedings of University of
Missouri's 13th annual conference on trace substances in environmental health]; June-
Columbia, MO. Columbia, MO: University of Missouri-Columbia; pp. 37-51. '
Mahaffey, K. R. ; Treloar, S. ; Banks, T. A.; Peacock, B. J.; Parekh, L. E. (1976) Differences
in dietary intake of calcium, phosphorus and iron of children having normal and elevated
blood lead concentrations. J. Nutr. 106(7): xxx.
Manton, W. I.; Cook, J. D. (1979) Lead content of cerebrospinal fluid and other tissue in
amyotrophic lateral sclerosis (ALS). Neurology 29: 611-612.
Marcus, A. H. (1979) The body burden of lead: comparison of mathematical models for accumu-
lation. Environ. Res. 19: 79-90.
Marcus, S. M. (1982) Experience with D-penicillamine in treating lead poisoning. Vet. Hum
Toxicol. 24: 18-20.
SRD10REF/A 10-70 7/1/83
-------
PRELIMINARY DRAFT
Markowitz, M. E. ; Rosen, J. F. (1981) Zinc (Zn) and copper (Cu) metabolism in CaNa2 EDTA-
treated children with plumbism. Pediatr. Res. 15: 635.
McLachlin, J. R-; Goyer, R. A.; Cherian, M. G. (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.
Mehani, S. (1966) Lead retention by the lungs of lead-exposed workers. Ann. Occup. Hyg. 9:
165-171.
Momcilovic, B. (1978) The effect of maternal dose on lead retention in suckling rats. Arch.
Environ. Health 33: 115-117.
Momcilovic, B.; Kostial, K. (1974) Kinetics of lead retention and distribution in suckling and
adult rats. Environ. Res. 8: 214-220.
Moore, M. R. ; Meredith, P. A.; Watson, W. S. ; Sumner, D. J. ; Taylor, M. K. ; Goldberg, A.
(1980) The percutaneous absorption of lead-203 in humans from cosmetic preparations
containing lead acetate, as assessed by whole-body counting and other techniques. Food
Cosmet. Toxicol. 18: 399-405.
Morgan, A.; Holmes, A. (1978) The fate of lead in petrol-engine exhaust particulates inhaled
by the rat. Environ. Res. 15: 44-56.
Morgan, A.; Holmes, A.; Evans, J. C. (1977) Retention, distribution, and excretion of lead by
the rat after intravenous injection. Br. J. Ind. Med. 34: 37-42.
Morrison, J. N.; Quarterman, J.; Humphries, W. R. (1977) The effect of dietary calcium and
phosphate on lead poisoning in lambs. J. Comp. Pathol. 87: 417-429.
Morrow, P. E. ; Beiter, H. ; Amato, F.; Gibb, F. R. (1980) Pulmonary retention of lead: an ex-
perimental study in man. Environ. Res. 21: 373-384.
Mortensen, R. A. (1942) The absorption of lead tetraethyl with radioactive lead as indicator.
J. Ind. Hyg. Toxicol. 24: 285-288.
Munro, I. C.; Willes, R. F.; Truelove, J. F. (1975) Absorption and tissue distribution of in-
organic lead in the developing infant monkey (Macaca irus). Toxicol. Appl. Pharmacol. 33:
128-129.
Mykkanen, H. M.; Dickerson, J. W. T.; Lancaster, M. C. (1979) Effect of age on the tissue
distribution of lead in the rat. Toxicol. Appl. Pharmacol. 51: 447-454.
Mvkka'nen, H. M.; Wasserman, R. H. (1981) Gastrointestinal absorption of lead (203Pb) in
chicks: influence of lead, calcium, and age. J. Nutr. Ill: 1757-1765.
National Academy of Sciences, National Research Council. (1974) Recommended dietary allow-
ances. 8th ED. Washington, DC: National Academy of Sciences.
National Academy of Sciences, National Research Council. (1976) Recommendations for the pre-
vention of lead poisoning in children. Washington, DC: National Academy of Sciences.
Available from: NTIS, Springfield, VA; PB 257645.
SRD10REF/A
10-71 7/1/83
-------
PRELIMINARY DRAFT
Needleman, H. L. ; Shapiro, S. M. (1974) Dentine lead levels in asymptomatic Philadelphia
school children: subclinical exposure in high and low risk groups. Environ. Health
Perspect. 7: 27-32.
Nielsen, T.; Jensen, K. A.; Grandjean, P. (1978) Organic lead in normal human brains. Nature
(London) 274: 602-603.
Niklowitz, W. J. ; Mandybur, T. I. (1975) Neurofibrillary changes following childhood lead
encephalopathy: case report. J. Neuropathol. Exp. Neurol. 34: 445-455.
Nozaki, K. (1966) Method for studies on inhaled particles in human respiratory system and
retention of lead fume. Ind. Health 4: 118-128.
Nygaard, S. P.; Ottosen, J.; Hansen, J. C. (1977) Whole-blood lead concentration in Danes:
relation to age and environment. Dan. Med. Bull. 24: 49-51.
O'Flaherty, E. J.; Hammond, P. B.; Lerner, S. I. (1982) Dependence of apparent blood lead
half-life on the length of previous lead exposure in humans. Fundam. Appl. Toxicol 2-
49-54.
Okazaki, H.; Aronson, S. M. ; DiMaio, D. J.; Olvera, J. E. (1963) Acute lead encephalopathy of
childhood: histologic and chemical studies, with particular reference to angiopathic
aspects. In: Yahr, M. D., ed. Transactions of the American Neurological Association
1963: vol. 88. New York, NY: Springer Publishing Company, Inc.; pp. 248-250. '
Ong, C. N. ; Lee, W. R. (1980a) Distribution of lead-203 in human peripheral blood in vitro
Br. J. Ind. Med. 37: 78-84. ~ •"
Ong, C. N. ; Lee, W. R. (1980b) High affinity of lead for fetal haemoglobin. Br. J. Ind. Med
37: 292-298.
Ong, C. N. ; Lee, W. R. (1980c) Interaction of calcium and lead in human erythrocytes. Br. J
Ind. Med. 37: 70-77.
Owen, G. ; Lippman, G. (1977) Nutritional status of infants and young children: U.S.A. Pediatr
Clin. North Am. 24: 211-227.
Pott, F.; Brockhaus, A. (1971) Vergleich der enteralen and pulmonalen Resorptionsquote von
Bleiverbindungen. [Comparison of the enteral and plumonary absorption rates of lead com-
pounds. ].Zentralb. Bakt. Hyg. Orig. B 155: 1-17.
Pounds, J. G.; Marlar, R. J.; Allen, J. R. (1978) Metabolism of lead-210 in juvenile and adult
Contam. Toxicol. 19: 684-691.
Pounds, J. G.; Wright, W. R.; Kodel, R. L. (1982) Cellular metabolism of lead: a kinetic anal-
ysis in the rat isolated hepatocyte. Toxicol. Appl. Pharmacol. 66: 88-101.
Prerovska", I.; Teisinger, J. (1970) Excretion of lead and its biological activity several
years after termination of exposure. Br. J. Ind. Med. 27: 352-355.
Prpi
-------
PRELIMINARY DRAFT
Quarter-man, J. ; Morrison, J. N. (1975) The effect of dietary calcium and phosphorus on the
retention and excretion of lead in rats. Br. J. Nutr. 34: 351-362.
Quarterman, J.; Morrison, J. N. ; Carey, L. F. (1973) The influence of dietary calcium and
phosphate on lead metabolism. In: Hemphill, D. D. , ed. Trace substances in environ-
mental health-VII: [proceedings of University of Missouri's 7th annual conference on
trace substances in environment health]; June; Columbia, MO. Columbia, MO: University of
Missouri-Columbia; pp. 347-352.
Quarterman, J.; Morrison, J. N. ; Humphries, W. R. (1978a) The influence of high dietary
calcium and phosphate on lead uptake and release. Environ. Res. 17: 60-67.
Quarterman, J. ; Morrison, E. ; Morrison, J. N.; Humphries, W. R. (1978b) Dietary protein and
lead retention. Environ. Res. 17: 68-77.
Quarterman, J. ; Morrison, E. (1978c) The effect of age on the absorption and excretion of
lead. Environ. Res. 17: 78-83.
Rabinowitz, M. B.; Needleman, H. L. (1982) Temporal trends in the lead concentrations of
umbilical cord blood. Science (Washington D.C.) 216: 1429-1432.
Rabinowitz, M. B.; Wetherill, G. W.; Kopple, J. D. (1973) Lead metabolism in the normal human:
stable isotope studies. Science (London) 182: 725-727.
Rabinowitz, M. ; Wetherill, G. W. ; Kopple, J. D. (1974) Studies of human lead metabolism by
using stable isotope tracers. Environ. Health Perspect. 7: 145-152.
Rabinowitz, M. B.; Wetherill, G. W. ; Kopple, J. D. (1976) Kinetic analysis of lead metabolism
in healthy humans. J. Clin. Invest. 58: 260-270.
Rabinowitz, M. B.; Wetherill, G. W.; Kopple, J. 0. (1977) Magnitude of lead intake from res-
piration by normal man. J. Lab. Clin. Med. 90: 238-248.
Rabinowitz, M. B.; Kopple, J. D.; Wetherill, G. W. (1980) Effect of food intake and fasting on
gastrointestinal lead absorption in humans. Am. J. Clin. Nutr. 33: 1784-1788.
Rader, J- I-J Peeler, J. T.; Mahaffey, K. R. (1981) Comparative toxicity and tissue distribu-
tion of lead acetate in weanling and adult rats. Environ. Health Perspect. 42: 187-195.
Rader, J- 1-5 Celesk, E. M.; Peeler, J. T.; Mahaffey, K. R. (1982) Tissue distribution and
retention of equivalent doses of lead acetate in young and adult rats. In: Hemphill, D.
D., ed. Trace substance in environmental health -XV: [proceedings of University of
Missouri's 15th annual conference on trace substances in environmental health]; June
1981; Columbia, MO. Columbia, MO: University of Missouri - Columbia; pp. PAGES. (IN
PRESS)
Raghavan, S. R. V. ; Gonick, H. C. (1977) Isolation of low-molecular-weight lead-binding pro-
tein from human erythrocytes. Proc. Soc. Exp. Biol. Med. 155: 164-167.
Raghavan, S. R. V.; Culver, B. D.; Gonick, H. C. (1980) Erythrocyte lead-binding protein after
occupational exposure. I: Relationship to lead toxicity. Environ. Res. 22: 264-270.
SRD10REF/A 10-73 7/1/83
-------
PRELIMINARY DRAFT
Raghavan, S. R. V.; Culver, B. D.; Gonick, H. C. (1981) Erythrocyte lead-bind.ing^protein after
occupational exposure. II: Influence on lead inhibition of membrane Na , K - adenosine-
triphosphatase. J. Toxicol. Environ. Health 7: 561-568.
Rastogi, S. C.; Clausen, J. (1976) Absorption of lead through the skin. Toxicology 6: 371-376.
Rendall, R. E. G.; Baily, P.; Soskolne, C. L (1975) The effect of particle size on absorption
of inhaled lead. Am. Ind. Hyg. Assoc. J. 36: 207-213.
Rosen, J. F. ; Markowitz, M. E. (1980) D-penicillamine: its actions on lead transport in bone
organ culture. Pediatr. Res. 14: 330-335.
Rosen, J. F. ; Zarate-Salvador, C. ; Trinidad, E. E. (1974) Plasma lead levels in normal and
lead-intoxicated children. J. Pediatr. (St. Louis) 84: 45-48.
Rosen, J. F. ; Chesney, R. W. ; Hamstra, A.; DeLuca, H. P.; Mahaffey, K. R. (1980) Reduction in
1,25-dihydroxyvitamin D in children with increased lead absorption. N. Engl. J. Med. 302-
1128-1131. Also in: Brown, S. S. ; Davis, D. S. , eds. (1981) Organ-directed toxicity:
chemical indices and mechanisms. New York, NY: Pergamon Press; pp. 91-95.
Ryu, J. E. ; Ziegler, E. E. ; Fomon, S. J. (1978) Maternal lead exposure and blood lead concen-
tration in infancy. J. Pediatr. (St. Louis) 93: 476-478.
Saenger, P.; Rosen, J. F.; Markowitz, M. E. (1982) Diagnostic significance of edetate disodium
calcium testing in children with increased lead absorption. Am. J. Dis. Child. 135-
312-315.
Sartor, F. A.; Rondia, D. (1981) Setting legislative norms for environmental lead exposure:
results of an epidemiological survey in the east of Belgium. Toxicol. Lett. 7: 251-25?!
Scanlon, J. (1971) Umbilical cord blood lead concentration. Am. J. Dis. Child. 121: 325-326.
Schroeder, H. A.; Tipton, I. H. (1968) The human body burden of lead. Arch. Environ. Health
17: 965-978.
Shapiro, I. M. ; Burke, A.; Mitchell, G.; Bloch, P. (1978) X-ray fluorescence analysis of lead
in teeth of urban children jn situ: correlation between the tooth lead level and the
concentration of blood lead and free erythroporphyrins. Environ. Res. 17: 46-52.
Shelling, D. H. (1932) Effect of dietary calcium and phosphorus on toxicity of lead in the
rat: rationale of phosphate therapy. Proc. Soc. Exp. Biol. Med. 30: 248-254.
Sherlock, J.; Smart, G.; Forbes, G. I.; Moore, M. R. ; Patterson, W. J. ; Richards, W. N.;
Wilson, T. S. (1982) Assessment of lead intakes and dose-response for a population in Ayr
exposed to a plumbsolvent water supply. Hum. Toxicol. 1: 115-122.
Singh, N.; Donovan, C. M. ; Hanshaw, J. B. (1978) Neonatal lead intoxication in a prenatally
exposed infant. J. Pediatr. (St. Louis) 93: 1019-1021.
Smith, C. M.; DeLuca, H. F.; Tanaka, Y.; Mahaffey, K. R. (1978) Stimulation of lead absorption
by vitamin D administration. J. Nutr. 108: 843-847.
SRD10REF/A 10-74 7/1/83
-------
PRELIMINARY DRAFT
Sobel, A. E.; Gawron, 0.; Kramer, B. (1938) Influence of vitamin D in experimental lead poi-
soning. Proc. Soc. Exp. Biol. Med. 38: 433-435.
Sobel, A. E.; Yuska, H.; Peters, D. D.; Kramer, B. (1940) The biochemical behavior of lead.
I: Influence of calcium, phosphorus, and vitamin D on lead in blood and bone. J. Biol.
Chem. 132: 239-265. Reprinted (1981) in Nutr. Rev. 39: 374-377.
Sorrell, M. ; Rosen, J. F.; Roginsky, M. (1977) Interactions of lead, calcium, vitamin D, and
nutrition in lead-burdened children. Arch. Environ. Health 32: 160-164.
Steenhout, A.; Pourtois, M. (1981) Lead accumulation in teeth as a function of age with dif-
ferent exposures. Br. J. Ind. Med. 38: 297-303.
Stephens, R. ; Waldron, H. A. (1975) The influence of milk and related dietary constituents on
lead metabolism. Food Cosmet. Toxicol. 13: 555-563.
Stevens, C. D. ; Feldhake, C. J.; Kehoe, R. A. (1960) Isolation of triethyllead ion from liver
after inhalation of tetraethyllead. J. Pharmacol. Exp. Ther. 128: 90-94.
Stowe, H. D.; Goyer, R. A.; Krigman, M. M.; Wilson, M.; Gates, M. (1973) Experimental oral
lead toxicity in young dogs: clinical and morphologic effects. Arch. Pathol. 95:
106-116.
Stuik, E. J. (1974) Biological response of male and female volunteers to inorganic lead. Int.
Arch. Arbeitsmed. 33: 83-97.
Thompson, J. A. (1971) Balance between intake and output of lead in normal individuals. Br. J.
Ind. Med. 28: 189-194.
Tola, S. ; Hernberg, S.; Asp, S.; Nikkanen, J. (1973) Parameters indicative of absorption and
biological effect in new lead exposure: a prospective study. Br. J. Ind. Med. 30:
134-141.
Trotter, M.; Hixon, B. B. (1974) Sequential changes in weight, density, and percentage ash
weight of human skeletons from an early fetal period through old age. Anat. Rec. 179:
1-18.
U $ Centers for Disease Control. (1978) Preventing lead poisoning in young children. J.
Pediatr. (St. Louis) 93: 709-720.
U S. Environmental Protection Agency, Health Effects Research Lab. (1977) Air quality criteria
for lead. Research Triangle Park, NC: U.S. Environmental Protection Agency, Criteria and
Special Studies Office; EPA report no. EPA-600/8-77-017. Available from: NTIS,
Springfield, VA; PB 280411.
United Kingdom Central Directorate on Environmental Pollution. (1982) The Glasgow duplicate
diet study (1979/1980): a joint survey for the Department of the Environment and the
Ministry of Agriculture Fisheries and Food. London, United Kingdom: Her Majesty's
Stationery Office; pollution report no. 11.
Watson, W. S.; Hume, R.; Moore, M. R. (1980) Oral absorption of lead and iron. Lancet 2(8188):
236-238.
SRD10REF/A 10-75 7/1/83
-------
PRELIMINARY DRAFT
Willoughby, R. A.; Thirapatsakun, T.; McSherry, B. J. (1972) Influence of rations low in
calcium and phosphorus on blood and tissue lead concentrations in the horse Am J Vet
Res. 33: 1165-1173.
Winneke, G.; Brockhaus, A.; Kra'mer, U.; Ewers, U.; Kujanek, G.; Lechner, H.; Janke, W. (1981)
Neuropsychological comparison of children with different tooth-lead levels: preliminary
report. In: International conference: heavy metals in the environment; Amsterdam The
Netherlands. Edinburgh, United Kingdom: CEP Consultants, Ltd.; pp. 553-556.
World Health Organization/United Nations Environmental Programme. (1977) Lead. Geneva
Switzerland: World Health Organization. (Environmental health criteria 3.) '
Yamamura, Y.; Takakura, J.; Hirayama, F.; Yamauchi, H.; Yoshida, M. (1975) Tetraethyl lead
poisoning caused by cleaning work in the aviation fuel tank. Jpn. J. Ind. Health 17-
223-235.
Yip, R.; Norris, T. N.; Anderson, A. S. (1981) Iron status of children with elevated blood
lead concentrations. J. Pediatr. (St. Louis) 98: 922-925.
Ziegler, E. E.; Edwards, B. B.; Jensen, R. L.; Mahaffey, K. R.; Fomon, S. J. (1978) Absorption
and retention of lead by infants. Pediatr. Res. 12: 29-34.
Zielhuis, R. L.; del Castilho, P.; Herber, R. F. M; Wibowo, A. A. E. (1978) Levels of lead and
other metals in human blood: suggestive relationships, determining factors. Environ
Health Perspect. 25: 103-109.
SRD10REF/A 10-76 7/1/83
-------
PRELIMINARY DRAFT
11. ASSESSMENT OF LEAD EXPOSURES AND ABSORPTION IN HUMAN POPULATIONS
11.1 INTRODUCTION
The purpose of this chapter is to describe effects on internal body burdens of lead in
human populations resulting from exposure to lead in their environment. This chapter dis-
cusses changes in various internal exposure indices that follow changes in external lead
exposures. The main index of internal lead exposure focused on herein is blood lead
levels, although other indices, such as levels of lead in teeth and bone are also briefly dis-
cussed. As noted in Chapter 10, blood lead levels most closely reflect recent exposures to
environmental lead. On the other hand, teeth and bone lead levels better reflect or index
cumulative exposures.
The following terms and definitions will be used in this chapter. Sources of lead are
those components of the environment (e.g., gasoline combustion, smelters) from which signifi-
cant quantities of lead are released into various environmental media of exposure. Environ-
mental media are direct routes by which humans become exposed to lead (e.g., air, soil, water,
dust). External exposures are levels at which lead is present in any or all of the environ-
mental media. Internal exposures are the amounts of lead present at various sites within the
body.
The present chapter is organizationally structured so as to achieve the following four
main objectives:
(1) Elucidation of patterns of absorbed lead in U.S. populations and identifi-
cation of important demographic covariates.
(2) Characterization of relationships between external and internal exposures
by exposure medium (air, food, water or dust).
(3) Identification of specific sources of lead which result in increased
internal exposure levels.
(4) Estimation of the relative contributions of various sources of lead in the
environment to total internal exposure.
The existing scientific literature must be examined in light of the investigators' own
objectives and the quality of the scientific investigations performed. Although all studies
need to be evaluated in regard to their methodology, the more quantitative studies are evalu-
ated here in greater depth. A discussion of the main types of methodological points con-
sidered in such evaluations is presented in Section 11.2.
After discussing methodological aspects, patterns of internal exposure to lead in human
populations are delineated in Section 11.3. This begins with a brief examination of the
PB11A2/B 11-1 7/29/83
-------
PRELIMINARY DRAFT
historical record of internal lead exposure in human populations. These data serve as a back-
drop against which recent U.S. levels can be contrasted and defines the relative magnitude of
external lead exposures in the past and present. The contrast is structured as follows: his-
torical data, recent data from populations thought to be isolated from urbanized cultures, and
then U.S. populations showing various degrees of urbanization and industrialization.
Recent patterns of internal exposure in U.S. populations are discussed in greater detail
Estimates of internal lead exposure and identification of demographic covariates are made
Studies examining the recent past for evidence of change in levels in internal exposure are
presented.' A discussion follows regarding exposure covariates of blood lead levels in urban
U.S. children, who are at special risk for increased internal exposure.
The statistical treatment of distributions of blood lead levels in human populations is
the next topic discussed. As part of that discussion, the empirical characteristics of blood
lead distributions in well defined homogeneous populations are denoted. Important issues
addressed include the proper choice of estimators of central tendency and dispersion, estima-
tors of percentile values and the potential influence of errors in measurement on statistical
estimation involving blood lead data.
Section 11.4 focuses on general relationships between external exposures and levels of
internal exposure. The distribution of lead in man is diagramatically depicted by the compo-
nent model shown in Figure 1. Of particular importance for this document is the relationship
between lead in air and lead in blood. If lead in air were the only medium of exposure, then
the interpretation of a statistical relationship between lead in air and lead in blood would
be relatively simple. However, this is not the case. Lead is present in a number of environ-
mental media, as described in Chapter 7 and summarized in Figure 11-1. There are relation-
ships between lead levels in air and lead concentrations in food, soil, dust and water. As
shown in Chapters 6, 7 and 8, lead emitted into the atmosphere ultimately comes back to con-
taminate the earth. However, only limited data are currently available that provide a quan-
titative estimate of the magnitude of this secondary lead exposure. The implication is that
an analysis involving estimated lead levels in all environmental media may produce an under-
estimate of the relationship between lead in blood and lead in air.
The discussion of relationships between external exposure and internal absorption com-
mences with air lead exposures. Both experimental and epidemiological studies are discussed.
Several studies are identified as being of most importance in determining the quantitative
relationship between lead in blood and lead in air. The shape of the relationship between
blood lead and air lead is of particular interest and importance.
P811A2/B 11-2 7/29/83
-------
PRELIMINARY DRAFT
INDUSTRIAL
EMISSIONS
CRUSTAL
WEATHERING
SURFACE AND
GROUND WATER
DRINKING
WATER
FECES URINE
Figure 11-1. Pathways of lead from the environment to man.
PB11A2/B
11-3
7/29/83
-------
PRELIMINARY DRAFT
After discussion of air lead vs. blood lead relationships, the chapter next discusses the
relationship of blood lead to atmospheric lead found in other environmental media. Section
11.5 describes studies of specific lead axposure situations useful in identifying specific
environmental sources of lead that contribute to elevated body burdens of lead. The chapter
concludes with a summary of key information and conclusions derived from the scientific evi-
dence reviewed.
11.2 METHODOLOGICAL CONSIDERATIONS
11.2.1 Analytical Problems
Internal lead exposure levels in human populations have been estimated by analyses of a
variety of biological tissue matrices (e.g., blood, teeth, bone, and hair). Lead levels in
each of these matrices have particular biological meanings with regard to external exposure
status; these relationships are discussed in Chapter 10. The principal internal exposure
index discussed in this chapter is blood lead concentration. Blood lead concentrations
are most reflective of recent exposure to lead and bear a consistent relationship to levels of
lead in the external environment if the latter have been stable. Blood lead levels are vari-
ously reported as ug/100 g, pg/100 ml, ug/dl, ppm, ppb, and um/1. The first four measures are
roughly equivalent, whereas ppb values are simply divisible by 1000 to be equivalent. Actual-
ly there is a small but not meaningful difference in blood lead levels reported on a per
volume vs. per weight difference. The difference results from the density of blood being
slightly greater than 1 g/rol. For the purposes of this chapter, data reported on a weight or
volume basis are considered equal. On the other hand, blood lead data reported on a umol/1
basis must be multiplied by 20.72 to get the equivalent ug/dl value. Data reported originally
as umol/1 in studies reviewed here are converted to ug/dl in subsequent sections of this
chapter.
As discussed in Chapter 9, the measurement of lead in blood has been accomplished via a
succession of analytical procedures over the years. The first reliable analytical methods
available were wet chemistry procedures that have been succeeded by increasingly automated in-
strumental procedures. With these changes in technology there has been increasing recognition
of the importance of controlling for contamination in the sampling and analytical procedures.
These advances, as well as institution of external quality control programs, have resulted in
markedly improved analytical results. Data summarized in Chapter 9 show that a generalized
improvement in analytical results across many laboratories occurred during Federal Fiscal
Years 1977 to 1979. No futher marked improvement was seen during Federal Fiscal Years 1979 to
1981.
As difficult as getting accurate blood lead determinations is, the achievement of accu-
rate lead isotopic determinations is even more difficult. Experience gained from the isotopic
PB11A2/B 11-4 7/29/83
-------
PRELIMINARY DRAFT
lead experiment (ILE) in Italy (reviewed in detail in Section 11.5.1.1.1) has indicated that
extremely aggressive quality control and contamination control programs must be implemented to
achieve acceptable results. With proper procedures, meaningful differences on the order of a
single nanogram are achievable.
11.2.2 Statistical Approaches
Many studies summarize the distribution of lead levels in humans. These studies usually
report measures of central tendency (means) and dispersion (variances). In this chapter, the
term "mean" refers to the arithmetic mean unless stated otherwise. This measure is always an
estimate of the average value, but it estimates the center of the distribution (50th percen-
tile) only for symmetric distributions. Many authors provide geometric means, which estimate
the center of the distribution if the distribution is lognormal. Geometric means are influ-
enced less by unusually large values than are arithmetic means. A complete discussion of the
lognormal distribution is given by Aitchison and Brown (1966), including formulas for conver-
ting ffom arithmetic to geometric means.
Most studies also give sample variances cr standard deviations in addition to the means.
jf geometric means are given, then the corresponding measure of dispersion is the geometric
standard deviation. Aitchison and Brown (1966) give formulas for the geometric standard devi-
ation and, also, explain how to estimate percentiles and construct confidence intervals. All
of the measures of dispersion actually include three sources of variation: population varia-
tion, measurement variation and variation due to sampling error. Values for these components
are needed in order to evaluate a study correctly.
A separate issue is the form of the distribution of blood lead values. Although the nor-
mal and lognormal distributions are commonly used, there are many other possible distribu-
tions. The form is important for two reasons: 1) it determines which is more appropriate,
the arithmetic or geometric mean, and 2) it determines estimates of the fraction of a popula-
tion exceeding given internal lead levels under various external exposures. Both of these
Questions arise in the discussion of the distribution of human blood lead levels.
Many studies attempt to relate blood lead levels to an estimate of dose such as lead
levels in air. Standard regression techniques should be used with caution, since they assume
that the dose variable is measured without error. The dose variable is an estimate of the
actual lead intake and has inherent inaccuracies. As a result, the slopes tend to be under-
estimated; however, it is extremely difficult to quantify the actual amount of this bias.
Multiple regression analyses have additional problems. Many of the covariates that measure
xternal exposures are highly correlated with each other. For example, much of the soil lead
nd house dust lead comes from the air. The exact effect of such high correlations with each
ther on the regression coefficients is not clear.
PB11A2/B 11-5 7/29/83
-------
PRELIMINARY DRAFT
11.3 LEAD IN HUMAN POPULATIONS
11.3.1 Introduction
This section is designed to provide insight into current levels of lead absorption in the
U.S. and other countries, and how they differ from "natural" levels, to examine the influence
of demographic factors, and to describe the degree of internal exposure in selected population
subgroups. This section will also examine time trend studies of blood lead levels.
11.3.2 Ancient and Remote Populations
A question of major interest in understanding environmental pollutants is the extent to
which current ambient exposures exceed background levels. Because lead is a naturally occur-
ring element it can be surmised that some level has been and will always be present in the
human body; the question of interest is what is the difference in the levels of current sub-
groups of the United States population from those "natural" levels. Information regarding this
issue has been developed from studies of populations that lived in the past and populations
that currently live in remote areas far from the influence of industrial and urban lead ex-
posures.
Man has used lead since antiquity for a variety of purposes. These uses have afforded
the opportunity for some segments of the human population to be exposed to lead and subse-
quently absorb it into the body. Because lead accumulates over a lifetime in bones and teeth
and because bones and teeth stay intact for extremely long times, it is possible to estimate
the extent to which populations in the past have been exposed to lead.
Because of the problems of scarcity of samples and little knowledge of how representative
the samples are of conditions at the time, the data from these studies provide only rough es-
timates of the extent of absorption. Further complicating the interpretation of these data
are debates over proper analytical procedures and the question of whether skeletons and teeth
pick up or release lead from or to the soil in which they are interred.
Despite these difficulties, several studies provide data by which to estimate internal
exposure patterns among ancient populations, and some studies have included data from both
past and current populations for comparisons. Figure 11-2, which is adapted from Angle (1982)
displays a historical view of the estimated lead usage and data from ancient bone and teeth
lead levels. There is a reasonably good fit. There appears to be an increase in both lead
usage and absorption over the time span covered. Specifics of these studies of bone and teeth
will be presented in Section 11.3.2.1. In contrast to the study of ancient populations using
bone and teeth lead levels, several studies have looked at the issue of lead contamination
from the perspective of comparing current remote and urbanized populations. These studies
have used blood lead levels as an indicator and found mean blood concentrations in remote
PB11A2/B 11-6 7/29/83
-------
PRELIMINARY DRAFT
w
o
Z
O
O
o
IU
1 1 1 1 T~7T
A PERU
O EGYPT
O NUBIA
• DENMARK
BRITAIN-ROMAN,
ANGLO SAXON
• U.S.
O BRITAIN
WORLDWIDE LEAD
PRODUCTION
LEAD CONCENTRATION
IN BONES
USE OF
SILVER COINS,
NEW
WORLD
SILVER
DEPLETION OF
ROMAN MINES
DOMINANCE OF
ATHENS ROME
£ 10' -
101 —
5500
BP
10
5000 4500 4000 3500 3000 2500 2000 1500 1000 500 PRESENT
YEARS BEFORE PRESENT
Figure 11-2. Estimate of world-wide lead production and lead concentrations
in bones (pglgm) from 5500 years before present to the present time.
Source: Adapted from Angle and Mclntire (1982).
PB11A2/B
11-7
7/29/83
-------
PRELIMINARY DRAFT
populations between 1 and 5 (jg/dl, which is an order of magnitude below current U.S. urban
population means. These studies are presented in detail in Section 11.3.2.2.
11.3.2.1 Ancient Populations. Table 11-1 presents summaries of several studies that analyzed
bones and teeth to yield approximate estimates of lead absorption in the- past. Some of these
studies also analyzed contemporary current samples so that a comparison between past and pre-
sent could be made.
Samples from the Sudan (ancient Nubians) were collected from several different periods
(Grandjean et al., 1978). The oldest sample (3300-2900 B.C.) averaged 0.6 ug/g for bone and
0.9 ug/g for teeth. Data from the later time of 1650-1350 B.C. show a substantial increase in
absorbed lead. Comparison of even the most recent ancient samples with a current Danish sam-
ple show a 4- to 8-fold increase over time.
Similar data were also obtained from Peruvian and Pennsylvania samples (Becker et al.
1968). The Peruvian and Pennsylvania samples were approximately from the same era (^1200-1400
A.D.). Little lead was used in these cultures as reflected by chemical analysis of bone lead
content. The values were less than 5 ug/g for both samples. In contrast, modern samples from
Syracuse, New York, ranged from 5 to 110 ug/g.
Fosse and Wesenberg (1981) reported a study of Norwegian samples from several eras. The
oldest material was significantly lower in lead than modern samples. Ericson et al. (1979)
also analyzed bone specimens from ancient Peruvians. Samples from 4500-3000 years ago to
about 1400 years ago were reasonably constant (<0.2 ug/g).
Aufderheide et al. (1981) report a study of 16 skeletons from colonial America. Two
social groups, identified as plantation proprietors and laborers, had distinctly different
diet exposures to lead as shown by the analyses of the skeletal samples. The proprietor
group averaged 185 ug/g bone ash while the laborer group averaged 35 ug/g.
Shapiro et al. (1975) report a study that contrasts teeth lead content of ancient popula-
tions with that of current remote populations and, also, with current urban populations. The
ancient Egyptian samples (1st and 2nd millenia) exhibited the lowest teeth lead levels, mean
of 9.7 ug/g. The more recent Peruvian Indian samples (12th Century) had similar levels
(13.6 ug/g). The contemporary Alaskan Eskimo samples had a mean of 56.0 ^g/g while
Philadelphia samples had a mean of 188.3 ug/g. These data suggest an increasing pattern of
lead absorption'from ancient populations to current remote and urban populations.
11.3.2.2 Remote Populations. Several studies have looked at the blood lead levels in current
remote populations (Piomelli et al., 1980; Poole and Smythe, 1980). These studies are impor-
tant in defining the baseline level of internal lead exposures found in the world today.
PB11A2/B 11-8 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-1. STUDIES OF PAST EXPOSURES TO LEAD
Population
Studied
Nubians1
vs. Modern
Danes
Nubians
A-group
C-group
pharonic
Merotic,
X-group &
Christians
Danei
~~ '
Ancient
Peruvians2
Ancient Penn-
sylvanian
Indians
Recent
Syracuse, NY
_—
Uvdal3
Modern
Buskend County
Bryggen
(medieval Bergen)
Norway
Age of
Sample
3300 B.C. to 750 A.D.
(5000 yrs. old)
3300 to 2900 B.C.
2000 to 1600 B.C.
1650 to 1350 B.C.
1 to 750 A.D.
Contemporary
500-600 yrs. old
500 yrs. old
Contemporary
Buried from before
1200 A.D. to 1804
Contemporary
?
Contemporary
Index of
Exposure Method of
Used Analysis
Teeth FASS
(circum- ASV
pupil
dentine)
Bone (temporal)
Bone Arc emission
(Tibia) spectroscopy
(Femur)
Teeth AAS
(Whole
teeth, but
values
corrected for
enamel and
dentine)
Lead
Levels
Pb
ug/g dry wt.
Bone Tooth
0.6 0.9
1.0 2.1
2.0 5.0
1.2 3.2
5.5 25.7
Bone
P9/g
Peru <5
Penn. N.D.
Modern 110, 75,
5, 45, 16
Tooth
M9/9
1.22
4.12
1.81
3.73
jgrandjean, P.; Nielsen, O.V.; Shapiro, I.M. (1978) Lead retention in ancient Nubian and
contemporary populations. J. Environ. Pathol. Toxicol. 2: 781-787.
2|Jecker, R.O.; Spadaro, J.A.; Berg, E.W. (1968) The trace elements in human bone. J. Bone
Jt. Surg. BOA: 326-334.
SFosse, G-; Wesenberg, G.B.R. (1981) Lead, cadmium, zinc and copper in deciduous teeth of
Norwegian children in the pre-industrial age. Int. J. Environ. Stud. 16: 163-170.
PB11A2/B
11-9
7/29/83
-------
PRELIMINARY DRAFT
Piomelli et al. (1980) report a study of blood lead levels of natives in a remote (far
from industrialized regions) section of Nepal. Portable air samplers were used to determine
the air lead exposure in the region. The lead content of the air samples proved to be less
than the detection limit, 0.004 ug/m . A later study by Davidson et al. (1981) from Nepal
confirmed the low air lead levels reported by Piomelli et al. (1980). Davidson et al. (1981)
found an average air lead concentration of 0.00086 ug/m .
Blood lead levels 'reported by Piomelli et al. (1980) for the Nepalese natives were low
the geometric mean blood lead for this population was 3.4 ug/dl. Adult males had a geometric
mean of 3.8 ug/dl and adult females, 2.9 ug/dl. Children had a geometric mean blood lead of
3.5 ug/dl. Only 10 of 103 individuals tested had a blood lead level greater than 10 ug/dl
The blood samples, which were collected on filter paper discs, were analyzed by a modification
of the Delves Cup Atomic Absorption Spectrophotometric method. Stringent quality control pro-
cedures were followed for both the blood and air samples.
To put these Nepalese values in perspective, Piomelli et al. (1980) reported analyses of
blood samples collected and analyzed by the same methods from Manhattan, New York. New York
blood leads averaged about 15 ug/dl, a 5-fold increase over the Nepalese values.
Poole and Smythe (1980) reported another study of a remote population, using contam-
ination-free micro-blood sampling and chemical analysis techniques. They reported acceptable
precision at blood lead concentrations as low as 5 ug/dl, using spectrophotometry. One hun-
dred children were sampled from a remote area of Papua, New Guinea. Almost all of the chil-
dren came from families engaging in subsistence agriculture. The children ranged from 7 to 10
years and included both sexes. Blood lead levels ranged from 1 to 13 ug/dl with a mean of
5.2. Although the data appear to be somewhat skewed to the right, they are in good agreement
with those of PiomeTIi for Nepalase subjects.
11.3.3 j.evels of Lead and Demographic Covariates in U.S. Populations
11.3.3.1 The NHANES II Study. The National Center for Health Statistics has provided the
best currently available picture of blood lead levels among United States residents as part of
the second National Health and Nutrition Examination Study (NHANES II) conducted from February
1976 to February 1980 (Mahaffey et al., 1982; McDowell et al., 1981; Annest et al., 1982).
These are the first national estimates of lead levels in whole blood from a representative
sample of the non-institutionalized U.S. civilian population aged 6 months to 74 years of age.
From a total of 27,801 persons identified through a stratified, multi-stage probability
cluster sample of households throughout the U.S., blood lead determinations were scheduled for
16,563 persons including all children ages 6 months to 6 years, and one-half of all persons
ages 7 to 74. Sampling was scheduled in 64 sampling areas over the 4-year period according to
PB11A2/B 11-10 7/29/83
-------
PRELIMINARY DRAFT
a previously determined itinerary to maximize operational efficiency and response of partici-
pants. Because of the constraints of cold weather, the examination trailers traveled in the
moderate climate areas during the winter, and the more northern areas during the summer
(McDowell et al., 1981).
All reported blood lead levels were based on samples collected by venipuncture. Blood
lead levels were determined by atomic absorption spectrophotometry using a modified Delves Cup
micro-method. Specimens were analyzed in duplicate, with both determinations done independ-
ently in the same analytical run. Quality control was maintained by two systems, a bench
system and a blind insertion of samples. If the NHANES II replicates differed by more than
7 ug/dl, the analysis was repeated for the specimen (about 0.3 percent were reanalyzed). If
the average of the replicate values of either "bench" or "blind" control specimens fell out-
side previously established 95 percent confidence limits, the entire run was repeated. The
estimated coefficient of variation for the "bench" quality control ranged from 7 to 15 percent
(Mahaffey et al., 1979).
The reported blood lead levels were based on the average of the replicates. Blood lead
levels and related data were reported as population estimates; findings for each person were
inflated by the reciprocal of selection probabilities, adjusted to account for persons who
were not examined and poststratified by race, sex and age. The final estimates closely
approximate the U.S. Bureau of Census estimates for the civilian non-institutionalized popula-
tion of the United States as of March 1, 1978, aged 1/2 to 74 years.
Participation rates varied across age categories; the highest non-response rate (51
percent) was for the youngest age group, 6 months through 5 years. Among medically examined
persons, those with missing blood lead values were randomly distributed by race, sex, degree
of urbanization and annual family income. These data are probably the best estimates now
available regarding the degree of lead absorption in the general United States population.
Forthofer (1983) has studied the potential effects of non-response bias in the NHANES II
survey and found no large biases in the health variables. This was based on the excellent
aareement of the NHANES II examined data, which had a 27 percent non-response rate, with the
National Health Interview Survey data, which had a 4 percent non-response rate.
The national estimates presented below are based on 9,933 persons whose blood lead levels
ranged from 2.0 to 66.0 ug/dl. The median blood lead for the entire U.S. population is 13.0
jn/dl. It 1S readily apparent that blacks have a higher blood lead level than whites (medians
for blacks and whites were 15.0 and 13.0 ug/dl, respectively).
Tables 11-2 through 11-4 display the observed distribution of measured blood lead levels
bv race, sex and age. The possible influence of measurement error on the percent distribution
estimates is discussed in Section 11.3.5. Estimates of mean blood lead levels differ sub-
stantially with respect to age, race and sex. Blacks have higher levels than whites, the
PB11A2/B 11-11 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-2. NHANES II BLOOD LEAD LEVELS OF PERSONS 6 MONTHS-74 YEARS, WITH WEIGHTED ARITHMETIC MEAN, STANDARD ERROR OF THE
MEAN, WEIGHTED GEOMETRIC MEAN, MEDIAN, AND PERCENT DISTRIBUTION, BY RACE AND AGE, UNITED STATES, 1976-80
Blood lead level (uq/dl)
Race and age
All racesc
Al 1 ages
6 months-5 years ....
6-17 years
18-74 years
£ White
ro
All ages
6 months-5 years ....
6-17 years
18-74 years
Black
All ages . .
6 months-5 years ....
6-17 years
18-74 years
Estimated
population
thousands
203,554
16,852
44,964
141,728
174,528
13.641
37.530
123,357
23,853
2,584
6.529
14.740
Number b
examined
9,933
2.372
1,720
5,841
8,369
1.876
1,424
5,069
1,332
419
263
650
Arith-
metic
Mean
13.9
16.0
12.5
14.2
13.7
14.9
12.1
14.1
15.7
20.9
14.8
15.5
Standard
error of
the mean
0.24
0.42
0.30
0.25
0.24
0.43
0.30
0.25
0.48
0.61
0.53
0.54
Geometric
Mean
12.8
14.9
11.7
13.1
12.6
14.0
11.3
12.9
14.6
19.6
14.0
14.4
Median
13.0
15.0
12.0
13.0
13.0
14.0
11.0
13.0
15.0
20.0
14.0
14.0
Less
than
10
22.1
12.2
27.6
21.2
23.3
14.5
30.4
21.9
13.3
2.5
12.8
14.7
10-19
Percent
62.9
63.3
64.8
62.3
62.8
67.5
63.4
63.3
63.7
45.4
70.9
62.9
20-29
30-39
40*
distribution
13.0
20.5
7.1
14.3
12.2
16.1
5.8
13.7
20.0
39.9
15.6
19.6
1.6
3.6
0.5
1.8
1.5
1.8
0.4
1.8
2.3
10.2
0.7
2.0
0.3
0.4
0.4
0.3
0.2
0.4
0.6
2.0
0.9
aAt the midpoint of the survey, March 1, 1978.
with lead determinations from blood specimens drawn by venipuncture.
Includes date for races not shown separately.
Numbers may not add to 100 percent due to rounding.
-------
PRELIMINARY DRAFT
TABLE 11-3. NHANES JI BLOOD LEAD LEVELS OF MALES 6 MONTHS-74 YLARS, WITH WEIGHTED ARITHMETIC MEAN, STANDARD ERRC* OF THE
MEAN, WEIGHTED GEOMETRIC MEAN. MEDIAN. AND PERCENT DISTRIBUTION. BY RACE AND AGE. UNITED STATES, 19/9-80
Blood lead level (pg/dl)
Race and age
All racesc
All ages
6 months-5 years ....
6-17 years
18-74 years
White
All ages
6 months-5 years ....
6-17 years
18-74 years
Black
All ages
6 months-5 years ....
6*17 years .... . .
18-74 years
Estimated
population
in
thousands
Number b
examined
Arith-
metic
Mean
Standard
error of
the mean
Geometric
Mean
Median
Less
than
10
10-19
20-29
30-39
40+
Percent distribution
99,062
8,621
22,887
67.555
85,112
6,910
19,060
59,142
11,171
1,307
3.272
6,592
4,945
1.247
902
2,796
4,153
969
753
2,431
664
231
129
304
16.1
16.3
13.6
16.8
15.8
15.2
13.1
16.6
18.3
20.7
16.0
19.1
0.26
0.46
0.32
0.28
0.27
0.46
0.33
0.29
0.52
0.74
0.62
0.70
15.0
15.1
12.8
15.8
14.7
14.2
12.4
15.6
17.3
19.3
15.3
18.1
15.0
15.0
13.0
16.0
15.0
14.0
13.0
16.0
17.0
19.0
15.0
18.0
10.4
11.0
19.1
7.6
11.3
13.0
21.4
8.1
4.0
2.7
8.0
2.3
65.4
63.5
70.1
64.1
66.0
67.6
69.5
64.8
59.6
48.8
69.9
56.4
20.8
21.2
10.2
24.2
19.6
17.3
8.4
23.3
31.0
35.1
21.1
34.9
2.8
4.0
0.7
3.4
2.6
2.0
0.7
3.3
4.1
11.1
1.0
4.5
0.5
0.3
0.6
0.4
0.1
0.6
1.3
2.4
1.8
aAt the midpoint of the survey, March 1, 1978.
With lead determinations from blood specimens drawn by venipuncture.
Includes date for races not shown separately.
Numbers may not add to 100 percent due to rounding.
-------
PRELIMINARY DRAFT
TABLE 11-4. NHANES II BLOOD LEAD LEVELS OF FEMALES 6 MONTHS-74 YEARS, WITH WEIGHTED ARITHEMETIC MEAN,
STANDARD ERROR OF THE MEAN. WEIGHTED GEOMETRIC MEAN, MEDIAN, AND PERCENT DISTRIBUTION, BY RACE AND AGE, UNITED STATES, 1976-80
Blood lead level
Race and age
All racesc
All ages
G months-5 years ....
6-17 years
18-74 years
White
Al 1 ages
6 months-5 years ....
6*17 years ....
18-74 years
Black
All ages
6 months-5 years ....
6-17 years
18-74 years
Estimated
population
in a
thousands
Number .
examined
Arith-
metic
Mean
Standard
error of
the mean
Geometric
Mean
Median
Less
than
10
(ugAU)
10-19
20-29
30-39
40+
Percent distribution
104,492
8.241
22,077
74,173
89,417
6,732
18,470
64.215
12,682
1,277
3,256
8.148
4,988
1,125
818
3,045
4,216
907
671
2.638
668
188
134
346
11.9
15.8
11.4
11.8
11.7
14.7
11.0
11.7
13.4
21.0
13.6
12.7
0.23
0.42
0.3Z
0.22
0.23
0.44
0.31
0.23
0.45
0.69
0.64
0.44
11.1
14.6
10.6
11.0
10.9
13.7
10.3
10.9
12.6
19.8
12.8
12.0
11.0
15.0
11.0
11.0
11.0
14.0
11.0
11.0
13.0
20.0
13.0
12.0
33.3
13.5
36.6
33.7
34.8
16.1
40.0
34.6
21.5
2.2
17.7
24.7
60.5
63.2
59.3
60.6
59.6
67.3
56.9
59.9
67.3
41.6
71.9
68.1
5.7
19.8
3.9
5.2
5.0
14.8
2.9
5.0
10.3
45.3
10.0
7.2
0.4
3.0
0.2
0.3
0.4
1.6
0.2
0.4
0.7
9.2
0.4
0.2
0.5
0.2
0.2
0.2
0.2
0.1
1.7
aAt the midpoint of the survey, March 1, 1978.
With lead determinations from blood specimens drawn
Includes date for races not shown separately.
Numbers may not add to 100 percent due to rounding.
by venipuncture.
-------
PRELIMINARY DRAFT
6-month to 5-year group is higher than the older age groups, and men are higher than women.
Overall, younger children show only a slight age effect, with 2- to 3-year-olds having slight-
ly higher blood lead levels than older children or adults (see Figure 11-3). In the 6-17 year
grouping there is a decreasing trend in lead levels with increasing age. Holding age constant,
there are significant race and sex differences; as age increases, the difference in mean blood
leads between males and females increases.
For adults 18-74 years, males have greater blood lead levels than females for both whites
and blacks. There is a significant relationship between age and blood lead, but it differs
for whites and blacks. Whites display increasing blood lead levels until 35-44 years of age
and then a decline, while blacks have increasing blood lead levels until 55-64.
This study showed a clear relationship between blood lead level and family income group.
For both blacks and whites, increasing family income is associated with lower blood lead level.
At the highest income level the difference between blacks and whites-is the smallest, although
blacks still have significantly higher blood lead levels than whites. The racial difference
was greatest for the 6-month to 5-year age range.
The NHANES II blood lead data were also examined with respect to the degree of urbaniza-
tion at the place of residence. The three categories used were urban areas with population
greater than one million, urban areas with population less than one million and rural areas.
Geometric mean blood lead levels increased with degree of urbanization for all race-age groups
except for blacks 18-74 years of age (see Table 11-5). Most importantly, urban black children
aged 6 months to 5 years appeared to have distinctly higher mean blood lead levels than any
other population subgroup.
11.3.3.2 The Childhood Blood Lead Screening Programs. In addition to the nationwide picture
presented by the NHANES II (Annest et al., 1982) study regarding important demographic corre-
lates of blood lead levels, Billick et al. (1979, 1982) provide large scale analyses of blood
lead values in specific cities that also address this issue.
Billick et al. (1979) analyzed data from New York City blood lead screening programs from
1970 through 1976. The data include age in months, sex, race, residence expressed as health
district, screening information and blood lead values expressed in intervals of 10 mg/dl.
Only the venous blood lead data (178,588 values), clearly identified as coming from the first
screening of a given child, were used. All blood lead determinations were done by the same
laboratory. Table 11-6 presents the geometric means of the children's blood lead levels by
age, race and year of collection. The annual means were calculated from the four quarterly
means which were estimated by the method of Hasselblad et al. (1980).
PB11A2/B 11-15 7/29/83
-------
PRELIMINARY DRAFT
25
20
5
2
_r
UJ
£ 15
_i
0
§
Q
O 10
3
CO
O
Black
White
AGE, years
Figure 11-3. Geometric mean blood lead levels by race and age for younger children in
the NHANES II study. The data were furnished by the National Center of Health
Statistics.
PB11A2/B
11-16
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-5. WEIGHTED GEOMETRIC MEAN BLOOD LEAD LEVELS
FROM NHANES II SURVEY BY DEGREE OF URBANIZATION OF PLACE OF
RESIDENCE IN THE U.S. BY AGE AND RACE, UNITED STATES 1976-80
Race and age
All races
All ages
6 months-5 years
6-17 years
18-74 years
Whites
All ages
6 months-5 years
6-17 years
18-74 years
Blacks
All ages
6 months-5 years
6-17 years
18-74 years
= J
14.0
16.8
13.1
14.1
14.0
15.6
12.7
14.3
14.4
20.9
14.6
13.9
Urban,
. million
(2,395)a
(544)
(414)
(1,437)
(1,767)
(358)
(294)
(1,115)
(570)
(172)
(111)
(287)
Degree
<1
Geometric
12.8
15.3
11.7
12.9
12.5
14.4
11.4
12.7
14.7
19.3
13.6
14.7
of urbanization
Urban,
mi 1 lion
mean (ug/dl)
(3,869)
(944)
(638)
(2,287)
(3,144)
(699)
(510)
(1,935)
(612)
(205)
(113)
(294)
Rural
11.9
13.1
10.7
12.2
11.7
12.7
10.5
12.1
14.4
16.4
12.9
14.9
(3,669)
(884)
(668)
(2,117)
(3,458)
(819)
(620)
(2,019)
(150)
(42)
(39)
(69)
aNumber with lead determinations from blood specimens drawn by venipuncture.
Source: Annest et al., 1982.
PB11A2/B
11-17
7/29/83
-------
TABLE 11-6. ANNUAL GEOMETRIC MEAN BLOOD LEAD LEVELS FROM THE NEW YORK BLOOD LEAD SCREENING STUDIES
OF BILLICK ET AL. (1979). ANNUAL "GEOMETRIC MEANS ARE CALCULATED FROM QUARTERLY
GEOMETRIC MEANS ESTIMATED BY THE METHOD OF HASSELBLAD ET AL. (1980)
oo
Geometric mean blood lead level, pg/100 ml
Ethnic group
Black
Hispanic
White
Year
1970
1971
1972
1973
1974
1975
1976
1970
1971
1972
1973
1974
1975
1976
1970
1971
1972
1973
1974
1975
1976
1-12 mo
25.2
24.0
22.2
22.9
22.0
19.8
16.9
20.8
19.9
18.7
20.2
19.8
16.3
16.0
21.1
22.5
20.1
21.5
20.4
19.3
15.2
13-24 mo
28.9
29.3
26.0
26.6
25.5
22.4
20.0
23.8
22.6
20.5
21.8
21.5
18.7
17.4
25.2
22.7
21.6
21.8
21.7
17.9
18.2
25-36 mo
30.1
29.9
26.3
26.0
25.4
22.4
20.6
24.5
24.6
21.8
22.5
22.7
19.9
18.1
26.0
22.7
20.7
21.7
21.3
16.1
17.1
37-48 mo
28.3
29.3
25.4
25.3
24.3
21.9
20.2
24.7
24.4
22.2
22.8
22.5
20.1
18.2
24.8
23.5
20.8
20.2
21.1
18.5
16.6
49-60 mo
27.8
28.2
24.7
24.4
23.4
21.2
19.5
23.8
23.9
21.8
22.0
21.9
19.8
18.0
26.0
21.6
21.0
21.3
20.6
16.8
16.2
61-72 mo
26.4
27.2
23.9
24.1
21.8
21.4
18.2
23.6
23.4
21.8
21.5
20.5
19.2
16.7
22.6
21.3
20.2
20.7
19.5
15.4
15.9
73- mo
25.9
26.5
23.3
23.3
21.9
18.9
18.4
23.0
23.5
21.0
21.7
20.2
17.2
17.2
21.3
19.5
17.3
18.4
17.3
15.9
8.8
All ages
27.5
27.7
24.5
24.6
23.4
21.1
19.1
23.4
23.1
21.1
21.8
21.3
18.7
17.4
23.8
21.9
20.2
20.8
20.2
17.1
15.1
-------
PRELIMINARY DRAFT
All racial/ethnic groups show an increase in geometric mean blood level with age for the
first two years and a general decrease in the older age groups. Figure 11-4 shows the trends
for all years (1970-1976) combined.
The childhood screening data described by Billick et al. (1979) show higher geometric
mean blood lead values for blacks than for Hispanics or for whites. Table 11-6 also presents
these geometric means for the three racial/ethnic groups for seven years. Using the method of
Hasselblad et al. (1980), the estimated geometric standard deviations were 1.41, 1.42 and 1.42
for blacks, Hispanics and whites, respectively.
11.3.4 Time Trends
In the past few years a number of reports have appeared that examined trends in blood
lead levels during the 1970's. In several of these reports some environmental exposure esti-
mates are available. - . -
11.3.4.1 Time Trends in the Childhood Lead Poisoning Screening Programs. Billick and col-
leagues have analyzed the results of blood lead screening programs conducted by the City of
New York (Billick et al. , 1979; Billick 1982). Most details regarding this data set were al-
ready described, but Table 11-7 summarizes relevant methodologic information for these
analyses and for analyses done on a similar data base from Chicago, Illinois. The discussion
of the New York data below is limited to an exposition of the time trend in blood lead levels
from 1970 to 1977.
Geometric mean blood lead levels decreased for all three racial groups and for almost all
age groups in the period 1970-76 (Table 11-6). Table 11-8 shows that the downward trend
covers the entire range of the frequency distribution of blood lead levels. The decline in
blood lead levels showed seasonal variability, but the decrease in time was consistent for
each season. The 1977 data were supplied to EPA by Dr. Billick.
In addition to this time trend observed in New York City, Billick (1982) examined similar
data from Chicago and Louisville. The Chicago data set was much more complete than the Louis-
ville one, and was much more methodologically consistent. Therefore, only the Chicago data
will be discussed here. The lead poisoning screening program in Chicago may be the longest
continuous program in the United States. Data used in this report covered the years 1967-1980.
Because the data set was so large, only a 1 in 30 sample of laboratory records was coded for
statistical analysis (similar to procedures used for New York described above).
The blood lead data for Chicago contains samples that may be repeats, confirmatory analy-
ses, or even samples collected during treatment, as well as initial screening samples. This
is a major difference from the New York City data, which had initial screening values only.
PB11A2/B 11-19 7/29/83
-------
PRELIMINARY DRAFT
30
25
3 20
J 15
O
O
2
CO
O 10
D Blacks
O Whites
A Hispanics
I
1 234567
AGE, years
Figure 11-4. Geometric mean blood lead values by race and age for younger children
in the New York City screening program (1970-1976).
PB11A2/B
11-20
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-7. CHARACTERISTICS OF CHILDHOOD LEAD POISONING SCREENING DATA
New York
Chicago
Time period
Sampling technique
Analytic technique
Laboratory
Screening status
Race classification
and total number of
samples used in
analysis*
Raw data
Gasoline data
1970 - 1979
Venous
AAS
(Hasel method)
In house
Avai1able/unknown
Unknown 69,658
White 5,922
Black 51,210
Hispanic 41,364
Other 4,398
TOTAL 172,552
Decade grouped
Tri-state (NY, NJ, CT)
1970 - 1979
SMSA 1974 - 1979
1967 - 1980 (QTR 2)
Venous
AAS
(Hasel method)
In house
Unavailable
Nonblack 6,459
Black 20,353
TOTAL 26,812
Ungrouped
SMSA
*New York data set only includes
confirmatory and repeat samples.
first screens while Chicago includes also
Year
TABLE 11-8. DISTRIBUTION OF BLOOD LEAD LEVELS FOR 13 TO 48
MONTH OLD BLACKS BY SEASON AND YEAR* FOR NEW YORK SCREENING DATA
January - March
Percent
<15(jg/dl 15 to 34ug/dl
>34(jg/d1
July - September
Percent
<15ug/dl 15 to 34|jg/dl
>34pg/dl
1970
1971
1972
1973
1974
1975
1976
1977
(insufficient sample size)
3.8
4.4
7.3
9.2
11.1**
21.1
28.4
69.5
76.1
80.3
73.8
77.5**
74.1
66.8
26.7
19.5
12.4
17.0
11.4**
4.8
4.8
3.4
1.3
4.3
2.7
8.2
7.3**
11.9
19.9
54.7
56.0
72.2
62.4
65.4
81.3**
75.8
72.9
42.0
42.7
23.4
34.9
26.4
11.4**
12.3
7.2
* data provided by I.H. Billick
**Percents estimated using interpolation assuming a lognormal distribution.
PB11A2/B
11-21
7/29/83
-------
PRELIMINARY DRAFT
Chicago blood lead levels were all obtained on venous samples and were analyzed by one labora-
tory, the Division of Laboratories, Chicago Department of Health. Lead determinations were
done by atomic absorption. Racial composition was described in more detail than for New York,
but analysis showed there was no difference among the non-blacks, so they were pooled in the
final analysis.
Table 11-7 displays important characteristics of the Chicago and New York screening pro-
grams, including the number of observations involved in these studies. From tables in the ap-
pendices of the report (Billick, 1982), specific data on geometric mean blood lead values,
race, sex and sampling data for both cities are available. Consistency of the data across
cities is depicted in Figure 11-5. The long-term trends are quite consistent, although the
seasonal peaks are somewhat less apparent.
11.3.4.2 Newark. Cause et al. (1977) present data "'from Newark, New Jersey, that reinforce
the findings of Billick and coworkers. Cause et al. studied the levels of blood lead among 5-
and 6-year-old children tested by the Newark Board of Education during the academic years
1973-74, 1974-75 and 1975-76. All Newark schools participated in all years. Participation
rates were 34, 33 and 37 percent of the eligible children for the three years, respectively.
Blood samples were collected by fingerstick onto filter paper. The samples were then
analyzed for lead by atomic absorption spectrophotometry. The authors point out that finger-
stick samples are more subject to contamination than venous samples; and that because erythro-
cyte protoporphyrin confirmation of blood lead values greater than 50 |jg/dl was not done until
1974, data from earlier years may contain somewhat higher proportions of false positives than
.later years.
Blood lead levels declined markedly during this 3-year period. In the three years covered
by the study the percentage of children with blood lead levels less than 30 ug/dl went from 42
percent for blacks in 1973-74 to 71 percent in 1975-76; similarly, the percentages went from
56 percent to 85 percent in whites. The percentage of high risk children (>49 ug/dl) dropped
from 9 to 1 percent in blacks and from 6 to 1 percent in whites during the study period.
Unfortunately, no companion analysis was presented regarding concurrent trends in en-
vironmental exposures. However, Foster et al. (1979) reported a study from Newark that exam-
ined the effectiveness of the city's housing deleading program, using the current blood lead
status of children who had earlier been identified as having confirmed elevated blood lead
levels; according to the deleading program, these children's homes should have been treated to
alleviate the lead problem. After intensive examination, the investigators found that 31 of
the 100 children studied had lead-related symptoms at the time of Foster's study. Examination
of the records of the program regarding the deleading activity indicated a serious lack of
compliance with the program requirements. Given the results of Foster's study, it seems un-
likely that the observed trend was caused by the deleading program.
PB11A2/B 11-22 7/29/83
-------
PRELIMINARY DRAFT
CHICAGO
NEW YORK
1970 1971 1972 1973 1974 1976 1976 1977 1978 1979 1980
YEAR (Beginning Jan. 1)
Figure 11-5. Time dependence of blood lead for blacks, aged 24 to 35
months, in New York City and Chicago.
Source: Adapted from Billick (1982).
PB11A2/B
11-23
7/29/83
-------
PRELIMINARY DRAFT
11.3.4.3 Boston. Rabinowitz and Needleman (1982) report a study of umbilical cord blood lead
levels from 11,837 births between April 1979 and April 1981 in the Boston area. These repre-
sent 97 percent of the births occurring in a hospital serving a diverse population. Blood
samples were analyzed for lead by anodic stripping voltammetry after stringent quality control
procedures were used. External quality control checks were done by participation in the Blood
Lead Reference Program, conducted by the Centers for Disease Control. The average difference
between the investigators' results and the reference lab was 1.4 M9/dl.
The overall mean blood lead concentration was 6.56 ± 3.19 (standard deviation) with a
range from 0.0 to 37.0 ug/dl. A downward trend in umbilical cord blood lead levels (-0.89
ug/dl/yr) was noted over the two years of the study (see Figure 11-6).
11.3.4.4 NHANES II. Blood lead data from NHANES II (see Section 11.3.3.1) also show a signi-
ficant downward trend over time (Annest et al., 1983). Predicted mean blood lead levels
dropped from 14.6 ug/dl in February 1976 to 9.2 ug/dl in February of 1980. Mean values from
these national data presented in 28 day intervals from February 1976 to February 1980 are dis-
played in Figure 11-7.
The decreases in average blood lead levels were found for both blacks and whites, all age
groups and both sexes. Further statistical analysis suggested that the decline was not en-
tirely due to season, income, geographic region or urban-rural differences. The analyses of
the quality control data showed no trend in the blind quality control data.
A review panel has examined this data, and a report of their findings is in Appendix 11-D.
The panel concluded that there was strong evidence of a downward trend during the period of
the study. The panel further stated that the magnitude of this drop could be estimated, and
that it appeared not only in the entire population, but in some major subgroups as well.
11.3.4.5 Other Studies. Oxley (1982) reported an English study that looks at the recent past
time trend in blood lead levels. Preemployment physicals conducted in 1967-69 and 1978-80
provided the subjects for the study. Blood samples were collected by venipuncture. Different
analytical procedures were used in the two surveys, but a comparison study showed that the
data from one procedure could be reliably adjusted to the other procedure. The geometric mean
blood lead levels declined from 20.2 to 16.6 ug/dl.
11.3.5 Distributional Aspects of Population Blood Lead Levels
The importance of the distribution form of blood lead levels was briefly discussed in
Section 11.2.3. The distribution form determines which measure of central tendency (arith-
metic mean, geometric mean, median) is most appropriate. It is even more important in esti-
mating percentiles in the upper tail of the distribution, an issue of much importance in esti-
mating percentages (or absolute numbers) of individuals in specific population groups likely
to be experiencing various lead exposure levels.
PB11A2/B 11-24 7/29/83
-------
PRELIMINARY DRAFT
12.0
10.0
8.0
6.0
4.0
Model Predicted
Actual Data
I I
I
T
4/79 7/79 10/79 1/80 4/80
TIME, days
7/80
10/80
1/81
4/81
Figure 11-6. Modeled umbilical cord blood lead levels by date of sample collection
for infants in Boston.
Source: Rabinowitz and Needleman (1982).
PB11A2/B
11-25
7/29/83
-------
00
25
_f
IU
IU
9 16
IU
O
O
O
CD
10
>-> O
7 <
WINTER 1976
(FEB.)
WINTER 1977
(FEB.)
WINTER 1978
(FEB.)
FALL 1978 WINTER 1979
(OCT.) (FEB.)
WINTER 1980
(FEB.)
I
I
10 15 20 25 30 35
CHRONOLOGICAL ORDER, 1 unit = 28 days
40
45
50
55
Figure 11-7. Average blood lead levels of U.S. population 6 months—74 years. United States,
February 1976—February 1980, based on dates of examination of NHANES II examinees with
blood lead determinations.
Source: Annest et al. (1983).
ro
!£>
00
co
-------
PRELIMINARY DRAFT
Distribution fitting requires large numbers of samples taken from a relatively homo-
geneous population. A homogeneous population is one in which the distribution of values
remains constant when split into subpopulations. These subpopulations could be defined by
demographic factors such as race, age, sex, income, degree of urbanization, and by degree of
exposure. Since these factors always have some effect, a relatively homogeneous population
will be defined as one with minimal effects from any factors that contribute to differences in
blood lead levels.
Several authors have suggested that the distribution of blood lead levels for any rela-
tively homogeneous population closely follows a lognormal distribution (Yankel et al., 1977;
Tepper and Levin, 1975; Azar et al., 1975). Lognormality has been noted for other metals,
such as 90Sr, 144Ce, Pu and Ti in various tissues of human populations (Cuddihy et al., 1979;
Schubert et al., 1967). Yankel et al. (1977), Tepper and Levin (1975) and Angle and Mclntire
(1979) all found their blood lead data to be lognormally distributed. Further analysis by EPA
of the Houston study of Johnson et al. (1974), the study of Azar et al. (1975) and the New
York children screening program reported by Billick et al. (1979) also demonstrated that a
lognormal distribution provided a good fit to the data.
The only nationwide survey of blood lead levels in the U.S. population is the NHANES II
survey (Ar.nest et al. , 1982). In order to obtain a relatively homogeneous subpopulation of
lower environmental exposure, the analysis was restricted to whites not living in an SMSA with
a family income greater than $6,000 per year, the poverty threshold for a family of four at
the midpoint of study as determined by the U.S. Bureau of Census. This subpopulation was
split into four subgroups based on age and sex. The summary statistics for these subgroups
are in Table 11-9.
Each of these four subpopulations were fitted to five different distributions: normal,
lognormal, gamma, Weibull and Wald (Inverse Gaussian) as shown in Table 11-10. Standard chi-
square goodness-of-fit tests were computed after collapsing the tails to obtain an expected
cell size of five. The goodness-of-fit test and likelihood functions indicate that the log-
normal distribution provides a better fit than the normal, gamma or Weibull. A histogram and
the lognormal fit for each of the four subpopulations appear in Figure 11-8. The Wald distri-
bution is quite similar to the lognormal distribution and appears to provide almost as good a
fit. Table 11-10 also indicates that the lognormal distribution estimates the 99th percentile
as well as any other distribution.
Based on the examination of the NHANES II data, as well as the results of several other
papers, it appears that the lognormal distribution is the most appropriate for describing the
distribution blood lead levels in homogeneous populations with relatively constant exposure
levels.
PB11A2/B 11-27 7/29/83
-------
PRELIMINARY DRAFT
The lognormal distribution appears to fit well across the entire range of the distribution
including the right tail. It should be noted, however, that the data being fitted are the
result of both measurement variation and population variation. The measurement variation
alone does not follow a lognormal distribution, as was shown by Saltzman et al., 1983.
TABLE 11-9. SUMMARY OF UNWEIGHTED BLOOD LEAD LEVELS IN WHITES
NOT LIVING IN AN SMSA WITH FAMILY INCOME GREATER THAN $6,000
Unweighted Mean
Subgroup
age 1/2 to 6
age 6 to 18
age 18+, men
age 18+, women
Sample
Size
752
573
922
927
Arith.
Mean
ug/dl
13.7
11.3
15.7
10.7
Geom.
Mean
ug/dl
12.9
10.6
14.7
10.0
Sample
Median
ug/dl
13.0
10.0
15.0
10.0
99th
fctile
ug/dl
32.0
24.0
35.8
23.0
Arith.
Std. Dev.
ug/dl
5.03
4.34
5.95
4.14
Geom.
Std. Dev.
ug/dl
1.43
1.46
1.44
1.46
It is obvious that even relatively homogeneous populations have considerable variation
among individuals. The estimation of this variation is important for determination of the
upper tail of the blood lead distribution, the group at highest risk. The NHANES 'II study
provides sufficent data to estimate this variation. In order to minimize the effects of loca-
tion, income, sex and age, an analysis of variance procedure was used to estimate the varia-
tion for several age-race groups. The variables just mentioned were used as main effects, and
the resulting mean square errors of the logarithms are in Table 11-11. The estimated
geometric standard deviations represent the estimated variances for subgroups with comparable
sex, age, income and place of residence. These are not necessarily representative of the
variances seen for specific subgroups described in the NHANES II study.
Analytical variation, which exists in any measurement of any kind, has an impact on the
bias and precision of statistical estimates. For this reason, it is important to estimate the
magnitude of variation. Analytical variation consists of both measurement variation (vari-
ation between measurements run at the same time) and variation created by analyzing samples at
different times (days). This kind of variation for blood lead determinations has been dis-
cussed by Lucas (1981).
The NHANES II survey is an example of a study with excellent quality control data. The
analytical variation was estimated specifically for this study by Annest et al. (1983). The
analytical variation was estimated as the sum of components estimated from the high and low
PB11A2/B
11-28
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-10. SUMMARY OF FITS TO NHANES II BLOOD LEAD LEVELS
OF WHITES NOT LIVING IN AN SMSA, INCOME GREATER THAN $6,000,
FOR FIVE DIFFERENT TWO-PARAMETER DISTRIBUTIONS
Normal
Lognormal
Gamma
Wei bull
Wald
Normal
Lognormal
Gamma
Weibull
Wald
Normal
Lognormal
Gamma
Weibull
Wald
Normal
Lognormal
Gamma
Weibull
Wald
Chi -square
75.52
14.75
17.51
66.77
15.71
Chi -square
39.58
3.22
4.88
24.48
2.77
Chi-square
156.98
12.22
34.26
132.91
14.42
Chi-square
66.31
7.70
11.28
56.70
10.26
Children
D.F.
8
10
9
8
10
Children
D.F.
6
8
7
6
8
Men £18
D.F.
10
13
12
11
13
Men 218
D.F.
5
8
7
6
8
<6 years
p-value
0.0000
0.1416
0.0413
0.0000
0.1083
6^ years £17
p-value
0.0000
0.9197
0.6745
0.0004
0.9480
years
p-value
0.0000
0.5098
0.0006
0.0000
0.3450
years
p-value
0.0000
0.4632
0.1267
0.0000
0.2469
log-
likelihood
-2280.32
-2210.50
-2216.51
-2271.57
-2211.83
log-
likelihood
-1653.92
-1607.70
-1609.33
-1641.35
-1609.64
log-
likelihood
-2952.85
-2854.04
-2864.79
-2934.14
-2855.94
log-
likelihood
-2631.67
-2552.12
-2553.34
-2611.78
-2556.88
deviation*
at
99 %tile
6.61
2.57
4.68
5.51
2.76
deviation*
at
99 %ti1e
2.58
-1.50
-0.64
1.72
-1.30
deviation*
at
99 %tile
6.24
1.51
4.00
4.88
1.72
deviation*
at
99 %tile
2.68
-1.18
0.90
1.73
-1.01
"observed ggth sample percent!le minus predicted 99th percentile
PB11A2/B
11-29
7/29/83
-------
PRELIMINARY DRAFT
15.5
23.5
31.5
7.5
15.5
23.5
31.5
BLOOD LEAD LEVELS l^g/dl)
FOR 6 MONTHS TO 6 YEAR OLD CHILDREN
BLOOD LEAD LEVELS
FOR 6 TO 17 YEAR OLD CHILDREN
U
ui
§
\
7.B
15.5
23.5
31.5
BLOOD LEAD LEVELS (/jg/dl)
FOR MEN »18 YEARS OLD
16.5
23.5 31.5
BLOOD LEAD LEVELS
FOR WOMEN £18 YEARS OLD
Figure 11-8. Histograms of blood lead levels with fitted lognormal curves for the NHAIMES II
study. All subgroups are white, non-SMSA residents with family incomes greater than $6000.
PB11A2/8
11-30
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-11. ESTIMATED MEAN SQUARE ERRORS RESULTING FROM
ANALYSIS OF VARIANCE ON VARIOUS SUBPOPULATIONS
OF THE NHANES II DATA USING UNWEIGHTED DATA
White,
Age Non SMSA
0.5 to 6 0.0916
(1.35)*
6 to 18 0.0814
(1.33)
18+, men 0.1155
(1.40)
18+, women 0.1083
(1.39)
White, SMSA,
not central city
0.0839
(1.34)
0.0724
(1-31)
0.0979
(1.37)
0.0977
(1-37)
White,
central city
0.1074
(1-39)
0.0790
(1.33)
0.1127
(1.40)
0.0915
(1.35)
Black,
central city
0.0978
(1-37)
0.0691
(1.30)
0.1125
(1.40)
0.0824
(1.33)
Note: Mean square errors are based on the logarithm of the blood lead levels.
^Estimated geometric standard deviations are given in parentheses.
blind pool and from the replicate measurements in the study of Griffin et al. (1975). The
overall estimate of analytical variation for the NHANES II study was 0.02083.
Analytical variation causes a certain amount of misclassification when estimates of the
percent of individuals above or below a given threshold are made. This is because the true
value of a person's blood lead could be below the threshold, but the contribution from analy-
tical variation may push the observed value over the threshold. The reverse is also possible.
These two types of misclassifications do not necessarily balance each other.
Annest et al. (1983) estimated this misclassification rate for several subpopulations in
the NHANES II data using a threshold value of 30 pg/dl. In general, the percent truly greater
than this threshold was approximately 24 percent less than the prevalence of blood lead levels
equal to or greater than 30 pg/dl, estimated from the weighted NHANES II data. This is less
tlian the values predicted by Lucas (1981) which were based on some earlier studies.
11.3.6 Exposure Covariates of Blood Lead Levels in Urban Children
Results obtained from the NHANES II study show that urban children generally have the
highest blood lead levels of any non-occupationally exposed population group. Furthermore,
black urban children have significantly higher blood lead levels than white urban children.
Several studies have been reported in the past few years that look at determinants of blood
PB11A2/B
11-31
7/29/83
-------
PRELIMINARY DRAFT
lead levels in urban children (Stark et al., 1982; Charney et al., 1980; Hammond et al., 1980;
Gilbert et al., 1979).
11.3.6.1 Stark Study. Stark et al. (1982) used a large scale lead screening program in New
Haven, Connecticut, during 1974-77 as a means of identifying study subjects. The screening
program had blood lead levels on 8289 children ages 1-72 months, that represented about 80
percent of the total city population in that age group. From this initial population, a much
smaller subset of children was identified for a detailed environmental exposure study. Using
the classifying criteria of residential stability and repeatable blood lead levels (multiple
measurements fell into one of three previously defined blood lead concentration categories), a
potential study population of 784 was identified. Change of residence following identifica-
tion and refusal to let sanitarians make inspections resulted in 407 children being dropped;
the final study population contained 377 children.
With the exception of dietary lead intake, each child's potential total lead exposure was
assessed. Information was obtained on lead in air, house dust, interior and exterior paint
and soil near and far from the home. A two percent sample of homes with children having
elevated lead levels had tap water lead levels assessed. No water lead levels above the
public health service standard of 50 ug/1 were found. Socioeconomic variables were also
obtained.
For all children in the study, micro blood samples were taken and analyzed for lead by
AAS with Delves cup attachment. Blood lead values were found to follow a lognormal distri-
bution. Study results were presented using geometric means and geometric standard deviation.
Among the various environmental measurements a number of significant correlation coefficients
were observed. However, air lead levels were independent of most of the other environmental
variables. Environmental levels of lead did not directly follow socioeconomic status. Most
of the children, however, were in the lower socioeconomic groups.
Multiple regression analyses were performed by Stark et al. (1982) and by EPA*. Stark
and coworkers derived a log-log model with R2 = 0.11, and no significant effects of race or
age were found. EPA fitted a linear exposure model in logarithmic form with results shown in
Table 11-12. Significant differences among age groups were noted, with considerably improved
predictability (R2 = 0.29, 0.30, 0.14 for ages 0-1, 2-3, and 4-7). Sex was not a significant
variable, but race equal black was significant at ages 4-7. Air lead did not significantly
improve the fit of the model when other covariates were available, particularly dust, soil,
paint and housekeeping quality. However, the range of air lead levels was small (0.7-1.3
pg/m3) and some of the inhalation effect may have been confounded with dust and soil inges-
tion. Seasonal variations were important at all ages.
*NOTE: The term EPA analyses refers to calculations done at EPA. A brief discussion of the
methods used is contained in Appendix 11-B; more detailed information is available at EPA
upon request.
PB11A2/B 11-32 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-12. MULTIPLE REGRESSION MODELS FOR BLOOD
OF CHILDREN IN NEW HAVEN, CONNECTICUT,
SEPTEMBER 1974 - FEBRUARY 1977
LEAD
Age group, years
Regression Coefficients and Standard Errors
0-1 2-3
4-7
Summer - Winter
Dust, ug/g
Housekeeping Quality
Soil near house, ug/g
Soil at curb, |jg/g
Paint, child's bedroom
Paint outside house
Paint quality
Race = Black
6.33 ± 2.11*
0.00402 ± 0.00170*
4.38 ± 2.02*
0.00223 ± 0.00091*
0.00230 ± 0.00190
0.0189 ± 0.0162
-0.0023 ± 0.0138
0.89 ± 1.71
2.16 ± 2.05
3.28 ± 1.30*
0.00182 ± 0.00066*
1.75 ± 1.17
-0.00016 ± 0.00042
0.00203 ± 0.00082*
0.0312 ± 0.0066*
0.0200 ± 0.0069*
3.38 ± 0.96*
0.07 ± 1.09
2.43 ± 1.38*
0.00022 ± 0.00077
-1.61 ± 1.12
0.00060 ± 0.00041
0.00073 ± 0.00079
0.0110 ± 0.0064*
± 0.0067*
± 1.15*
0.0172
4.14
5.81
± 1.00*
Residual Standard Deviations 0.1299
Multiple R2 0.289
Sample size (blood samples) 153
0.0646
0.300
334
0.1052
0.143
439
*Significant positive coefficient, one-tailed p <0.05
11.3.6.2 Charney Study. Charney et al. (1980) conducted a case control study of children 1.5
to 6 years of age with highly elevated and non-elevated blood lead levels. Cases and controls
were initially identified from the lead screening programs of two Rochester, New York, health
facilities. Cases were defined as children who had at least two blood lead determinations
between 40 and 70 pg/dl and FEP values greater than 59 ug/dl during a 4-month period. Con-
trols were children who had blood lead levels equal to or less than 29 pg/dl and FEP equal to
or less than 59 pg/dl. High level children were selected first and low level children were
group matched on age, area of residence, and social class of the family. Home visits were
made to gain permission as well as to gather questionnaire and environmental data. Lead anal-
yses of the various environmental samples were done at several different laboratories. No
specification was provided regarding the analytical procedures followed.
The matching procedure worked well for age, mother's educational level and employment
status. There were more blacks in the high lead group as well as more Medicaid support. These
factors were then controlled in the analysis; no differences were noted between the high and
P811A2/B
11-33
7/29/83
-------
PRELIMINARY DRAFT
low blood lead groups regarding residence on high traffic density streets (>10,000 vehicles/
day) or census tract of residence.
The two groups differed regarding mean house dust lead levels (1265 ug/sample for high
and 123 (jg/sample for low). Median values also differed, 149 vs. 55 ug/sample. One-third of
the children in the low blood lead group had house dust lead samples with more lead than those
found in any middle class home previously investigated.
There were considerably greater quantities of lead on the hands of the high blood lead
group compared with the low group (mean values were 49 (jg/sample and 21 pg/sample, respective-
ly). Hand and house dust lead levels were correlated (r = 0.25) but the relationship was not
linear. At the low end of the house dust lead values, hand dust was always low but the con-
verse was not true: not every child exposed to high house dust lead had high hand dust levels.
In addition to hand and house dust lead, other factors differentiated the high and low
blood lead groups. Although both groups had access to peeling paint in their homes (~2/3),
paint lead concentrations exceeding 1 percent were found more frequently in the high as oppo-
sed to the low group. Pica (as defined in Chapter Seven) was more prevalent in the high lead
group as opposed to the low lead group.
Since the data suggested a multifactorial contribution of lead, a multiple regression
analysis was undertaken. The results suggest that hand lead level, house dust lead level,
lead in outside soil, and history of pica are very important in explaining the observed vari-
ance in blood lead levels.
11.3.6.3 Hammond Study. Hammond et al. (1980) conducted a study of Cincinnati children with
the dual purpose of determining whether inner city children with elevated blood lead levels
have elevated fecal lead and whether fecal lead correlates with lead-base paint hazard in the
home or traffic density as compared with blood lead.
Subjects were recruited primarily to have high blood lead levels. Some comparison chil-
dren with low blood lead levels were also identified. The three comparison children had to be
residentially stable so that their low blood lead levels were reflective of the lead intake of
their current environment. The subjects from the inner city were usually from families in
extremely depressed socio-economic circumstances.
Stool samples were collected on a daily basis for up to 3 weeks, then analyzed for lead.
2
Fecal lead levels were expressed both as mg/kg day and as mg/m day.
An environmental assessment was made at the home of each child. Paint lead exposure was
rated on a three-point scale (high, medium and low) based on paint lead level and integrity of
the painted wall. Air lead exposure was assessed by the point scale (high, medium and low)
based on traffic density, because there are no major point sources of lead in the Cincinnati
area.
PB11A2/B 11-34 7/29/83
-------
PRELIMINARY DRAFT
Blood samples were collected on an irregular basis but were taken sufficiently often to
havp at least one sample from a child from every house studied. The blood samples were ana-
lyzed for lead by two laboratories that had different histories of performance in the CDC pro-
ficiency testing program. All blood lead levels used in the statistical analysis were ad-
justed to a common base. Because of the variable number of fecal and blood lead levels, the
data were analyzed using a nested analysis of variance.
The homes of the children were found to be distributed across the paint and traffic lead
exposure categories. Both fecal lead levels and blood lead levels were positively associated
with interior paint lead hazard. A marginal association between fecal lead levels and
exterior paint hazard was also obtained. Neither fecal lead or blood lead was found to be
associated with traffic density; the definition of the high traffic density category, however,
began at a low level of traffic flow (7500 cars/day).
Examination of fecal and blood lead levels by sex and race showed that black males had
the highest fecal lead excretion rates followed by white males and black females. White fe-
males were only represented by two subjects, both of whom had high fecal lead excretion.
Blood lead levels were more influenced by race than by sex. The results suggested that chil-
dren in high and medium paint hazard homes (high = at least 1 surface >0.5 percent Pb, peeling
or loose) were probably ingesting paint in some form. This could not be confirmed, however,
by finding physical evidence in the stools.
Long term stool collection in a subset of 13 children allowed a more detailed examination
of the pattern of fecal lead excretion. Two patterns of elevated fecal lead excretion were
noted. The first was a persistent elevation compared with controls; the second was markedly
elevated occasional spikes against a normal background.
One family moved from a high hazard home to a low one during the course of the study.
This allowed a detailed examination of the speed of deleading of fecal and blood lead level.
The fecal levels decreased faster than the blood lead levels. The blood leads were still
elevated at the end of the collection.
11.3.6.4 Gilbert Study. Gilbert et al. (1979) studied a population of Hispanic youngsters in
Springfield, Massachusetts, in a case control study designed to compare the presence of
sources of lead in homes of lead poisoned children and appropriately matched controls. Cases
were defined as children having two consecutive blood lead levels greater than 50 ug/dl. Con-
trols were children with blood lead levels less than or equal to 30 ug/dl who had no previous
history of lead intoxication and were not siblings of children with blood lead levels greater
than 30 ug/dl. Study participants had to be residentially stable for at least 9 months and
not have moved into their current home from a lead contaminated one. All blood lead levels
were analyzed by Delves cup method of AAS. Cases and controls were matched by age (±3 months),
PB11A2/B 11-35 7/29/83
-------
PRELIMINARY DRAFT
sex and neighborhood area. The study population consisted of 30 lead intoxication cases and
30 control subjects.
Home visits were undertaken to gather interview information and conduct a home in-
spection. Painted surfaces were assessed for integrity of the surface and lead content. Lead
content was measured by X-ray fluorimetry. A surface was scored as positive if the lead con-
tent exceeded 1.2 mg/cm2. Drinking water lead was assessed for each of the cases and was
found to contain less than 50 ug/1, sufficiently low so as not to constitute a hazard. Tap
water samples were not collected in the homes of the controls. Soil samples were collected
from three sites in the yard and analyzed for lead by X-ray fluorometry.
Cases and controls were compared on e-.ivironment.al lead exposures and interview data using
McNemar's test for pair samples. The odds ratio was calculated as an estimator of the rela-
tive risk on all comparisons. Statistically significant differences between cases and con-
trols were noted for lead in paint and the presence of loose paint. Large odds ratios (>10)
were obtained; there appeared to be little influence of age or sex on the odds ratios.
Significant differences between cases and controls were obtained for both intact and
loose paint by individual surfaces within specific living areas of the home. Surfaces acces-
sible to children were significantly associated with lead poisoning status while inaccessible
surfaces generally were not. Interestingly, the odds ratios tended to be larger for the in-
tact surface analysis than for the loose paint one.
Median paint lead levels in the homes of cases were substantially higher than those in
the homes of controls. The median paint lead for exterior surfaces in cases was about 16-20
mg/cm2 and about 10 mg/cm2 for interior surfaces. Control subjects lived in houses in which
the paint lead generally was less than 1.2 mg/cm2 except for some exterior surfaces.
.Soil lead was significantly associated with lead poisoning; the median soil lead level
for homes of cases was 1430 ug/g, while the median soil lead level for control homes was
440 ug/g.
11.4 STUDIES RELATING EXTERNAL DOSE TO INTERNAL EXPOSURE
The purpose of this section is to assess the importance of environmental exposures In
determining the level of lead in human populations. Of prime interest are those studies that
yield quantitative estimates of the relationship between air lead exposures and blood lead
levels. Related to this question is the evaluation of which environmental sources of airborne
lead play a significant role in determining the overall impact of air lead exposures on blood
lead levels.
A factor that complicates the analysis presented here is that lead does not remain sus-
pended in the atmosphere but rather falls to the ground, is incorporated into soil, dust, and
PB11A2/B 11-36 7/29/83
-------
PRELIMINARY DRAFT
water, and enters the food chain over time (see Figure 11-1). Since man is exposed to lead
from all of these media, as will be demonstrated below, studies that relate air lead levels to
blood lead levels (especially experimental exposure studies) may underestimate the overall
impact of airborne lead on blood lead levels. In observational studies, the effects of air
lead will thus be confounded with lead exposures from other pathways. The simultaneous pre-
sence of lead in multiple environmental media requires the use of multiple variable analysis
techniques or surrogate assessment of all other external exposures. Virtually no assessments
of simultaneous exposures to all media have been done.
Although no study is ever done perfectly, there are several key factors that are present
in good studies relating external exposure to internal exposure of lead:
(1) The study population is well-defined.
(2) There is a good measure of the exposure of each individual.
(3) The response variable (blood lead) is measured with adequate quality control,
preferably with replicates.
(4) The statistical analysis model is biologically plausible and is consistent with
the data.
(5) The important covariates are either controlled for or measured.
Even studies of considerable importance do not address all of these factors adequately.
We have selected as key studies (for discussion below) those which address enough of these
factors sufficiently well to establish meaningful relationships.
11.4.1 Air Studies
The studies emphasized in this section are those most relevant to answering the following
question: If there is moderate change in average ambient air lead concentrations due to
changes in environmental exposure (at or near existing EPA air lead standards), what changes
are expected in blood lead levels of individual adults and children in the population? Longi-
tudinal studies in which changes in blood lead can be measured in single individuals as
responses to changes in air lead are discussed first. The cross-sectional relationship
between blood lead and air lead levels in an exposed population provides a useful but differ-
ent kind of information, since the population "snapshot" at some point in time does not direc-
tly measure changes in blood lead levels or responses to changes in air lead exposure. We
have also restricted consideration to those individuals without known excessive occupational
or personal exposures (except, perhaps, for some children in the Kellogg/Silver Valley study).
The previously published analyses of relevant studies have not agreed on a single form
for the relationship between air lead and blood lead. All of the experimental studies have at
least partial individual air lead exposure measures, as does the cross-sectional observational
PB11A2/B 11-37 7/29/83
-------
PRELIMINARY DRAFT
study of Azar et al. (1975) The 1974 Kellogg/Silver Valley study (Yankel et al., 1977) has
also been analyzed using several models. Other population cross-sectional studies have been
analyzed by Snee (1981). The most convenient method for summarizing these diverse studies and
their several analyses is by use of the blood lead-air lead slope (p), where p measures the
change in blood lead that is expected for a unit change in air lead. If determined for indi-
vidual subjects in a study population, this slope is denoted p.. If the fitted equation is
linear, then p or p. is the slope of the straight line relationship at any air lead level. If
the fitted relationship is nonlinear, then the slope of the relationship measures the expected
effect on blood lead of a small change in air lead at some given air lead value and thus will
be somewhat different at different air lead levels. It is necessary to compare the slopes of
the nonlinear relationships at the same value of air lead or change in air lead. A discussion
of the linear, nonlinear and compartment models is in Appendix 11A-B.
Snee (1982b,c) has indicated that inclusion of additional sources of lead exposure im-
proves biological plausibility of the models. It is desirable that these sources be as spe-
cific to site, experiment and subject as possible.
11.4.1.1 The Griffin et al. Study. In two separate experiments conducted at the Clinton
Correctional Facility in 1971 and 1972, adult male prisoner volunteers were sequestered in a
prison hospital unit and exposed to approximately constant levels of lead oxide (average
10.9 ug/m3 in the first study and 3.2 ug/m3 in the second). Volunteers were exposed in an ex-
posure chamber to an artificially generated aerosol of submicron-sized particles of lead
dioxide. All volunteers were introduced into the chamber 2 weeks before the initiation of the
.exposure; the lead exposures were scheduled to last 16 weeks, although the volunteers could
drop out whenever they wished. Twenty-four volunteers, including 6 controls, participated in
the 10.9 ug/m3 exposure study. Not all volunteers completed the exposure regimen. Blood lead
levels were found to stabilize after approximately 12 weeks. Among 8 men exposed to 10.9
ug/m3 for at least 60 days, a stabilized mean level of 34.5 ±5.1 ug/dl blood was obtained, as
compared with an initial level of 19.4 ± 3.3 ug/dl. All but two of the 13 men exposed at 3.2
ug/m3 for at least 60 days showed increases and an overall stabilized level of 25.6 ± 3.9
ug/dl was found, compared with an initial level of 20.5 ± 4.4 ug/dl. This represented an in-
crease of about 25 percent above the base level.
The aerosols used in this experiment were somewhat less complex chemically, as well as
somewhat smaller, than those found in the ambient environment. Griffin et al. (1975),
however, pointed out that good agreement was achieved on the basis of the comparison of their
observed blood lead levels with those predicted by Goldsmith and Hexter's (1967) equation;
that is, Iog10 blood lead = 1.265 + 0.2433 log,- atmospheric air lead. The average diet con-
tent of lead was measured and blood lead levels were observed at 1- or 2-week intervals for
PB11A2/B 11-38 7/29/83
-------
PRELIMINARY DRAFT
several months. Eight subjects received the maximum 4-month exposure to 10.9 ug/m3; nine sub-
jects were exposed for 1 to 3 months. Six subjects had the maximum 4-month exposure to 3.2
ug/m3, and eight others had shorter exposures.
Compartmental models have been fitted to these data by 0'Flaherty et al. (1982) and by
EPA. The basis of these models is that the mass of lead in each of several distinct pools or
compartments within the body changes according to a system of coupled first-order linear dif-
ferential equations with constant fractional transfer rates (Batschelet et al., 1979; Rabino-
witz et al., 1976). Such a model predicts that when the lead intake changes from one constant
level to another, then the relationship between the mass of lead in each compartment and time
with constant intake has a single exponential term.
The subjects at 3.2 ug/m3 exhibited a smaller increase in blood lead, with corres-
pondingly less accurate estimates of the parameters. Several of the lead-exposed subjects
failed to show an increase.
Figure 11-9 shows a graph of the blood lead levels for the 10.9 ug/m3 exposure by length
of exposure. Each person's values are individually normalized, and then averaged across
5 80-
CO
K
C
o
UJ
s
u
oc
UJ
Q.
20 —
0 —
10
20 30
40 SO
60
70
80
90 100 110 120
DAY OF EXPOSURE
Figure 11-9. Graph of the average normalized increase in blood lead for subjects exposed to
10.9 M9'm3 of lead in Griffin et al. study (1975).
PB11A2/B
11-39
7/29/83
-------
PRELIMINARY DRAFT
subjects for each time period. The smooth curve shows a fitted one-compartment model, assum-
ing pre-exposure equilibrium and constant lead intake during exposure.
EPA has reanalyzed these data using a two-compartment model for two reasons:
(1) Semi logarithmic plots of blood lead vs. time for most subjects showed a two-
component exponential decrease of blood lead during the postexposure or washout
phase of the experiments. Rabinowitz et al. (1977) show that at least two
pools are necessary to model blood lead kinetics accurately. The first pool is
tentatively identified with blood and the most labile soft tissues. The second
pool probably includes soft tissues and labile bone pools.
(2) Kinetic models are needed to account for the subjects' lead burdens not being
in equilibrium at any phase of the experiments.
The pre-exposure decline in Figure 11-9 is apparently real and suggests a low pre-exposure
lead intake. The deviation from the fitted curve after about 50 days suggests a possible
change in lead intake at that time.
Previously published analyses have not used data for all 43 subjects, particularly for
the same six subjects (labeled 15 to 20 in both experiments) who served as controls both
years. These subjects establish a baseline for non-inhalation exposures to lead, e.g., in
diet and water, and allow an independent assessment of within-subject variability over time.
EPA analyzed data for these subjects as well as others who received lead exposures of shorter
duration.
The estimated blood lead inhalation slope, p, was calculated for each individual subject
according to the formula:
(Change in intake, ug/day) x (mean residence time in blood, day)
p = 5
(Change in air exposure, ug/m ) x (Volume of distribution, dl)
The mean values of these parameters are given in Tables 11-13 through 11-15. The changes In
air exposure were 10.9 - 0.15 = 10.75 pg/m for 1970-71 and 3.2 - 0.15 = 3.05 ug/m3 -jn 1971-
72. Paired sample t-tests of equal means were carried out for the six controls and five sub-
jects with exposure both years, and independent sample t-tests were carried out comparing the
remaining 12 subjects the first year and nine different subjects the next year. All standard
error estimates include within-subject parameter estimation uncertainties as well as between
subject differences. The following are observations.
(1) Non-inhalation lead intake of the control subjects varied substantially during the
second experiment at 3.2 ug/m , with clear indication of low intake during the 14-day pre-
exposure period (net decrease of blood lead), see Figure 11-10. There was an increase in lead
intake (either equilibrium or net increase of blood lead) during the exposure period.
PB11A2/B 11-40 7/29/83
-------
PRELIMINARY DRAFT
KEY
O Subject 15
A Subject 16
A Subject 17
• Subject 18
O Subject 19
• Subject 20
1
0
1 1 1 1 1
20
40 60 80
100
1
120
1 1 1 1 1
140 160 180 200
• ...i,, i nr*T"T rvpofi IDF.,, &
TIME, days
Figure 11-10. Control subjects in Griffin experiment at 3.2
PB11A/B
11-41
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-13. GRIFFIN EXPERIMENTS - SUBJECTS EXPOSED TO AIR LEAD BOTH YEARS
Subject
At 3.2
1
2
'P3
4
5
Mean
Mean w/o
At 10.9
3
13
14
7
4
Mean Residence Time.d.
At 3.2 At 10.9
42.1 ±
47.6 ±
48.0 ±
42.5 ±
43.6 ±
44.7 ±
17.4
21.4
21.7
17.6
18.2
8.7
55.2
38.4
40.1
50.1
35.9
43.9
± 27.2
± 14.5
± 15.8
± 22.5
± 12.8
± 9.4
Change in Intake,
Post-Pre-exposure, pq/d*
At 3.2 At" 10. 9
-4.4 + 13.8
3.1 ± 14.1
3.3 ± 13.1
12.0 ± 14.2
0.6 ± 19.3
2.9 ± 7.2
-3.0 ±
3.8 ±
11.6 ±
5.1 ±
-9.5 ±
1.6 ±
12.2
14.6
13.4
13.6
14.3
7.1
At
0.92
3.96
2.50
3.36
3.76
2.90
3.39
Inhalation slope,
ug/d£ per ug/m *
3.2 At
± 1.94
± 3.44
± 2.20
± 2.49
± 2.93
± 1.31
± 1.44
1.09
1,27
1.88
1.57
1.29
1.42
10.9
± 0.80
i 0.79
± 1.03
± 0.99
± 0.68
1 0.41
subject 1 at 3.2
*Assumed volume of blood pool is 75 dl.
i
-e.
tv
TABLE 11-14. GRIFFIN EXPERIMENTS - SUBJECTS EXPOSED TO AIR LEAD BOTH YEARS
Subject Mean
At 3.2 At 3.2
15 28.61
16 36.2 ±
17 33.5 ±
18 34.4 ±
19 36.8 ±
20 34.0 ±
10.
14.
14.
15.
19.
17.
Residence Time.d.
At 10.9
.4
6
.0
7
6
8
38.
35.
44.
36.
49.
47.
3 ±
2 ±
2 ±
3 ±
1 ±
5 i
21
16
20
18
27
24
.8
.8
.7
.2
.3
.0
Change in Intake,
Post-Pre-exposure, pg/d*
At 3.2 At 10.9
18.6 ± 11.3 - 1.
5.0 ± 11.6 4.8 ± 11.8 1.
7.9 ± 12.1 -8.6 ± 13.5 1.
2.1 ± 12.1 0.
-3.1 ± 15.6 0.
-7.2 ± 14.5 2.
At 3.
76 ±
57 1
25 ±
67 ±
73 ±
90 i
Inhalation slope,
ug/d£ per ug/m *
2 At 10.9
1.17
1.31
1.43
1.11
2.82
2.46
-0.16
0.14
-0.75
0.09
-0.25
-0.29
±0.46
± 0.35
± 0.68
± 0.38
± 0.73
± 0.70
Mean ± s.e.m. 34.6 ±6.5 41.8 ±9.2
•Assumed volume of blood pool is 75 d£.
•10.5 ± 7.9
-2.4 ± 6.6
1.48 ± 0.84
-0.20 ± 0.27
-------
PRELIMINARY DRAFT
TABLE 11-15. GRIFFIN EXPERIMENT - SUBJECTS EXPOSED TO AIR LEAD ONE YEAR ONLY
OJ
At 3.2 (second ^ear only)
Subject
6
7
8
9
10
11
12
14
21
Mean
Mean w/o
subject 6
Time, d.
49.4 ±
34.6 ±
38.0 ±
29.7 ±
40.4 ±
37.5 ±
43.3 ±
37.9 ±
36.8 +
38.6 +
26.1
11.9
15.2
9.7
16.9
15.3
17.3
14.7
15.6
5.8
Intake Change [iq/d.
3.9
7.0
9.4
3.3
5.7
7.4
-1.4
-7.7
3.5
± 20.1
± 15.6
± 15.6
± 14.8
± 13.9
-
± 14.6
± 16.6
± 22.5
± 6.3
Slope
O.b2 ± 3.29
4.35 t 2.48
3.33 ± 2.33
3.26 i 1.59
2.08 ± 1.95
3.93 ± 2.50
4.62 i 2.81
3.32 ± 2.25
2.06 i 3.19
3.05 i 0.95
3.37 i 0.92
Subject
1
2
5
6
8
9
10
11
12
21
23
24
Mean
Time, d.
35.3 ±
32.6 ±
25.7 ±
45.5 ±
52.0 ±
38.1 +
36.9 ±
30.1 ±
38.5 ±
62.9 ±
43.2 ±
30.3 ±
39.3 ±
15.4
13.9
9.3
17.5
22.3
14.1
15.8
14.3
15.7
37.2
15.8
8.3
6.0
At 10.9
(first year
only)
Intake Difference, yg/d Slope
5.2
8.2
3.0
-6.4
1.5
7.2
-3.9
10.3
0.5
18.6
5.2
12.6
5.2
± 20.0
± 19.7
t 18.6
± 12.4
± 12.9
± 13.7
± 22.5
± 15.9
± 23.6
± 16.9
± 14.1
± 13.0
± 5.4
2.17 ± 1.22
1.57 ± 0.95
1.08 ± 0.62
1.42 ± 0.76
1.90 ± 1.05
1.67 ± 0.84
0.65 ± 1.06
1.36 ± 1.05
2.09 ± 1.39
1.80 ± 1.40
2.04'± 0.97
1.80 ± 0.65
1.63 ± 0.32
-------
PRELIMINARY DRAFT
Subjects 16 and 20 had substantial increases, subjects 15 and 19 had moderate increases and
subject 18 had no increase in blood lead during exposure. Subject 17 had a marked decline in
blood lead, but the rate of decrease was much faster in the pre-exposure period, suggesting an
apparent increase of intake during exposure periods even for this subject. These subjects had
not apparently achieved equilibrium in either blood or tissue compartments. Even though these
subjects were not exposed to air lead, the estimated difference between blood lead intake be-
fore and during exposure of the other subjects was used to calculate the apparent inhalation
slope at that exposure. The pooled inhalation slope estimated for all six controls (1.48 ±
0.82 s.e.) was significantly positive (Z = 1.76, one-tailed p <0.05), as shown in Table 11-16.
No explanation for the increased lead intake during the winter of 1971-72 can be advanced at
this time, but factors such as changes in diet or changes in resorption of bone lead are
likely to have had equal effect on the lead-exposed subjects.
No statistically significant changes in the controls were found during the first experi-
ment at 10.9 ug/m .
(2) Among the controls, the estimated mean residence time in pool 1 was slightly longer
for the first year than the second year, 41.8 ± 9.2 days vs. 34.6 ± 6.5 days, but a paired
sample Z-test found that the mean difference for the controls (7.2 ± 11.2 days) was not signi-
ficantly different from zero (see Table 11-17).
3 3
(3) Among the five subjects exposed to 10.9 ug/m the first year and 3.2 ug/m the
second year, the mean residence time in blood was almost identical (43.9 ±9.4 vs. 44.7 ±8.7
days).
(4) The average inhalation slope for all 17 subjects exposed to 10.9 ug/m is 1.77 ±
0.37 when the slope for the controls is subtracted. The corrected inhalation slope for all 14
subjects exposed to 3.2 ug/m3 is 1.52 ± 1.12, or 1.90 ± 1.14 without subjects 1 and 6 who were
"non-responders." These are not significantly different. The pooled slope estimate for all
subjects is 1.75 ± 0.35. The pooled mean residence time for all subjects is 39.9 ± 2.5 days.
Thus, in spite of the large estimation variability at the lower exposure level, the aver-
age inhalation slope estimate and blood lead half-life are not significantly different at the
two exposure levels. This suggests that blood lead response to small changes in air lead in-
halation is approximately linear at typical ambient levels.
11.4.1.2 The Rabinowitz et al. Study. The use of stable lead isotopes avoids many of the
difficulties encountered in the analysis of whole blood lead levels in experimental studies.
Five adult male volunteers were housed in the metabolic research wards of the Sepulveda and
Wadsworth VA hospitals in Los Angeles for extended periods (Rabinowitz et al., 1974; 1976;
1977). For much of the time they were given low-lead diets with controlled lead content, sup-
plemented by tracer lead salts at different times.
PB11A/B 11-44 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-16. INHALATION SLOPE ESTIMATES
Group
Controls
All exposed
Difference
(Exposed-
controls)
Without sub-
jects 1, 6
At 3.2 ug/m3
1.48 ± 0.82
3.00 ± 0.76
1.52 ± 1.12
3.38 ± 0.79
At 10.9 pg/m3
-0.20 ± 0.27
1.57 ± 0.26
1.77 ± 0.37
Difference
(Exposed w/o
1,6 - control)
Pooled: (all subjects)
(without subjects 1,6)
1.90 ± 1.14
1.75 ± 0.35
1.78 ± 0.35
TABLE 11-17. MEAN RESIDENCE TIME IN BLOOD
3.2 pg/m
Experiment
10.9 ug/mj
Experiment
Control
Exposed
34.6 ±6.5 days
40.8 ±4.4 days
41.8 ±9.2 days
40.6 ±3.6 days
Four subjects were initially observed in the ward for several weeks. Each subject was in
the semi-controlled ward about 14 hours per day and was allowed outside for 10 hours per day,
allowing the blood lead concentration to stabilize.
Subjects B, D and E then spent 22 to 24 hours per day for 40, 25 and 50 days, respec-
tively, in a low lead room with total particulate and vapor lead concentrations that were much
Tower than in the metabolic wards or outside (see Table 11-18). The subjects were thereafter
exposed to Los Angeles air with much higher air lead concentrations than in the ward.
The calculated changes in lead intake upon entering and leaving the low-lead chamber are
shown in Table 11-19. These were based on the assumption that the change in total blood lead
was proportional to the change in tracer lead. The change in calculated air lead intakes
(other than cigarettes) due to removal to the clean room were also calculated independently by
the lead balance and labeled tracer methods (Rabinowitz et al., 1976) and are consistent with
these direct estimates.
PB11A/B
11-45
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-18. AIR LEAD CONCENTRATIONS (ug/m3) FOR TWO SUBJECTS
IN THE RABINOWITZ STUDIES
Average Range
Subject A outside (Sepulveda VA) 1.8 (1.2-2.4)
inside (Sepulveda VA,
airconditioned without
filter) 1.5 (1.0-2.7)
inside (Wadsworth VA,
open air room) 2.1 (1.8-2.6)
Subject B (Wadsworth VA)
outside 2.0 (1.6-2.4)
in room (air conditioner
with filter, no purifier) 0.91 (0.4-2.1)
in room (with purifiers,
"clean air") 0.072 (0.062-0.087)
open-air room 1.9 (1.8-1.9)
organic vapor lead
outside 0.10
"clean air" 0.05
* 5-20 days exposure for each particulate lead filter
Rabinowitz and coworkers assumed that the amount of lead in compartments within the body
evolved as a coupled system of first-order linear differential equations with constant frac-
tional transfer rates. This compartmental model was fitted to the data. This method of
analysis is described in Appendix 11A.
Blood lead levels calculated from the three compartment model adequately predicted the
observed blood lead levels over periods of several hundred days. There was no evidence to
suggest homeostasis or other mechanisms of lead metabolism not included in the model. There
was some indication (Rabinowitz et al., 1976) that gut absorption may vary from time to time.
The calculated volumes of the pool with blood lead (Table 11-19) are much larger than the
body mass of blood (about 7 percent of body weight, estimated respectively as 4.9, 6.3, 6.3,
4.6 and 6.3 kg for subjects A-E). The blood lead compartment must include a substantial mass
of other tissue.
PB11A/B 11-46 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-19. ESTIMATES OF INHALATION
SLOPE FOR RABINOWITZ STUDIES
Changes in
Intake*,
Subject ug/day
A 17 ± 5*
B 16 ± 3
C 15 ± 5*
D 9 ± 2
E 12 ± 2
Volume**,
kg
7.4 ± 0.6
10.0 ± 0.8
10.1 ± 1**
9.9 ± 1.2
11.3 ± 1.4
Residence!
Time, days
34 ± 5
40 ± 5
37 ± 5
40 ± 5
27 ± 5
Changes in
Air Leadtt
ug/m3
2.5tt
2.0
2.2tt
2.0
2.0
Inhalationt
Slope ug/d£
per ug/m3
2.98 ± 1.06
3.56 ± 0.93
2.67 ± 1.04
2.02 ± 0.60
1.59 ± 0.47
Maximum
Slope
4.38 ± 1.55
5.88 ± 1.54
4.16 ± 1.62
3.34 ± 0.99
2.63 ± 0.78
*From (Rabinowitz et al., 1977) Table VI. Reduced intake by low-lead method for subjects
B, D, E, tracer method for A, balance method for C. Standard error for C is assumed by EPA
to be same as A.
**From (Rabinowitz et al., 1976) Table II. EPA has assumed standard error with coefficient
of variation same as that for quantity of tracer absorbed in Table VI, except for subject C.
tEstimates from (Rabinowitz et al., 1976) Table II. Standard error estimate from combined
sample.
ttSee text, For A and C, estimated from average exposure. For B, D, E reduced by 0.2 ug/m3
for clean room exposure. Coefficient of variation assumed to be 10%.
^Assumed density of blood 1.058 g/cm3.
++Assuming outside air exposure is 2.1 ug/m3 rather than 4 ug/m3 for 10 hours.
The mean residence time in blood in Table 11-19 includes both loss of lead from blood to
urine and transfer of a fraction of blood lead to other tissue pools. This parameter reflects
the speed with which blood lead concentrations approach a new quasi-equilibrium level. Many
years may be needed before approaching a genuine equilibrium level'that includes lead that can
be mobilized from bones.
One of the greatest difficulties in using these experiments is that the air lead ex-
posures of the subjects were not measured directly, either by personal monitors or by restric-
ting the subjects to the metabolic wards. The times when the subjects were allowed outside
the wards included possible exposures to ground floor and street level air, whereas the outside
air lead monitor was mounted outside the third-floor window of the ward. The VA hospitals are
not far from major streets and the subjects' street level exposures could have been much
higher than those measured at about 10 m elevation (see Section 7.2.1.3). Some estimated
ratios between air concentrations at elevated and street level sites are given in Table 7.6.
A second complication is that the inside ward value of 0.97 ug/m (Rabinowitz et al.,
1977) used for subject B may be appropriate for the Wadsworth VA hospital, but not for subject
PB11A/B
11-47
7/29/83
-------
PRELIMINARY DRAFT
A in the Sepulveda VA hospital (see Table 11-18). The change in air lead values shown in
Table 11-19 is thus nominal, and is likely to have systematic inaccuracies much larger than
the nominal 10 percent coefficients of variation stated. The assumption is that for subjects
B, D and E, the exposure to street level air for 10 hours per day was twice as large as the
measured roof level air, i.e., 4 ug/m ; and the remaining 14 hours per day was at the ward
level of 0.97 pg/m ; thus the time-averaged level was (10 x 4 + 14 x 0.97)/24 = 2.23
The average controlled exposure during the "clean room" part of the experiment was 23, 22 and
24 hours respectively for subjects B, D, E; thus averaged exposures were 0.19, 0.28, and 0.12
ug/m , and reductions in exposure were about 2.0 ug/m . This value is used to calculate the
slope. For subject A, the total intake due to respired air is the assumed indoor average of
1.5 ug/m for the Sepulveda VA hospital, combining indoor and outdoor levels (10 x 4 + 14 x
1.5)/24 = 2.54 ug/m . For subject C we use the Wadsworth average. Apart from uncertainties in
the air lead concentration, the inhalation slope estimates for Rabinowitz's subjects have less
internal uncertainty than those calculated for subjects in Griffin's experiment.
The inhalation slopes thus calculated are the lowest that can be reasonably derived from
this experiment, since the largest plausible air lead concentrations have been assumed. The
third-floor air monitor average of 2.1 \ig/n\3 is a plausible minimum exposure, leading to the
higher plausible maximum inhalation slopes in the last column of Table 11-19. These are based
on the assumption that the time-averaged air lead exposure is smaller by 10x(4-2.1)/24 = 0.79
ug/m3 than assumed previously. It is also possible that some of this difference can be
attributed to dust ingestion while outside the metabolic ward.
11.4.1.3 The Chamberlain et al. Study. A series of investigations were carried out by
A.C. Chamberlain et al . (1975a,b; 1978) at the U.K. Atomic Energy Research Establishment in
Harwell, England. The studies included exposure of up to 10 volunteer subjects to inhaled,
ingested and injected lead in various physical forms. The inhalation exposures included labo-
ratory inhalation of lead aerosols generated in a wind tunnel, or box, of various particle
sizes and chemical compositions (lead oxide and lead nitrate). Venous blood samples were
taken at several times after inhalation of 203Pb. Three subjects also breathed natural high-
way exhaust fumes at various locations for times up to about 4.5 hours.
The natural respiratory cycles in the experiments varied from 5.7 to 17.6 seconds (4 to
11 breaths per minute) and tidal volumes from 1.6 to 2.3 liters. Lung deposition of lead-
bearing particles depended strongly on particle size anci composition, with natural exhaust
particles being more efficiently retained by the lung (30 to 50 percent) than were the chem-
ical compounds (20 to 40 percent).
The clearance of lead from the lungs was an extended process over time and depended on
particle size and composition, leaving only about 1 percent of the fine wind tunnel aerosols
PB11A/B 11-48 7/29/83
-------
PRELIMINARY DRAFT
in the lung after 100 hours, but about 10 percent of the carbonaceous exhaust aerosols. The
203Pb isotope reached a peak blood level about 30 hours after inhalation, the blood level then
representing about 60 percent of the initial lung burden.
A substantial fraction of the lead deposited in the lung appears to be unavailable to the
blood pool in the short term, possibly due to rapid transport to and retention in other tis-
sues including skeletal tissues. In long term balance studies, some of this lead in deep tis-
sue compartment would return to the blood compartment.
Lead kinetics were also studied by use of injected and ingested tracers, which suggested
that in the short term, the mean residence time of lead in blood could be calculated from a
one-pool model analysis.
Chamberlain et al. (1978) extrapolated these high level, short term exposures to longer
term ones. The following formula and data were used to calculate a blood-to-air level ratio:
[T, /?] [% Deposition] [% Absorption] [Daily ventilation]
P = [Blood volume] [0.693]
where: - _ biolog1-cal haif
With an estimated value of T.. ,„ = 18 days (mean residence time T., /_/0.693 = 26 days), with 50
percent for deposition in lung for ordinary urban dwellers, and 55 percent of the lung lead
retained in the blood lead compartment (all based on Chamberlain's experiments), with an
3
assumed ventilation of 20 m /day over blood volume 5400 ml (Table 10.20 in Chamberlain et
al. , 1978), then
o - 26 day X 0.50 X 0.55 X 20 m3/day , 7 3, .,
p — - « - «- = i. i m /dl
54 dl
This value of p could vary for the following reasons,
1. The absroption from lung to blood used here, 0.55, refers to short term kine-
tics. In the long term, little lead is lost through biliary or pancreatic
secretions, nails, hair and sweat, so that most of the body lead is available to
the blood pool even if stored in the skeleton from which it may be resorbed.
Chamberlain suggests an empirical correction to 0.55 X 1.3 = 0.715 absorption.
2. The mean residence time, 26 days, is shorter than in Rabinowitz's subjects, and
the blood volume is less, 54 dl. It is possible that in the Rabinowitz study,
PB11A/B 11-49 7/29/83
-------
PRELIMINARY DRAFT
the mean times are longer and the blood pool size (100 dl) is larger than here
because Rabinowitz et al. included relatively less labile tissues such as kidney
and liver in the pool. Assuming 40 days mean residence time and 100 dl blood
volume the slope can be recalculated,
„ 40 d X 0.50 X 0.55 X 20 m3/d _ , , 3/n
- - --- 2'2 m /dl
3. The breathing rate could be much less, for inactive people.
11.4.1.4 The Kehoe Study. Between 1950 and 1971, Professor R. A. Kehoe exposed 12 subjects
to various levels of air lead under a wide variety of conditions. Four earlier subjects had
received oral Pb during 1937-45. The inhalation experiments were carried out in an inhalation
chamber at the University of Cincinnati, in which the subjects spent varying daily time
periods over extended intervals. The duration was typically 112 days for each exposure level
in the inhalation studies, at the end of this period it was assumed the blood lead concentra-
tion had reached a near equilibrium level. The experiments are described by Kehoe (1961a,b,c)
and the data and their analyses by Gross (1981) and Hammond et al. (1981). The studies most
relevant to this document are those in which only particles of lead sesquioxide aerosols in
the submicron range were used, so that there was at least one air lead exposure (other than
control) for which the time-averaged air lead concentration did not exceed 10 ug/m3. Only six
subjects met these criteria: LD (1960-63), JOS (1960-63), NK (1963-66), SS (1963-68), HR
(1966-67) and DH (1967-69). Subject DH had a rather high initial lead concentration (30
ug/dl) that fell during the course of the experiment to 28 ug/dl; apparently daily detention
in the inhalation chamber altered DH's normal pattern of lead exposure to one of lesser total
exposure. The Kehoe studies did not measure non-experimental airborne lead exposures, and
did not measure lead exposures during "off" periods. Subject. HR received three exposure
levels from 2.4 to 7.5 ug/m3, subject NK seven exposure levels from 0.6 to 4.2 ug/m3 and sub-
jects SS 13 exposure levels from 0.6 to 7.2 ug/m3. LD and JOS were each exposed to about 9,
19, 27 and 36 ug/m3 during sequential periods of 109-113 days.
A great deal of data on lead content in blood, feces, urine and diet were obtained in
these studies and are exhibited graphically in Gross (1979) (see Figure 11-11). Apart from
the quasi-equilibrium blood lead values and balances reported in Gross (1979; 1981), there has
been little use of these data to study the uptake and distribution kinetics of lead in man.
EPA analyses used only the summary data in Gross (1981).
Data from Gross (1981) were fitted by least squares linear and quadratic regression
models. The quadratic models were not significantly better than the linear model except for
PB11A/B 11-50 7/29/83
-------
—
—
SUBJECT - SS
BALANCE
-
-
•
<
Q
£
a
I
£
0.
c
<
Q
ffi
Q.
ci
5
>
<
c
a
a
a
5
5
a
§
E
a
(9
5
200. 300. 400. -500. 600.
900. 1000. 1100. 1200. 1300. 1400
-
-
-
TIME (days)
Figure 11-11. Data plots for individual subjects with time for kehoe data as presented by Gross.
-------
PRELIMINARY DRAFT
subjects LD and JOS, who were exposed to air levels above 10 ug/m3. The linear terms predomi-
nate in all models for air lead concentrations below 10 ug/m3 and are reported in Table 11-20.
These data represent most of the available experimental evidence in the higher range of
ambient exposure levels, approximately 3 to 10 ug/m3.
Data for the four subjects with statistically significant relationships are shown in
Figure 11-12, along with the fitted regression curve and its 95 percent confidence band.
TABLE 11-20. LINEAR SLOPE FOR BLOOD LEAD VS. AIR LEAD AT
LOW AIR LEAD EXPOSURES IN KEHOE'S SUBJECTS
SUBJECT
DH3
HRa
Jos'5
LD
NKC
ssc
LINEAR SLOPES
LINEAR MODEL
-0.
0.
0.
0.
2.
1.
34 ± 0.
70 ± 0.
67 ± 0.
64 ± 0.
60 ± 0.
31 ± 0.
28
46
07
11
32
20
B, mVdl, ± s.e.
QUADRATIC MODEL
0.
0.
1.
1.
1.
1.
14 ± 1.
20 ± 2.
01 ± 0.
29 ± 0.
55 ± 1.
16 ± 0.
25
14
19
06
28
78
5.
2.
9.
9.
0.
0.
RANGE
AIR* BLOOD
6 -
4 -
4 -
3 -
6 -
6 -
8.8
7.5
35.7
35.9
4.0
7.2
26 -
21 -
21 -
18 -
20 -
18 -
31
27
46
41
30
29
aNo statistically significant relationship between air and blood lead.
''High exposures. Use linear slope from quadratic model.
cLow exposures. Use linear slope from linear model.
11.4.1.5 The Azar et al. Study. Thirty adult male subjects were obtained from each of five
groups: 1) Philadelphia cab drivers; 2) DuPont employees in Starke, Florida; 3) DuPont
employees in Barksdale, Wisconsin; 4) Los Angeles cab drivers; and 5) Los Angeles office
workers (Azar et al., 1975). Subjects carried air lead monitors in their automobiles and in
their breathing zones at home and work. Personal variables (age, smoking habits, water
samples) were obtained from all subjects, except for water samples from Philadelphia cab
drivers. Blood lead, ALAD urine lead and other variables were measured. From two to eight
blood samples were obtained from each subject during the air monitoring phase. Blood lead
determinations were done in duplicate. Table 11-21 presents the geometric means for air lead
and blood lead.for the five groups. The geometric means were calculated by EPA from the raw
data presented in the authors' report (Azar et al., 1975).
The Azar study has played an important role in setting standards because of the care used
in measuring air lead in the subjects' breathing zone. Blood lead levels change in response
to air lead levels, with typical time constants of 20 to 60 days. One must assume that the
subjects' lead exposures during preceding months had been reasonably similar to those during
PB11A/B
11-52
7/29/83
-------
PRELIMINARY DRAFT
I I I i I i / r
Iff I I I I I I
01 234567
0 5 10 15 20 25 30 35
AIR LEAD,
0 5 10 15 20 26 30 35
AIR LEAD,
PB11A/B
Figure 11-12. Blood level vs. air lead relationships for kehoe inhalation studies: lineu' rela-
tion for low exposures, quadratic for high exposures, with 95% confidence bands
11-53 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-21. GEOMETRIC MEAN AIR AND BLOOD LEAD LEVELS (ug/100 g)
FOR FIVE CITY-OCCUPATION GROUPS (DATA CALCULATED BY EPA)
Group
Geometric mean Geometric mean
air lead, blood lead, Sample
|jg/m3 GSD pg/100 g GSD size Code
Cab drivers
Philadelphia, PA
Plant employees
Starke, FL
Plant employees
Barksdale, WI
Cabdrivers
Los Angeles, CA
Office workers
Los Angeles, CA
2.59
0.59
0.61
6.02
2.97
1.16
2.04
2.39
1.18
1.29
22.1
15.4
12.8
24.2
18.4
1.16
1.41
1.43
1.20
1.24
30
29
30
30
30
Source: Azar et al. (1975).
the study period. Models have been proposed for these data by Azar et al. (1975), Snee (1981-
198;1;) and Hammond et al. (1981) including certain nonlinear models.
Azar et al. (1975) used a log-log model for their analysis of the data. The model in-
cluded dummy variables, C,, C,
C3' C4'
C,-, which take on the value 1 for subjects in that
group and 0 otherwise (see Table 11-21 for the definitions of these dummy variables). The
fitted model using natural logarithms was
log (blood Pb) = 2.951 ^ + 2.818 C£ +
2.627 C3 + 2.910 C4 + 2.821 Cg + 0.153 log (air Pb)
This model gave a residual sum of squares of 9.013, a mean square error of 0,63 (143 degrees
of freedom), and a multiple R2 of 0.502. The air lead coefficient had a standard error of
0.040. The fitted model is nonlinear in air lead, and so the slope depends on both air lead
and the intercept. Using an average intercept value of 1.226, the curve has a slope ranging
from 10.1 at an air lead level of 0.2 ug/m to 0.40 at an air lead level of 9 ug/m .
PB11A/B
11-54
7/29/83
-------
PRELIMINARY DRAFT
Snee (1982b) reanalyzed the same data and fitted the following power function model,
log (blood Pb) = log [12.1 (air Pb + 6.00 (^ + 1.46 C^
+ 0.44 Cg* 2.23 C4 '+ 6.26 C5)°'2669]
This model gave a residual sum of squares of 9.101, a mean square error of 0.064 (142 degrees
of freedom) and a multiple R2 of 0.497. Using an average constant value of 3.28, the slope
ranges from 1.29 at an air lead of 0.2 to 0.51 at an air lead of 9.
An important extension in the development of models for the data was the inclusion of
separate non-air contributions or background exposures for each separate group. The coeffi-
cients of the group variables, C., in the lead exposure model may be interpreted as measures
J
of total exposure of that group to non-air external sources (cigarettes, food, dust, water)
and to endogenous sources (lead stored in skeleton). Water and smoking variables were used to
estimate some external sources. (This required deleting another observation for a subject
with unusually high water lead.) The effect of endogenous lead was estimated using subject
age as a surrogate measure of cumulative exposure, since lead stored in skeleton is known to
increase approximately linearly with age, for ages 20 to 60 (Gross et al., 1975; Barry, 1975;
Steenhout, 1982) in homogeneous populations.
In order to facilitate comparison with the constant p ratios calculated from the clinical
studies, EPA fitted a linear exposure model to the Azar data. The model was fitted on a loga-
rithmic scale to facilitate comparison of goodness of fit with other exposure models and to
produce an approximately normal pattern of regression residuals. Neither smoking nor water
lead provided significantly better fits to the log (blood lead) measurements after the effect
of age was removed.
Age and air lead may be confounded to some extent because the regression coefficient for
age may include the effects of prior air lead exposures on skeletal lead buildup. This would
have the effect of reducing the estimated apparent slope p.
Geometric mean regressions of blood lead on air lead were calculated by EPA for several
assumptions: (i) A linear model analogous to Snee's exposure model, assuming different non-
air contributions in blood lead for each of the five subgroups; (ii) linear model in which age
of the subject is also used as a surrogate measure of the cumulative body burden of lead that
provides an endogenous source of blood lead; (iii) linear model similar to (ii), in which the
change of blood lead with age is different in different subgroups, but it is assumed that the
non-air contribution is the same in all five groups (as was assumed in the 1977 criteria docu-
ment); (iv) linear model in which both the non-air background and the change in blood lead
PB11A/B 11-55 7/29/83
-------
PRELIMINARY DRAFT
with age may differ by group; and (v) nonlinear model similar to (iv). None of the fitted
models are significantly different from each other using statistical tests of hypotheses about
parameter subsets in nonlinear regression (Gallant, 1975).
11.4.1.6 Silver Valley/Kellogg, Idaho Study. In 1970, EPA carried out a study of a lead
smelter in Kellogg, Idaho (Hammer et al. , 1972; U.S. Environmental Protection Agency, 1972).
The study was part of a national effort to determine the effects of sulfur dioxide, total sus-
pended particulate and suspended sulfates, singly and in combination with other pollutants, on
human health. It focused on mixtures of the sulfur compounds and metals. Although it was
demonstrated that children had evidence of lead absorption, insufficient environmental data
were reported to allow further quantitative analyses.
In 1974, following the hospitalization of two children from Kellogg with suspected acute
lead poisoning, the CDC joined the State of Idaho in a comprehensive study of children in the
Silver Valley area of Shoshone County, Idaho, near the Kellogg smelter (Yankel et al., 1977-
Landrigan et al., 1976).
The principal source of exposure was a smelter whose records showed that emissions aver-
aged 8.3 metric tons per month from 1955 to 1964 and 11.7 metric tons from 1965 to September
1973. After a September 1973 fire extensively damaged the smelter's main emission filtration
facility, emissions averaged 35.3 metric tons from October 1973 to September 1974 (Landrigan
et al., 1976). The smelter operated during the fall and winter of 1973-74 with severely limited
air pollution control capacity. Beginning in 1971, ambient concentrations of lead in the
vicinity of the smelter were determined from particulate matter collected by Hi-Vol air
samples. Data indicated that monthly average levels measured in 1974 (Figure 11-13) were
three to four times the levels measured in 1971 (von Lindern and Yankel, 1976). Individual
exposures of study participants to lead in the air were estimated by interpolation from these
data. Air lead exposures ranged from 1.5 ug/m3 to 30 (jg/m3 monthly average (see Figure 11-13).
Soil concentrations were as high as 24,000 |jg/g and averaged 7000 ug/g within one mile of the
smelter. House dusts were found to contain as much as 140,000 ug/g and averaged 11,000 ug/g
in homes within one mile of the complex.
The study was initiated in May of 1974 and the blood samples were collected in August
1974 from children 1 to 9 years old in a door-to-door survey (greater than 90 percent partici-
pation). Social, family and medical histories were conducted by interview. Paint, house
dust, yard and garden soils, grass, and garden vegetable samples were collected. At that
time, 385 of the 919 children examined (41.9 percent) had blood lead levels in excess of 40
ug/dl, 41 children (4.5 percent) had levels greater than 80 ug/dl. All but 2 of the 172
children living within 1.6 km of the smelter had levels greater than or equal to 40 ug/dl.
Those two children had moved into the area less than six months earlier and had blood lead
PB11A/B 11-56 7/29/83
-------
PRELIMINARY DRAFT
1971
1972
1973
TIME, year
1974
1975
Figure 11-13. Monthly ambient air lead concentrations in Kellogg, Idaho,
1971 through 1975.
PB11A/B
11-57
7/29/83
-------
PRELIMINARY DRAFT
levels greater than 35 |jg/dl. Both the mean blood lead concentration and the number of chil-
dren classified as exhibiting excess absorption, decreased with distance from the smelter
(Table 11-22). Blood lead levels were consistently higher in 2- to 3-year-old children than
they were in other age groups (Table 11-23). A significant negative relationship between
blood lead level and hematocrit value was found. Seven of the 41 children (17 percent) with
blood lead, levels greater than 80 |jg/dl were diagnosed as being anemic on the basis of
hematocrit less than 33 percent, whereas only 16 of 1006 children (1.6 percent) with blood
lead levels less than 80 pg/dl were so diagnosed. Although no overt disease was observed in
children with higher lead intake, differences were found in nerve conduction velocity.
Details of this finding are discussed in chapter 12.
Yankel et al. (1977) fitted the data to the following model:
In (blood lead) = 3.1 + 0.041 air lead + 2.1 x 10* soil lead
+ 0.087 dustiness - 0.018 age
+ 0.024 occupation
where air lead was in [ig/m3; soil lead was in |jg/g; dustiness was 1, 2 or 3; age was in years;
and occupation was a Hollingshead index. The analysis included 879 subjects, had a multiple
R2 of 0.622 and a residual standard deviation of 0.269 (geometric standard deviation of 1.31).
Walter et al. (1980) used a similar model to examine age specific differences of the re-
gression coefficients for the different variables. Those coefficients are summarized in Table
11-24. The variable that was most significant overall was air lead; its coefficient was ap-
proximately the same for all ages, corresponding to a change in blood lead of about 1 pg/dl
per unit increase of air lead (in MS/1"3) at an air exposure of 1 ug/m3 and about 2.4 ug/dl per
unit increase in air at an air exposure of 22 (jg/m3.
The next most important variable that attained significance at a variety of ages was the
household dustiness level (coded as low = 0, medium = 1 or high = 2), showing a declining ef-
fect with age and being significant for ages 1 to 4 years. This suggested age-related hygiene
behavior and a picture of diminishing home orientation as the child develops. For ages 1 to 4
years, the coefficient indicates the child in a home with a "medium" dust level would have a
blood lead level ~ 10 percent higher than a child in a home with a "low" dust level, other
factors being comparable.
The coefficients for soil lead-blood lead relationships exhibited a fairly regular pat-
tern, being highly significant (p <0.01) for ages 3 to 6 years, and significant (p <0.05) at
ages 2 to 6 years. The maximum coefficient (at age 6) indicates a 4 percent increase in blood
lead per 1000 ug/g increase in soil lead.
PB11A/B 11-58 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-22. GEOMETRIC MEAN BLOOD LEAD LEVELS BY AREA COMPARED WITH
ESTIMATED AIR-LEAD LEVELS FOR 1- TO 9-YEAR-OLD CHILDREN
LIVING NEAR IDAHO SMELTER. (GEOMETRIC STANDARD DEVIATIONS
SAMPLE SIZES AND DISTANCES FROM SMELTER ARE ALSO GIVEN)3
Area
1
2
3
4
5
6
Geometric mean
blood lead,
ug/dl
65.9
47.7
33.8
32.2
27.5
21.2
GSD
1.30
1.32
1.25
1.29
1.30
1.29
Sample
size
170
192
174
156
188
90
% blood
lead
>40(jg/dl
98.9
72.6
21.4
17.8
8.8
1.1
Estimated
air lead,
ug/m3
18.0
14.0
6.7
3.1
1.5
1.2
Distance from
smelter, km
0- 1.6
1.6- 4.0
4.0-10.0
10.0-24.0
24.0-32.0
about 75
JEPA analysis of data from Yankel et al. (1977).
TABLE 11-23. GEOMETRIC MEAN BLOOD LEAD LEVELS BY AGE AND AREA FOR
SUBJECTS LIVING NEAR THE IDAHO SMELTER
Area
1
2
3
4
5
6
7
1
69*
50
33
31
27
21
28
2
72
51
36
35
35
25
30
3
75
55
36
34
29
22
28
4
75
46
35
31
29
23
32
5
68
49
35
31
29
20
30
Age Group
678
66
50
35
35
28
22
26
63
47
31
30
25
20
37
60
42
32
32
27
22
30
9
57
40
32
30
24
17
20
Teenage
39
33
28
35
Adult
37
33
30
34
32
32
*error in original publication (Yankel et al., 1977).
PB11A/B
11-59
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-24. AGE SPECIFIC REGRESSION COEFFICIENTS FOR THE ANALYSIS OF
LOG-BLOOD-LEAD LEVELS IN THE IDAHO SMELTER STUDY
Age Ai r
1
2
3
4
5
6
7
8
9
* P
t P
0.0467*
0.0405*
0.0472*
0.0366*
0.0388*
0.0361*
0.0413*
0.0407*
0.0402*
<0.01
<0.05
Dust
0.119T
0.106T
o.ioet
0.107T
0.052
0.070
0.053
0.051
0.081T
Occupation
0.0323
0.0095
0. 0252
0.0348
0.0363t
0.0369t
0.0240
0.0422T
0.0087
Pica
0.098
0.225*
0.077
0.117
0.048
0.039
0.106
0.010
0.108
Sex Soil (xlO4)
0.055
0.002
0.000
0.032
-0.081
-0.092
-0.061
-0.106f
-0.158*
3.5
20. 6t
24.2*
32.1*
23.4*
38.4*
21. 3f
16.2
11.6
Intercept
3.017
3.567
3.220
3.176
3.270
3.240
3.329
3.076
3.477
N
98
94
115
104
130
120
113
105
104
Pica (coded absent = 0, present = 1) had a significant effect at age 2 years, but was in-
significant elsewhere; at age 2 years, an approximate 25 percent elevation in blood lead is
predicted in a child with pica, compared with an otherwise equivalent child without pica.
Occupation was significant at ages 5, 6 and 8 years; at the other ages, however, the sign
of the coefficient was always positive, consistent with a greater lead burden being introduced
into the home by parents working in the smelter complex.
Finally, sex (coded male = 0; female = 1) had a significant negative coefficient for ages
8 and 9 years, indicating that boys would have lead levels 15 percent higher than girls at
this age, on the average. This phenomenon is enhanced by similar, but nonsignificant, nega-
tive coefficients for ages 5 to 7 years.
Snee (1982c) also reanalyzed the Idaho smelter data using a log-linear model. He used
dummy variables for age, work status of the father, educational level of the father, and
2
household dust level (cleanliness). The resulting model had a multiple R of 0.67 and a resi-
dual standard deviation of 0.250 (geometric standard deviation of 1.28). The model showed
that 2-year-olds had the highest blood lead levels. The blood lead inhalation slope was es-
sentially the same as that of Yankel et al. (1977) and Walter et al. (1980).
The above non-linear analyses of the Idaho smelter study are the only analyses which sug-
gest that the blood lead to air lead slope increases with increasing air lead, a finding in
counterdistinction to the findings of decreasing slopes seen at high air lead exposures in
other studies. An alternative to this would be to attempt to fit a linear model as described
in Appendix 11-B. Exposure coefficients were estimated for each of the factors shown in
Table 11-25. The results for the different covariates are similar to those of Snee (1982c)
and Walter et al. (1980).
PB11A/B 11-60 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-25. ESTIMATED COEFFICIENTS* AND STANDARD
ERRORS FOR THE IDAHO SMELTER STUDY
Factor
Coefficient
Asymptotic
Standard Error
Intercept (ug/dl)
Air lead (pg/m )
Soil lead (1000 pg/g)
Sex (male=l, female=0)
Pica (eaters=l , noneaters=0)
Education (graduate training=0)
At least high school
No high school
Cleanliness of home (clean=0)
Moderately clean
Dirty
Age (1 year olds=0)
2 years olds
3 years olds
4 years olds
5 years olds
6 years olds
7 years olds
8 years olds
9 years olds
Work status (no exposure=0)
Lead or zinc worker
Residual standard deviation = 0.2576
Multiple R2 = 0.662
Number of observations = 860
13.19
1.53
1.10
1.31
2.22
-
3.45
4.37
-
3.00
6.04
-
4.66
5.48
3.16
2.82
2.74
0.81
-0.19
-1.50
-
3.69
(geometric standard
1.90
0.064
0.14
0.59
0.90
1.44
1.51
0.65
1.06
1.48
1.32
1.32
1.25
1.24
1.23
1.28
1.21
0.61
deviation = 1.29)
"Calculations made by EPA
Because the previous analyses noted above indicated a nonlinear relationship, a similar
model with a quadratic air lead term added was also fitted. The coefficients for the other
factors remained about the same, and the improvement in the model was marginally significant
(p = 0.05). This model gave a slope of 1.16 at an air lead of 1 ug/m , and 1.39 at an air
lead of 2 pg/m . Both the linear and quadratic models, along with Snee's (1982) model are
shown in Figure 11-14. The points represent mean blood lead levels adjusted for the factors
in Table 11-25 (except air lead) for each of the different exposure subpopulations.
PB11A/B 11-61 7/29/83
-------
PRELIMINARY DRAFT
Yankel et al. (1977), Walter et al. (1980) and Snee (1982c) make reference to a follow-up
study conducted in 1975. The second study was undertaken to determine the effectiveness of
control and remedial measures instituted after the 1974 study. Between August 1974 and August
1975, the mean annual air lead levels decreased at all stations monitored. In order of in-
creasing distance from the smelter, the annual mean air lead levels for the one year preceding
each drawing were 18.0 to 10.3 pg/m3, 14.0 to 8.5 pg/m3, 6.7 to 4.9 pg/m3 and, finally 3.1 to
2.5 pg/m3 at 10 to 24 km. Similar reductions were noted in house dust lead concentrations.
In a separate report, von Lindern and Yankel (1976) described reductions in blood lead levels
of children for whom determinations were made in both years. The results demonstrated that
significant decreases in blood lead concentration resulted from exposure reductions.
80
70
v
1 60
uj 50
O
§ 40
(O
| M
-»
9 20
10
J I I I I I I I I I I I I I II I I I I I I I...F
LINEAR (EPA)
QUADRATIC (EPA)
LOG-LINEAR (SNEE)
I I I I I I I I I I I I I I I I I I I I I I I f
10 15
AIR LEAD, M9/m'
20
25
Figure 11-14. Fitted equations to Kellogg Idaho/Silver Valley
adjusted blood lead data.
PB11A/B
11-62
7/29/83
-------
PRELIMINARY DRAFT
11.4.1.7 Omaha, Nebraska Studies. Exposure from both a primary and secondary smelter in the
inner city area of Omaha, Nebraska, has been reported in a series of publications (Angle et
al., 1974; Angle and Mclntire, 1977; Mclntire and Angle, 1973). During 1970 to 1977 children
were studied from: an urban school at a site immediately adjacent to a small battery plant
and downwind from two other lead emission sources; from schools in a mixed commercial-residen-
tial area; and from schools in a suburban setting. Children's blood lead levels were obtained
by macro technique for 1970 and 1971, but Delves micro assay was used for 1972 and later. The
differences for the change in techniques were taken into account in the presentation of the
data. Air lead values were obtained by Hi-Vol samplers and dustfall values were also moni-
tored. Table 11-26 presents the authors' summary of the entire data set, showing that as air
lead values decrease and then increase, dustfall and blood lead values follow. The authors
used regression models, both log-linear and semilog, to calculate (air lead)/(blood lead).
Specific reports present various aspects of the work. Black children in the two ele-
mentary schools closest to the battery plant had higher blood leads (34.1 ug/dl) than those in
elementary and junior high schools farther away (26.3 ug/dl). Best estimates of the air ex-
posures were 1.65 and 1.48 ug/m , respectively (Mclntire and Angle, 1973). The latter study
compared three populations: urban vs. suburban high school students, ages 14 and 18; urban
black children, ages 10 to 12, vs. suburban whites, age 10 to 12; and blacks ages 10 to 12
with blood lead levels over 20 ug/dl vs. schoolmates with blood lead levels below 2"0 ug/dl
(Angle et al., 1974). The urban vs. suburban high school children did not differ significan-
tly, 22.3 ±1.2 and 20.2 ±7.0 ug/dl, respectively, with mean values of air lead concentra-
tions of 0.43 and 0.29 ug/m . For 15 students who had environmental samples taken from their
homes, correlation coefficients between blood lead levels and soil and housedust lead levels
were 0.31 and 0.29, respectively.
Suburban 10-to-12-year-o1ds had lower blood lead levels than their urban counterparts,
17.1 ±0.7 versus 21.7 ±0.5 ug/dl (Angle et al., 1974). Air lead exposures were higher in
the urban than in the suburban-population, although the average exposure remained less than 1
T 2
ug/m . Dustfall lead measurements, however, were very much higher; 32.96 mg/m /month for
2
urban 10-to-12-year-olds vs. 3.02 mg/m /month for suburban children.
Soil lead and house dust lead exposure levels were significantly higher for the urban
black high lead group than for the urban low lead group. A significant correlation (r = 0.49)
between blood lead and soil lead levels was found.
Angle has reanalyzed the Omaha study using all of the data on children. There were 1075
samples from which blood lead (ug/dl), air (ug/m3), soil (ug/g) and house dust (ug/g) lead
were available. The linear regression model, fitted in logarithmic form, was
PB11A/B 11-63 7/29/83
-------
PRELIMINARY DRAFT
Pb-Blood = 15/67 + 1.92 Pb-Air + 0.00680 Pb-Soil + 0.00718 Pb-House Dust
(±0.40) (±0.60) (±0.00097) (±0.00090)
(N = 1075, R2 = 0.20, S2 = 0.0901, GSD = 1.35)
Similar models fitted by age category produced much more variable results, possibly due to
small ranges of variation in air lead within certain age categories.
TABLE 11-26. AIR, DUSTFALL AND BLOOD LEAD CONCENTRATIONS IN
OMAHA, NE STUDY, 1970-19773
Group
All urban children
1970-71
1972-73
1974-75
1976-77
Children at school
1970-71
1972-73
1974-75
1976-77
Air
ug/m3 (N)b
, mixed commercial and
1.48
0.43
0.10
0.52
± 0
± 0
± 0
± 0
in a commercial
1.69
0.63
0.10
0.60
± 0
± 0
± 0
± 0
.14(7;
.08(8;
.03(10
.07(12
site
.11(7;
.15(8;
.03(10
.10(12
Dustfall,
ug/m - mo
(N)C
Blood,
ug/dl (N)d
residential site
65)
72)
;72)
;47)
67)
74)
;70)
;42)
10.
6.
8.
25.
14.
33.
6
0
8
9
3
9
--
± 0.
± 0.
—
± 0.
± 4.
3(6)
1(4)
(7)
6(5)
1(4)
(7)
31.4 ±
23.3 ±
20.4 ±
22.8 ±
34.6 ±
21.9 ±
19.2 ±
22.8 ±
0.
0.
0.
0.
1.
0.
0.
0.
7(168)
3(211)
1(284)
7(38)
5(21)
6(54)
9(17)
7(38)
All suburban children in a residential site
1970-71
1972-73
1974-75
1976-77
0.
0.
0.
79
29
12
± 0.
± 0.
± 0.
M •*
06(7;65)
04(8; 73)
05(10;73)
4.
2.
6
9
—
± I.
± 0.
...
1(6)
9(4)
19.6
14.4
18.2
-•
±
±
±
0.5(81)
0.6(31)
0.3(185)
Blood lead 1970-71 1s by the macro technique, corrected for an established
laboratory bias of 3 ug/dl, macro-micro; all other values are by Delves micro
assay.
bN = Number of months; number of 24-hour samples.
CN = Number of months.
dN = Number of blood samples.
Source: Adapted from Angle and Mclntire, 1977.
PB11A/B
11-64
7/29/83
-------
PRELIMINARY DRAFT
11.4.1.8 Roels et al. Studies. Roels et al. (1976, 1978, 1980) have conducted a series of
studies in the vicinity of a lead smelter in Belgium. Roels et al. (1980) reports a follow-up
study (1975) that included study populations from a rural-nonindustrialized area as well as
from the lead smelter area. The rural group consisted of 45 children (11-14 years). The
smelter area group consisted of 69 school children from three schools. These children were
divided into two groups; group A (aged 10-13) lived less than 1 km from the smelter and their
schools were very close to the smelter; group B consisted of school children living more than
1.5 km from the smelter and attending a school more distant from the smelter.
In 1974 the smelter emitted 270 kg of lead and the air lead levels were 1 to 2 orders of
magnitude greater than the current Belgian background concentration for air lead (0.23 ug/m ).
Soil and vegetation were also contaminated with lead; within 1 km the soil lead level was
12,250 ug/g. The concentration of lead in drinking water was less than 5 ug/1.
Environmental assessment included air, soil and dust. Air monitoring for lead had been
continuous since September 1973 at two sites, one for each of the two groups. In the rural
area, air monitoring was done at two sites for five days using membrane pumps. Lead was ana-
lyzed by flameless atomic absorption spectrophotometry. Dust and soil samples were collected
at the various school playgrounds. The soil sample was analyzed by flameless atomic absorp-
tion.
A 25 ml blood sample was collected from each child and immediately divided among three
tubes. One tube was analyzed for lead content by flameless atomic absorption with background
correction. Another tube was analyzed for ALA-D activity while the third was analyzed for FEP.
FEP was determined by the Roels modification of the method of Sassa. ALA-D was assayed by the
European standard method.
Air lead levels decreased from area A to area B. At both sites the airborne lead levels
declined over the two years of monitoring. The amount of lead produced at this smelter during
this time remained constant, about 100,000 tons/year. The median air lead level at the closer
site (A) dropped from 3.2 to 1.2 ug/m , while at the far site (B) the median went from 1.6 to
0.5-0.8 ug/m . The rural area exposure levels did not vary over the study period, remaining
rather constant at about 0.30 ug/m .
Both smelter vicinity groups showed signs of increased lead absorption relative to the
rural population. Blood lead levels for group A were about three times those for the rural
population (26 ug/dl vs. 9 ug/dl). The former blood lead levels were associated with about a
50 percent decrease in ALA-D activity and a 100 percent increase in FEP concentration. How-
ever, FEP levels were not different for group B and rural area residents.
Later surveys of children (Roels et al., 1980) were conducted in 1976, 1977 and 1978; the
former two in autumn, the latter in spring. In total there were five surveys conducted yearly
from 1974 to 1978. A group of age-matched controls from a rural area was studied each time
except 1977. In 1976 and 1978 an urban group of children was also studied.
PB11A/B 11-65 7/29/83
-------
PRELIMINARY DRAFT
The overall age for the different groups ranged from 9 to 14 years (mean 11-12). The
length of residence varied from 0.5 to 14 years (mean 7-10 years). The subjects were always
recruited from the same five schools: one in the urban area, one in the rural area and three
in the smelter area (two <1 km and one, 2.5 km away). Air lead levels decreased from 1977 to
1378. However, the soil lead levels in the vicinity of the smelter were still elevated (
-------
PRELIMINARY DRAFT
TABLE 11-27. MEAN AIRBORNE AND BLOOD LEAD LEVELS RECORDED DURING FIVE DISTINCT SURVEYS
(1974 to 1978) FOR STUDY POPULATIONS OF 11-YEAR-OLD CHILDREN LIVING LESS THAN 1 km
OR 2.5 km FROM A LEAD SMELTER, OR LIVING IN A RURAL OR URBAN AREA
Study
populations
1 Survey
(1974)
2 Survey
(1975)
3 Survey
(1976)
4 Survey
(1977)
5 Survey
(1978)
Pb-Ai
<1 km
2.5 km
Rural
<1 km
2.5 km
Rural
<1 km
2.5 km
Urban
Rural
<1 km
2.5 km
1 km
2.5 km
Urban
Rural
r
4.06
1.00
0.29
2.94
0.74
0.31
3.67
0.80
0.45
0.30
3.42
0.49
2.68
0.54
0.56
0.37
lotal
n
37
—
92
40
29
45
38
40
26
44
56
50
43
36
29
42
Blood lead
Population
Mean ±
30.1 ±
--
9.4 ±
26.4 ±
13.6 ±
9.1 ±
24.6 ±
13.3 ±
10.4 ±
9.0 ±
28.9 ±
14.8 ±
27.8 ±
16.0 ±
12.7 ±
10.7 ±
SD
5.7
2.1
7.3
3.3
3.1
8.7
4.4
2.0
2.0
6.5
4.7
9.3
3.8
3.1
2.8
n
14
14
28
19
17
14
18
24
17
21
27
34
20
26
18
17
concentration
Boys
Mean ±
31.0
21.1
9.7
27.4
14.8
8.2
28.7
15.6
10.6
9.2
31.7
15.7
29.3
16.6
13.4
11.9
SD
± 5.5
± 3.4
±1.6
± 6.5
± 3.6
± 2.1
±8.0
± 2.9
± 2.0
± 2.3
± 9.5
+ 4.8
± 9.8
± 3.5
± 2.3
± 3.0
(uq/di
n
23
—
64
21
12
31
20
16
9
23
29
16
23
10
11
25
)
Girls
Mean
29.6 ±
--
9.3 ±
25.4 ±
11.9 ±
9.5 ±
20.8 ±
9.8 ±
9.9 ±
8.7 ±
26.4 ±
13.0 ±
26.5 ±
14.3 ±
11.5 ±
10.0 ±
± SD
5.9
2.2
8.1
1.9
3.4
7.6
3.8
2.0
1.7
8.7
4.3
8.9
4.2
4.0
2.4
Source: Roels et al. 1980.
statistically significant regression but not statistically significant coefficients. Thus the
playground dust measurement was dropped and the following model obtained with almost as small
a residual sum of squares,
In(Pb-Blood) = 1n(7.37 + 2.46 Pb-Air
(±.45) (±.58)
0.0195 Pb-Hand + 2.10 Male)
(±.0062) (±0.56)
The fitted model for the 148 observations gave an R2 of 0.654 and a mean square error (S2) of
0.0836 (GSD = 1.335). The significance of the estimated coefficient establishes that intake
of lead-bearing dust from the hands of children does play a role in childhood lead absorption
over and above the role that can be assigned to inhalation of air lead. Individual habits of
mouthing probably also affect lead absorption along this pathway. Note too that the estimated
inhalation slope, 2.46, is somewhat larger than most estimates for adults. However, the ef-
fect of ingestion of hand dust appears to be almost as large as the effect of air lead in-
halation in children of this age (9-14 years). Roels et al. (1980), using group means,
PB11A/B 11-67 7/29/83
-------
PRELIMINARY DRAFT
concluded that the quantitative contribution of hand lead to children's blood lead levels was
far greater than that of air lead.
The high mutual correlations among air, hand, and dust lead suggest the use of their
principal components or principal factors as predictors. Only the first principal component
(which accounted for 91% of the total variance in lead exposure) proved a statistically sig-
nificant covariate of blood lead. In this form the model could be expressed as
In(Pb-Blood) = ln(7.42 + 1.56Pb-Air + 0.0120Pb-Hand + 0.00212Pb-Dust + 2.29 Male)
The estimated standard error on the inhalation slope is ±0.47. The difference between these
inhalation slope and hand lead coefficients is an example of the partial attribution of the
effects of measured lead exposure sources to those sources that are not measured.
11.4.1.9 Other Studies Relating Blood Lead Levels to Air Exposure. The following studies
also provide information on the relationship of blood lead to air lead exposures, although
they are less useful in accurately estimating the slope at lower exposure levels. The first
group of studies are population studies with less accurate estimates of individual exposures.
The second group of studies represent industrial exposures at very high air lead levels in
which the response of blood lead appears to be substantially different than at ambient air
levels.
The Tepper and Levin (1975) study included both air and blood lead measurements. House-
wives were recruited from locations in the vicinity of air monitors. Table 11-28 presents the
geometric mean air lead and adjusted geometric mean blood lead values for this study. These
values were calculated by Hasselblad and Nelson (1975). Geometric mean air lead values ranged
from 0.17 to 3.39 ug/m3, and geometric mean blood lead values ranged from 12.7 to 20.1 ug/dl.
Nordman (1975) reported a population study from Finland in which data from five urban and
two rural areas were compared. Air lead data were collected by stationary samplers. All
levels were comparatively low, particularly in the rural environment, where a concentration of
0.025 ug/m3 was seen. Urban-suburban levels ranged from 0.43 to 1.32 ug/m .
A study was undertaken by Tsuchiya et al. (1975) in Tokyo using male policemen who
worked, but not necessarily lived, in the vicinity of air samplers. In this study, five zones
were established, based on degree of urbanization, ranging from central city to suburban. Air
monitors were established at various police stations within each zone. Air sampling was con-
ducted from September 1971 to September 1972; blood and urine samples were obtained from 2283
policemen in August and September 1971. Findings are presented in Table 11-29.
Goldsmith (1974) obtained data for elementary school (9- and 10-year-olds) and high
school students in 10 California communities. Lowest air lead exposures were 0.28 ug/m3 and
highest were 3.4 ug/m . For boys in elementary school, blood lead levels ranged from 14.3 to
PB11A/B 11-68 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-28. GEOMETRIC MEAN AIR LEAD AND ADJUSTED BLOOD LEAD
LEVELS FOR 11 COMMUNITIES IN STUDY OF
TEPPER AND LEVIN (1975) AS REPORTED BY
HASSELBLAD AND NELSON (1975)
Geometric mean
air lead,
Community (jg/m3
Los Alamos, NM
Okeana, OH
Houston, TX
Port Washington, NY
Ardmore, PA
Lombard, IL
Washington, DC
Philadelphia, PA
Bridgeport, IL
Greenwich Village, NY
Pasadena, CA
0.17
0.32
0.85
1.13
1.15
1.18
1.19
1.67
1.76
2.08
3.39
Age and smoking
adjusted geometric
mean blood lead,
Mg/dl
15.1
16.1
12.7
15.3
17.9
14.0
18.7
20.1
17.6
16.5
17.6
Sample
size
185
156
186
196
148
204
219
136
146
139
194
Multiple R = 0.240
Residual standard deviation = 0.262 (geometric standard deviation = 1.30)
TABLE 11-29. MEAN AIR AND BLOOD LEAD VALUES FOR
FIVE ZONES IN TOKYO STUDY
Zones
1
2
3
4
5
Air lead,
pg/m3
0.024
0.198
0.444
0.831
1.157
Blood lead,
pg/100 g
17.0
17.1
16.8
18.0
19.7
Source: Tsuchiya et al. 1975.
PB11A/B
11-69
7/29/83
-------
PRELIMINARY DRAFT
23.3 ug/dl; those for girls ranged from 13.8 to 20.4 M9/dl for the same range of air lead ex-
posures. The high school student population was made up of only males from some of the 10
towns. The air lead range was 0.77 to 2.75 pg/m , and the blood lead range was 9.0 to 12.1
|jg/dl. The high school students with the highest blood lead levels did not come from the town
with the highest air lead value. However, a considerable lag time occurred between the col-
lection and analysis of the blood samples. In one of the communities the blood samples were
refrigerated rather than frozen.
Another California study (Johnson et al., 1975, 1976) examined blood lead levels in rela-
tion to exposure to automotive lead in two communities, Los Angeles and Lancaster (a city in
the high desert). Los Angeles residents studied were individuals living in the vicinity of
heavily traveled freeways within the city. They included groups of males and females, aged 1
through 16, 17 through 34, and 34 and over. The persons selected from Lancaster represented
similar age and sex distributions. On two consecutive days, blood, urine and fecal samples
were collected. Air samples were collected from one Hi-Vol sampler in Los Angeles, located
near a freeway, and two such samplers in Lancaster. The Los Angeles sampler collected for 7
days; the two in Lancaster operated for 14 days. Soil samples were collected in each area in
the vicinity of study subjects.
Lead in ambient air along the Los Angeles freeway averaged 6.3 ± 0.7 ng/m and, in the
3
Lancaster area, the average was 0.6 ± 0.2 ug/m . The mean soil lead in Los Angeles was 3633
ug/g, whereas that found in Lancaster was 66.9 ug/g. Higher blood lead concentrations were
found in Los Angeles residents than in individuals living in the control area for all age
.groups studied. Differences between Los Angeles and Lancaster groups were significant with
the sole exception of the older males. Snee (1981) has pointed out a disparity between blood
samples taken on consecutive days from the same child in the study. This calls into question
the validity of using this study to quantify the air lead to blood lead relationship.
Daines et al. (1972) studied black women living near a heavily traveled highway in New
Jersey. The subjects lived in houses on streets paralleling the highway at three distances:
3.7, 38.1 and 121.9 m. Air lead as well as blood lead levels were measured. Mean annual air
lead concentrations were 4.60, 2.41 and 2.24 ug/m , respectively, for the three distances.
The mean air lead concentration for the area closest to the highway was significantly dif-
ferent from that in both the second and third, but the mean air lead concentration of the
third area was not significantly different from that of the second. The results of the blood
lead determinations paralleled those of the air lead. Mean blood lead levels of the three
groups of women, in order of increasing distance, were 23.1, 17.4 and 17.6 ug/dl, respec-
tively. Again, the first group showed a significantly higher mean than the other two, but the
second and third groups' blood lead levels were similar to each other. Daines et al. (1972),
in the same publication, reported a second study in which the distances from the highway were
33.5 and 457 meters and in which the subjects were white upper middle class women. The air
PB11A/B H-70 7/29/83
-------
PRELIMINARY DRAFT
lead levels were trivially different at these two distances, and the blood lead levels did not
differ either. Because the residents nearest the road were already 33 m from the highway, the
differences in air lead may have been insufficient to be reflected in the blood lead levels.
(See Chapter 7)
A summary of linear relationships for other population studies has been extracted from
Snee (1981) and is shown in Table 11-30. The Fugas study is described later in Section
11.5.2.3. There is a large range of slope values (-0.1 to 3.1) with most studies in the range
of 1.0 to 2.0. Additional information on the more directly relevant studies is given in the
Summary Section 11.4.1.10.
TABLE 11-30. BLOOD LEAD-AIR LEAD SLOPES FOR SEVERAL POPULATION
STUDIES AS CALCULATED BY SNEE
Study
Tepper & Levin
(1975)
Johnson et al .
(1975)
Nordman (1975)
Tsuchiya et al.(1975)
Goldsmith (1974)
Fugas (1977)
Daines et al. (1972)
Johnson et al .
(1975)
Goldsmith (1974)
No.
Subjects
1935
65
96
536
478
537
89
79
352
61
88a
37a
43
486
Sex
Female
Male
Female
Male
Female
Male
Male
Female
Male
Female
(spring)
Female (fall
Male
(children)
Female
(children)
Slope
1.1
0.8
0.8
1.2
0.6
3.1
-0.1
0.7
2.2
1.6
) 2.4
1.4
1.1
95% confidence
Intervals
±1.8
±0.7
±0.6
±1.0
±0.9
±2.2
±0.7
±0.7
±0.7
±1.7
±1.2
±0.6
±0.6
Male & Female
(children)
2.0
±1.3
aOutlier results for four subjects deleted.
Source: Snee, 1981.
There is a great deal of information on blood lead responses to air lead exposures of
workers in lead-related occupations. Almost all such exposures are at air lead levels far in
excess of typical non-occupational exposures. The blood lead vs. air lead slope p is very
much smaller at high blood and air levels. Analyses of certain studies are shown in Table
11-31.
PB11A/B
11-71
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-31. A SELECTION OF RECENT ANALYSES ON OCCUPATIONAL
8-HOUR EXPOSURES TO HIGH AIR LEAD LEVELS
Analysis
Ashford et al.
(1977)
King et al.
(1979)
Gartside
et al. (1982)
Bishop and
Hill (1983)
Study
Williams et al.(1969)
Globe Union
Delco-Remy
Factory 1, 1975
Factory 2a, 1975
Factory 3a, 1975
Delco-Remy,
1974-1976
Battery plants A
1975-1981 B
C
D
E
F
Air Lead
|jg/m3
50-300
35-1200
10-350
20-170
2-200
7-170
7-195
20-140
4-140
Blood Lead
H9/dl
40-90
25-90
22-72
12-50
18-72
22-60
24-75
18-60
15-53
P
(slope)
0.19
0.10
0.05
0.032
0.07
0.068
0.0514
• — —
Nonlinear:
at 50:
0.081
0.045
0.048
0.022
0.045
0.101
Any of several equally plausible nonlinear curves could be used to extrapolate from the
linear low-exposure blood lead relation whose slope is 1.0 to 2.0 to the linear high-exposure
relation whose slope is 0.03 to 0.20, but the correct form of the curve has not yet been es-
tablished.
11.4.1.10 Summary of Blood Lead vs. Inhaled Air Lead Relations. Any summary of the relation-
ship of blood lead level and air lead exposure is complicated by the need for reconciling the
results of experimental and observational studies. Further, defining the form of the
statistical relationship is problematical due to the lack of consistency in the range and ac-
curacy of the air lead exposure measures in the various studies.
EPA has chosen to emphasize the results of studies that relate lead in air and lead in
blood under ambient conditions. At low air lead exposures there is no statistically signif-j-
cant difference between curvilinear and linear blood lead inhalation relationships. Therefore
EPA has fitted linear relationships (Tables 11-32, 11-33 and 11-34) to blood lead levels in
the studies to be described next with the explicit understanding that the fitted relationships
are intended only to describe changes in blood lead due to modest changes in air lead amona
individuals whose blood lead levels do not exceed 30 ug/dl.
PB11A/B
11-72
7/29/83
-------
TABLE 11-32. CROSS-SECTIONAL OBSERVATIONAL STUDY WITH HEASUREO INDIVIDUAL AIR LEAD EXPOSURE
Study
Azar et al.
Study done
(1975)
in
Analysis
Azar et al. (1975)
In (PBB)
Model
= 0.153 In (PBA) + separate intercepts for each group
0
R2
.502
Model
d.f.
6
=— ^--
(1
Slope at
1.0 Mg/ro
2.57
.23, 3.91)
an air lead of
i3 2.0
1.43
(0.64,
Mg/mJ
2.30)
1970-1971 in five
U.S. cities
sample size
Blood leads
, total
= 149.
ranged
froB 8 to 40 ug/dl.
Air leads ranged
from 0.2 to
ug/H3
i— •
i— •
i
—j
CO
9.1
Snee (19626)
NawMnd et al .
(1981)
EPA
EPA
EPA
EPA
EPA
EPA
In (PBB)
- 1 A
(PBB) l-U
In(PBB) =
In(PBB) =
In(PBB) =
In(PBB) =
In(PBB) =
In(PBB) =
= 0.2669 in (PBA > separate background for each group)
* 1.0842
19 0 104
= 0. 179 (PBA + separate background for each group) *
-0.098
?n(l. 318 PBA * separate background for each group)
?
ln(2.902 PBA - 0.257 PBA* + separate background
for each group)
ln(1.342 PBA * separate background * age slope x age)
ln(1.593 PBA = coiMon intercept + age x separate age
slope)
ln(1.255 PBA * separate background + age + separate
age slope)
0.25 In- (PBA * separate background * age x separate
age slope)
0
0
0
.497
.49
.491
7
8
6
(0
1.12
.29, 1.94)
1.08
1.32
(0.46. 2.17)
0
0
0.
0.
0.
.504
.499
,489
521
514
7
7
7
11
12
(0.
(0.
(0.
2.39
1.34
32, 2.37)
1.59
76, 2.42)
1.26
46, 2.05)
about 1.0
(varies by
city)
0.96
(0.25, 1
1.07
1.32
(0.46, 2
1.87
1.34
.66)
^
y
r
r
i-
K
.17) =
J
?
C
(0.32, 2.37) *
1.59
-i
(0.76, 2.42) -
1.26
(0.46, 2.05)
about 1.0
(varies
city)
by
Note: PBB stands for blood lead (ug/dl); PBA stands for air lead (ug/M2); slope means rate of change of blood lead per unit change in air lead at the
stated air lead value. The 95 percent confidence intervals for the slope are given in parentheses. These are approximate and should be used
with caution. The analyses labelled "EPA" are calculated fron the original authors' data.
-------
TABLE 11-33. CROSS-SECTIONAL OBSERVATIONAL STUDIES ON CHILDREN WITH ESTIMATED AIR EXPOSURES
Model Slope at an air lead of
Study
Kellogg Idaho/Silver
Valley study conducted
in 1974 based on about
880 children. Air
leads ranged from
0.5 to 22 ug/m"3'-
Blood leads ranged
from 11 to 164
Kellogg Idaho/Silver
Valley study as above
restricted to 537 chil-
dren with air leads
I—- below 10 uq/m3
' Roels et al.
*. (1980)
Angle and Mclntire
(1979)
Analysis
Vankel et al.
(1977)
Snee (1982c)
EPA
EPA
Walter et al .
(1980)
Snee (1982a)
Roels et al.
(1980) based
on 8 groups
EPA analysis
on 148 subjects
Angle and
Mclntire (1979)
on 832 samples
ages 6-18
Angle et al.
(1983) on 1074
samples for ages
1-18
832 samples ages
6 to 18
In(PBB)
In(PBB)
In(PBB)
In(PBB)
In(PBB)
In(PBB)
PBB = 0
In(PBB)
In(PBB)
In(PBB)
In(PBB)
Model
= 0.041 PBA + 2.1xlO~b soil + 0.087 dust
- 0.018 age + 0.024 occupation + 3.14
= 0.039 PBA + 0.065 In (soil) * terms for sex,
occupation, cleanliness, education, pica
= ln(1.52 PBA to 0.0011 soil + terms for sex,
occupation, cleanl iness, -education, pica)
= ln(1.13 PBA + 0.026 PBA » terms for soil, sex,
occupation, cleanliness, education, pica)
= separate slopes for air, dust, occupation, pica 0.
sex and soil by age
= 0.039 PBA + 0.055 In (soil) + terms for sex, occupation
cleanliness, education, pica
.007 PBA + 11.50 log(PB-Hand) - 4.27
- 4.27
= ln(2.46 PBA + 0.0195 (Pb-Hand) +2.1 (Male) * 7.37)
= ln(8.1) + 0.03 In (PBA) + 0.10 In (PB-Soil)
+ 0.07 In (Pb-House Dust)
= ln(1.92 PBA + 0.00680 Pb-Soil
«• 0.00718 Pb-House Dust + 15.67)
= In (4.40 PBA to .00457 Pb-Soil
+ 0.00336 Pb-House Dust + 16.21)
R2
0.622
0.666
0.655
0.656
56 to 0.70
0.347
0.65
0.654
0.21
0.199
0.262
d.f. 1.0 ug/m-'
6 1.16
(1.09, 1.23) (1
25 1.13
(1.06, 1.20) (1
18 1.52
19 1.16
7 1.01 to 1.26 1.
25 1.07
(0.89, 1.25) (1
3 0.007
4 2.46
(1.31.3.61)
4 0.6
4 l.St
(0.74,3.10)
4 4.40
(3.20,5.60)
5. 0 (jg/ma
1.37
.27, 1.46)
1.32
.23, 1.42)
1.52
1.39
18 to 1.48
1.25
.01, 1.50)
0.007
2.46
(1.31,3.61)
0.14
1.92
(0.74,3.10)
4.40
(3.20,5.60)
TO
z
ya
-n
— l
Note: PBB stands for blood lead (ug/dl); PBA stands for air lead (ug/m3); slope means rate of change of blood lead per unit change in air lead at the
stated air lead value. The 95 percent confidence intervals for the slope are given in parentheses. These are approximate and should be used
with caution. The analyses labelled "EPA" are calculated from the original authors' data.
-------
TABLE 11-34. LONGITUDINAL EXPERIMENTAL STUDIES WITH MEASURED INDIVIDUAL AIR LEAD EXPOSURE
Experiment
Kehoe 1950-1971
1960-1969
^ Griffin et al.
^ 1971-1972
en
Chaaberlaln et
al. 1973-1978
Rabinmrftz
et al. 1973-1974
Analysis
Gross (1981)
Haaaond et a 1. (1981)
Snee (1981)
EPA
Keelson et al.(1973)
NaMMnd et al.(1981)
Snee (1981)
EPA
Chamberlain et al.
(1978)
EPA
Snee (1981)
EPA
A PBB =
A PBB =
A PBB =
PBB =
A PBB =
A PBB =
A PBB =
A PBB =
A PBB =
A PBB =
A PBB =
A PBB =
Model Air Lead
ug/m3
0.57 A PBA 0.6 to 36
B{A PBA, Pi by subject fron -0.6 to 2.94
Pn-A PBA, pf by subject froa 0.* to 2.4
p. PBA + background, p. by subject fron -.34 to 2.60 0.6 to 9
0.327 PBA + 3.236 + (2.10 PBA + 1.96) (In P8A * p.) by subject 0.15, 3.2
P A PBA, p = 1.90 at 3.2 and p = 1.54 at 10.9 0.15, 10.9
p. A PBA, p. by subject, B = 2.3 at 3.2 and p = 1.5 at 10.9
Pj A PBA, p. by subject, mean p = 1.52 at 3.2
and p = 1.77 at 10.9
P APBA, p = 1.2 calculated
p APBA, p - 2.7 calculated
Pi APBA, p. by subject fron 1.7 to 3.9 0.2 to 2
p. APBA, 0. by subject froa 1.59 to 3.56
Blood Lead
ng/dl
18 to 41
ii
16 to 29
11 to 32
14 to 43
14 to 28
-------
PRELIMINARY DRAFT
The blood lead inhalation slope estimates vary appreciably from one subject to another in
experimental and clinical studies, and from one study to another. The weighted slope and
standard error estimates from the Griffin study in Table 11-16 (1.75 ± 0.35) were combined
with those calculated similarly for the Rabinowitz study in Table 11-19 (2.14 ± 0.47) and the
Kehoe study in Table 11-20 (1.25 ± 0.35 setting DH = 0), yielding a pooled weighted slope es-
timate of 1.64 ± 0.22 ug/dl per (jg/m . There are some advantages in using these experimental
studies on adult males, but certain deficiencies need to be acknowledged. The Kehoe study ex-
posed subjects to a wide range of exposure levels while in the exposure chamber, but did not
control air lead exposures outside the chamber. The Griffin study provided reasonable control
of air lead exposure during the experiment, but difficulties in defining the non-inhalation
baseline for blood lead (especially in the important experiment at 3.2 ug/m ) add much uncer-
tainty to the estimate. The Rabinowitz study controlled well for diet and other factors and
since they used stable lead isotope tracers, they had no baseline problem. However, the
actual air lead exposure of these subjects outside the metabolic ward was not well determined.
Among population studies, only the Azar study provides a slope estimate in which air lead
exposures are known for individuals. However, there was no control of dietary lead intake or
other factors that affect blood lead levels, and slope estimates assuming only air lead and
location as covariables (1.32 ± 0.38) are not significantly different from the pooled experi-
mental studies.
Snee and Pfeifer (1983) have extensively analyzed the observational studies, tested the
equivalence of slope estimates using pooled within-study and between-study variance com-
ponents, and estimated the common slope. The result of five population studies on adult males
(Azar, Johnson, Nordman, Tsuchiya, Fugas) was an inhalation slope estimate ±95 percent
confidence limits of 1.4 ± 0.6. For six populations of adult females [Tepper-Levin, Johnson,
Nordman, Goldsmith, Daines (spring), Daines (fall)], the slope was 0.9 ± 0.4. For four
populations of children [Johnson (male), Johnson (female), Yankel, Goldsmith], the slope
estimate was 1.3 ± 0.4. The between-study variance component was not significant for any
group so defined, and when these groups were pooled and combined with the Griffin subjects,
the slope estimate for all subjects was 1.2 ± 0.2.
The Azar slope estimate was not combined with the experimental estimates because of the
lack of control on non-inhalation exposures. Similarly, the other population studies in Table
11-30 were not pooled because of the uncertainty about both inhalation and non-inhalation lead
exposures. These studies, as a group, have lower slope estimates than the individual experi-
mental studies.
There are no experimental inhalation studies on adult females or on children. The inha-
lation slope for women should be roughly the same as that for men, assuming proportionally
smaller air intake and blood volume. The assumption of proportional size is less plausible
PB11A/B H-76 7/29/83
-------
PRELIMINARY DRAFT
for children. Slope estimates for children from population studies have been used in which
some other important covariates of lead absorption were controlled or measured, e.g., age,
sex, dust exposure in the environment or on the hands. Inhalation slopes were estimated for
the studies of Angle and Mclntire (1.92 ± 0.60), Roels (2.46 ± 0.58) and Yankel et al. (1.53 ±
0.064). "iiie standard error of the Yankel study is extremely low and a weighted pooled slope
estimate for children would reflect essentially that study alone. In this case the small
standard error estimate is attributable to the very large range of air lead exposures of
children in the Silver Valley (up to 22 ug/m ). The relationship is in fact not linear, but
increases more rapidly in the upper range of air lead exposures. The slope estimate at lower
air lead concentrations may not wholly reflect uncertainty about the shape of the curve at
higher concentrations. The unweighted mean slope of the three studies and its standard error
estimate are 1.97 ± 0.39.
This estimate was not combined with the child population studies of Johnson or Goldsmith.
The Johnson study slope estimate used air lead measured at only two sites and is sensitive to
assumptions about data outliers (Snee, 1981), which adds a large non-statistical uncertainty
to the slope estimate. The Goldsmith slope estimate for children (2.0 ± 0.65) is close to
the estimate derived above, but was not used due to non-statistical uncertainties about blood
lead collection and storage.
One can summarize the situation briefly:
(1) The experimental studies at lower air lead levels, 3.2 ug/m or less, and lower
blood levels, typically 30 pg/dl or less, have linear blood lead inhalation
relationships with slopes p. of 0 to 3.6 for most subjects. A typical value of
1.64 ± 0.22 may be assumed for adults.
(2) Population cross-sectional studies at lower air lead and blood lead levels are
approximately linear with slopes p of 0.8 to 2.0.
(3) Cross-sectional studies irw occupational exposures in which air lead levels are
higher (much above 10 pg/m ) and blood lead levels are higher (above 40 ^g/dl),
show a much more shallow linear blood lead inhalation relation. The slope p is
in the range 0.03 to 0.2.
(4) Cross-sectional and experimental studies at levels of air lead somewhat above
the higher ambient exposures (9 to 36 pg/m ) and blood leads of 30 to 40 pg/dl
can be described either by a nonlinear relationship with decreasing slope or by
a linear relationship with intermediate slope, approximately p = 0.5. Several
biological mechanisms for these differences have been discussed (Hammond et
al., 1981; O'Flaherty et al., 1982; Chamberlain, 1983; Chamberlain and Heard,
1981). Since no explanation for the decrease in steepness of the blood lead
inhalation response to higher air lead levels has been generally accepted at
this time, there is little basis on which to select an interpolation formula
from low air lead to high air lead exposures. The increased steepness of the
inhalation curve for the Silver Valley/ Kellogg study is inconsistent with the
PB11A/B 11-77 7/29/83
-------
PRELIMINARY DRAFT
other studies presented. It may be that smelter situations are unique and must
be analyzed differently, or it may be that the curvature is the result of
imprecise exposure estimates.
(5) The blood-lead inhalation slope for children is at least as steep as that for
adults, with an estimate of 1.97 ± 0.39 from three major studies (Yankel et
al., 1977; Roels, et al. (1980); Angle and Mclntire, 1979).
11.4.2 Dietary Lead Exposures Including Water
Another major pathway by which lead enters the body is by ingestion. As noted in Chap-
ters 6 and 7, the recycling of both natural and anthropogenic lead in the environment results
in a certain amount of lead being found in the food we eat and the water we drink. Both of
these environmental media provide external exposures to lead that ultimately increase internal
exposure levels in addition to internal lead elevations caused by direct inhalation of lead in
air. The Nutrition Foundation Report (1982) presents a compilation of recent estimates of
dietary intakes in the United States and Canada. The report gives information on relation-
ships between external lead exposures and blood lead levels. The mechanisms and absorption
rates for uptake of lead from food and water are described in Chapter 10. The purpose of the
present section is to establish (analogously to Section 11.4.1) the relationships between
external exposures to lead in food and drinking water and resulting internal lead exposures.
The establishment of these external and internal lead exposure relationships for the en-
vironmental media of food and water, however, is complicated by the inherent relationship be-
tween food and water. First, the largest component of food by weight is water. Second,
drinking water is used for food preparation and, as shown in Section 7.3.1.3 provides addi-
tional quantities of lead that are appropriately included as part of external lead exposures
ascribed to food. Third, the quantity of liquid consumed daily by people varies greatly and
substitutions are made among different sources of liquid: soft drinks, coffee, tea, etc., and
drinking water. Therefore, at best, any values of water lead intake used in drinking water
calculations are somewhat problematic.
A further troubling fact is the influence of lead in the construction of plumbing facil-
ities. Studies discussed in Section 7.3.2.1.3 have pointed out the substantial lead exposures
in drinking water that can result from the use of lead pipes in the delivery of water to the
tap. This problem is thought to occur only in limited geographic areas in the U.S. However,
where the problem is present, substantial water lead exposures occur. In these areas one can-
not make a simplifying assumption that the lead concentration in the water component of food
is similar to that of drinking water. But rather one is adding a potentially major additional
lead exposure to the equation.
PB11A/B H-78 7/29/83
-------
PRELIMINARY DRAFT
Studies that have attempted to relate blood lead levels to ingested lead exposures have
used three approaches to estimate the external lead exposures involved: duplicate meals, fe-
cal lead determinations, and market basket surveys. In duplicate diet studies, estimated lead
exposures are assessed by having subjects put aside a duplicate of what they eat at each meal
for a limited period of time. These studies probably provide a good, but short term, estimate
of the ingestion intake. However, the procedures available to analyze lead in foods have his-
torically been subject to inaccuracies. Hence, the total validity of data from this approach
has not been established. Studies relying on the use of fecal lead determinations face two
major difficulties. First, this procedure involves the use of a mathematical estimate of the
overall absorption coefficient from the gut to estimate the external exposure. Until re-
cently, these estimates have not been well documented and were assumed to be relatively con-
stant. Newer data discussed later show a much wider variability in the observed absorption
coefficients than was thought to be true. These new observations cloud the utility of studies
using this method to establish external/internal exposure relationships. Second, it is dif-
ficult to collect a representative sample.
The last approach is the market basket approach. This approach uses the observed lead
concentrations for a variety of food items coupled with estimated dietary consumption of the
particular food items. Some studies use national estimates of typical consumption patterns
upon which to base the estimated exposures. Other studies actually record the daily dietary
intakes. This approach faces similar analytic problems to those found in the duplicate diet
pproach. It also faces the problem of getting accurate estimates of dietary intakes. The
irist current total diet study (Pennington, 1983) is described in Section 7.3.1.2.
Exposures to lead in the diet are thought to have decreased from the 1940's. Estimates
•om that period were in the range of 400-500 ug/day for U.S. populations. Current estimates
or U.S. populations are under 100 ug/day for adults. Unfortunately, a good historical record
regarding the time course of dietary exposures is not available. In the years 1978-82, ef-
forts have been made by the American food canning industry in cooperation with the FDA to re-
duce the lead contamination of canned food. Data presented in-'SMtton 7.3.1.2.5 confirm the
success of this effort.
The specific studies available for review regarding dietary exposures will be organized
into three major divisions: lead ingestion from typical diets, lead ingestion from experi-
mental dietary supplements and inadvertent lead ingestion from lead plumbing.
11.4.2.1 Lead Ingestion from Typical Diets.
11.4.2.1.1 Ryu study on infants and toddlers. Ryu et al. (1983) reported a study of four
breast-fed infants and 25 formula-fed infants from 8 days to 196 days of age. After 112 days,
the formula-fed infants were separated into a group of 10 who received carton milk and a
PB11B/A 11-79 7/29/83
-------
PRELIMINARY DRAFT
second group of seven who received either canned formula or heat-treated milk in cans. In ad-
dition to food concentrations, data were collected on air, dust and water lead. Hemoglobin
and FEP were also measured.
The trends in blood lead for the formula-fed infants are shown in Table 11-35. The re-
sults up to day 112 are averaged for all 25 infants. The estimated average intake was 17
ug/day for this time period. After day 112, the subgroup of seven infants fed either canned
formula or heat-treated cow's milk in cans (higher lead), had average estimated lead intake of
61 ug/day. This resulted in an increase of 7.2 pg/dl in the average blood lead for an in-
crease of 45 ug/day in lead intake. The estimated slope from this data is 0.16.
TABLE 11-35. BLOOD LEAD LEVELS AND LEAD INTAKE VALUES
FOR INFANTS IN THE STUDY OF RYU ET AL.
Age in
Days
8
28
56
84
112
140
168
196
Blood lead of combined
group (ug/dl)
Lower
6.
7.
7.
Lead
2
0
2
8
5
5
5
6
.9
.8
.1
.4
.1
Higher Lead
9.3
12.1
14.4
Average
combined
Lower Lead
16
16
16
lead intake of
group (ga/dav)
17
17
17
17
17
Higher
61
61
61
Lead
Source: Ryu et al. (1983).
11.4.2.1.2 Rabinowitz study. This study on male adults was described in Section 11.4.1 and
in Chapter 10, where ingestion experiments were analyzed in more detail (Rabinowitz et al.,
1980). As in other studies, the fraction of ingested stable isotope lead tracers absorbed
into the blood was much lower when lead was consumed with meals (10.3 ± 2.2 percent) than
between meals (35 ± 13 percent). Lead nitrate, lead sulfide and lead cysteine as carriers
made little difference. The much higher absorption of lead on an empty stomach implies
greater significance of lead ingestion from leaded paint and from dust and soil when consumed
between meals, as seems likely to be true for children.
11.4.2.1.3 Hubermont study. Hubermont et al. (1978) conducted a study of pregnant women
living in rural Belgium because their drinking water was suspected of being lead contaminated.
This area was known to be relatively free of air pollution. Seventy pregnant women were
PB11B/A
11-80
7/29/83
-------
PRELIMINARY DRAFT
recruited and were asked to complete a questionnaire. Information was obtained on lifetime
residence history, occupational history, smoking and drinking habits. First flush tap water
samples were collected from each home with the water lead level determined by fTameless atomic
absorption spectrophotometry. Biological samples for lead determination were taken at
delivery. A venipuncture blood sample was collected from the mother as was a fragment of the
placenta. An umbilical cord blood sample was used to estimate the newborn's blood lead
status.
For the entire population, first flush tap water samples ranged from 0.2 to 1228.5 ug/1.
The mean was 109.4 while the median was 23.2. The influence of water lead on the blood lead
of the mother and infants was examined by categorizing the subjects on the basis of the lead
level of the water sample, below or above 50 M9/1- Table 11-36 presents the results of this
study. A significant difference in blood lead levels of mothers and newborns was found for
the water lead categories. Placenta lead levels also differed significantly between water
lead groups. The fitted regression equation of blood lead level for mothers is given in
summary Table 11-42.
11.4.2.1.4 Sherlock study. Sherlock et al. (1982) reported a study from Ayr, Scotland, which
considered both dietary and drinking water lead exposures for mothers and children living in
the area. In December 1980, water lead concentrations were determined from kettle water from
114 dwellings in which the mother and child lived less than 5 years. The adult women had
venous blood samples taken in early 1981 as part of a European Economic Community (EEC) survey
on blood lead levels. A duplicate diet survey was conducted on a random sample of these 114
women stratified by kettle water lead levels.
A study population of 11 mothers with infants less than 4 months of age agreed to
participate in the infant survey. A stratified sample of 31 of 47 adult volunteers was
selected to participate in the duplicate diet study.
Venous blood samples for adults were analyzed for lead immediately before the duplicate
diet study; in some instances additional samples were taken to give estimates of long term ex-
posure. Venous samples were taken from the infants immediately after the duplicate diet week.
Blood lead levels were determined by AAS with graphite furnace under good quality control.
Two other laboratories analyzed each sample by different methods. The data reported are based
on the average value of the three methods.
Dietary intakes for adults and children were quite different; adults had higher intakes
than children. Almost one third of the adults had intakes greater than 3 mg/week while only
20 percent of the infants had that level of intake. Maximum values were 11 mg/week for adults
and 6 mg/week for infants.
PB11B/A 11-81 7/29/83
-------
PRELIMINARY DRAFT
The observed blood lead values in the dietary study had the following distributions:
Adults
Infants
EEC Directive
>20 ug/dl
55%
100%
50%
>30 ug/dl
16%
55%
10%
>35 ug/dl
2%
36%
2%
TABLE 11-36. INFLUENCE OF LEVEL OF LEAD IN WATER
ON BLOOD LEAD LEVEL IN BLOOD AND PLACENTA
Comparison
Group
Age (Years)
Pb-B mother
(ug/dl )
Pb-B newborn
(ug/dl)
Pb placenta
(ug/100 g)
Water Pb
(ug/n
Water
Level
Low**
High***
Low
High
Low
High
Low
High
Low
High
Mean
25.6
26.3
10.6
13.8
8.8
12.1
9.7
13.3
11.8
247.4
Median
24
25
9.9
13.1
8.5
11.9
8.2
12.0
6.3
176.8
Range
18-41
20-42
5.1-21.6
5.3-26.3
3.4-24.9
2.9-22.1
4.4-26.9
7.1-28
0.2-43.4
61.5-1228.
Significance
NS*
<0.005
<0.001
<0.005
5
Source: Hubermont et al. (1978)
*NS means not significant
**Water Lead <50 ug/1
***Water Lead >50 ug/1
Table 11-37 presents the crosstabulation of drinking water lead and blood lead level for
the 114 adult women in the study. A strong trend of,increasing blood lead levels with in-
creasing drinking water lead levels is apparent. A curvilinear regression function fits the
data better than a linear one. A similar model including weekly dietary intake was fitted to
the data for adults and infants. These models are in summary Tables 11-41 and 11-44.
The researchers also developed a linear model for the relationship between dietary intake
and drinking water lead. The equation indicates that, when the concentration of lead in water
was about 100 ug/1, approximately equal amounts of lead would be contributed to the total
PB11B/A
11-82
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-37. BLOOD LEAD AND KETTLE WATER LEAD CONCENTRATIONS
FOR ADULT WOMEN LIVING IN AYR
Water lead (ug/1)
Rlnnri IpAff
ug per 100 ml
<10
11-15
16-20
21-25
26-30
31-35
36-40
>40
Total
<10 11-
99
8 5
4 7
1 3
4
13 19
100-
299
3
12
9
2
2
28
300-
499
2
3
7
4
1
1
1
19
500-
999
3
5
4
2
1
4
19
1000-
1499
2
2
1
3
8
>1500
1
3
1
3
8
Total
13
17
22
25
12
10
4
11
114
week's intake from water and from the diet; as water lead concentrations increase from this
value, the principal contributor would be water.
11.4.2.1.5 Central Directorate on Environmental Pollution study. The United Kingdom Central
Directorate on Environmental Pollution (1982) studied the relationship between blood lead
level and dietary and drinking water lead in infants. Subjects were first recruited by
soliciting participation of all pregnant women attending two hospitals and residing within a
single water distribution system. Each woman gave a blood sample and a kettle water sample.
The women were then allocated 'to one of six potential study groups based on the concentration
of water lead.
At the start of the second phase (duplicate diet) a total of 155 women volunteered
(roughly 17 to 32 per water lead level category). During the course of the study, 24 mothers
withdrew; thus a final study population of 131 mothers was achieved.
At 13 weeks of age, duplicate diet for a week's duration was obtained for each infant.
Great care was exerted to allow collection Of the most accurate sample possible. Also, at
this time a variety of water samples were collected for subsequent lead analysis.
Blood samples were collected by vem'puncture from mothers before birth, at delivery, and
about the time of the duplicate diet. A specimen was also collected by venipuncture from the
infant at the time of the duplicate diet. The blood samples were analyzed for lead by graph-
ite furnace AAS with deuterium background correction. Breast milk was analyzed analogously to
the blood sample after pretreatment for the different matrix. Water samples were analyzed by
flame atomic absorption. Food samples were analyzed after ashing by flameless atomic absorp-
tion.
PB11B/A 11-83 7/29/83
-------
PRELIMINARY DRAFT
Both mothers and infants exhibited increased lead absorption by EEC directive standards.
The infants generally had higher blood leads than the mothers. However, in neither population
was there evidence of substantial lead absorption.
Water lead samples ranged from less than 50 ug/1 to greater than 500 ug/1, which was ex-
pected due to the sampling procedure used. First draw samples tended to be higher than the
other samples. The composite kettle samples and the random daytime samples taken during the
duplicate diet week were reasonably similar: 59 percent of the composite kettle samples con-
tained up to 150 ug/1 as did 66 percent of the random daytime samples.
Lead intakes from breast milk were lower than from duplicate diets. The lead intakes
estimated by duplicate diet analysis ranged from 0.04 mg/week to 3.4 mg/week; about 1/4 of the
diets had intakes less than 1.0 mg/week. The minimum intakes were truncated, as the limit of
detection for lead was 10 ug/kg and the most common diets weighed 4 kg or more.
The authors used both linear and cube root models to describe their data. Models rela-
ting blood lead levels of infants to dietary intake are in Table 11-41. Models relating blood
lead levels for both mothers and infants to first flush water lead levels and running water
lead levels are in Tables 11-43 and 11-44, respectively. In most cases, the nonlinear (cubic)
model provided the best fit. Figure 11-15 illustrates the fit for the two models showing
infant blood lead levels vs. dietary lead intake.
11.4.2.1.6 Pocock study. Pocock et al. (1983) have recently reported an important study ex-
amining the relationship in middle aged men of blood lead level and water lead levels. Men
aged 40 to 59 were randomly selected from the registers of general practices located in 24
British towns. Data were obtained between January 1978 and June 1980.
Blood lead levels were obtained on 95 percent of the 7378 men originally selected. The
levels were determined by microatomic absorption spectrophotometry. A strict internal and ex-
ternal quality control program was maintained on the blood lead determinations for the entire
study period. Tap water samples were obtained on a small subset of the population. About 40
men were chosen in each of the 24 towns to participate in the water study. First draw samples
were collected by the subjects themselves, while a grab daytime and flushed sample were col-
lected by study personnel. These samples were analyzed by several methods of AAS depending
on the concentration range of the samples.
Blood lead and water lead levels were available for a total of 910 men from 24 towns.
Table 11-38 displays the association between blood lead levels and water lead levels. Blood
lead levels nearly doubled from the lowest to highest water lead category.
The investigators analyzed their data further by examining the form of the relationship
between blood and water lead. This was done by categorizing the water lead levels into nine
intervals of first draw levels. The first group (<6 pg/D had 473 men while the remaining
PB11B/A 11-84 7/29/83
-------
PRELIMINARY DRAFT
50,
E
s
O)
o —
O
o
2
00
2-0
LEAD INTAKE, mg/wk
3-0
Figure 11-15. Blood-lead concentrations versus weekly lead
intake for bottle-fed infants.
PB11B/A
11-85
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-38. RELATIONSHIP OF BLOOD LEAD
AND WATER LEAD (ug/dl) IN 910 MEN AGED 40-59 FROM 24 BRITISH TOWNS
First Draw
Water Lead
(Mg/D
<50
50-99
100-299
£300
Total
Number of
Men
. 789
69
40
12
910
Mean Blood
Lead
(ug/dl)
15.06
18.90
21.65
34.19
15.89
Standard
Deviation
5.53
7.31
7.83
15.27
6.57
% with
Blood Lead
>35 ug/dl
0.7
4.3
7.5
41.7
1.9
' —
Daytime
Water Lead
(ug/D
<50
50-99
100-299
£300
Total
845
36
23
5
909
15.31
19.62
24.78
39.78
15.85
5.64
7.89
9.68
15.87
6.44
0.7
8.3
17.4
60.0
1.8
Source: Pocock et al. (1983).
eight intervals had ~ 50 men each. Figure 11-16 presents the results of this analysis. "The
impression is that mean blood lead increases linearly with first draw water lead except for
the last group with very high water concentrations." The regression line shown in the figure
is only for men less than 100 ug/1, and is given in Table 11-43. A separate regression was
done for the 49 men whose water lead exposures were greater than 100 ug/1. The slope for the
second line was only 23 percent of the first line.
PB11B/A
11-86
7/29/83
-------
'PRELIMINARY DRAFT
Additional analyses were done examining the possible influence of water hardness on blood
lead levels. A strong negative relationship (r = -0.67) was found between blood lead level
and water hardness. There is a possibility that the relationship between blood lead and water
hardness was due to the relationship of water hardness and water lead. It was found that a
relationship with blood lead and water hardness still existed after controlling for water lead
level.
The authors come to the following conclusion regarding the slope of the relationship
between blood lead and water lead:
This study confirms that the relation is not linear at higher levels. Previous
research had suggested a power function relationship--for example, blood lead in-
creases as the cube root of water lead. Our data, based on a large and more
representative sample of men, do not agree with such a curve, particularly at low
concentrations of water lead.
1.25
0 50 100 320 350
FIRST DRAW WATER LEAD (M9/D
i mi iii i —-
61 52
473 60 51 50 65 49 49
Figure 11-16. Mean blood lead for men grouped by first draw water concentra-
tion.
Source: Pocock et al. (1983).
PB11B/A
11-87
7/29/83
-------
PRELIMINARY DRAFT
11.4.2.2. Lead Ingestion from Experimental Dietary Supplements.
11.4.2.2.1 Kehoe study. Experimental studies have been used to study the relationship of
food lead and blood lead levels. Gross (1981) reanalyzed the results of Kehoe. Oral doses of
lead included 300, 1000, 2000, and 3000 ug/day. Each subject had a control period and an ex-
posure period. Some also had a post-exposure period. Blood samples were collected by veni-
puncture and analyzed by spectrographic and dithizone methods during the study years. The
ingestion doses were in addition to the regular ingestion of lead from the diet. The results
of the dose response analysis for blood lead concentrations are summarized in Table 11-39.
Both subjects MR and EB had long exposure periods, during which time their blood lead
levels increased to equilibrium averages of 53 and 60 ug/dl, respectively. The exposure for
IF was terminated early before his blood lead had achieved equilibrium. No response in blood
lead was seen for subject SW whose supplement was 300 ug/day.
TABLE 11-39. DOSE RESPONSE ANALYSIS FOR BLOOD LEAD LEVELS IN THE KEHOE STUDY
AS ANALYZED BY GROSS (1981)
Subject
SW
MR
EB
IF*
Added lead
(ug/day)
300
1000
2000
3000
Diet
(ug/day)
308
1072
1848
2981
Difference fron
Feces
(ug/day)
208
984
1547
2581
n control
Urine
(ug/day) 1
3
55
80
49
Blood
fog/dl)
-1
17
33
19
*Subject did not reach equilibrium.
11.4.2.2.2 Stuik study. Stuik (1974) administered lead acetate in two dose levels (20 and 30
ug/kg body weight-day) to volunteers. The study was conducted in two phases. The first
phase was conducted for 21 days during February-March 1973. Five males and five females aged
p+
18-26 were exposed to a daily dose of 20 ug Pb /kg of body weight. Five males served as
2+
controls. In the second phase, five females received 20 ug Pb /kg body weight and five males
2+
received 30 ug Pb /kg body weight. Five females served as controls. Pre-exposure values
were established during the week preceding the exposures in both phases. Blood lead levels
were determined by Hessel's method.
The results of phase I for blood lead levels are presented in Figure 11-17. Blood lead
levels appeared to achieve an equilibrium after 17 days of exposure. Male blood lead levels
went from 20.6 M9/9 *° 40.9 ug/g while females went from 12.7 to 30.4 ug/g. The males seemed
to respond more to the same body weight dose.
PB11B/A 11-88 7/29/83
-------
PRELIMINARY DRAFT
600
a
a
to 300
a.
100
I I
MM! M I i
_— CONTROL GROUP
— — — EXPOSED
— • — EXPOSED
_^ «,
/
/.
**'
I I I
MALE SUBJECTS: 20 Hg(kg(day —
FEMALE SUBJECTS: 20 Mfl/kg/day
^"*'**" \
s N
— '^' ""•^."-^^
^^* • ,^^^^ ^^^
^^^* * ^^^B. "^™
Ca EDTA Ca EDTA —
. .. | | MALE GROUP FEMALE GROUP
13 8 10 15 17 22
DAYS
29 31
Figure 11-17. Average PbB levels, Exp. I.
Source: Stuik (1974).
38
46
500
00
£
300
100
\\
PB11B/A
I I
I I
CONTROL GROUP
EXPOSED MALE SUBJECTS: 30 Mg'kg/day
EXPOSED FEMALE SUBJECTS: 20 ^kg/day
•Pb EXPOSURE-
II
Ca EDTA
I »
MALE GROUP
-20 47 11 14 18 21 25 27
DAYS
Figure 11-18. Average PbB levels, Exp. II.
Source: Stuik (1974).
11-89
34
7/29/83
-------
PRELIMINARY DRAFT
In phase II, males were exposed to a higher lead dose (30 ug/kg-day). Figure 11-19 dis-
plays these results. Male blood lead rose higher than in the first study (46.2 vs. 40.9
u9/9)'> furthermore, there was no indication of a leveling off. Females also achieved a higher
blood lead level (41.3 vs. 30.4), which the author could not explain. The pre-exposure level,
however, was higher for the second phase than the first phase (12.7 vs. 17.3 ug/g).
11.4.2.2.3 Cools study. Cools et al. (1976) extended the research of Stuik (1974) by ran-
domly assigning 21 male subjects to two groups. The experimental group was to receive a 30
pg/kg body weight dose of oral lead acetate long enough to achieve a blood lead level of 30.0
M9/g, when the lead dose would be adjusted downward to attempt to maintain the subjects at a
blood lead level of 40.0 ug/g. The other group received a placebo.
In the pre-exposure phase, blood lead levels were measured three times, while during ex-
posure they were measured once a week, except for the first three weeks when they were deter-
mined twice a week. Blood lead was measured by flame AAS according to the Westerlund modifi-
cation of Hessel's method.
Pre-exposure blood lead values for the 21 volunteers averaged 172 ppb. The effect of
ingestion of lead acetate on blood lead is displayed in Figure 11-19. After 7 days mean blood
lead levels had increased from 17.2 to 26.2 |jg/g. The time to reach a blood lead level of
35.0 |jg/g took 15 days on the average (range 7-40 days).
11.4.2.2.4 Schlegel study. Schlegel and Kufner (1979) report an experiment in which two sub-
a •£. +2
jects received daily oral doses of 5 mg Pb as an aqueous solution of lead nitrate for 6 and
13 weeks, respectively. Blood and urine samples were taken. Blood lead uptake (from 16 to 60
(jg/dl in 6 weeks) and washout were rapid in subject HS, but less so in subject GK (from 12 to
29 ug/dl in 6 weeks). Time series data on other heme system indicators (FEP, 6-ALA-D,
6-ALA-U, coproporphyrin III) were also reported.
11.4.2.2.5 Chamberlain study. This study (Chamberlain et al., 1978) was described in Section
11.4.1, and in Chapter 10. The ingestion studies on six subjects showed that the gut absorp-
tion of lead was much higher when lead was ingested between meals. There were also dif-
ferences in absorption of lead chloride and lead sulfide.
11.4.2.3 Inadvertent Lead Ingestion from Lead Plumbing.
11.4.2.3.1 Early studies. Although the use of lead piping has been largely prohibited in
recent construction, occasional episodes of poisoning from this lead source still occur.
These cases most frequently involve isolated farms or houses in rural areas, but a surprising
urban episode was revealed in 1972 when Beattie et al. (1972a,b) showed the seriousness of the
situation in Glasgow, Scotland, which had very pure but soft drinking water as its source.
The researchers demonstrated a clear association between blood lead levels and inhibition of
the enzyme ALA-D in children living in houses with (1) lead water pipes and lead water tanks,
PB11B/A 11-90 7/29/83
-------
PRELIMINARY DRAFT
450
400
350
A
300
00
200
100
I 1 11 II 11 I I
f) EXPOSED (n =
O CONTROLS 30
8
DAYS
49
Figure 11-19. Lead in blood (mean values and range) in volunteers. In
the lower curve the average daily lead dose of the exposed group is
shown.
Source: Cools (1976).
PB11B/A
11-91
7/29/83
-------
PRELIMINARY DRAFT
(2) no lead water tank but with more than 60 ft of lead piping and (3) less than 60 ft of lead
piping. The mean lead content of the water as supplied by the reservoir was 17.9 H9/1; those
taken from the faucets of groups 1, 2 and 3 were 934, 239 and 108 ug/1, respectively.
Another English study (Crawford and Crawford, 1969) showed a clear difference between the
bone lead contents of the populations of Glasgow and London, the latter having a hard, nonsol-
vent water supply.
In a study of 1200 blood donors in Belgium (DeGraeve et al., 1975), persons from homes
with lead piping and supplied with corrosive water had significantly higher blood lead levels.
11.4.2.3.2 Moore studies. M. R. Moore and colleagues have reported on several studies rela-
ting blood lead levels to water lead levels. Moore (1977) studied the relationship between
blood lead level and drinking water lead in residents of a Glasgow tenement. The tenement was
supplied with water from a lead-lined water tank carried by lead piping. Water samples were
collected during the day. Comparative water samples were collected from houses with copper
pipes and from 15 lead plumbed houses. Blood samples were taken wherever possible from all
inhabitants of these houses. The data indicated that if a house has lead lined pipes, it is
almost impossible to reach the WHO standard for lead in water. Linear regression equations
relating blood lead levels to first flush and running water lead levels are in Tables 11-43
and 11-44.
Moore et al. (1977) also reported the analysis of blood lead and water lead data col-
lected over a four year period for different sectors of the Scottish population. The combined
data showed consistent increases in blood lead levels as a function of first draw water lead,
but the equation was nonlinear at the higher range. The water lead values were as high as
2000 pg/1. The fitted regression equation for the 949 subjects is in Table 11-43.
Moore et al. (1981a,b) reported a study of the effectiveness of control measures for
plumbosolvent water supplies. In autumn and winter of 1977, they studied 236 mothers aged 17
to 37 in a post-natal ward of a hospital in Glasgow with no historical occupational exposure.
Blood lead and tap water samples from the home were analyzed for lead by AAS under a quality
control program.
A skewed distribution of blood lead levels was obtained with a median value of 16.6
ug/dl; 3 percent of the values exceeding 41 ug/dl. The geometric mean was 14.5 ug/dl. A
curvilinear relationship between blood lead level and water lead level was found. The log of
the maternal blood lead varied as the cube root of both first flush and running water lead
concentrations. In Moore et al. (1979) further details regarding this relationship are
provided. Figure 11-20 presents the observed relationship between blood lead and water lead.
In April 1978 a closed loop lime dosing system was installed. The pH of the water was
raised from 6.3 to 7.8. Before the treatment, more than 50 percent of random daytime water
samples exceeded 100 ug of Pb/1, the WHO standard. After the treatment was implemented, 80
PB11B/A 11-92 7/29/83
-------
PRELIMINARY DRAFT
Q
O
O
23.5
25
24 26 25
24
NO. IN
GROUP
Figure 11-20. Cube root regression of blood lead on first flush water lead.
This shows mean ± S.D. of blood lead for pregnant women grouped in 7
intervals of first flush water lead.
PB11B/A
11-93
7/29/83
-------
PRELIMINARY DRAFT
percent of random samples were less than 100 ug/1. It was found, however, that the higher pH
was not maintained throughout the distribution system. Therefore, in August 1980, the pH was
raised to 9 at the source, thereby maintaining the tap water at 8. At this time more than 95
percent of random daytime samples were less than 100 ug/1.
In the autumn and winter of 1980, 475 mothers from the same hospital were studied. The
median blood lead was 6.6 ug/dl and the geometric mean was 8.1 ug/dl. Comparison of the fre-
quency distributions of blood lead between these two blood samplings show a remarkable drop.
No other source of lead was thought to account for the observed change.
11.4.2.3.3 Thomas study. Thomas et al. (1979) studied women and children residing on two
adjacent housing estates. One estate was serviced by lead pipes for plumbing while the other
was serviced by copper pipe. In five of the homes in the lead pipe estate, the lead pipe had
been replaced with copper pipe. The source water is soft, acidic and lead-free.
Water samples were collected from the cold tap in the kitchen in each house on three oc-
casions at two-week intervals. The following water samples were collected: daytime - first
water out of tap at time of visit; running - collected after tap ran moderately for 5 minutes
after the daytime sample; and first flush - first water out of tap in morning (collected by
residents). Lead was analyzed by a method (unspecified in report) that was reportedly under
quality control.
Blood samples were collected from adult females (2.5 ml venipuncture) who spent most of
the time in the home and from the youngest child (capillary sample). Blood samples were ana-
lyzed for lead by a quality controlled unspecified method. Blood lead levels were higher in
.the residents of the lead estate homes than in the residents of the copper estate homes.
Median levels for adult females were 39 ug/dl and 14.5 ug/dl for the lead and copper estate
homes, respectively. Likewise, children's blood lead levels were 37 ug/dl and 16.6 ug/dl,
respectively. Water lead levels were substantially higher for the lead estate than for the
copper estate. This was true for all three water samples.
The researchers then monitored the effectiveness of replacing the lead pipe on reducing
both exposure to lead in drinking water and ultimately blood lead levels. This monitoring was
done by examining subsamples of adult females for up to 9 months after the change was
implemented. Water lead levels became indistinguishable from those found in the copper estate
homes. Blood lead levels declined about 30 percent after 3 to 4 months and 50 percent at 6
and 9 months. At 6 months the blood lead levels reached those of women living in the copper
estates. A small subgroup of copper estate females was also followed during this time. No
decline was noted among them. Therefore, it was very likely that the observed reduction in
blood lead levels among the other women was due to the changed piping.
PB11B/A 11-94 7/29/83
-------
PRELIMINARY DRAFT
The researchers then analyzed the form of the relationship between blood lead levels and
water lead levels. They tried several different shapes for the regression line. Curvilinear
models provided better fits. Figure 11-21 depicts the scatter diagram of blood lead and water
lead. An EPA analysis of the data is in Table 11-43.
A later publication by Thomas (1980) extended his earlier analysis. This more extensive
analysis was limited to lead estate residents. Subjects who did not consume the first drawn
water from the tap had significantly lower blood lead levels than those who did (10.4 ug/dl
difference). No gradient was noted in blood lead levels with increasing water consumption.
Furthermore, no gradient in blood lead levels was noted with total beverage consumption (tea
ingestion frequency).
11.4.2.3.4 Worth study. In Boston, Massachusetts an investigation was made of water distri-
bution via lead pipes. In addition to the data on lead in water, account was taken of socio-
economic and demographic factors as well as other sources of lead in the environment (Worth et
al., 1981). Participants, 771 persons from 383 households, were classified into age groups of
less than 6, 6 to 20, and greater than 20 years of age for analysis. A clear association
between water lead and blood lead was apparent (Table 11-40). For children under 6 years of
age, 34.6 percent of those consuming water with lead above the U.S. standard of 50 ug/1 had a
blood lead value greater than or equal to 35 ug/dl, whereas only 17.4 percent of those con-
suming water within the standard had blood lead values of greater than or equal to 35 ug/dl.
Worth et al. (1981) have published an extensive regression analysis of these data. Blood
lead levels were found to be significantly related to age, education of head of household, sex
and water lead exposure. Of the two types of water samples taken, standing grab sample and
running grab sample, the former was shown to be more closely related to blood lead levels than
the latter. Regression equations are given in Tables 11-43 and 11-44.
11.4.2.4 Summary of Dietary Lead Exposures Including Water. It is difficult to obtain accu-
rate dose-response relationships between blood lead levels and lead levels in food or water.
Dietary intake must be estimated by duplicate diets or fecal lead determinations. Water lead
levels can be determined with some accuracy, but the varying amounts of water consumed by dif-
ferent individuals adds to the uncertainty of the estimated relationships.
Studies relating blood lead levels to dietary lead intake are compared in Table 11-41.
Most of the subjects in the Sherlock et al. (1982) and United Kingdom Central Directorate on
Environmental Pollution (1982) studies received quite high dietary lead levels (>300 ug/day).
The fitted cubic equations givexhigh slopes at lower dietary lead levels. On the other hand,
the linear slope of the United Kingdom Central Directorate on Environmental Pollution (1982)
study is probably an underestimate of the slope at lower dietary lead levels. For these
PB11B/A 11-95 7/29/83
-------
PRELIMINARY DRAFT
4.0
3.0
I
3.
Q
Q
O
O
2.0
1.0
I
MAXIMUM WATER LEAD
LEVELS ON 'COPPER' ESTATE
I I
MEDIAN WATER LEAD
LEVELS ON 'LEAD' ESTATE
9ft
5*
39
I
I
I
1.0 2.0
FIRST FLUSH WATER LEAD, mg/liter
3.0
Figure 11-21. Relation of blood lead (adult female) to first flush water lead in
combined estates. (Numbers are coincidental points: 9 = 9 or more.) Curve a,
present data; curve b, data of Moore etal.
PB11B/A
11-96
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-40. BLOOD LEAD LEVELS OF 771 PERSONS IN RELATION
TO LEAD CONTENT OF DRINKING WATER, BOSTON, MA
Persons consuming water (standing grab samples)
x2
p <
Blood lead
levels, (jg/dl
<35
>35
Total
= 14.35; df = 1.
CO. 01.
Source: Worth et al .
<50 ug Pb/1
No. Percent
622 91
61 9
683 100
(1981).
£50 jjg Pb/1
No. Percent Total
68 77.3 690
20 22.7 81
88 100.0 771
reasons, the Ryu et al. (1983) study is the most believable, although it only applies to in-
fants. Estimates for adults should be taken from the experimental studies or calculated from
assumed absorption and half-life values.
The experimental studies are summarized in Table 11-42. Most of the dietary intake sup-
plements were so high that many of the subjects had blood lead concentrations much in excess
of 30 ug/dl for a considerable part of the experiment. Blood lead levels thus may not
completely reflect lead exposure, due to the previously noted nonlinearity of blood lead re-
sponse at high exposures. The slope estimates for adult dietary intake are about 0.02 pg/dl
increase in blood lead per M9/day intake, but consideration of blood lead kinetics may in-
crease this value greatly. Such values are a bit lower than those estimated from the popu-
lation studies extrapolated to typical dietary intakes in Table 11-41, about 0.05 |jg/dl per
jag/day. The value for infants is much larger.
The studies relating first flush and running water lead levels to blood lead levels are
in Tables 11-43 and 11-44, respectively. Many of the authors chose to fit cube root models to
their data, although polynomial and logarithmic models were also used. Unfortunately, the
form of the model greatly influences the estimated contributions to blood lead levels from
relatively low water lead concentrations.
The models producing high estimated contributions are the cube root models and the loga-
rithmic models. These models have a slope that approaches infinity as water lead concentra-
tions approache zero. All other are polynomial models, either linear, quadratic or cubic.
The slopes of these models tend to be relatively constant at the origin.
PB11B/A
11-97
7/29/83
-------
TABLE 11-41. STUDIES RELATING BLOOD LEAD LEVELS
TO DIETARY INTAKES
*— *
1— •
VO
CD
Study
Sherlock et al.
(1982) study of
31 adult women
in Ayr
Sherlock et al.
(1982) study of
infants in Ayr
combined with U.K.
Central Directorate
Study
U.K. Central
Directorate
(1982) Study
of infants in
Glasgow
Model
Analysis Model R2 O.F.
3
Sherlock et al. PBB = -1.4 + 3.6 V PBD 0.52 2
(1982)
Sherlock et al. PBB = 2.5 + 5.0 J~PBO - 2
(1982)
U.K. Central PBB = 17.1 > .056(PBD) 0.39 2
Directorate or 3
on Environmental PBB = 3.9 + 4.6 V PBD 0.43 2
Pollution
(1982)
Estimated
Blood
lead at
0 H20 Pb
-1.4
2.5
17.1
3.9
Predicted blood lead
contribution (yg/dl ) for
a given dietary intake
(pg/day)
100 200 300
16.7 21.1 24.1
23.2 29.2 33.5
\
5.6 11.2 16.8
21.4 26.9 30.8
Slope from 100 to 200
(jg/d. , pg/dl per ug/d.
0.034
•o
m
r—
0.060 ^
_<
z
TO
-C.
•yo
0.056 -n
-H
0.053
Ryu et al. (1983)
study of infants
EPA
PBB = A + .16PBD
16.0 32.0 4fl.O
0.16
-------
TABLE 11-42. STUDIES INVOLVING BLOOD LEAD LEVELS (ug/dl)
AND EXPERIMENTAL DIETARY INTAKES
Study
Stuik (1974)
Study 1
Study II
i Cools et al.
tg (1976)
Schlegel and
Kufner (1979)
Gross (1979)
analysis of
Kehoe's
experiments
* Exposure
** Corrected
Subjects
5 adult male students
5 adult female students
5 adult nale students
5 adult female students
5 adult mate students
5 adult female students
11 adult males
10 "lult males
1 adult male
1 adult nale
1 adult mate
1 adult male
1 adult male
1 adult male
(ug/d) = Exposure (ug/kg/day)
for decrease of 2.2 ug/dl in
*** Assumed mean life 40d. This increases
Exposure
20 ug Pb/ kg/day - 21 d.
20 ug Pb/kg/day - 21 d.
Controls - 21 d.
20 ug Pb/kg/day
30 ug Pb/kg/day
Controls
30 ug Pb/kg/day -7 days
Controls
SO ug Pb/kg/day - 6 wk.
70 fjg Pb/kg/day -13 wk.
300 pg/day
1000 ug/day
2000 ug/day
3000 ug/day
x 70 kg for males, 55 kg for
control males.
slope estimate for short-tern
Form of Lead
Lead acetate
Lead acetate
Placebo
Lead acetate
Lead acetate
Placebo
Lead acetate
Placebo
Lead nitrate
Lead nitrate
Lead acetate
Lead acetate
Lead acetate
Lead acetate
females. Slope = (Final
studies. Stuik Study I
Blood Lead
Initial
20.6
12.7
20.6
17.3
16.1
-17.0
17.2
16.5
12.4
- Initial
would be
Final
40.9
30.4
18.4
41.3
46.2
-17.0
26.2
-19.0
64.0
30.4
-1
+ 17
+33
+19
Blood Lead)/Exposure
Slope" ug/dl
per ug/d.
0 017** ***
0.018**,***
0.022
0.014
0.027***
0.014
0.004****
[0]
0.017
0.016
0.006*****
(ug/d).
73
m
i—
t~t
73
0
70
0.042, 0.044 respectively for males, females.
**** Assumed limited absorption of lead.
***** Removed from exposure before equilibrium.
-------
TABLE 11-43. STUDIES RELATING BLOOD LEAD LEVELS (ug/dl) TO FIRST-FLUSH WATER LEAD (ug/1)
Study
Analysis
Model
Estimated Predicted blood lead
Blood contribution (pg/dl for
Model lead at a given water lead (ug/1)
R2 O.F. 0 H20 Pb 5 10 25 50
c
Worth et al. (1981) study of 524
subjects In greater Boston. Water
leads (standing water) ranged fron
<13 to 1108 ug/1. Blood leads
ranged fro* 6 to 71.
Moore et al. (1979) study of 949
subjects from different areas of
Scotland. Water leads were as
high as 2000 ug/1.
Hubermont et al. (1978) study of
70 pregnant women in rural Belgium.
Water leads ranged from 0.2 to
1228.5 ug/1. Blood leads ranged
>_ fron 5.1 to 26.3 ug/dl.
U.K. Central Directorate (1982)
study of 128 mothers in greater
Glasgow. Water leads ranged from
under 50 ug/1 (35X) to over 500
ug/1 (11X). Blood leads ranged
from under 5 pg/dl (2X) to over
35 |ig/dl (5X).
U.K. Central Directorate (1982)
study of 126 infants (as above).
Blood leads ranged from under 5
ug/dl (4X) to over 40 ug/dl (4X).
Thomas et al. (1979) study of 115
adult Welsh females. Water leads
ranged fro* <10 to 2800 ug/dl.
Blood leads ranged from S to 65
ug/dl.
Moore (1977) study of 75 residents
of a Glasgow tenement
Pocock et al. (1983) study of 7735
men aged 40-59 in Great Britain.
Water leads restricted to '100 pg/1.
Worth et al. (1981) In (PBB) = 2.729 PBW - 4.699 (PBW)2 +
2.116 (PBW)3 + other terms for age,
sex, education, dust (PBW is in mg/1)
EPA In(PBB) = In (40.69 PBW - 21.89 (PBW)2
+ other terms for age, sex, education,
dust) (PBW is in mg/1)
Moore et al. (1979) PBB = 11.0 + 2.36 (PBW)
Hubermont et al.
(1978)
U.K. Central
Directorate on
Environmental
Pollution
(1982)
U.K. Central
Directorate on
Environmental
Pollution
(1982)
EPA
PBB = 9.62 + 0.756 in (PBW)
PBB =13.2+1.8 (PBW)
PBB = 18.0 + 0.009 PBW
1/3
PBB = 9.4 * 2.4 (PBW)
PBB = 17.1 + 0.018 PBW
1/3
Moore (1977) PBB = 15.7 «• 0.015 PBW
Pocock et al. (1983) PBB = 14.48 + 0.062 PBW
0.18 14
0.18 11
0.14
0.11
0.05
0.17
0.12
0.34
In (PBB) = [14.9 + 0.041 PBW - 0.000012 0.61 3
(PBW)2]
20.5
21.1
11.0
8.4*
13.2
18.0
9.4
17.1
14.9
15.7
14.5
0.3 0.6 1.4 2.7
0.2 0.4 1.0 2.1
4.0 5.1 6.9 8.7
2.4 3.0 3.7 4.2
3.1 3.9 5.3 6.6
0.0 0.1 0.2 0.4
4.1 5.2 7.0 8.8
0.1 0.2 0.4 0.9
0.2 0.4 1.0 2.0
0.1 0.2 0.4 0.8
0.3 0.6 1.6 3.1
"minimum water lead of 0.2 ug/dl used instead of 0.
-------
TABLE 11-44. STUDIES RELATING BIOOD LEAD LEVELS
TO RUNNING WATER LEAD 61 ug/dl. Kettle water leads
ranged from <10 to >2570 p9/l-
U.S. EPA (1980)
EPA
U.S. EPA (1980)
EPA
EPA
U.K. Central
Directorate on
Environmental
Pollution
(1982)
U.K. Central
Directorate on
Environmental
Pollution
(1982)
Moore (1977)
Sherlock et al.
(1982)
PBB = 14.33 + 2.541 (PBW)i/J
EPA In (PBB) = In (18 6 - 0.071 PBW)
In (PBB) = In (0.073 PBW * other terms
for sex, education, and dust)
PBB = 13.38 * 2.487 (PBW)1/3 '
In (PB8) = ?n (17.6 * 0.067 PBW)
In (PBB) = (0.067 PBW + other terms
for education and dust)
PBB = 12.8 + 1.8 (PBW)1/3
PBB = IB. 1 + . 014 PBW
PBB = 7.6 * 2.3 (PBW)1/3
PBB = 16.7 + 0.033 PBW
PBB = 16.6 + 0.02 PBW
PBB =4.7+2.78 (P6W)1/3
0.023
0.028
0.153
0.030
0.032
0.091
0.12
0.06
0.22
0.12
0.27
0.56
2
2
7
2
2
6
2
2
2
2
2
2
14.3
18.6
18.8
13.4
17.6
17.6
12.8
18.1
7.6
16.7
16.6
4.7
4.4
0.4
0.4
4.3
0.3
0.3
3.1
0.1
3.9
0.2
0.1
4.8
5.4 7.4
0.7 1.8
0.7 1.8
5.4 7.3
0.7 1.7
0.7 1.7
3.9 5.3
0.1 0.4
5.0 6.7
0.3 0.8
0.2 0.5
6.0 8.1
9.4
3.6
3.7
9.2
3.4
3.4
6.6
0.7
8.5
1.6
1.0
10.2
-------
PRELIMINARY DRAFT
The problem of determining the most appropriate model(s) is essentially equivalent to the
low dose extrapolation problem, since most data sets estimate a relationship that is primarily
based on water lead values from 50 to 2000 ug/dl. The only study that determines the re-
lationship based on lower water lead values (<100 pg/1) is the Pocock et al. (1983) study.
The data from this study, as well as the authors themselves, suggest that in this lower range
of water lead levels, the relationship is linear. Furthermore, the estimated contributions to
blood lead levels from this study are quite consistent with the polynomial models from other
studies, such as the Worth et al. (1981) and Thomas et al. (1979) studies. For these reasons,
the Pocock et al. (1983) slope of 0.06 is thought to represent the current best estimate. The
possibility still exists, however, that the higher estimates of the other studies may be cor-
rect in certain situations, especially at higher water lead levels (>100 pg/1).
11.4.3 Studies Relating Lead in Soil and Dust to Blood Lead
The relationship of exposure to lead contained in soil and house dust, and the amount of
lead absorbed by humans, particularly children, has been the subject of scientific investi-
gation for some time (Ouggan and Williams, 1977; Barltrop, 1975; Creason et al., 1975; Barl-
trop et al., 1974; Roberts et al., 1974; Sayre et al., 1974; Ter Haar and Aronow, 1974; Fairey
and Gray, 1970). Duggan and Williams (1977) published an assessment of the risk of increased
blood lead resulting from the ingestion of lead in dust. Some of these studies have been con-
cerned with the effects of such exposures (Barltrop, 1975; Creason et al., 1975; Barltrop et
al., 1974; Roberts et al., 1974; Fairey and Gray, 1970); others have concentrated on the means
by which the lead in soil and dust becomes available to the body (Sayre et al., 1974; Ter Haar
and Aronow, 1974).
11.4.3.1 Omaha Nebraska Studies. The Omaha studies were described in Section 11.4.1.7. Soil
samples were 2-inch cores halfway between the building and the lot line. Household dust was
collected from vacuum cleaner bags. The following analysis was provided courtesy of Or.
Angle. The model is also described in Section 11.4.1.8, and provided the following coeffi-
cients and standard errors:
Asymptotic
Factor Coefficient Standard Error
Intercept (ug/dl) 15.67 0.398
Air lead (M9/m3) 1.92 0.600
Soil lead (mg/g) 6.80 0.966
House dust (mg/g) 7.18 0.900
Multiple R2 = 0.198
Sample size = 1075
Residual standard deviation = 0.300 (geometric standard deviation = 1.35)
PB11B/A 11-102 7/29/83
-------
PRELIMINARY DRAFT
11.4.3.2 The Stark Study. EPA analyses of data from children in New Haven (Stark et al.,
1982) found substantial evidence for dust and soil lead contributions to blood lead, as well
as evidence for increased blood lead due to decreased household cleanliness. These factors
are somewhat correlated with each other, but the separate roles of increased concentration and
Llc-oi.l i.ie--. could be distinguished. The fitted models were summarized earlier (Section
11.3.6.1).
11.4.3.3 The Silver Valley/Kellogg Idaho Study. The Silver Valley Kellogg Idaho study was
discussed in section 11.4.1.6. Yankel et al . (1977) showed _that lead in both soil and dust
was independently related to blood lead levels. In their opinion, 1000 ug/g soil lead ex-
posure was cause for concern. Walter et al . (1980) showed that children aged 3 through 6
showed the strongest relationship between soil lead and blood lead, but 2-year olds and 7-year
olds also had a significant relationship (Table 11-24). The slope of 1.1 for soil lead (1000
ug/g) to blood lead (ug/dl) represents an average relationship for all ages.
The Silver Valley-Kel logg Idaho K§tudy also gave some information on house dust lead, al-
though this data was less complete cttvan the other information. Regression coefficients for
these data are in Tables 11-24 and 11-25. In spite of the correlation of these predictors,
significant regression coefficients could be estimated separately for these effects.
11.4.3.4 Charleston Studies. In one of the earliest investigations, Fairey and Gray (1970)
conducted a retrospective study of lead poisoning cases in Charleston, South Carolina. Two-
inch core soil samples were collected from 170 randomly selected sites in the city and were
compared with soil samples taken from homes where 37 cases of lead poisoning had occurred.
Jhe soil lead values obtained ranged from 1 to 12,000 ug/g, with 75 percent of the samples
containing less than 500 ug/g. A significant relationship between soil lead levels and lead
poisoning cases was established; 500 ug/g was used as the cutpoint in the chi-square contin-
gency analysis. Fairey and Gray were the first to examine this complex problem and, although
their data support the soil lead hypothesis, the relationship between soil lead and blood lead
levels could not be quantified. Furthermore, because no other source of lead was measured,
any positive association could have been confounded by additional sources of lead, such as
paint or air.
A later study by Galke et al. (1975), in Charleston, used a house-to-house survey to re-
cruit 194 black preschool children. Soil, paint and air lead exposures as measured by traffic
density were established for each child. When the population was divided into two groups
based on the median soil lead value (585 ug/g), a 5 ug/dl difference in blood lead levels was
obtained. Soil lead exposure for this population ranged from 9 to 7890 ug/g- Vehicle traffic
patterns were defined by area of recruitment as being high or low. A multiple regression
analysis of the data showed that vehicle traffic patterns, lead level in exterior siding
PB11B/A 11-103 7/29/83
-------
PRELIMINARY DRAFT
paint, and lead in soil were all independently and significantly related to blood lead levels.
Using the model described in Appendix 11B, the following coefficients and standard errors were
obtained:
Asymptotic
Factor Coefficient Standard Error
Intercept (ug/dl) 25.92 1.61
Pica (1 = eater, 0 = otherwise) 7.23 1.60
Traffic Pattern (1 = high, 0 = low) 7.11 1.48
Siding paint (mg/cm2) 0.33 0.11
Door paint (mg/cm2) 0.18 0.12
Soil lead (mg/g) 1.46 0.59
Multiple R2 = 0.386
Residual standard deviation = 0.2148 (geometric standard deviation = 1.24)
11.4.3.5 Barltrop Studies. Barltrop et al. (1974) described two studies in England investi-
gating the soil lead to blood lead relationship. In the first study, children aged 2 and 3
and their mothers from two towns chosen for their soil lead content had their blood lead
levels determined from a capillary sample. Hair samples were also collected and analyzed for
lead. Lead content of the suspended particulate matter and soil was measured. Soil samples
for each home were a composite of several 2-inch core samples taken from the yard of each
home. Chemical analysis of the lead content of soil in the two towns showed a 2- to 3-fold
difference, with the values in the control town about 200 to 300 ug/g compared with about 700
to 1000 ug/g in the exposed town. A difference was also noted in the mean air lead content of
the two towns, 0.60 ug/m3 compared with 0.29 (jg/m3. Although this difference existed, both
air lead values were thought low enough not to affect the blood level values differentially.
Mean surface soil lead concentrations for the two communities were statistically different,
the means for the high and low community being 909 and 398 ug/g, respectively. Despite this
difference, no statistically significant differences in maternal blood lead levels or chil-
dren's blood or hair lead levels were noted. Further statistical analysis of the data, using
correlational analysis on either raw or log-transformed blood lead data, likewise failed to
show a statistical relationship of soil lead with either blood lead or hair lead.
The second study was reported in both preliminary and final form (Barltrop et al., 1974;
Barltrop, 1975). In the more detailed report (Barltrop, 1975), children's homes were clas-
sified by their soil lead content into three groups, namely: less than 1,000; 1,000 to
10,000; and greater than 10,000 ug/g. As shown in Table 11-45, children's mean blood lead
levels increased correspondingly from 20.7 to 29.0 ug/dl. Mean soil lead levels for the low
and high soil exposure groups were 420 and 13,969 ug/g, respectively. Mothers' blood levels,
PB11B/A H-104 7/29/83
-------
PRELIMINARY DRAFT
however, did not reflect this trend; nor were the children's fecal lead levels different
across the soil exposure areas.
An analysis of the data in Table 11-45 gives the following model:
blood lead ((jg/dl) = 0.64 soil lead (1000 pg/g) + 20.98
No confidence intervals were calculated since the calculations were based on means.
TABLE 11-45. MEAN BLOOD AND SOIL LEAD
CONCENTRATIONS IN ENGLISH STUDY
Category
of soil lead,
M9/9
<1000
1000-10000
>10000
Sample
size
29
43
10
Children's
blood lead,
pg/dl
20.7
23.8
29.0
Soil lead,
pg/g
420
3390
13969
Source: Barltrop, 1975.
11.4.3.6 The British Columbia Studies. Neri et al. (1978) studied blood lead levels in chil-
dren living in Trail, British Columbia. These blood lead measurements were made by the
capillary method. An episode of poisoning of horses earlier had been traced to ingestion of
lead. Environmental monitoring at that time did not suggest that a human health risk existed.
However, it was later thought wise to conduct a study of lead absorption in the area.
Trail had been the site of a smelter since the turn of the century. The smelter had
undergone numerous changes for reasons of both health and productivity. At the time of the
blood lead study, the smelter was emitting 300 pounds of lead daily, with ambient air lead
3
levels at aboyt 2 pg/m in 1975. Nelson, BC was chosen as the control city. The cities are
reasonably close (~3Q miles distant), are similar in population, and served by the same water
basin. The average air lead level in Nelson during the study was 0.5 ug/m .
Initial planning called for the sampling of 200 children in each of three age groups (1-3
years, 1st grade and 9th grade) from each of the two sites. A strike at the smelter at the
onset of the study caused parts of the Trail population to move. Hence, the recruited sample
deviated from the planned one. School children were sampled in May 1975 at their schools
while the 1- to 3-year olds were sampled in September 1975 at a clinic or home. This delayed
sampling was intentional to allow those children to be exposed to the soil and dust for the
entire summer. Blood and hair samples were collected from each child.
PBUB/A 11-105 7/29/83
-------
PRELIMINARY DRAFT
Blood samples were analyzed for lead by anodic stripping voltammetry. The children in
the younger age groups living in Trail had higher blood lead levels than those living in
Nelson. An examination of the frequency distributions of the blood lead levels showed that
the entire frequency of the distribution shifted between the residents of the two cities.
Interestingly, there was no difference in the ninth grade children.
Table 11-46 displays the results of the soil lead levels along with the blood lead levels
obtained in the earlier study. Blood lead levels were higher for 1- to 3-year olds and first
graders in the two nearest-to-smelter categories than in the far-fronrsmelter category.
Again, no difference was noted for the ninth graders.
An EPA analysis of the Neri et al. (1978) data gives the following models for children 1-
to 3-years old:
Blood lead ((jg/dl) = 0.0076 soil lead (fjg/g) + 15.43, and
Blood lead (pg/dl) = 0.0046 soil lead (pg/g) + 16.37
for children in grade one. No confidence intervals were calculated since the analysis was
based on means.
TABLE 11-46. LEAD CONCENTRATION OF SURFACE SOIL AND CHILDREN'S
BLOOD BY RESIDENTIAL AREA OF TRAIL, BRITISH COLUMBIA
Residential
area(s)
1 and 2
5
9
3, 4, and 8
6 and 7
Total
Mean
soil lead
concentration (ug/g)
± standard error
(and no. of samples)
225 ± 39 (26)
777 ± 239 (12)
570 ± 143 (11)
1674 ± 183 (53)
1800 ± 212 (51)
1320 ± 212 (153)
Blood lead concentration
(ug/dl), mean ± standard
error (and no. of children)
1- to 3-
year olds
17.2 ± 1.1 (27)
19.7 ± 1.5 (11)
20.7 ± 1.6 (19)
27.7 ± 1.8 (14)
30.2 ± 3.0 (16)
22.4 + 1.0 (87)
Grade one
children
18.0 ± 1.9 (18)
18.7 ± 2.3 (12)
19.7 ± 1.0 (16)
23.8 ± 1.3 (31)
25.6 ± 1.5 (26)
21.9 ± 0.7 (103)
Source: Schmitt et al., 1979.
11.4.3.7 Other Studies of Soil and Dusts. Lepow et al. (1975) studied the lead content of
air, house dust and dirt, as well as the lead content of dirt on hands, food and water, to
determine the cause of chronically elevated blood lead levels in 10 children 2- to 6-years-old.
PB11B/A 11-106 7/29/83
-------
PRELIMINARY DRAFT
in Hartford, Connecticut. Lead-based paints had been eliminated as a significant source of
lead for these children. Ambient air lead concentrations varied from 1.7 to 7.0 ug/m . The
mean lead concentration in dirt was 1,200 ug/g and in dust, 11,000 ug/g. The mean concentra-
tion of lead in dirt on children's hands was 2,400 ug/g. The mean weight of samples of dirt
from hands was 11 mg, which represented only a small fraction of the total dirt on hands. Ob-
servation of the mouthing behavior in these young children led to the conclusion that the
hands-in-mouth exposure route was the principal cause of excessive lead accumulation.
Several studies have investigated the mechanism by which lead from soil and dust gets in-
to the body (Sayre et al., 1974; Ter Haar and Aronow, 1974). Sayre et al. (1974) in
Rochester, New York, demonstrated the feasibility of house dust as a source of lead for chil-
dren. Two groups of houses, one inner city and the other suburban, were chosen for the study.
Lead-free sanitary paper towels were used to collect dust samples from house surfaces and the
hands of children (Vostal et al., 1974). The medians for the hand and household samples were
used as the cutpoints in the chi-square contingency analysis. A statistically significant
difference between the urban and suburban homes for dust levels was noted, as was a relation-
ship between household dust levels and hand dust levels (Lepow et al., 1975).
Ter Haar and Aronow (1974) investigated lead absorption in children that can be at-
tributed to ingestion of dust and dirt. They reasoned that because the proportion of the
naturally occurring isotope of 2l°Pb varies for paint chips, airborne particulates, fallout
dust, house dust, yard dirt and street dirt, it would be possible to identify the sources of
ingested lead. They collected 24-hour excreta from eight hospitalized children on the first
.day of hospitalization. These children, 1- to 3-years old, were suspected of having elevated
body burdens of lead, and one criterion for the suspicion was a history of pica. Ten children
of the same age level, who lived in good housing in Detroit and the suburbs, were selected as
controls and 24-hour excreta were collected from them. The excreta were dried and stable lead
as well as 210Pb content determined. For seven hospitalized children, the stable lead mean
value was 22.43 ug/g dry excreta, and the eighth child had a value of 1640 ug/g. The con-
trols' mean for stable lead was 4.1 ug/g dry excreta. However, the respective means for 210Pb
expressed as pCi/g dry matter were 0.044 and 0.040. The authors concluded that because there
is no significant difference between these means for 210Pb, the hypothesis that young children
with pica eat dust is not supported. The authors further concluded that children with
evidence of high lead intake did not have dust and air suspended particulate as the sources of
their lead. It is clear that air suspended particulate did not account for the lead levels in
the hospitalized children. However, the 2l°Pb concentrations in dust and feces were similar
for all children, making it difficult to estimate the dust contribution.
Heyworth et al. (1981) studied a population of children exposed to lead in mine tailings.
These tailings were used in foundations and playgrounds, and had a lead content ranging from
PB11B/A 11-107 7/29/83
-------
PRELIMINARY DRAFT
10,000 to 15,000 ug/g. In December 1979 venous blood samples and hair were collected from 181
of 346 children attending two schools in Western Australia. One of the schools was a primary
school; the other was a combined primary and secondary school. Parents completed question-
naires covering background information as well as information regarding the children's expo-
sure to the tailings. Blood lead levels were determined by the AAS method of Farrely and
Pybos. Good quality control measures were undertaken for the study, especially for the blood
lead levels. Blood lead levels were higher in boys vs. girls (mean values were 14.0 and 10.4
ug/dl, respectively). This difference was statistically significant. Five percent of the
children (n = 9) had blood lead levels greater than 25 ug/dl. Five of the children had blood
lead levels greater than 30 ug/dl. Blood lead levels decreased significantly with age and
were slightly lower in children living on properties on which tailings were used. However,
they were higher for children attending the school that used the tailings in the playground.
Landrigan et al. (1982) studied the impact on soil and dust lead levels on removal of
leaded paint from the Mystic River Bridge in Masschusetts. Environmental studies in 1977 in-
dicated that surface soil directly beneath the bridge had a lead content ranging from 1300 to
1800 ug/g. Analysis of concomitant trace elements showed that the lead came from the bridge.
A concurrent survey of children living in Chelsea (vicinity of bridge) found that 49 percent
of 109 children had blood lead levels greater than or equal to 30 ug/dl. Of children living
more distant from the bridge "only" 37 percent had that level of blood lead.
These findings prompted the Massachusetts Port Authority to undertake a program to delead
the bridge. Paint on parts of the bridge that extended over neighborhoods was removed by
abrasive blasting and replaced by zinc primer. Some care was undertaken to minimize both the
occupational as well as environmental exposures to lead as a result of the blasting process.
Concurrently with the actual deleading work, a program of air monitoring was established
to check on the environmental lead exposures being created. In June 1980 four air samples
taken at a point 27 meters from the bridge had a mean lead content of 5.32 ug/m . As a result
of these findings air pollution controls were tightened; mean air lead concentrations 12
o
meters from the bridge in July were 1.43 ug/m .
Samples of the top 1 cm of soil were obtained in July 1980 from within 30, 30 to 80, and
100 meters from the bridge. Comparison samples from outside the area were also obtained.
Samples taken directly under the bridge had a mean lead content of 8127 ug/g.
Within 30 meters of the bridge, the mean content was 3272 ug/g, dropping to 457 ug/g at 30 to
80 meters. At 100 meters the soil lead level dropped to 197 ug/g. Comparison samples ranged
from 83 to 165 ug/g depending on location.
Fingerstick blood samples were obtained on 123 children 1-5 years of age living within
0.3 km of the bridge in Charlestown. Four children (3.3 percent) had blood lead levels
PB11B/A 11-108 7/29/83
-------
PRELIMINARY DRAFT
greater than 30 ug/dl with a maximum of 35 pg/dl. All four children lived within two blocks
of the bridge. Two of the four had lead paint in their homes but it was intact. None of the
76 children living more than two blocks from the bridge had blood leads greater than or equal
to 30 (jg/dl, a statistically significant difference.
Sheii Dear's (1973) case report from New Zealand ascribes a medically diagnosed case of
lead poisoning to high soil lead content in the child's home environment. Shellshear et al.
(1975) followed up his case report of increased lead absorption resulting from exposure to
lead contaminated soil with a study carried out in Christchurch, New Zealand. Two related
activities comprised the study. First, from May 1973 to November 1973, a random study of
pediatric admissions to a local hospital was made. Blood samples were taken and analyzed for
lead. Homes were visited and soil samples were collected and analyzed for lead. Lead anal-
yses for both soil and blood were conducted by AAS. Second, a soil survey of the area was
undertaken. Whenever a soil lead value greater than 300 ug/g was found and a child aged one
to five was present, the child was referred for blood testing.
The two methods of subject recruitment yielded a total of 170 subjects. Eight (4.7 per-
cent) of the children had blood lead equal to or greater than 40 ug/dl, and three of them had
a blood lead equal to or greater than 80 ug/dl. No correlation with age was noted. The mean
blood lead of the pediatric admissions was 17.5 ug/dl with an extremely large range (4 to 170
ug/dl). The mean blood lead for soil survey children was 19.5 ug/dl.
Christchurch was divided into two sections based on the date of development of the area.
The inner area had developed earlier and a higher level of lead was used there in the house
paints. The frequency distribution of soil lead levels showed that the inner zone samples had
much higher soil lead levels than the outer zone. Furthermore, analysis of the soil lead
levels by type of exterior surface of the residential unit showed that painted exteriors had
higher soil lead values than brick, stone or concrete block exteriors.
Analysis of the relationship between soil lead and blood lead was restricted to children
from the sampled hospital who had lived at their current address for at least 1 year. Table
11-47 presents the analysis of these results. Although the results were not statistically
significant, they are suggestive of an association.
Analysis of the possible effect of pica on blood lead levels showed the mean blood lead
for children with pica to be 32 ug/dl while those without pica had a mean of 16.8 ug/dl. The
pica blood lead mean was statistically significantly higher than the non-pica mean.
Wedeen et al. (1978) reported a case of lead nephropathy in a black female who exhibited
geophagia. The patient, who had undergone chelation therapy, eventually reported that she had
a habit of eating soil from her garden in East Orange, New Jersey. During spring and summer,
she continuously kept soil from her garden in her mouth while gardening. She even put a sup-
ply away for winter. The soil was analyzed for lead and was found to contain almost 700 ug/g.
PB11B/A 11-109 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-47. ANALYSIS OF RELATIONSHIP BETWEEN SOIL LEAD AND BLOOD
LEAD IN CHILDREN
Soil lead (ug/g) Blood lead ug/dl)
Area of city
Inner
Outer
zone
zone
Mean
1950
150
Range
30-11000
30-1100
n
21
47
Mean
25.
18.
4
3
Range
4-170
5-84
Source: Shellshear (1973).
The authors estimated that the patient consumed 100 to 500 mg of lead each year. One month
after initial hospitalization her blood lead level was 70 ug/dl.
11.4.3.8 Summary of Soil and Dust Lead . Studies relating soil lead to blood lead levels are
difficult to compare. The relationship obviously depends on depth of soil lead, age of the
children, sampling method, cleanliness of the home, mouthing activities of the children, and
possibly many other factors. Table 11-48 gives some estimated slopes taken from several dif-
ferent studies. The range of these values is quite large, ranging from 0.6 to 7.6. The
values from the Stark et al. (1980) study of about 2 ug/dl per mg/g represent a reasonable
median estimate.
The relationship of house dust lead to blood lead is even more difficult to obtain.
Table 11-49 contains some values for three studies that give data permitting such caculations.
The median value of 1.8 ug/dl per mg/g for 2-3 years old in the Stark study may also represent
a reasonable value for use here.
PB11B/A 11-110 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-48. ESTIMATES OF THE CONTRIBUTION OF SOIL LEAD
TO BLOOD LEAD
Study
Anqle and Mclntire
(198?) study of
children in
Omaha, NE
Stark et al.
(1982) study
of children
New Haven, CT
Range of soil
lead values
(pg/g)
16 to 4792
30 to 7000
(age 0-1)
30 to 7600
(age 2-3)
Depth of
sample
2"
V
Estimated .,
slope (X10J)
6.8
2.2
2.0
Sample
size
1075
153
334
R2
.198
.289
.300
Yankel et al.
(1977) study
of children
in Kellogg, ID
Galke et al.
(1975)
study of
chilren in
Charleston, SC
50 to 24,600
9 to 7890
3/4"
2"
1.1
1.5
860
194
.662
.386
Barltrop et
al. (1975)
study of
children in
England
Neri et al.
(1978) study
of children
in British
Columbia
420 to 13,969 2"
(group means)
225-1800 NA
(group means ,
age 1-3)
225-1800 NA
(group means,
age 2-3)
0.6 82 NA*
7.6 87 NA
4.6 103 NA
*NA means Not Available.
PB11B/A
11-111
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-49. ESTIMATES OF THE CONTRIBUTION OF
HOUSEDUST TO BLOOD LEAD IN CHILDREN
Range of dust
Study Lead values (ug/g)
Angle and Mclntire
(1979) study in
Omaha, NE
Stark et al. (1982)
study in New Haven,
CT
Yankel et al. (1977)
study in Kellogg,
ID
18-5571
70-7600
40-7600
9-4900
50-35,600
Age range
in years
1-18
6-18
0-1
2-3
4-7
0-4
5-9
Estimated , Sample
slope (X10 ) Size
7.18
3.36
4.02
1.82
0.02
0.19
0.20
1074
832
153
334
439
185
246
R2
.198
.262
.289
.300
.143
.721
.623
11.4.4 Paint Lead Exposures
A major source of environmental lead exposure for the general population comes from lead
contained in both interior and exterior paint on dwellings. The amount of lead present, as
well as its accessibility, depends upon the age of the residence (because older buildings
contain paint manufactured before lead content was regulated) and the physical condition of
the paint. It is generally accepted by the public and by health professionals that lead-based
paint is one major source of overtly symptomatic pediatric lead poisoning in the United States
(Lin-Fu, 1973).
The level and distribution of lead paint in a dwelling is a complex function of history,
geography, economics, and the decorating habits of its residents. Lead pigments were the
first pigments produced on a large commercial scale when the paint industry began its growth
in the early 1900's. In the 1930's lead pigments were gradually replaced with zinc and other
opacifiers. By the 1940's, titanium dioxide became available and is now the most commonly
used pigment for residential coatings. There was no regulation of the use of lead in house
paints until 1955, when the paint industry adopted a voluntary standard that limited the lead
content in interior paint to no more than 1 percent by weight of the nonvolatile solids. At
about the same time, local jurisdictions began adopting codes and regulations that prohibited
the sale and use of interior paints containing more than 1 percent lead (Berger, 1973a,b).
In spite of the change in paint technology and local regulations governing its use, and
contrary to popular belief, interior paint with significant amounts of lead was still availa-
ble in the 1970's. Studies by the National Bureau of Standards (1973) and by the U.S.
PB11B/A
11-112
7/29/83
-------
PRELIMINARY DRAFT
Consumer Product Safety Commission (1974) showed a continuing decrease in the number of in-
terior paints with lead levels greater than 1 percent. By 1974, only 2 percent of the in-
terior paints sampled were found to have greater than 1 percent lead in the dried film (U.S.
Consumer Product Safety Commission, 1974).
The level of lead in paint in a residence that should be considered hazardous remains in
question. Not only is the total amount of lead in paint important, but also the accessibility
of the painted surface to a child, as well as the frequency of ingestion must be considered.
Attempts to set an acceptable lead level, jn situ, have been unsuccessful, and preventive con-
trol measures of lead paint hazards has been concerned with lead levels in currently manu-
factured paint. In one of its reviews, the NAS concluded: "Since control of the lead paint
hazard is difficult to accomplish once multiple layers have been applied in homes over two to
three decades, and since control is more easily regulated at the time of manufacture, we re-
commend that the lead content of paints be set and enforced at time of manufacture" (National
Academy of Sciences, 1976).
Legal control of lead paint hazards is being attempted by local communities through
health or housing codes and regulations. At the Federal level, the Department of Housing and
Urban Development has issued regulations for lead hazard abatement in housing units assisted
or supported by its programs. Generally, the lead level considered hazardous ranges from 0.5
2
to 2.5 mg/cm , but the level of lead content selected appears to depend more on the sensiti-
vity of field measurement (using X-ray fluorescent lead detectors) than on direct biological
dose-response relationships. Regulations also require lead hazard abatement when the paint is
loose, flaking, peeling or broken, or in some cases when it is on surfaces within reach of a
child's mouth.
Some studies have been carried out to determine the distribution of lead levels in paint
in residences. A survey of lead levels in 2370 randomly selected dwellings in Pittsburgh pro-
vides some indication of the lead levels to be found (Shier and Hall, 1977). Figure 11-22
shows the distribution curves for the highest lead level found in dwellings for three age
groupings. The curves bear out the statement often made that paint with high levels of lead
is most frequently found in pre-1940 residences. One cannot assume, however, that high lead
paint is absent in dwellings built after 1940. In the case of the houses surveyed in
Pittsburgh, about 20 percent of the residences built after 1960 have at least one surface with
2
more than 1.5 mg/cm .
The distribution of lead within an individual dwelling varies considerably. Lead paint
2
Is most frequently found on doors and windows where lead levels greater than 1.5 mg/cm were
found on 2 percent of the surfaces surveyed, whereas only about 1 percent of the walls had
2
lead levels greater than 1.5 mg/cm (Shier and Hall, 1977).
PB11B/A 11-113 7/29/83
-------
PRELIMINARY DRAFT
0.8 —
0.7 —
X
Al
UJ 0.6
Q
<
UJ
O
Z
0.5
I
9 04
o
CC
U.
0.3
0.2
0.1
N = 2525
LEAD LEVEL (X), mg/cm*
Figure 11-22. Cumulative distribution of lead levels in dwelling units.
PB11B/A
11-114
7/29/83
-------
PRELIMINARY DRAFT
In a review of the literature (Lin-Fu, 1973) found general acceptance that the presence
of lead In paint is necessary but not sufficient evidence of a hazard. Accessibility in
terms of peeling, flaking or loose paint also provide evidence for the presence of a hazard.
Of the total samples surveyed, about 14 percent of the residences had accessible paint with a
2
lead content greater than 1.5 mg/cm . As discussed in Section 7.3.2.1.2, one must note that
lead oxides of painted surfaces contribute to the lead level of house dust.
It is not possible to extrapolate the results of the Pittsburgh survey nationally; how-
ever, additional data from a pilot study of 115 residences in Washington, DC, showed similar
results (Hall 1974).
An attempt was made in the Pittsburgh study to obtain information about the correlation
between the quantity and condition of lead paint in buildings, and the blood lead of children
who resided there (Urban, 1976). Blood lead analyses and socioeconomic data for 456 children
were obtained, along with the information about lead levels in the dwelling. Figure 11-23 is
a plot of the blood lead levels vs. the fraction of surfaces within a dwelling with lead
2
levels of at least 2 mg/cm . Analysis of the data shows a low correlation between the blood
p
lead levels of the children and fraction of surfaces with lead levels above 2 mg/cm , but
there is a stronger correlation between the blood lead levels and the condition of the painted
surfaces in the dwellings in which children reside. This latter correlation appeared to be
independent of the lead levels in the dwellings.
Two other studies have attempted to relate blood lead levels and paint lead as determined
by X-ray fluorescence. Reece et al. (1972) studied 81 children from two lower socioeconomic
communities in Cincinnati. Blood leads were analyzed by the dithizone method. There was con-
siderable lead in the home environment, but it was not reflected in the children's blood lead.
Analytical procedures used to test the hypothesis were not described; neither were the raw
data presented.
Galke et al. (1975), in their study of inner city black children measured the paint lead,
both interior and exterior, as well as soil and traffic exposure. In a multiple regression
analysis, exterior siding paint lead was found to be significantly related to blood lead
levels.
Evidence indicates that a source of exposure in childhood lead poisoning is peeling lead
paint and broken lead-impregnated plaster found in poorly maintained houses. There are also
reports of exposure cases that cannot be equated with the presence of lead paint. Further,
the analysis of paint in homes of children with lead poisoning has not consistently revealed a
hazardous lead content (Lin-Fu, 1973). For example, one paper reported 5466 samples of paint
obtained from the home environment of lead poisoning cases in Philadelphia between 1964 and
1968. Among these samples of paint, 67 percent yielded positive findings, i.e., paint with
more than 1 percent lead (Tyler, 1970).
PB11B/A 11-115 7/29/83
-------
PRELIMINARY DRAFT
i
g
X
u
T3
*
CO
>
Q
UJ
_J
O
O
O
00
30
25
20
15
I I
I
SURFACES IN
_ CHALKING,
-
••• • • "™
••fl •—B^^1™^^^^"^^
"•
—
«f I I
I
BAD
I I I I
CONDITION, i.e., PEELING,
I
OR POOR SUBSTRATE _
SURFACES
• — """o"
Q
I
•-— •
|
0
*j n u
• o
I I I I
f*
9 -
*^™
—
4
I 1
0.1 0.2 0.3 0.4 0.5
FRACTIONS OF SURFACES WITH LEAD >2 mg/cm
0.6 0.7 0.8 0.9
2
1.0
Figure 11-23. Correlation of children's blood lead levels with fractions of surfaces
within a dwelling having lead concentrations > 2 mg Pb/cm2.
s
PB11B/A
11-116
7/29/83
-------
PRELIMINARY DRAFT
Data published or made available by the Centers for Disease Control also show that a sig-
nificant number of children with undue lead absorption occupy buildings that were inspected
for lead-based paint hazards, but in which no hazard could be demonstrated (U.S. Centers for
Disease Control, 1977a; Hopkins and Houk, 1976). Table 11-50 summarizes the data obtained
frc::. the HEW funded lead-based paint poisoning control projects for Fiscal Years 1981, 1979,
1978, 1975, and 1974. These data show that in Fiscal Years 1974, 1975, and 1978, about 40 to
50 percent of confirmed cases of elevated blood lead levels, a possible source of lead paint
hazard could not be located. In fiscal year 1981, the U.S. Centers for Disease Control
(1982a,b), screened 535,730 children and found 21,897 with lead toxicity. Of these, 15,472
dwellings were inspected and 10,666 or approximately 67 percent were found to have leaded
paint. The implications of these findings are not clear. The findings are presented in order
to place in proper perspective both the concept of total lead exposure and the concept that
lead paint is one source of lead that contributes to the total body load. The background con-
tribution of lead from other sources is still not known, even for those children for whom a
potential lead paint hazard has been identified; nor is it known what proportion of lead came
from which source.
TABLE 11-50. RESULTS OF SCREENING AND HOUSING INSPECTION IN CHILDHOOD LEAD
POISONING CONTROL PROJECT BY FISCAL YEAR
Fiscal Year
Results
Children screened
Children with elevated
lead exposure
Dwellings inspected
Dwellings with
lead hazard
1981
535,730
21,897
15,472
10,666
1979
464,751
32,537
17,911
12,461
1978
397,963
25,801
36,138
18,536
1975
440,650
28,597a
30,227
17,609
1974
371,955
16,228a
23,096
13,742
Confirmed blood lead level >40 pg/dl.
Source: U.S. Centers for Disease Control (1977a, 1979, 1980, 1982a,b);
Hopkins and Houk, 1976.
PB11B/A 11-117 7/29/83
-------
PRELIMINARY DRAFT
11.5 SPECIFIC SOURCE STUDIES
The studies reviewed in this section all provide important information regarding specific
environmental sources of airborne lead that play a significant role in population blood lead
levels. These studies also illustrate several interesting approaches to this issue.
11.5.1 Combustion of Gasoline Antiknock Compounds
11.5.1.1 Isotope Studies. Two field investigations have attempted to derive estimates of
the amount of lead from gasoline that is absorbed by the blood of individuals. Both of these
investigations used the fact that non-radioactive isotopes of lead are stable. The varying
proportions of the isotopes present in blood and environmental samples can indicate the source
of the lead. The Isotopic Lead Experiment (ILE) is an extensive study that attempted to use
differing proportions of the isotopes in geologic formations to infer the proportion of lead
in gasoline that is absorbed by the body. The other study utilized existing natural shifts in
isotopic proportions in an attempt to do the same thing.
11.5.1.1.1 Italy. The ILE is a large scale community study in which the geologic source of
lead for antiknock compounds in gasoline was manipulated to change the isotopic composition of
the atmosphere (Garibaldi et al., 1975; Facchetti, 1979). Preliminary investigation of the
environment of Northwest Italy, and the blood of residents there, indicated that the ratio of
lead 206/207 in blood was a constant, about 1.16, and the ratio in gasoline was about 1.18.
This preliminary study also suggested that it would be possible to substitute for the curren-
tly used geologic sources of lead for antiknock production, a geologically distinct source of
lead from Australia that had an isotopic 206/207 ratio of 1.04. It was hypothesized that the
resulting change in blood lead 206/207 ratios (from 1.16 to a lower value) would indicate the
proportion of lead in the blood of exposed human populations attributable to lead in the air
contributed by gasoline combustion in the study area.
Baseline sampling of both the environment and residents in the geographic area of the
study was conducted in 1974-75. The sampling included air, soil, plants, lead stock, gasoline
supplies, etc. Human blood sampling was done on a variety of populations within the area.
Both environmental and human samples were analyzed for lead concentrations as well as isotopic
206/207 composition.
In August 1975 the first switched (Australian lead labelled) gasoline was introduced;
although it was originally intended to get a 100 percent substitution, practical and logisti-
cal problems resulted in only a 50 percent substitution being achieved by this time. By May
1977, these problems were worked out and the substitution was practically complete. The sub-
stitution was maintained until the end of 1979, when a partial return to use of the original
sources of lead began. Therefore, the project had four phases: phase zero - background;
phase one - partial switch; phase two - total switch; and phase three - switchback.
PB11C/A 11-118 7/29/83
-------
PRELIMINARY DRAFT
Airborne lead measurements were collected in a number of sites to generate estimates of
the lead exposure that was experienced by residents of the area. Turin, the major city of the
region, was found to have a much greater level of atmospheric lead than the surrounding
countryside. There also appeared to be fairly wide seasonal fluctuations.
The isotopic lead ratios obtained in the samples analyzed are displayed in Figure 11-24.
It can easily be seen that the airborne particulate lead rapidly changed its isotope ratio in
line with expectations. Changes in the isotope ratios of the blood samples appeared to lag
somewhat behind. Background blood lead ratios for adults were 1.1591 ± 0.0043 in rural areas
and 1.1627 ± 0.0022 in Turin in 1975. For Turin adults, a mean isotopic ratio of 1.1325 was
obtained in 1979, clearly less than background. Isotopic ratios for Turin schoolchildren,
obtained starting in 1977, tended to be somewhat lower than the ratios for Turin adults.
Preliminary analysis of the isotope ratios in air lead allowed for the estimation of the
fractional contribution of gasoline in the city of Turin, in small communities within 25 km of
Turin, and in small communities beyond 25 km (Facchetti and Geiss, 1982). At the time of
maximal use of Australian lead isotope in gasoline (1978-79), about 87.3 percent of the air
lead in Turin and 58.7 percent of the air lead in the countryside was attributable to
gasoline. The determination of lead isotope ratios was essentially independent of air lead
concentrations. During that time, air lead averaged about 2.0 M9/m in Turin (from 0.88 to
3 3
4.54 ug/m depending on location of the sampling site), about 0.56 ug/m in the nearby com-
3 3
munities (0.30 to 0.67 ug/m ) and about 0.30 ug/m in more distant (> 25 km) locations.
Blood lead concentrations and isotope ratios for 35 adult subjects were determined on two
or more occasions during phases 0-2 of the study (see Appendix C). Their blood lead isotope
ratios decreased over time and the fraction of lead in their blood attributable to the
Australian lead-labelled gasoline could be estimated independently of blood lead concentration
(see Appendix C for estimation method). The mean fraction of blood lead attributable to the
Australian lead-labelled gasoline ranged from 23.7 ± 5.4 percent in Turin to 12.5 ±7.1 per-
cent in the nearby (< 25 km) countryside and 11.0 ± 5.8 percent in the remote countryside.
These likely represent minimal estimates of fractions of blood lead derived from gasoline due
to: (1) use of some non-Australian lead-labelled gasoline brought into the study area from
outside; (2) probable insufficient time to have achieved steady-state blood lead isotope
ratios by the time of the switchback; (3) probable insufficient time to fully reflect delayed
movement of the Australian lead from gasoline via environmental pathways in addition to air.
These results can be combined with the actual blood lead concentrations to estimate the
fraction of gasoline uptake attributable or not attributable to direct inhalation. The
results are shown in Table 11-51 (based on a suggestion by Or. Facchetti). From Section
11.4.1, we conclude that an assumed value of 8=1.6 is plausible for predicting the amount of
PB11C/A 11-119 7/29/83
-------
PRELIMINARY DRAFT
I I I I I I I I I I I I I I I I
*) BASED ON A LIMITED NUMBER OF SAMPLES
Pb 206/Pb 207
• ADULTS < 25 km
BLOOD & ADULTS > 25 km
O ADULTS TURIN
D TRAFFIC WARDENS-TURIN
• SCHOOL CHILDREN TURIN
1.20
1.18
1.16
1.14
1.12
1.10
1.08
1.06
AIRBORNE
PARTICULAR
• TURIN
A COUNTRYSIDE
O PETROL
Phase 0
Phase 1
Phase 2
Phase 3
I l_l I I I I I I I I I I I I I I
74
75
76
77
78
79
80
81
Figure 11-24. Change in Pb-206/Pb-207 ratios in petrol, airborne particulate,
and blood from 1974 to 1981.
Source: Facchetti and Geiss (1982).
PB11C/A
11-120
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-51. ESTIMATED CONTRIBUTION OF LEADED GASOLINE TO BLOOD LEAD
BY INHALATION AND NON-INHALATION PATHWAYS
Air Lead Blood Pb Blood PB Non-
Fraction Mean Fraction Mean PB From Inhaled
From Air From Blood From Gaso- Pb From
Gaso? x Lead ,. xGaso? >. Lead , ,,. Gaso/ ^ line ,f^ 63507 x
line(a) Conc.(b)line(c) Conc.(d) line(e) In Air(f) line(g)
(ug/m3) (ug/dl) (ug/di) (ug/di) (ug/di)
Location
Turin 0.873 2.0 0.237 21.77 5.16 2.79 2.37
<25 km 0.587 0.56 0.125 25.06 3.13 0.53 2.60
>25 km 0.587 0.30 0.110 31.78 3.50 0.28 3.22
Estimated
Fraction
Gas- Lead
Inhalar
tion
-------
PRELIMINARY DRAFT
Lead uptake may also be associated with occupation, sex, age, smoking and drinking
habits. The linear exposure model used in Section 11.4 was also used here to estimate the
fraction of labelled blood lead from gasoline attributable to exposure via direct inhalation
and other pathways. EPA used blood lead measurements in Phase 2 for the 35 subjects for whom
repeated measurements allowed estimation of the change in isotope ratios in the blood. Their
blood lead concentrations in Phase 2 were also determined, allowing for estimation of the total
gasoline contribution to blood lead. Possible covariates included sex, age, cigarette
smoking, drinking alcoholic beverages, occupation, residence location and work location. In
order to obtain some crude comparisons with the inhalation exposure studies of Section 11.4.1,
EPA analysis assigned the air lead values listed in Table 11-52 to various locations. Lower
values for air lead in Turin would increase the estimated blood lead inhalation slope above
the estimated value 1.70. Since the fraction of time subjects were exposed to workplace air
was not known, this was also estimated from the data as about 41 percent (i.e., 9.8 hours/day).
The results are shown in Figure 11-25 and Table 11-53. Of all the available variables, only
location, sex and inhaled air lead from gasoline proved statistically significant in predic-
2
ting blood lead attributable to gasoline. The model predictability is fairly good, R = 0.654.
It should be noted that a certain amount of confounding of variables was unavoidable in this
small set of preliminary data, e.g., no female subjects in Turin or in occupations of traffic
wardens, etc. There was a systematic increase in estimated non-inhalation contribution from
gasoline increase for remote areas, but the cause is unknown. Nevertheless, the estimated
non-inhalation contribution of gasoline to blood lead in the ILE study is significant (i.e.
1.8 to 3.4 |jg/dl).
TABLE 11-52. ASSUMED AIR LEAD CONCENTRATIONS FOR MODEL
Residence or workplace code 1-4 5 6
Location outside Turin Turin residential Turin central
Air lead concentration (a) 1.0 pg/m3^ 2.5
(a) Use value for community air lead, 0.16 to 0.67 pg/m .
(b) Intermediate between average traffic areas (1.71 g/m ) and low traffic areas (0.88 g/m )
in Turin.
(c) Intermediate between average traffic areas (1.71 |jg/m3) and heavy traffic areas (4.54
g/m ) in Turin.
PB11C/A 11-122 7/29/83
-------
PRELIMINARY DRAFT
The preliminary linear analysis of the overall ILE data set (2161 observations) found
tnat total blood lead levels depended on other covariates for which there were plausible
mechanisms of lead exposure, including location, smoking, alcoholic beverages, age and occu-
pation (Facchetti and Geiss 1982). The difference between total blood lead uptake and blood
Total contribution of
gasoline lead to
blood lead in
Italian men.
Non-inhalation contribution
of gasoline to blood lead
in Italian men.
Contribution to blood lead
by direct inhalation from
air lead attributable to
gasoline.
>25km <25km
AVERAGE AIR LEAD CONCENTRATION ATTRIBUTABLE TO GASOLINE
Figure 11-25. Estimated direct and indirect contributions of lead in
gasoline to blood lead in Italian men, based on EPA analysis of
ILE data (Table 11-53).
lead uptake attributable to gasoline lead has yet to be analyzed in detail, but these analyses
suggest that certain important differences may be found. Some reservations have been expres-
sed about the ILE study, both by the authors themselves and by Elwood (1983). These include
unusual conditions of meteorology and traffic in Turin, and demographic characteristics of the
35 subjects measured repeatedly that may restrict the general izability of the study. However,
it is clear that changes in air lead attributable to gasoline were tracked by changes in blood
PB11C/A
11-123
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-53. REGRESSION MODEL FOR BLOOD LEAD ATTRIBUTABLE TO GASOLINE
Variable Coefficient ± Standard Error
Air lead from gas 1.70 ± 1.04 ug/dl per ug/m
LOCATION
Turin 1.82 ± 2.01 pg/dl
<25 km 2.56 ± 0.59 ug/dl
>25 km 3.42 ± 0.85 ug/dl
Sex -2.03 ± 0.48 pg/dl for women
lead in Turin residents. The airborne particulate lead isotope ratio quickly achieved new
equilibrium levels as the gasoline isotope ratio was changed, and maintained that level during
the 2k years of Phase 2. The blood lead isotope ratios fell slowly during the changeover
period, and rose again afterwards as shown in Figure 11-24. Equilibrium .was not clearly
achieved for blood lead isotope ratios, possibly due to large endogenous pools of old lead
stored in the skeleton and slowly mobilized over time. Even with such reservations, this
study provides a useful basis for relating blood lead and air lead derived from gasoline com-
bustion.
11.5.1.1.2 United States. Manton (1977) conducted a long term study of 10 subjects whose
blood lead isotopic composition was monitored for comparison with the isotopic composition of
the air they breathed. Manton had observed that the ratio of 266Pb/264Pb in the air varied
with seasons in Dallas, Texas; therefore, the ratio of those isotopes should vary in the
blood. By comparing the observed variability, estimates could then be made of the amount of
lead in air that is absorbed by the blood.
Manton took monthly blood samples from all 10 subjects from April 1974 until June 1975.
The blood samples were analyzed for both total lead and isotopic composition. The recruited
volunteers included a mix of males and females, and persons highly and moderately exposed to
lead. However, none of the subjects was thought to be exposed to more than 1 ug/m of lead in
air. Lead in air samples was collected by Hi-Vol samplers primarily from one site in Dallas.
That site, however, had been shown earlier to vary in isotopic composition paralleling another
site some 16 miles away. All analyses were carried out under clean conditions with care and
caution being exercised to avoid lead contamination.
The isotope ratio of lead 206Pb/204Pb increased linearly with time from about 18.45 to
19.35, approximately a 6 percent increase. At least one of the two isotopic lead ratios in-
creased linearly in 4 of the 10 subjects. In one other, they increased but erratically. In
PB11C/A 11-124 7/29/83
-------
PRELIMINARY DRAFT
the remainder of the subjects, the isotopic ratios followed smooth curves showing inflection
points. The curves obtained for the two subjects born in South Africa were 6 months out of
phase with the curves of the native-born Americans. The fact that the isotope ratios in 9 of
the 10 subjects varied regularly was thought to indicate that the non-airborne sources of lead
varied in isotopic composition very slowly.
The blood lead levels exhibited a variety of patterns, although none of the subjects
showed more than a 25 percent change from initial levels. This suggests a reasonably steady
state external environment.
Manton carried his analyses further to estimate the percentage of lead in blood that
comes from air. He estimated that the percentage varied from 7 to 41 percent, assuming that
dietary sources of lead had a constant isotopic ratio while air varied. He calculated the
percent contribution according to the following equation:
—9— = — where
100+q a
b = rate of change of an isotope ratio in blood,
a = rate of change of the same ratio in the air,
q = constant - the number of atoms of the isotope in the denominator
of the airborne lead ratio mixed with 100 atoms of the same iso-
tope of lead from non-airborne sources.
The results are shown in Table 11-54. Slopes were obtained by least squares regression.
Percentages of airborne lead in blood varied between 7±3 and 41±3.
TABLE 11-54. RATE OF CHANGE OF 206Pb/2(MPb AND 206Pb/207Pb
IN AIR AND BLOOD, AND PERCENTAGE OF AIRBORNE
LEAD IN BLOOD OF SUBJECTS 1, 3, 5, 6 AND 9
Subject
(Air)
1
3
5
6
9*
Rate of Change per Day
206pb/204Pb 206pb/207pb
X 10~4 X 10"5
17.60 ± 0.77 9.97 ± 0.42
... 0.70 ± 0.30
5.52 ± 0.55 ...
... 3.13 ± 0.34
6.53 ± 0.49 4.10 ± 0.25
3.25 2.01
Percentage of Airborne Lead in Blood
From 206Pb/204Pb
31.4 ±3.4
37.1 ± 2.8
18.5
From 206Pb/207pb
7 ± 3
31.4+3.7
41.1 ± 3.0
20.0
Note: Errors quoted are one standard deviation
*From slope of tangent drawn to the minima of subject's blood curves. Errors
cannot realistically be assigned.
PB11C/A 11-125 7/29/83
-------
PRELIMINARY DRAFT
Stephens (1981) has extended the analysis of data in Manton's study (Table 11-55). He
used the observed air lead concentrations based on actual 24-hour air lead exposures in three
adults. He assumed values for breathing rate, lung deposition and absorption into blood to
estimate the blood lead uptake attributable to 2°4Pb by the direct inhalation pathway. Sub-
jects 5, 6 and 9 absorbed far more air lead in fact than was calculated using the values in
Table 11-54. The total air lead contribution was 8.4, 4.4 and 7.9 times larger than the
direct inhalation. These estimates are sensitive to the assumed parameter values.
In summary, the direct inhalation pathway accounts for only a fraction of the total air
lead contribution to blood, the direct inhalation contribution being on the order of 12 to 23
percent of the total uptake of lead attributable to gasoline, using Stephen's assumptions.
This is consistent with estimates (i.e. 8 to 54 percent) from the ILE study, taking into
account the much higher air lead levels in Turin.
11.5.1.2 Studies of Childhood Blood Lead Poisoning Control Programs. Billick et al. (1979)
presented several possible explanations for the observed decline in blood lead levels in New
York City children as well as evidence supporting and refuting each. The suggested contribu-
ting factors include the active educational and screening program of the New York City Bureau
of Lead Poisoning Control, and the decrease in the amount of lead-based paint exposure as a
result of rehabilitation or removal of older housing or changes in environmental lead exposure.
Information was only available to partially evaluate the last source of lead exposure and
particularly only for ambient air lead levels. Air lead measurements were available during
the entire study period for only one station which was located on the west side of Manhattan
at a height of 56 m. Superposition of the air lead and blood lead levels indicated a
similarity in cycle and decline. The authors cautioned against overinterpretation by assuming
that one air monitoring site was representative of the air lead exposure of New York City
residents. With this in mind, the investigators fitted a multiple regression model to the
data to try to define the important determinants of blood lead levels for this population.
Age, ethnic group and air lead level were all found to be significant determinants of blood
lead levels. The authors further point out the possibility of a change in the nature of the
population being screened before and after 1973. They reran this regression analysis sepa-
rately for years both before and after 1973. The same results were still obtained, although
the exact coefficients varied.
Billick et al. (1980) extended their previous analysis of the data from the single moni-
toring site mentioned earlier. The investigators examined the possible relationship between
blood lead level and the amount of lead in gasoline used in the area. Figures 11-26 and 11-27
present illustrative trend lines in blood leads for blacks and Hispanics, vs. air lead and
PB11C/A 11-126 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-55. CALCULATED BLOOD LEAD UPTAKE FROM AIR LEAD
USING MANTON ISOTOPE STUDY
Blood Uptake from Air
Sub-
ject
5
6
9
Concen-
tration
0.22 Mg/m3
3
1.09 Mg/ra
3
0.45 (jg/m
Expo-
sure*
15 m3/day
15 m /day
15 m3/day
Deposi-
tion
37%
37%
37%
Absorp-
tion
50%
50%
50%
Calcu-
lated
Inhala-
tion
0.61 ug/d
3.0 pg/d
1.2 pg/d
Observed
5.1 MQ/d
13.2 pg/d
9.9 ug/d
Fraction of lead
uptake from gasoline
by direct inhalation
0.120
0.229
0.126
^assumed rather than measured exposure, deposition and absorption.
Source: Stephens, 1981, based on Wanton, 1977; Table III.
gasoline lead, respectively. Several different measures of gasoline lead were tried: mid-
Atlantic Coast (NY, NJ, Conn), New York, New York plus New Jersey and New York plus
Connecticut. The lead in gasoline trend line appears to fit the blood lead trend line better
than the air lead trend, especially in the summer of 1973.
Multiple regression analyses were calculated using six separate models. The best fitting
2
model had an R = 0.745. Gasoline lead content was included rather than air lead. The gaso-
line lead content coefficient was significant for all three racial groups. The authors state
a number of reasons for gasoline lead providing a better fit than air lead, including the fact
that the single monitoring site might not be representative.
Nathanson and Nudelman (1980) provide more detail regarding air lead levels in New York
City. In 1971, New York City began to regulate the lead content of gasoline sold. Lead in
gasoline was to be totally banned by 1974, but supply and distribution problems delayed the
effect of the ban. Ultimately regulation of lead in gasoline was taken over by the U.S.
Environmental Protection Agency.
New York City measured air lead levels during the periods June 1969 to September 1973 and
during 1978 at multiple sites. The earlier monitoring was done by 40 rooftop samples using
cel.lulose filters analyzed by AAS. The latter sampling was done by 27 rooftop samplers using
glass fiber filters analyzed by X-ray fluorescence (XRF). There was excellent agreement
between the XRF and atomic absorption analyses for lead (r = 0.985). Furthermore, the XRF
analyses were checked against EPA AAS and again excellent agreement was found. The authors
did, however, point out that cellulose filters are not as efficient as glass fiber filters.
Therefore, the earlier results tend to be underestimates of air lead levels.
PB11C/A 11-127 7/29/83
-------
PRELIMINARY DRAFT
8
7»
ul 30
§
O
^ 25
O
O
3
m
i 20
O
£
| 15
O
UJ
O
10
I I I | I I I | II I | I Tip 1 I p I I | I II
BLACK
_ __ HISPANIC
_ ... AIR LEAD
r\
• • t \ • • s* /•
\\ \\ - \ /\ A
- V V v/ x/.^-
I I I t I I I I I M I I I I I I I I I I I I I I I I
m
2.0 >
30
1.5
1.0
0.0
g
5
m
1970 1971 1972 1973 1974 197S 1976
QUARTERLY SAMPLING DATE
Figure 11-26. Geometric mean blood lead levels of New York City
children (aged 25-36 months) by ethnic group, and ambient air lead
concentration versus quarterly sampling period, 1970-1976.
Source: Billick (1980).
PB11C/A
11-128
7/29/83
-------
PRELIMINARY DRAFT
E
8
35
30
"J 25
Q
o
o
o
£
fc
i
(9
20
15
10
I I j TTI J ill I I I 1 I I I
__— - BLACK
— — — HISPANIC
— ' — GASOLINE LEAD
I I I
f \ V \ / . x * ' \ ^—'
y \ ' /v v' V /\
7 \ *. /\ / \ A Ns' _
v V x/ \ / \
v \ I \
6.0
5.0
m
4.0 ff
3.0
01 I I 1 I I I I I I I I I 1 I I I I I I I I I I I I I I 11
S
1
0.0
1970 1971 1972 1973 1974 1975 1976
QUARTERLY SAMPLING DATE
Figure 11-27. Geometric mean blood lead levels of New York City
children (aged 25-36 months) by ethnic group, and estimated
amount of lead present in gasoline sold in New York, New Jersey.
and Connecticut versus quarterly sampling period, 1970-1976.
Source: Bitlick (1980).
PB11C/A
11-129
7/29/83
-------
PRELIMINARY DRAFT
Quarterly citywide air lead averages generally declined during the years 1969-1978. The
3
maximum quarterly citywide average obtained was about 2.5 ug/m for the third quarter of 1970.
The citywide trend corresponds to the results obtained from the single monitoring site used in
Billick et al.'s analysis. The citywide data suggest that the single monitoring site in Man-
hattan is a responsible indicator of air lead level trends. The graph in Figure 11-28 rein-
forces this assertion by displaying the geometric mean blood lead levels for blacks and
Hispanics in the 25 to 36-month age groups and the quarterly citywide air lead levels for the
periods of interest. A good correspondence was noted.
As part of a detailed investigation of the relationship of blood lead levels and lead in
gasoline covering three cities, Billick (1982) extended the time trend analyses of New York
City blood lead data. Figure 11-29 presents the time trend line for geometric mean blood
leads for blacks age 24-35 months extended to 1979. The downward trend noted earlier was
still continuing, although the slopes for both the blood and gasoline lead seem to be somewhat
shallower toward the most recent data. A similar picture is presented by the percent of chil-
dren with blood lead levels greater than 30 ug/dl. In the early 70's, about 60 percent of the
screened children had these levels; by 1979 the percent had dropped between 10 and 15 percent.
11.5.1.3 NHANES II. Blood lead data from the second National Health and Nutrition Examina-
tion Survey has been described in sections 11.3.3.1 and 11.3.4.4. The report by Annest et al.
(1983) found highly significant associations between amounts of lead used in gasoline produc-
tion in the U.S. and blood lead levels. The associations persisted after adjusting for race,
sex, age, region of the country, season, income and degree of urbanization.
Various analyses of the relationship between blood lead values in the NHANES II sample
and estimated gasoline lead usage were also reviewed by an expert panel (see Appendix 11-0).
They concluded that the correlation between gasoline lead usage and blood lead levels was con-
sistent with the hypothesis that gasoline lead is an important causal factor, but the analyses
did not actually confirm the hypothesis.
11.5.1.4 Frankfurt. West Germany. Sinn (1980; 1981) conducted a study specifically examining
the environmental and biological impact of the gasoline lead phasedown implemented in West
Germany on January I, 1976. Frankfurt am Main provided a good setting for such a study
because of its physical character.
Air and dustfall lead levels at several sites in and about the city were determined be-
fore and after the phasedown was implemented. The mean air lead concentrations obtained
during the study are presented in Table 11-56. A substantial decrease in air lead levels was
noted for the low level high traffic site (3.18 ug/m3 in 1975-76 to 0.68 ug/m3 in 1973-79).
No change was noted for the background site while only minor changes were observed for the
other locations. Dustfall levels fell markedly (218 mg/cm2-day for 1972-73 to 128 mg/cm2•day
PBUC/A 11-130 7/29/83
-------
PRELIMINARY DRAFT
I I I I I I I I II I I 1 M ] M I [ I T I I
IT
E
S
I
UJ
§
O
§
O
8
I
ui
c
t
i
O
35 —
BLACK
— — — - HISPANIC
— . —. AIR LEAD
. I I I 1 I | I i I I I I I I I I I I I I I I I I I I I
1970 1971 1972 1973 1974 1975 1976
2.5
2.0
1.5
1.0
nn
O
a
i~
m
a
n
m
D
§
m
r
QUARTERLY SAMPLING DATE
Figure 11-28. Geometric mean blood levels for blacks and
Hispanics in the 25-to-36-month age group and rooftop
quarterly averages for ambient citywide lead levels.
Source: Nathanson and Nudelman (1980).
PB11C/A
11-131
7/29/83
-------
PRELIMINARY DRAFT
E
8
D
O
O
E
50
40
30
20
10
*
oo
I II I I I I I I
GEO. MEAN BLOOD Pb
— — — — GAS LEAD
TRISTATE X 4
— SMSAX20
66 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81
YEAR
Figure 11-29. Time dependence of blood lead and gas lead for blacks, aged 24 to
35 months, in New York.
So -co: Billick (1982).
Source: Billick (1982).
PB11C/A
11-132
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-56. MEAN AIR LEAD CONCENTRATIONS DURING THE VARIOUS BLOOD SAMPLING
PERIODS AT THE MEASUREMENT SITES DESCRIBED IN THE TEXT (ug/m3)
1975-1976
1976-1977
1977-1978
1978-1979
Residential
Low Traffic
0.57
0.39
0.32
0.39
High Traffic
(>20m)
0.59
0.38
0.31
0.31
High Traffic
(3m)
3.18
1.04
0.66
0.68
Background
Site
0.12
0.09
0.10
0.12
Source: Sinn (1980, 1981).
for 1977-78). Traffic counts were essentially unchanged in the area during the course of
study.
A number of population groups were included in the study of the blood lead levels; they
were selected for having either occupational or residential exposure to high density automo-
bile traffic. Blood samples were taken serially throughout the study (three phases in
December-January 1975-76, December-January 1976-77 and December-January 1977-78). Blood
samples were collected by venipuncture and analyzed by three different laboratories. All the
labs used AAS although sample preparation procedures varied. A quality control program across
the laboratories was conducted. Due to differences in laboratory analyses, attrition and loss
of sample, the number of subjects who could be examined throughout the study was considerably
reduced from the initial number recruited (124 out of 300).
Preliminary analyses indicated that the various categories of subjects had different
blood lead levels, and that males and females within the same category differed. A very com-
plicated series of analyses then ensued that made it difficult to draw conclusions because the
various years' results were displayed separately by each laboratory performing the chemical
analysis and by different groupings by sex and category. In Sinn's later report (1981) a
downward trend was shown to exist for males and females who were in all years of the study and
whose blood levels were analyzed by the same laboratory.
11.5.2 Primary Smelters Populations
Most studies of nonindustry-employed populations living in the vicinity of industrial
sources of lead pollution were triggered because evidence of severe health impairment had been
found. Subsequently, extremely high exposures and high blood lead concentrations were found.
The following studies document the excessive lead exposure that developed, as well as some of
the relationships between environmental exposure and human response.
PB11C/A
11-133
7/29/83
-------
PRELIMINARY DRAFT
11.5.2.1 El Paso, Texas. In 1972, the Centers for Disease Control studied the relationships
between blood lead levels and environmental factors in the vicinity of a primary smelter lo-
cated in El Paso, Texas emitting lead, copper and zinc. The smelter had been in operation
since the late 1800's (Landrigan et al., 1975; U.S. Centers for Disease Control, 1973). Daily
Hi-Vol samples collected on 86 days between February and June 1972 averaged 6.6 ug/m . These
air lead levels fell off rapidly with distance, reaching background values approximately 5 km
from the smelter. Levels were higher downwind, however. High concentrations of lead in soil
and house dusts were found, with the highest levels occurring near the smelter. The geometric
means of 82 soil and 106 dust samples from the sector closest to the smelter were 1791 and
4022 ug/g, respectively. Geometric means of both soil and dust lead levels near the smelter
were significantly higher than those in study sectors 2 or 3 km farther away.
Sixty-nine percent of children 1- to 4-years old living near the smelter had blood lead
levels greater than 40 |jg/dl, and 14 percent had blood lead levels that exceeded 60 pg/dl.
Concentrations in older individuals were lower; nevertheless, 45 percent of the children 5- to
9-years old, 31 percent of the individuals 10- to 19-years old and 16 percent of the in-
dividuals above 19 had blood lead levels exceeding 40 pg/dl. The data presented preclude cal-
culations of means and standard deviations.
Data for people aged 1 to 19 years of age living near the smelter showed a relationship
between blood lead levels and concentrations of lead in soil and dust. For individuals with
blood lead levels greater than 40 pg/dl, the geometric mean concentration of lead in soil at
their homes was 2587 pg/g, whereas for those with a blood lead concentration less than 40
ug/dl, home soils had a geometric mean of 1419 pg/g. For house dust, the respective geometric
means were 6447 and 2067 pg/g. Length of residence was important only in the sector nearest
the smelter.
Additional sources of lead were also investigated. A relationship was found between
blood lead concentrations and lead release from pottery, but the number of individuals exposed
to lead-glazed pottery was very small. No relationships were found between blood lead levels
and hours spent out of doors each day, school attendance, or employment of a parent at the
smelter. The reported prevalence of pica also was minimal.
Data on dietary intake of lead were not obtained because there was no food available from
sources near the smelter since the climate and proximity to the smelter prevented any farming
in the area. It was unlikely that the dietary lead intakes of the children from near the
smelter or farther away were significantly different. It was concluded that the primary
factor associated with elevated blood lead levels in the children was ingestion or inhalation
of dust containing lead.
Morse et al. (1979) conducted a follow-up investigation of the El Paso smelter to deter-
mine whether the environmental controls instituted following the 1972 study had reduced the
PBUC/A 11-134 7/29/83
-------
PRELIMINARY DRAFT
lead problem described. In November 1977, all children 1- to 18-years old living within
1.6 km of the smelter on the U.S. side of the border were surveyed. Questionnaires were ad-
ministered to the parents of each participant to gather background data.
Venous blood samples were drawn and analyzed for lead by modified Delves cup spectropho-
tometry. House dust and surface soil samples, as well as sample pottery items were taken from
each participant's residence. Dust and soil samples were analyzed for lead by AAS. Pottery
lead determinations were made by the extraction technique of Klein. Paint, food, and water
specimens were not collected because the earlier investigations of the problem had demon-
strated these media contributed little to the lead problem in El Paso.
Fifty-five of 67 families with children (82 percent) agreed to participate in the study.
There were 142 children examined in these homes. The homes were then divided into two groups.
Three children lived in homes within 0.8 km of the smelter. Their mean blood lead level in
1977 was 17.7 ug/dl. By contrast, the mean blood lead level of 160 children who lived within
0.8 km of the smelter in 1972 had been 41.4 ug/dl. In 1977, 137 children lived in homes lo-
cated 0.8 to 1.6 km from the smelter. Their mean blood lead level was 20.2 ug/dl. The mean
blood level of 96 children who lived in that same area in 1972 had been 31.2 ug/dl.
Environmental samples showed a similar improvement. Dust lead fell from 22,191 ug/g to
1,479 ug/g while soil lead fell from 1,791 ug/g to 427 ug/g closest to the smelter. The mean
air lead concentration at 0.4 km from the smelter decreased from 10.0 to 5.5 pg/m and at 4.0
3
km from 2.1 to 1.7 ug/m . Pottery was not found to be a problem.
11.5.2.2 CDC-EPA Study. Baker et al. (1977b), in 1975, surveyed 1774 children 1 to 5 years
old, most of whom lived within 4 miles of lead, copper or zinc smelters located in various
parts of the United States. Blood lead levels were modestly elevated near 2 of the 11 copper
and 2 of the 5 zinc smelters. Although blood lead levels in children were not elevated in the
vicinity of three lead smelters, their FEP levels were somewhat higher than those found in
controls. Increased levels of lead and cadmium in hair samples were found near lead and zinc
smelters; this was considered evidence of external exposure. No environmental determinations
were made for this study.
11.5.2.3 Meza Valley, Yugoslavia. A series of Yugoslavian studies investigated exposures to
lead from a mine and a smelter in the Meza Valley over a period of years (Fugas et al., 1973;
Graovac-Leposavic et al. 1973; Milic et al., 1973; Djuric et al., 1971, 1972). In 1967,
24-hour lead concentrations measured on 4 different days varied from 13 to 84 ug/m in the
3
village nearest the smelter, and concentrations of up to 60 ug/m were found as far as 5 km
from the source. Mean particle size in 1968 was less than 0.8 urn. Analysis of some common
foodstuffs showed concentrations that were 10 to 100 times higher than corresponding food-
stuffs from the least exposed area (Mezica) (Djuric et al., 1971). After January 1969, when
PB11C/A 11-135 7/29/83
-------
PRELIMINARY DRAFT
partial control of emissions was established at the smelter, weighted average weekly exposure
was calculated to be 27 ug/m in the village near the smelter. In contrast to this, the city
of Zagreb (Fugas et al., 1973), which has no large stationary source of lead, had an average
2
weekly air lead level of 1.1 (jg/m .
In 1968, the average concentration of ALA in urine samples from 912 inhabitants of 6 vil-
lages varied by village from 9.8 to 13 mg/1. A control group had a mean ALA of 5.2 mg/1.
Data on lead in blood and the age and sex distribution of the villagers were not given (Djuric
et al., 1971).
Of the 912 examined, 559 had an ALA level greater than 10 mg/1 of urine. In 1969, a more
extensive study of 286 individuals with ALA greater than 10 mg/1 was undertaken (Graovac-
Leposavic et al. 1973). ALA-U increased significantly from the previous year. When the
published data were examined closely, there appeared to be some discrepancies in inter-
pretation. The exposure from dust and from food might have been affected by the control de-
vices, but no data were collected to establish this. In one village, Zerjua, ALA-U dropped
from 21.7 to 9.4 mg/1 in children 2 to 7 years of age. Corresponding ALA-U values for 8- to
15-year-olds and for adult men and women were reduced from 18.7 to 12.1, from 23.9 to 9.9 and
from 18.5 to 9.0 mg/1, respectively. Because lead concentrations in air (Fugas et al . , 1973),
even after 1969, indicated an average exposure of 25 vg/m , it is possible that some other
explanation should be sought. The author indicated in the report that the decrease 1n ALA-U
showed "the dependence on meteorologic, topographic, and technological factors" (Graovac-
Leposavic et al. , 1973).
Fugas (1977) in a later report estimated the time-weighted average exposure of several
populations studied during the course of this project. Stationary samplers as well as
personal monitors were used to estimate the exposure to airborne lead for various parts of the
day. These values were then coupled with estimated proportions of time at which these expo-
sure held. In Table 11-57, the estimated time-weighted air lead values as well as the ob-
served mean blood lead levels for these studied populations are presented. An increase in
blood lead values occurs with increasing air lead exposure.
11.5.2.4 Kosovo Province. Yugoslavia. Residents living in the vicinity of the Kosovo smelter
were found to have elevated blood lead levels (Popovac et al., 1982). In this area of
Yugoslavia, five air monitoring stations had been measuring air lead levels since 1973. Mean
air lead varied from 7.8 to 21.7 ug/m in 1973; by 1980 the air lead averages ranged from 21.3
to 29.2 ug/ro • In 1978 a Pilot study suggested that there was a significant incidence of ele-
vated blood lead levels in children of the area. Two major surveys were then undertaken.
In August 1978 letters were sent to randomly selected families from the business commu-
nity, hospitals or lead-related industries in the area. All family members were asked to come
PB11C/A 11-136 7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-57. MEAN BLOOD LEAD LEVELS IN SELECTED YUGOSLAVIAN
POPULATIONS, BY ESTIMATED WEEKLY TIME-WEIGHTED AIR LEAD EXPOSURE
Population
Rural I
Rural II
Rural III
Postmen
Customs officers
Street car drivers
Traffic policemen
N
49
47
45
44
75
43
24
Time-weighted,
air lead, pg/m
0.079
0.094
0.146
1.6
1.8
2.1
3.0
Blood lead
ug/di
7.9
11.4
10.5
18.3
10.4
24.3
12.2
level ,
SD
4.4
4.8
4.0
9.3
3.3
10.5
5.1
Source: Fugas, 1977.
to a hospital for primary screening by erythrocyte protoporphyrin. A central population of
comparable socioeconomic and dietary background was collected from a town without lead emis-
sions. Blood levels were determined primarily for persons with greater than pg/g Hgb. EP was
measured by a hematof1uorimeter, while blood lead was determined by the method of Fernandez
using atomic absorption with graphite furnace and background correction.
Mean EP values were higher in the 1978 survey for exposed residents compared to controls
in the average age group. EP values seemed to decline with age. Similar differences were
noted for blood lead levels. The observed mean blood leads, ranging from 27.6 in the greater
than 15-year age group to 50.9 ug/dl in the 5- to 10-year group, suggest substantial lead ex-
posure of these residents. In the control group the highest blood lead level was 19 ug/dl.
In December 1980 a second survey was conducted to obtain a more representative sample of
persons residing in the area. Letters were sent again, and 379 persons responded. EP levels
were higher in all ages in 1980 vs. 1978, although the differences were not statistically sig-
nificant. The air lead levels Increased from 14.3 ug/m in 1978 to 23.8 ug/m in 1980.
Comparing the 1980 blood lead results with the 1978 control group shows that the 1980
levels were higher in each age group. Males older than 15 years had higher mean blood lead
levels than the females (39.3 vs. 32.4 \ig/
-------
PRELIMINARY DRAFT
aged 8 to 11. The control population was 25 nursery school children aged 3 to 6 and 64 pri-
mary school children aged 8 to 11. Since the smelter had installed filters 8 years before the
study, the older children living in the smelter area had a much higher lifetime exposure.
Blood lead analysis was performed on venous samples using anodic stripping voltammetry by
Morrell's method. Precision was checked over the range 10 to 100 ug/dl. Reported reproduci-
bility was also good. All samples were subsequently reanalyzed by AAS using graphite furnace
and background correction by the method of Volosen. The average values obtained by the
second method were quite similar to those of the first (average difference 1.4 ug/dl; cor-
relation coefficient, 0.962).
Air was sampled for lead for 1 month at three sampling sites. The sites were located at
150 m, 300 m and 4 km from the wall of the lead smelter. The average air lead levels were
2.32, 3.43 and 0.56 ug/m , respectively.
A striking difference in blood lead levels of the exposed and control populations was ob-
served; levels in the exposed population were almost twice that in the control population.
There was no significant difference between nursery school and primary school children. The
geometric mean for nursery school children was 15.9 and 8.2 for exposed and control, respecti-
vely. For primary school it was 16.1 and 7.0 ug/dl. In the exposed area 23 percent of the
subjects had blood lead levels between 21 and 30 ug/dl and 3 percent greater than 31 ug/dl.
No control children had PbB greater than 20 ug/dl. The air leads were between 2 to 3 ug/m in
the exposed and 0.56 ug/m in the control cases.
11.5.3 Battery Plants
Studies of the effects of storage battery plants have been reported from France and Italy
(Dequidt et al., 1971; De Rosa and Qobbato, 1970). The French study found that children from
an industrialized area containing such a plant excreted more ALA than those living in a diffe-
rent area (Dequidt et al., 1971). Increased urinary excretion of lead and coproporphyrins was
found in children living up to 100 m from a battery plant in Italy (De Rosa and Gobbato,
1970). Neither study gave data on plant emissions or lead in air.
Zielhius et al. (1979) studied children living in the vicinity of the Arnhem secondary
lead smelter. In 1976 they recruited children to serve as subjects and controls. The chil-
dren chosen were 2 and 3 years old. Parents were asked to complete a questionnaire for back-
ground information. Two ml venous samples were collected from 17 children living less than 1
km, from 54 children living 1 to 2 km, and from 37 children living greater than 2 km from the
smelter (control group). Blood samples were analyzed for lead by graphite furnace AAS and for
FEP by the method of Piomelli. Air measurements for lead were made in autumn 1976. Samples
were established about 2 km northeast and about 0.4 km north of the plant. Air lead levels
3 3
ranged from 0.8 to 21.6 ug/m northeast and from 0.5 to 2.5 ug/m north of the plant.
PB11C/A 11-138 7/29/83
-------
PRELIMINARY DRAFT
Blood leads were statistically significantly higher closer to the smelter. For all chil-
dren the mean blood lead level was 19.7 pg/dl for the less than 1 km and 11.8 ug/dl for the
controls (>2 km). Similarly, FEP levels were higher for the closer (41.9 pg/100 ml RBC)
children as opposed to the control (32.5 ng/100 ml RBC). Higher blood levels were associated
with lower socioeconomic status.
Further investigation of this smelter was undertaken by Brunekreef et al. (1981) and
Diemel et al. (1981). In May 1978 venipuncture blood samples were collected from 95 one- to
three-year old children living within 1 km of the smelter. Blood leads were determined by
graphite AAS.
Before the blood sampling, an environmental sampling program was conducted. The samples
collected are listed in Table 11-58. Questionnaires were administered to collect background
and further exposure information. A subset of 39 children was closely observed for 1 or 2
days for mouthing behavior. Table 11-58 also presents the overall results of the environ-
mental sampling. As can be readily seen, there is a low exposure to airborne lead (G.M.
3 3
0.41 \ig/m with a range of 0.28 to 0.52 pg/m ). Soil exposure was moderate, although high.
Interior dust was high in lead, geometric mean of 967 pg/g with a maximum of 4741 ug/g. In a
few homes, high paint lead levels were found. Diemel et al. (1981) extended the analysis of
the environmental samples. They found that indoor pollution was lower than outside. In
Arnhem, it was found that lead is carried into the homes in particulate form by sticking to
shoes. Most of the lead originated from soil from gardens and street dust.
Simple correlation coefficients, were calculated to investigate the relationship between
log blood lead and the independent variables. Significantly, correlations were found with
quantity of house dust, quantity of deposited lead indoors, observational score of dustiness,
age of child and the average number of times an object is put in the mouth. Multiple regre-
ssion analyses were calculated on four separate subpopulations. Among children living in
houses with gardens, the combination of soil lead level and educational level of the parents
explained 23 percent of the variations of blood lead. In children without gardens, the amount
of deposited lead indoors explained 26 percent of the variance. The authors found that an
increase in soil lead level from 100 to 600 pg/g results in an increase in blood lead of
63 ug/dl.
PBUC/A 11-139 7/29/83
-------
TABLE 11-58. ENVIRONMENTAL PARAMETERS AND METHODS: ARNHEM LEAD STUDY, 1978
Parameter
Method
Geometric Mean
Range
I
I—1
o
1. Lead in,ambient air
(M9/nO
2. Lead irudustfall
(ug/m -day)
3. Lead in soil
(pg/g)
4. Lead in street dust
(M9/g)
5. Lead in-indoor air
6. Lead in dustfall
indoors (ug/m -day)
7. Lead in floor dust
8. Easily available
lead indoors
9. Lead in tapwater
10. Dustiness of homes
High volume samples; 24-hr measurements
at 6 sites, continuously for 2 months
Standard deposit gauges; 7-day measurements
at 22 sites, semicontinuously for 3 months
Sampling in gardens of study populations;
analysis of layers from 0 to 5 cm and
5 to 20 cm
Samples at 31 sites, analysis of fraction
<0.3mm
Low volume samples; 1-month measurements
in homes of study population, continuously
for 2 months
Greased glass plates of 30 x 40 cm; 1-month
measurements in homes of study population,
continuously for 3 months
Vacuum cleaner with special filter
holder; 5 samples, collected on 3 different
occasions; with intervals of approximately
1 month, in homes of study populations
Wet tissues, 1 sample in homes of study
population
Proportional samples, during 1 week in
homes of study population
Visual estimation, on a simple scale ranging
from 1 (clean) to 3 (dusty); 6 observations
in homes of study population
0.41
467
240
690
0.26
7.34
fine 957
course 282
0.28-0.52
108-2210
21-1126
77-2667
0.13-0.74
1.36-42.35
463-4741
117-5250
z
>
•<
85% of samples <20 ug Pb/tissue
5.0 (arthimetic)
mean
<0.5-90.0
All lead analyses were performed by atomic absorption spectrophotometry, except part of the tapwater analysis,
which was performed by anodic stripping voltametry. Lead in tapwater analyzed by the National Institute of
Drinking Water Supply in Leidscherdam. Soil and street dust analyzed by the Laboratory of Soil and Plant
Research in Oosterbeek. (Zielhuis, et. al., 1979; Diemel, et. al., 1981)
-------
PRELIMINARY DRAFT
11.5.4 Secondary Smelters
In a Dallas, Texas, study of two secondary lead smelters, the average blood lead levels
of exposed children was found to be 30 ug/dl vs. an average of 22 ug/dl in control children
(Johanson and Luby, 1972). For the two study populations, the air and soil lead levels were
3.5 and 1.5 ug/m and 727 and 255 ug/g, respectively.
In Toronto, Canada the effects of two secondary lead smelters on the blood and hair lead
levels of nearby residents have been extensively studied (Ontario Ministry of the Environment,
1975; Roberts et al . , 1974). In a preliminary report, Roberts et al. (1974) stated that blood
and hair lead levels were higher in children living near the two smelters than in children
living in an urban control area. Biologic and environmental lead levels were reported to de-
crease with increasing distance from the base of the smelter stacks.
A later and more detailed report identified a high rate of lead fallout around the two
secondary smelters (Ontario Ministry of the Environment, 1975). Two groups of children living
within 300 iS of each of the smelters had geometric mean blood lead levels of 27 and 28 M9/dl »
respectively; the geometric mean for 1231 controls was 17 ug/dl. Twenty-eight percent of the
sample children tested near one smelter during the summer and 13 percent of the sample chil-
dren tested near the second smelter during the winter had blood lead levels greater than 40
ug/dl. Only 1 percent of the controls had blood lead levels greater than 40 ug/dl. For chil-
dren, blood lead concentrations increased with proximity to both smelters, but this trend did
not hold for adults, generally. The report concluded that soil lead levels were the main de-
terminant of blood lead levels; this conclusion was disputed by Horn (1976).
Blood lead levels in 293 Finnish individuals, aged 15 to 80, were significantly cor-
related with distance of habitation from a secondary lead smelter (Nordman et al . , 1973). The
geometric mean blood lead concentration for 121 males was 18.1 ug/dl ; for 172 females, it was
14.3 ng/dl . In 59 subjects who spent their entire day at home, a positive correlation was
found between blood lead and distance from the smelter up to 5 km. Only one of these 59 in-
dividuals had a blood lead greater than 40 ug/dl , and none exceeded 50 ug/dl.
11.5.5 Secondary Exposure of Children
Excessive intake and absorption of lead on the part of children can result when parents
who work in a dusty environment with a high lead content bring dust home on their clothes,
shoes or even their automobiles. Once they are home, their children are exposed to the dust.
Landrigan et al. (1976) reported that the 174 children of smelter workers who lived with-
in 24 km of the smelter had significantly higher blood lead levels, a mean of 55.1 ug/dl , than
the 511 children of persons in other occupations who lived in the same areas whose mean
PB11C/A 11-141 7/29/83
-------
PRELIMINARY DRAFT
blood lead levels were 43.7 ug/dl. Analyses by EPA of the data collected in Idaho showed that
employment of the father at a lead smelter, at a zinc smelter, or in a lead mine resulted in
higher blood lead levels in the children living in the same house as opposed to those children
whose fathers were employed in different locations (Table 11-59). The effect associated with
parental employment appears to be much more prominent in the most contaminated study areas
nearest to the smelter. This may be the effect of an intervening socioeconomic variable: the
lowest paid workers, employed in the highest exposure areas within the industry, might be ex-
pected to live in the most undesirable locations, closest to the smelter.
TABLE 11-59. GEOMETRIC MEAN BLOOD LEAD LEVELS FOR CHILDREN
BASED ON REPORTED OCCUPATION OF FATHER, HISTORY
OF PICA, AND DISTANCE OF RESIDENCE FROM SMELTER
Area
1
2
3
.4
5
6
Distance
from
smelter, km
1.6
1.6 to 4.0
4.0 to 10.0
10.0 to 24.0
24.0 to 32.0
75
Lead
smelter
worker
No
Pica Pica
78.7 74.2
50.2 52.2
33.5 33.3
30.3
24.5
-
Lead/zinc mine
worker
Pica
75.3
46.9
36.7
38.0
31.8
-
No
Pica
63.9
46.9
33.5
32.5
27.4
-
Zinc smelter
worker
Pica
69.7
62.7
36.0
40.9
-
-
No
Pica
59.1
50.3
29.6
36.9
-
-
Other
occupations
No
Pica Pica
70.8
37.2
33.3
-
28.0
17.3
59.9
46.3
32.6
39.4
26.4
21.4
Source: Landrigan et al. 1976.
Landrigan et al. (1976) also reported a positive history of pica for 192 of the 919 chil-
dren studied in Idaho. This history was obtained by physician and nurse interviews of
parents. Pica was most common among 2-year old children and only 13 percent of those with
pica were above age 6. Higher blood lead levels were observed in children with pica than in
those without pica. Table 11-59 shows the mean blood lead levels in children as they were af-
fected by pica, occupation of the father and distance of residence from the smelter. Among
the populations living nearest to the smelter environmental exposure appears to be sufficient
at times to more than overshadow the effects of pica, but this finding may also be caused by
inadequacies inherent in collecting data on pica.
PB11C/A 11-142 7/29/83
-------
PRELIMINARY DRAFT
These data indicate that in a heavily contaminated area, blood lead levels in children
may be significantly increased by the intentional ingestion of nonfood materials having a high
lead content.
Data on the parents' occupation are, however, more reliable. It must be remembered also
that the study areas were not homogeneous socioeconomically. In addition, the specific type
of work an individual does in an industry is probably much more important than simply being
employed in a particular industry. The presence in the home of an industrial employee exposed
occupationally to lead may produce increases in the blood lead levels ranging from 10 to 30
percent.
The importance of the infiltration of lead dusts onto clothing, particularly the under-
garments, of lead workers and their subsequent transportation has been demonstrated in a
number of studies on the effects of smelters (Martin et a!., 1975). It was noted in the
United Kingdom that elevated blood lead levels were found in the wives and children of
workers, even though they resided some considerable distance from the facility. It was most
prominent in the workers themselves who had elevated blood lead levels. Quantities of lead
dust were found in workers' cars and homes. It apparently is not sufficient for a factory
merely to provide outer protective clothing and shower facilities for lead workers. In
another study in Bristol, from 650 to 1400 M9/9 of lead was found in the undergarments of
workers as compared with 3 to 13 ug/g in undergarments of control subjects. Lead dust will
remain on the clothing even after laundering: up to 500 mg of lead has been found to remain
on an overall garment after washing (Lead Development Association, 1973).
Baker et al. (1977a) found blood lead levels greater than 30 ug/dl in 38 of 91 children
whose fathers were employed at a secondary lead smelter in Memphis, TN, House dust, the only
source of lead in the homes of these children, contained a mean of 2687 |jg/g compared with 404
pg/g in the homes of a group of matched controls. Mean blood lead levels in the workers'
children were significantly higher than those for controls and were closely correlated with
the lead content of household dust. In homes with lead in dust less than 1000 ug/g, 18 chil-
dren had a mean blood lead level of 21.8 ± 7.8 ug/dl, whereas in homes where lead in dust was
greater than 7000 ug/g, 6 children had mean blood lead levels of 78.3 ± 34.0 ug/dl. See
Section 7.3.2.1.6 for a further discussion of household dust.
Other studies have documented increased lead absorption in children of families where at
least one member was occupationally exposed to lead (Fischbein et al., 1980a). The occupa-
tional exposures involved battery operations (Morton et al., 1982; U.S. Centers for Disease
Control, 1977b; Dolcourt et al., 1978, 1981; Watson et al., 1978; Fergusson et al., 1981) as
well as other occupations (Snee, 1982b; Rice et al., 1978).
PB11C/A 11-143 7/29/83
-------
PRELIMINARY DRAFT
In late summer of 1976, a battery plant in southern Vermont provided the setting for the
first documented instance of increased lead absorption in children of employees in the battery
industry. The data were first reported by U.S. Centers for Disease Control (1977b) and more
completely by Watson et al. (1978).
Reports of plant workers exposed to high levels of lead stimulated a study of plant
employees and their children in August and September 1975. In the plant, lead oxide powder is
used to coat plates in the construction of batteries. Before the study, the work setting of
all 230 employees of the plant had been examined and 62 workers (22 percent) were identified
as being at risk for high lead exposure. All of the high risk workers interviewed reported
changing clothes before leaving work and 90 percent of them reported showering. However, 87
percent of them stated that their work clothes were washed at home.
Of the high risk employees, 24 had children between the ages of 1 and 6 years. A case-
control study was conducted in the households of 22 of these employees. Twenty-seven children
were identified. The households were matched with neighborhood controls including 32 control
children. None of the control family members worked in a lead industry. Capillary blood
specimens were collected from all children and the 22 battery plant employees had venous spec-
imens taken. All blood samples were analyzed for lead by AAS. Interviewers obtained back-
ground data, including an assessment of potential lead exposures.
About 56 percent of the employees' children had blood leads greater than 30 ug/dl com-
pared with about 13 percent of the control children. Mean blood lead levels were stat-
istically significantly different, 31.8 ug/dl and 21.4 ug/dl, respectively. Blood lead levels
in children were significantly correlated with employee blood lead levels.
House dust lead levels were measured in all children's homes. Mean values were 2239.1
ug/g and 718.2 ug/g for employee and control homes, respectively; this was statistically sig-
nificant. Examination of the correlation coefficient between soil lead and blood lead levels
in the two sets of homes showed a marginally significant coefficient in the employee household
but no correlation in the control homes. Tap water and paint lead levels did not account for
the observed difference in blood leads between children of workers and neighborhood controls.
It is significant that these findings were obtained despite the changing of clothes at the
plant.
Morton et al. (1982) conducted their study of children of battery plant workers and con-
trols during February-March 1978. Children were included in the study if one parent had at
least 1 year of occupational exposure, if they had lived at the same residence for at least 6
months, and if they were from 12-83 months of age. Children for the control group had to have
no parental occupational exposure to lead for 5 years, and had to have lived at the same ad-
dress at least 6 months.
PB11C/A 11-144 7/29/83
-------
PRELIMINARY DRAFT
Thirty-four children were control matched to the exposed group by neighborhoods and age
(±1 year). No matching was thought necessary for sex because in this age group blood lead
levels are unaffected by sex. The selection of the control population attempted to adjust for
both socioeconomic status as well as exposure to automotive lead.
Capillary blood specimens were collected concurrently for each matched pair. Blood lead
levels were measured by the CDC lab using a modified Delves cup AAS procedure. Blood lead
levels for the employees for the previous year were obtained from company records. Question-
naires were administered at the same time as the blood sampling to obtain background informa-
tion. The homemaker was asked to complete the interview to try to get a more accurate picture
of the hygiene practices followed by the employees.
Children's blood lead levels differed significantly between the exposed and control
groups. Fifty-three percent of the employees' children had blood lead levels greater than 30
ug/dl, while no child in the control population had a value greater than 30 ug/dl. The mean
blood lead for the children of the employees was 49.2 ug/dl with a standard deviation of 8.3
ug/dl. These data represent the population average for yearly individual average levels. The
employees had an average greater than 60 MQ/dl- Still, this is lower than the industry
average. Of the eight children with blood levels greater than 40 pg/dl, seven had fathers
with blood lead greater than 50 ug/dl. Yet there was not a significant correlation between
children's blood lead level and father's blood lead level.
Investigations were made into the possibility that other lead exposures could account for
the observed difference in blood lead levels between children of employees and control chil-
dren. In 11 of the 33 pairs finally included in the study, potential lead exposures other
than fathers' occupations were found in the employee child of the matched pair. These in-
cluded a variety of lead sources such as automobile body painting, casting of lead, and
playing with spent shell casings. The control and exposed populations were again compared
after removing these 11 pairs from consideration. There was still a statistically significant
difference in blood lead level between the two groups of children.
An examination of personal hygiene practices of the workers showed that within high ex-
posure category jobs, greater compliance with recommended lead containment practices resulted
in lower mean blood lead levels in children. Mean blood leads were 17.3, 36.0 and 41.9 ug/dl
for good, moderately good and poor compliance groups, respectively. In fact, there was only a
small difference between the good hygiene group within the high exposure category and the mean
of the control group (17.3 ug/dl vs. 15.9 ug/dl). Insufficient sample sizes were available to
evaluate the effect of compliance on medium and low lead exposures for fathers.
Dolcourt et al. (1978) investigated lead absorption in children of workers in a plant
that manufactures lead-acid storage batteries. The plant became known to these researchers as
a result of finding an elevated blood lead level in a 20-month-old child during routine
PB11C/A 11-145 7/29/83
-------
PRELIMINARY DRAFT
screening. Although the child was asymptomatic, his mother proved not to be. Two siblings
were also found to have elevated blood lead levels. The mother was employed by the plant; her
work involved much hard labor and brought her into continual contact with powdery lead oxide.
No uniforms or garment covers were provided by the company. As a result of these findings,
screening was offered to all children of plant employees.
During February to May 1977, 92 percent of 63 eligible children appeared for screening.
Age ranged from 10 months to 15 years. About equal numbers of girls and boys underwent
screening. Fingerstick blood samples were collected on filter paper and were analyzed for
lead by AAS. Children with blood lead levels equal to or greater than 40 ug/dl were referred
for more detailed medical evaluation including an analysis of a venous blood specimen for
lead. Dust samples were collected from carpeting in each home and analyzed for lead by gra-
phite furnace AAS. Home tap water was analyzed for lead by AAS, and house paint was analyzed
for lead by XRF.
Of the 58 children who had the initial fingerstick blood lead elevation, 69 percent had
blood lead levels equal to or greater than 30 ug/dl. Ten children from six families had blood
lead levels equal to or greater than 40 (jg/dl, and blood lead levels were found to vary
markedly with age. The 0- to 3-year old category exhibited the highest mean with the 3- to
6-year-olds the next highest (39.2 ug/dl). Lowest mean values were found in the equal to or
greater than 10-year-old group (26.7 ug/dl).
More detailed investigation of the six families with the highest blood lead levels in
their children revealed the following: five of the six lived in rural communities, with no
pre-existing source of lead from water supply, house paint, industrial emissions or heavy
automobile traffic. However, dust samples from the carpets exhibited excessively high lead
concentrations. These ranged from 1700 to 84,050 ug/g.
Fergusson et al. (1981) sampled three population groups: general population, employees
of a battery plant, and children of battery plant employees, using hair lead levels as indices
of lead. Hair lead levels ranged from 1.2 to 110.9 ug/g in the 203 samples from the general
population. The distribution of hair lead levels was nearly lognormal. Employees of the bat-
tery factory had the highest hair lead levels (median ~250 ug/g) while family members (median
~40 M9/g) had a lesser degree of contamination and the general population (median ~5 ug/g)
still less.
Analysis of variance results indicated a highly significant difference between mean lead
levels of the general survey and family members of the employees, and a significant difference
between the mean lead levels in the hair of the employees and their families. No significant
differences were found comparing mean hair lead levels among family members in terms of age
and sex. The analyses of the house dust suggested that the mechanism of exposure of family
PB11C/A 11-146 7/29/83
-------
PRELIMINARY DRAFT
members is via the lead in dust that is carried home. Mean dust lead levels among the homes
of factory employees was 5580 ug/g while the dust inside of houses along a busy road was only
1620 |jg/g. Both of these concentrations are for particles less than 0.1 mm.
Dol court et al. (1981) reported two interesting cases of familiar exposure to lead caused
by recycling of automobile storage batteries. The first case was of a 22 member, 4 generation
family living in a three bedroom house in rural eastern North Carolina. The great grandfather
of the index case worked at a battery recycling plant. He had two truckloads of spent casings
delivered to the home to serve as fuel for the wood stove; the casings were burned over a 3-
month period.
The index case presented with classic signs of acute lead encephalopathy, the most severe
and potentially fatal form of acute lead poisoning. The blood lead level was found to be 220
pg/dl . Three months after initial diagnosis and after chelation therapy, she continued to
have seizures and was profoundly mentally retarded. Dust samples were obtained by vacuum
cleaner and analyzed for lead by flameless AAS. Dust from a sofa near the wood stove con-
tained 13,283 pg/g lead while the kitchen floor dust had 41,283 \tg/Q. There was no paint
lead. All other members of the family had elevated blood lead levels ranging from 27-256
The other case involved a truck driver working in a low exposure area of a battery re-
cycling operation in rural western North Carolina. He was operating an illegal battery re-
cycling operation in his home by melting down reclaimed lead on the kitchen stove. No family
member was symptomatic for lead symptoms but blood lead levels ranged from 24 to 72 MQ/dl-
Soil samples taken from the driveway, which was paved with fragments of the discarded battery
casing, contained 12-13 percent lead by weight.
In addition to families being exposed as a result of employment at battery plants, stu-
dies have been reported recently for smelter worker families (Rice et al., 1978; Snee, 1982c).
Rice et al . studied lead contamination in the homes of secondary lead smelters. Homes of em-
ployees of secondary smelters in two separate geographic areas of the country were examined to
determine whether those homes had a greater degree of lead contamination than homes of workers
in the same area not exposed to lead. Both sets of homes ( area I and II) were examined at
the same time of the year.
Thirty-three homes of secondary smelter employees were studied; 19 homes in the same or
similar neighborhoods were studied as controls. Homes studied were in good condition and were
one or two family dwellings. Blood lead levels were not obtained for children in these homes.
In the homes of controls, a detailed occupational history was obtained for each employed
person. Homes where one or more residents were employed in a lead contaminated environment
were excluded from the analysis.
PB11C/A 11-147 7/29/83
-------
PRELIMINARY DRAFT
House dust samples were collected by Vostal's method and were analyzed for lead by AAS.
In one of the areas, samples of settled dust were collected from the homes of employees and
controls. Dust was collected over the doorways. In homes where the settled dust was col-
lected, zinc protoporphyrin (ZPP) determinations were made in family members of the lead
workers and in the controls.
In both areas the wipe samples' were statistically significantly higher in the homes of
employees compared to controls (geometric mean 79.3 ± 61.8 ug/g vs. 28.8 ± 7.4 ug/g Area I;
112.0 ± 2.8 ug/g vs. 9.7 ± 3.9 ug Area II). No significant differences were found between
workers' homes or controls between Area I and Area II. Settled dust lead was significantly
higher in the homes of employees compared to controls (3300 vs. 1200 ug/g). Lead content of
particulate matter collected at the curb and of paint chips collected in the home was not sig-
nificantly different between employee homes and controls. Zinc protoporphyrin determinations
were done on 15 children, 6 years or younger. ZPP levels were higher in employee children
than in control children. Mean levels were 61.4 ug/ml and 37.6 ug/ml, respectively.
It should be noted again that the wipe samples were not different between employee homes
in the two areas. Interviews with employees indicated that work practices were quite similar
in the two areas. Most workers showered and changed before going home. Work clothes were
washed by the company. Obviously much closer attention needs to be paid to other potential
sources of lead introduction into the home (e.g., automobile surfaces).
11.5.6 Miscellaneous Studies
11.5.6.1 Studies Using Indirect Measures of Air Exposure.
11.5.6.1.1 Studies in the United States. A 1973 Houston study examined the blood lead levels
of parking garage attendants, traffic policemen, and adult females living near freeways
(Johnson et al., 1974). A control group for each of the three exposed populations was selec-
ted by matching for age, education and race. Unfortunately, the matching was not altogether
successful; traffic policemen had less education than their controls, and the garage employees
were younger than their controls. Females were matched adequately, however. It should be
noted that the mean blood lead values for traffic policemen and parking garage attendants, two
groups regularly exposed to higher concentrations of automotive exhausts, were significantly
higher than the means for their relevant control groups. Statistically significant dif-
ferences in mean values were not found, however, between women living near a freeway, and con-
trol women living at greater distances from the freeway.
PB11C/A 11-148 7/29/83
-------
PRELIMINARY DRAFT
A study of the effects of lower level urban traffic densities on blood lead levels was
undertaken in Dallas, Texas, in 1976 (Johnson et al., 1978). The study consisted of two
phases. One phase measured air lead values for selected traffic densities and conditions,
ranging from equal to or less than 1,000 to about 37,000 cars/day. The second phase consisted
of an epidemiological study of traffic density and blood lead levels among residents. Figure
11-30 shows the relationship between arithmetic means of air lead and traffic density. As can
be seen from the graph, a reasonable fit was obtained.
In addition, for all distances measured (1.5 to 30.5 m from the road), air lead concen-
trations declined rapidly with distance from the street. At 15 m, concentrations were about
55 percent of the street concentrations. In air lead collections from 1.5 to 30.5 m from the
street, approximately 50 percent of the airborne lead was in the respirable range (<1 urn), and
the proportions in each sire class remained approximately the same as the distance from the
street increased.
Soil lead concentrations were higher in areas with greater traffic density, ranging from
73.6 ug/g at less than 1,000 cars per day to a mean of 105.9 at greater than 19,500 cars per
day. The maximum soil level obtained was 730 ug/g.
Dustfall samples for 28 days from 9 locations showed no relationship to traffic
densities, but outdoor levels were at least 10 times the indoor concentration in nearby
residences.
In the second phase, three groups of subjects, 1- to 6-years-old, 18- to 49-years old and
50 years and older, were selected in each of four study areas. Traffic densities selected
were less than 1,000, 8,000 to 14,000, 14,000 to 20,000 and 20,000 to 25,000 cars/day. The
study groups averaged about 35 subjects, although the number varied from 21 to 50. The
smallest groups were from the highest traffic density area. No relationship between traffic
density and blood lead levels in any of the age groups was found (Figure 11-31). Blood lead
levels were significantly higher in children, 12 to 18 ug/dl, than in adults, 9 to 14 ug/dl.
Caprio et al. (1974) compared blood lead levels and proximity to major traffic arteries
•jo a study reported in 1971 that included 5226 children in Newark, New Jersey. Over 57 per-
cent of the children living within 30.5 m of roadways had blood lead levels greater than 40
ug/dl. For those living between 30.5 and 61 m from the roadways, more than 27 percent had
such levels, and at distances greater than 61 m, 31 percent exceeded 40 ug/dl. The effect of
automobile traffic was seen only in the group that lived within 30.5 m of the road.
No other sources of lead were considered in this study. However, data from other studies
on mobile sources indicate that it is unlikely that the blood lead levels observed in this
study resulted entirely from automotive exhaust emissions.
P&11C/A 11-149 7/29/83
-------
PRELIMINARY DRAFT
o>
g
<
oc
i-
z
LU
U
o
U
O
2.0
1.6
1.2
0.8
0.4
I I
I I I I T
Y = 0.6598 + 0.0263 X
X = TRAFFIC COUNT/1000
0 4,000 8,000 12,000 16,000 20,000 24,000 28,000 32,000 36,000 38,000
TRAFFIC VOLUME, cars/day
Figure 11-30. Arithmetic mean of air lead levels by traffic volume,
Dallas, 1976.
In 1964, Thomas et al. (1967) investigated blood lead levels in 50 adults who had lived
for at least 3 years within 76 m of a freeway (Los Angeles) and those of 50 others who had
lived for a similar period near the ocean or at least 1.6 km from a freeway. Mean blood lead
levels for those near the freeway were 22.7 ±5.6 for men and 16.7 ±7.0 ug/dl for women.
These concentrations were higher than for control subjects living near the ocean: 16.0 ±8.4
ug/dl for men and 9.9 ± 4.9 ug/dl for women. The higher values, however, were similar to
those of other Los Angeles populations. Measured mean air concentrations of lead in Los
Angeles for October 1964 were 12.25 ± 2.70 ug/m3 at a location 9 m from the San Bernardino
freeway; 13.25 ± 1.90 ug/m3 at a fourth floor location 91.5 m from the freeway; and 4.60 ±
1.92 ug/m3 1.6 km from the nearest freeway. The investigators concluded that the differences
observed were consistent with coastal inland atmospheric and blood lead gradients in the Los
Angeles basin and that the effect of residential proximity to a freeway (7.6 to 76 m) was not
demonstrated.
Ter Haar and Chadzynski report a study of blood lead levels of children living near three
heavily travelled streets in Detroit (Ter Haar, 1981; Ter Haar and Chadzynski, 1979). Blood
lead levels were not found to be related to distance from the road but were related to condi-
tions of housing and age of the child after multiple regression analyses.
PB11C/A 11-150 7/29/83
-------
PRELIMINARY DRAFT
25
20
Z
01
O
O
O
111
_l
Q
O
o
_l
CO
10
FEMALES <9
MALES <9
-o
MALES> 49
FEMALES 19-49
Irrs
FEMALES >49
1
1
I
< 1,000 1,00013,500 13,500- 19,500-
19,500 38,000
TRAFFIC DENSITY, cars/day
Figure 11-31. Blood lead concentration and traffic density by sex and
age, Dallas, 1976.
PBUC/A
11-151
7/29/83
-------
PRELIMINARY DRAFT
11.5.6.1.2 British Studies. In a Birmingham, England study, mean blood lead levels in 41
males and 58 females living within 800 m of a highway interchange were 14.41 and 10.93 ug/dl,
respectively, just before the opening of the interchange in May 1972 (Waldron, 1975). From
October 1972 to February 1973, the respective values for the same individuals were 18.95 and
14.93 ug/dl. In October 1973 they were 23.73 and 19.21 pg/dl. The investigators noted dif-
ficulties in the blood collection method during the baseline period and changed from capillary
to venous blood collection for the remaining two samples. To interpret the significance of
the change in blood collection method, some individuals gave both capillary and venous blood
at the second collection. The means for both capillary and venous bloods were calculated for
the 18 males and 23 females who gave both types of blood samples (Barry, 1975). The venous
blood mean values for both these males and females were lower by 0.8 and 0.7 ug/dl, respec-
tively. If these differences were applied to the means of the third series, the mean for
males would be reduced to 24.8 |jg/dl and that for the females to 18.7 ug/dl. These adjusted
means still show an increase over the means obtained for the first series. Comparing only the
means for venous bloods, namely series two and three, again shows an increase for both groups.
The increase in blood lead values was larger than expected following the model of Knelson et
al. (1973), because air lead values near the road were approximately 1 ug/m3. The investi-
gators concluded that either the lead aerosol of very small particles behaved more like a gas
so that considerably more than 37 percent of inhaled material was absorbed or that ingestion
of lead contaminated dust might be responsible.
Studies of taxicab drivers have employed different variables to represent the drivers'
lead exposure (Flindt et al., 1976; Jones et al., 1972): one variable was night vs. dayshift
drivers (Jones et al., 1972); the other, mileage driven (Flindt et al., 1976). No difference
was observed, in either case.
The studies reviewed show that automobiles produce sufficient emissions to increase air
and nearby soil concentrations of lead as well, as increase blood lead concentrations in chil-
dren and adults. The problem is of greater importance when houses are located within 100 ft
(30 m) of the roadway.
11.5.6.2 Miscellaneous Sources of Lead. The habit of cigarette smoking is a source of lead
exposure. Shaper et al. (1982) report that blood lead concentration is higher for smokers
than nonsmokers and that cigarette smoking makes a significant independent contribution to
blood lead concentration in middle-aged men in British towns. A direct increase in lead in-
take from cigarettes is thought to be responsible. Hopper and Mathews (1983) comment that
current smoking has a significant effect on blood lead level, with an average increase of 5.8
percent in blood lead levels for every 10 cigarettes smoked per day. They also report that
PB11C/A 11-152 7/29/83
-------
PRELIMINARY DRAFT
past smoking history had no measurable effect on blood lead levels. Hasselblad and Nelson
(1975) report an average increase in women's blood lead levels of 1.3 pg/dl in the study of
Tepper and Levin (1975).
Although no studies are available, it is conceivable that destruction of lead-containing
plastics (to recover copper), which has caused cattle poisoning, also could become a source of
lead exposure for humans. Waste disposal is a more general problem because lead-containing
materials may be incinerated and may thus contribute to increased air lead levels. This
source of lead has not been studied in detail. Tyrer (1977) cautions of the lead hazard in
the recycling of waste.
The consumption of illicitly distilled liquor has been shown to produce clinical cases of
lead poisoning. Domestic and imported earthenware (De Rosa et al., 1980) with improperly
fired glazes have also been related to clinical lead poisoning. This source becomes important
when foods or beverages high in acid are stored in earthenware containers, because the acid
releases lead from the walls of the containers.
Particular cosmetics, popular among some Oriental and Indian ethnic groups, contain high
percentages of lead that sometimes are absorbed by users in quantities sufficient to be toxic.
Ali et al. (1978) and Attenburrow et al. (1980) discuss the practice of surma and lead poison-
ing. Other sources of lead are presented in Table 11-60.
TABLE 11-60. SOURCES OF LEAD
Source References
Gasoline Sniffing Kaufman and Wiese (1978)
Coodin and Boeckx (1978)
Hansen and Sharp (1978)
Colored Gift Wrapping Bertagnolli and Katz (1979)
Gunshot Wound Dillman et al. (1979)
Drinking Glass Decorations Anonymous (1979)
Electric Kettles Wigle and Charlebois (1978)
Hair dye Searle and Harnden (1979)
Snuff use Filippini and Simmler (1980)
Firing ranges Fischbein et al. (1979, 1980b)
PBUC/A 11-153 7/29/83
-------
PRELIMINARY DRAFT
11.6 SUMMARY AND CONCLUSIONS
Studies of ancient populations using bone and teeth show that levels of internal exposure
of lead today are substantially elevated over past levels. Studies of current populations
living in remote areas far from urbanized cultures show blood lead levels in the range of 1 to
5 ug/dl. In contrast to the blood lead levels found in remote populations, data from current
U.S. populations have geometric means ranging from 10 to 20 ug/dl depending on age, race, sex
and degree of urbanization. These higher current exposure levels appear to be associated with
industrialization and widespread commercial use of lead, e.g. in gasoline combustion.
Age appears to be one of the single most important demographic covariates of blood lead
levels. Blood lead levels in children up to six years of age are generally higher than those
in non-occupationally exposed adults. Children aged two to three years tend to have the high-
est levels as shown in Figure 11-32. Blood lead levels in non-occupationally exposed adults
may increase slightly with age due to skeletal lead accumulation.
Sex has a differential impact on blood lead levels depending on age. No significant dif-
ferences exist between males and females less than seven years of age. Males above the age of
seven generally have higher blood lead levels than females.
Race also plays a role, in that blacks generally have higher blood lead levels than
either whites or Hispanics and urban black children (aged 6 mo. to 5 yr.) have markedly higher
blood lead concentrations than any other racial or age group. Possible genetic factors asso-
ciated with race have yet to be fully disentangled from differential exposure levels as im-
portant determinants of blood lead levels.
Blood lead levels also generally increase with degree of urbanization. Data from NHANES
II show blood lead levels in the United States, averaged from 1976 to 1980, increasing from a
geometric mean of 11.9 ug/dl in rural populations to 12.8 ug/dl in urban populations less than
one million, increasing again to 14.0 ug/dl in urban populations of one million or more.
Recent U.S. blood lead levels show a downward trend occurring consistently across race
age and geographic location. The downward pattern commenced in the early part of the 1970's
and has continued into 1980. The downward trend has occurred from a shift in the entire dis-
tribution and not through a truncation in the high blood lead levels. This consistency sug-
gests a general causative factor, and attempts have been made to identify the causative ele-
ment. Reduction in lead emitted from the combustion of leaded gasoline is a prime suspect, but
at present no causal relationship has been established.
Blood lead levels, examined on a population basis, have similarly skewed distributions.
Blood lead levels, from a population thought to be homogenous in terms of demographic and lead
exposure characteristics, approximately follow a lognormal distribution. The geometric stan-
dard deviations, an estimation of dispersion, for four different studies are shown in Table
11-61. The values, including analytic error, are about 1.4 for children and possibly somewhat
PB11C/A 11-154 7/29/83
-------
PRELIMINARY DRAFT
40
36
30
26
\
6
uj 20
0
O
O
15
10
IDAHO STUDY
NEW YORK SCREENING - BLACKS
NEW YORK SCREENING - WHITES
NEW YORK SCREENING - HISPANICS
- — NHANES II STUDY - BLACKS
— NHANES II STUDY - WHITES
I
I
I
I
10
AGE IN YEARS
Figure 11-32. Geometric mean blood lead levels by race and age for younger children in the
NHANES II study, and the Kellogg/Silver Valley and New York Childhood Screening Studies.
PB11C/A
11-155
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-61. SUMMARY OF POOLED GEOMETRIC STANDARD
DEVIATIONS AND ESTIMATED ANALYTIC ERRORS
Study
NHANES II
N.Y. Childhood
Pooled Geometric Standard Deviations
Inner City
Black Children
1.37
1.41
Inner City
White Children
1.39
1.42
Adults
Females
1.36a
Adult
Males
1.40a
Estimated
Analytic
Error
0.021
(b)
Screening Study
Tepper-Leven
Azar et al.
1.30
1.29
0.056
0.042C
Note: To calculate an estimated person-to-person GSD, compute Exp [((In(GSD)) -
Analytic Error)*s]
pooled across areas of differing urbanization
not known, assumed to be similar to NHANES II
ctaken from Lucas (1981).
smaller for adults. This allows an estimation of the upper tail of the blood lead distri-
bution, the group at higher risk.
Because the main purpose of this chapter is to examine relationships of lead in air and
lead in blood under ambient conditions, the results of studies most appropriate to this area
have been emphasized. A summary of the most appropriate studies appears in Table 11-62. At
air lead exposures of 3.2 |jg/m or less, there is no statistically significant difference be-
tween curvilinear and linear blood lead inhalation relationships. At air lead exposures of 10
ug/m or more, either nonlinear or linear relationships can be fitted. Thus, a reasonably
consistent picture emerges in which the blood-lead air-lead relationship by direct inhalation
was approximately linear in the range of normal ambient exposures of 0.1 - 2.0 ug/m (as dis-
cussed in Chapter 7). Differences among individuals in a given study (and among several
studies) are large, so that pooled estimates of the blood lead inhalation slope depend upon
the the weight given to various studies. Several studies were selected for analysis, based
upon factors described earlier. EPA analyses* of experimental and clinical studies (Griffin
et al. 1975; Rabinowitz et al., 1974, 1976, 1977; Kehoe 1961a,b,c; Gross 1981; Hammond et al.,
1981) suggest that blood lead in adults increases by 1.64 + 0.22 ug/dl from direct inhalation
*Note: The term EPA analyses refers to calculations done at EPA. A brief discussion of the
methods used is contained in Appendix 11-B; more detailed information is available at EPA upon
request.
PB11C/A
11-156
7/29/83
-------
PRELIMINARY DRAFT
TABLE 11-62. SUMMARY OF BLOOD INHALATION SLOPES, (p)
pg/dl per |jg/m
Population Study
Children Angle and
Me I nt ire, 1979
Omaha, NE
Roels et al.
(1980)
Belgium
Yankel et al.
(1977); Walter
et al. (1980)
Idaho
Adult Males Azar et al.
(1975). Five
groups
Griffin et al.
(1975), NY
prisoners
Gross
(1979)
Rabinowitz et
al. (1973,1976,
1977)
Study (p) Model Sensitivity
Type N Slope Of Slope*
pg/dl per pg/m3
Population 1074 1.92 (1.40 - 4.40)1'2'3
Population 148 2.46 (1.55 - 2.46)1'2
Population 879 1.52 (1.07 - 1.52)1'2'3
Population 149 1.32 (1.08 - 2.39)2'3
Experiment 43 1.75 (1.52 - 3.38)4
Experiment 6 1.25 (1.25 - 1.55)2
Experiment 5 2.14 (2.14 - 3.51)5
*Selected from among the most plausible statistically equivalent models. For nonlinear models,
slope at 1.0
Sensitive to choice of other correlated predictors such as dust and soil lead.
Sensitive to linear vs. nonlinear at low air lead.
Sensitive to age as a covariate.
^Sensitive to baseline changes in controls.
Sensitive to assumed air lead exposure.
PB11C/A
11-157
7/29/83
-------
PRELIMINARY DRAFT
of each additional (jg/m of air lead. EPA analyses of population studies (Yankel et a!.,
1977; Roels et al., 1980; Angle and Mclntire, 1979) suggest that, for children, the blood lead
increase is 1.97 ± 0.39 ug/dl per ug/m for air lead. EPA anaylsis of Azar's population study
(Azar et al., 1975) yields a slope of 1.32 ± 0.38 for adult males.
These slope estimates are based on the assumption that an equilibrium level of blood lead
is achieved within a few months after exposure begins. This is only approximately true, since
lead stored in the skeleton may return to blood after some years. Chamberlain et al. (1978)
suggest that long term inhalation slopes should be about 30 percent larger than these
estimates. Inhalation slopes quoted here are associated with a half-life of blood lead in
adults of about 30 days. 0'Flaherty et al. (1982) suggest that the blood-lead half-life may
increase slightly with duration of exposure, but this has not been confirmed (Kang et al.,
1983).
One possible approach would be to regard all inhalation slope studies as equally infor-
mative and to calculate an average slope using reciprocal squared standard error estimates as
weights. This approach has been rejected for two reasons. First, the standard error estima-
tes characterize only the internal precision of an estimated slope, not its representativeness
(i.e., bias) or predictive validity. Secondly, experimental and clinical studies obtain more
information from a single individual than do population studies. Thus, it may not be appro-
priate to combine the two types of studies.
Estimates of the inhalation slope for children are only available from population
studies. The importance of dust ingestion as a non-inhalation pathway for children is estab-
lished by many studies. A slope estimate has been derived for air lead inhalation based on
those studies (Angle and Mclntire 1979; Roels et al., 1980; Yankel et al., 1977) from which
the air inhalation and dust ingestion contributions can both be estimated.
While direct inhalation of air lead is stressed, this is not the only air lead contribu-
tion that needs to be considered. Smelter studies allow partial assessment of the air lead
contributions to soil, dust and finger lead. Conceptual models allow preliminary estimation
of the propagation of lead through the total food chain as shown in Chapter 7. Useful mathe-
matical models to quantify the propagation of lead through the food chain need to be
developed. The direct inhalation relationship does provide useful information on changes in
blood lead as responses to changes in air lead on a time scale of several months. The in-
direct pathways through dust and soil and through the food chain may thus delay the total
blood lead response to changes in air lead, perhaps by one or more years. The Italian ILE
study facilitates partial assessment of this delayed response from leaded gasoline as a
source.
Dietary absorption of lead varies greatly from one person to another and depends on the
physical and chemical form of the carrier, on nutritional status, and on whether lead is
PB11C/A 11-158 7/29/83
-------
PRELIMINARY DRAFT
ingested with food or between meals. These distinctions are particularly important for con-
sumption by children of leaded paint, dust and soil. Typical values of 10 percent absorption
of ingested lead into blood have been assumed for adults and 25 to 50 percent for children.
It is difficult to obtain accurate dose-response relationships between blood lead levels
and lead levels in food or water. Dietary intake must be estimated by duplicate diets or
fecal lead determinations. Water lead levels can be determined with some accuracy, but the
varying amounts of water consumed by different individuals adds to the uncertainty of the es-
timated relationships.
Quantitative analyses relating blood lead levels and dietary lead exposures have been re-
ported. Studies on infants provide estimates that are in close agreement. Only one indi-
vidual study is available for adults (Sherlock et al. 1982); another estimate from a number of
pooled studies is also available. These two estimates are in good agreement. Most of the
subjects in the Sherlock et al. (1982) and United Kingdom Central Directorate on Environmental
Pollution (1982) studies received quite high dietary lead levels (>300 ug/day). The fitted
cube root equations give high slopes at lower dietary lead levels. On the other hand, the
linear slope of the United Kingdom Central Directorate on Environmental Pollution (1982) study
is probably an underestimate of the slope at lower dietary lead levels. For these reasons,
the Ryu et al. (1983) study is the most believable, although it only applies to infants.
Estimates for adults should be taken from the experimental studies or calculated from assumed
absorption and half-life values. Most of the dietary intake supplements were so high that
many of the subjects had blood lead concentrations much in excess of 30 ug/m for a considera-
ble part of the experiment. Blood lead levels thus may not completely reflect lead exposure,
due to the previously noted nonlinearity of blood lead response at high exposures. The slope
estimates for adult dietary intake are about 0.02 ng/dl increase in blood lead per ug/day in-
take, but consideration of blood lead kinetics may increase this value to about 0.04. Such
values are a bit lower than slopes of about 0.05 ug/dl per ug/day estimated from the popula-
tion studies extrapolated to typical dietary intakes. The value for infants is larger.
The relation between blood lead and water lead is not clearly defined and is often de-
scribed as nonlinear. Water lead intake varies greatly from one person to another. It has
been assumed that children can absorb 25 to 50 percent of lead in water. Many authors chose
to fit cube root models to their data, although polynomial and logarithmic models were also
used. Unfortunately, the form of the model greatly influences the estimated contributions to
blood leads from relatively low water lead concentration.
Although there is close agreement in the quantitative analyses of the relationship bet-
ween blood lead level and dietary lead, there is a larger degree of variability in results of
the various water lead studies. The relationship is curvilinear, but its exact form is yet to
be determined. At typical levels for U.S. populations, the relationship appears linear. The
PB11C/A 11-159 7/29/83
-------
PRELIMINARY DRAFT
only study that determines the relationship based on lower water lead values (<100 M9/1) is
the Pocock et al. (1983) study. The data from this study, as well as the authors themselves,
suggest that in this lower range of water lead levels, the relationship is linear. Further-
more, the estimated contributions to blood lead levels from this study are quite consistent
with the polynomial models from other studies. For these reasons, the Pocock et al. (1983)
slope of 0.06 is considered to represent the best estimate. The possibility still exists,
however, that the higher estimates of the other studies may be correct in certain situations,
especially at higher water lead levels (>100 pg/1).
Studies relating soil lead to blood lead levels are difficult to compare. The relation-
ship obviously depends on depth of soil lead, age of the children, sampling method, cleanli-
ness of the home, mouthing activities of the children, and possibly many other factors. Var-
ious soil sampling methods and sampling depths have been used over time, and as such they may
not be directly comparable and may produce a dilution effect of the major lead concentration
contribution from dust which is located primarily in the top 2 cm of the soil. Increases in
soil dust lead significantly increase blood lead in children. From several studies (Yankel et
al., 1977; Angle and Mclntire, 1979) EPA estimates an increase of 0.6 to 6.8 pg/dl in blood
lead for each increase of 1000 M9/9 in soil lead concentration. Values of about 2.0 [jg/dl per
1,000 H9/9 soil lead from the Stark et al. (1982) study may represent a reasonable median
estimate. The relationship of housedust lead to blood lead is difficult to obtain. House-
hold dust also increases blood lead, children from the cleanest homes in the Silver Valley/
Kellogg Study having 6 pg/dl less lead in blood, on average, than those from the households
with the most dust.
A number of specific environmental sources of airborne lead have been evaluated for pot-
ential direct influence on blood lead levels. Combustion of leaded gasoline appears to be the
largest contributor to airborne lead. Two studies used isotope ratios of lead to estimate the
relative proportion of lead in the blood coming from airborne lead. From one study, by
Manton, it can be estimated that between 7 and 41 percent of the blood lead in study subjects
in Dallas resulted from airborne lead. Additionally, these data provide a means of estimating
the indirect contribution of air lead to blood lead. By one estimate, only 10 to 20 percent
of the total airborne contribution in Dallas is from direct inhalation.
From the ILE data in Facchetti and Geiss (1982), as shown in Table 11-63, the direct in-
halation of air lead may account for 54 percent of the total adult blood lead uptake from
leaded gasoline in a large urban center, but inhalation is a much less important pathway in
suburban parts of the region (17 percent of the total gasoline lead contribution) and in the
rural parts of the region (8 percent of the total gasoline lead contribution). EPA analyses
of the preliminary results from the ILE study separated the inhalation and non-inhalation con-
tributions of leaded gasoline to blood lead into the following three parts: (1) An increase
PB11C/A 11-160 7/29/83
-------
PRELIMINARY DRAFT
of about 1.7 ug/dl in blood lead per ug/m of air lead, attributable to direct inhalation of
the combustion products of leaded gasoline; (2) a sex difference of about 2 |jg/dl attributable
to lower exposure of women to indirect (non-inhalation) pathways for gasoline lead; and (3) a
non-inhalation background attributable to indirect gasoline lead pathways, such as ingestion
of dust and food, increasing from about 2 ug/dl in Turin to 3 jjg/dl in remote rural areas.
The non-inhalation background represents only two to three years of environmental accumulation
at the new experimental lead isotope ratio. It is not clear how to extrapolate numerically
these estimates to U.S. subpopulations; but it is evident that even in rural and suburban
parts of a metropolitan area, the indirect (non-inhalation) pathways for exposure to leaded
gasoline make a significant contribution to blood lead. This can be seen in Table 11-63. It
should also be noted that the blood lead isotope ratio responded fairly rapidly when the lead
isotope ratio returned to its pre-experimental value, but it is not yet possible to estimate
the long term change in blood lead attributable to persistent exposures to accumulated envi-
ronmental lead.
Studies of data from blood lead screening programs suggest that the downward trend in
blood lead levels noted earlier is due to the reduction in air lead levels, which has been at-
tributed to the reduction of lead in gasoline.
TABLE 11-63. ESTIMATED CONTRIBUTION OF LEADED GASOLINE TO BLOOD LEAD
BY INHALATION AND NON-INHALATION PATHWAYS
Air Lead
Fraction
From , N
Gasoline^ ;
Blood
Lead
Fraction
From f. x
Gasoline^0'
Blood Pb
From
Gasoline,
In Air[C}
ug/di
Blood Lead
Not Inhaled
From,Gaso-
line01"
ug/dl
Estimate
Fraction
Gas- Lead . v
Inhalation1 ;
Location
(a)
(b)
(c)
(d)
(e)
Turi n
<25 km
>25 km
0.873
0.587
0.587
Fraction of air
Mean fraction of
Estimated
Estimated
Fraction
blood
blood
lead
in Phase
0.237
0.125
0.110
2 attributable
2.79
0
0
to
.53
.28
lead in
2.37
2.60
3.22
0.
0.
0:
54
17
08
gasoline.
blood lead in Phase 2 attributable to lead
lead
lead
from gas
from gas
of blood lead uptake
inhalation = p
K
, non- inhalation =
from gasoline
(a) x (b)
(f)-(e)
, P
attributable to
in gasoline
= 1.6.
direct inhalation = (f)/(e)
Source: Facchetti and Geiss (1982), pp. 52-56.
PB11C/A
11-161
7/29/83
-------
PRELIMINARY DRAFT
Primary lead smelters, secondary lead smelters and battery plants emit lead directly into
the air and ultimately increase soil and dust lead concentrations in their vicinity. Adults,
and especially children, have been shown to exhibit elevated blood lead levels when living
close to these sources. Blood lead levels in these residents have been shown to be related to
air, as well as to soil or dust exposures.
PB11C/A 11-162 7/29/83
-------
PRELIMINARY DRAFT
11.7 REFERENCES
Aitchison, J. ; Brown J. A. C. (1966) The lognormal distribution. London, United Kingdom: Cam-
bridge University Press.
Ali, A. R. ; Smales, 0. R. C.; Aslam, M. (1978) Surma and lead poisoning. Br. Med. J. 2(6142):
915-916.
Angle, C. R. (1982) Calculation of PbB/PbS recent references, prepared for "correlation of soil
lead and blood lead." Presented at: University of Texas Southwest Medical Center; April;
Dallas, TX.
Angle, C. R.; Mclntire, M. S. (1977) Is busing good for your blood lead? Presented at: 2nd an-
nual French-American congress on clinical and analytical toxicology; August; St. Adele,
Canada. Washington, DC: National Institute of Environmental Health Sciences.
Angle, C. R.; Mclntire, M. S. (1979) Environmental lead and children: the Omaha study. J.
Toxicol. Environ. Health 5: 855-870.
Angle, C. R.; Mclntire, M. S. (1982) Children, the barometer of environmental lead. Adv.
Pediatr. 27: 3-31.
Angle, C. R.; Mclntire, M. S. ; Colucci, A. V. (1974) Lead in air, dustfall, soil, housedust,
milk and water: correlation with blood lead of urban and suburban school children. In:
Hemphill, D. D. , ed. Trace substances in environmental health-VIII: [proceedings of
University of Missouri's 8th annual conference on trace substances in environmental
health]; June; Columbia, MO. Columbia, MO: University of Missouri-Columbia; pp. 23-29.
Annest, J. L.; Casady, R. J.; White, A. A. (1983) The NHANES II study: analytic error and its
effects on national estimates of blood lead levels: United States, 1976-80. Available for
inspection at: U.S. Environmental Protection Agency, Environmental Criteria Assessment
Office, Research Triangle Park, NC.
Annest, J. L.; Mahaffey, K. R.; Cox, D. H.; Roberts, J. (1982) Blood lead levels for persons
6 months - 74 years of age: United States, 1976-80. Hyattsville, MD: [U.S. Department of
Health and Human Services. (Advance data from vital and health statistics of the National
Center for Health Statistics: no. 79.)]
Anonymous. (1979) Lead and cadmium may leach from drinking glass decorations. J. Am. Med.
Assoc. 241: 544.
Ashford, N. A.; Gecht, R. D., Hattis, D. B.; Katz, J. I. (1977) The effects of OSHA medical
removal protection on labor costs of selected lead industries. Cambridge, MA:
Massachusetts Institute of Technology, Center for Policy Alternatives; report no.
CPA-77/11. Available from: NTIS, Springfield, VA; PB 278653.
Attenburrow, A. A.; Campbell, S. ; Logan, R. W.; Goel, K. M. (1980) Surma and blood lead levels
in Asian children in Glasgow. Lancet 1(8163): 323.
Aufderheide, A. C. ; Neiman, F. D. ; Wittmers, L. E. , Jr.; Rapp, G. (1981) Lead in bone. II:
Skeletal-lead content as an indicator of lifetime lead ingestion and the social correlates
in an archaeological population. Am. J. Phys. Anthro. 55: 285-291.
I11REF/A 11-163 7/29/83
-------
PRELIMINARY DRAFT
Azar, A.; Snee, R. D. ; Habibi, K. (1975) An epidemiologic approach to community air lead ex-
posure using personal air samplers. In: Lead. Environ, qual. Saf. Suppl. 2: 254-288.
Baker, E. L. , Jr.; Folland, D. S.; Taylor, T. A.; Frank, M.; Peterson, W.; Lovejoy, G.; Cox,
D.; Housworth, J. ; Landrigan, P. J. (1977a) Lead poisoning in children of lead workers;
house contamination with industrial dust. N. Engl. J. Med. 296: 260-261.
Baker, E. L., Jr.; Hayes, C. G.; Landrigan, P. J.; Handke, J. L. ; Leger, R. T. ; Housworth, W.
J.; Harrington, J. M. (1977b) A nationwide survey of heavy metal absorption in children
living near primary copper, lead, and zinc smelters. Am. J. Epidemiol. 106: 261-273.
Barltrop, D. (1975) Significance of lead-contaminated soils and dusts for human populations.
Arh. High. Rada Toksikol. Suppl. 26: 81-93.
Barltrop, D.; Strehlow, C. D. ; Thorton, I.; Webb, J. S. (1974) Significance of high soil lead
concentrations for childhood lead burdens. Environ. Health Perspect. 7: 75-82.
Barry, P. S. I. (1975) Lead levels in blood. Nature (London) 258: 775.
Batschelet, E.; Brand, L. ; Steiner, A. (1979) On the kinetics of lead in the human body. J.
Math. Biol. 8: 15-23.
Beattie, A. D.; Dagg, J. H.; Goldberg, A.; Wang, L; Ronald, J. (1972a) Lead poisoning in rural
Scotland. Br. Med. J. 2(5182): 488-491.
Beattie, A. 0.; Moore, M. R.; Devenay, W. T.; Miller, A. R.; Goldberg, A. (1972b) Environmental
lead pollution in an urban soft-water area. Br. Med. J. 2(5182): 491-493.
Becker, R. 0.; Spadaro, J. A.; Berg, E. W. (1968) The trace elements of human bone. J. Bone Jt.
Surg. BOA: 326-334.
Berger, H. W. (1973) The NBS lead paint poisoning project: housing and other aspects.
Washington, DC: U.S. Department of Commerce, National Bureau of Standards; NBS technical
note 759.
Bernard, S. R. (1977) Dosimetric data and metabolic model for lead. Health Phys. 32: 44-46.
Bertagnolli, J. F.; Katz, S. A. (1979) Colored gift wrapping papers as a potential source of
toxic metals. Int. J. Environ. Anal. Chem. 6: 321-326.
Billick, I. H. (1982) Prediction of pediatric blood lead levels from gasoline consumption [sub-
mitted to docket for public hearing on lead phasedown proposed rulemaking, April 15.]
Available from: U.S. Environmental Protection Agency, Central Docket Section Washington,
DC; docket no. A-81-36; document no. IVA.4.
Billick, I. H.;-Cm-ran, A. S. ; Shier, D. R. (1979) Analysis of pediatric blood lead levels in
New York City for 1970-1976. Environ. Health Perspect. 31: 183-190.
Billick, I. H.; Curran, A. S.; Shier, D. R. (1980) Relation of pediatric blood lead levels to
lead in gasoline. Environ. Health Perspect. 34: 213-217.
Billick, I. H. ; Shier, D. R. ; Spiegelman, C. H. (1982) Sensitivity of trends in geometric mean
blood levels to random measurement errors. Sci. Total Environ. 24: 233-248.
I11REF/A 11-164 7/29/83
-------
PRELIMINARY DRAFT
Bishop, R. L. ; Hill, W. J. (1983) Modelling workplace lead-in-air and lead-in-blood relation-
ships. Proceedings of the seventh symposium on statistics and the environment. In press.
Executive summary supplied by E. I. DuPont de Nemours & Company to U.S. Environmental
Protection Agency, Environmental Criteria and Assessment Office, Research Triangle Park,
HC.
Brunekreef, B.; Veenstra, S. J.; Biersteker, K.; Boleij, J. S. M. (1981) The Arnhem lead study.
1: Lead uptake by 1- to 3-year-old children living in the vicinity of a secondary lead
smelter in Arnhem, The Netherlands. Environ. Res. 25: 441-448.
Castellino, N.; Aloj, S. (1964) Kinetics of the distribution and excretion of lead in the rat.
Br. J. Ind. Med. 21: 308-314.
Cavalleri, A.; Minoia, C. ; Pozzoli, L.; Baruffini, A. (1978) Determination of plasma lead
levels in normal subjects and in lead-exposed workers. Br. J. Ind. Med. 35: 21-26.
Cerklewski, F. L. (1979) Influence of dietary zinc on lead toxicity during gestation and
lactation in the female rat. J. Nutr. 109: 1703-1709.
Cerklewski, F. L. (1980) Reduction in neonatal lead exposure by supplemental dietary iron
during gestation and lactation in the rat. J. Nutr. 110: 1453-1457.
Cerklewski, F. L.; Forbes, R. M. (1976) Influence of dietary zinc on lead toxicity in the rat.
J. Nutr. 106: 689-696.
Chamberlain, A. C. (1983) Effect of airborne lead on blood lead. Atmos. Environ. 17:
(IN PRESS)
Chamberlain, A. C.; Heard, M. J. (1981) Lead tracers and lead balances. In: Lynam, D. R.;
Piantainida, L. G.; Cole, J. F., eds. Environmental lead. New York, NY: Academic Press;
pp. 175-198.
Chamberlain, A. C.; Heard, M. J.; Little P.; Newton, D.; Wells, A. C.; Wiffen, R. D. (1978)
Investigations into lead from motor vehicles. Harwell, United Kingdom: United Kingdom
Atomic Energy Authority; report no. AERE-R9198.
Chesney, R. W. ; Rosen, J. F. ; Hamstra, A. J.; Smith, C.; Mahaffey, K. ; DeLuca, H. F. (1981)
Absence of seasonal variation in serum concentrations of 1,25-dihydroxyvitamin D despite a
rise in 25-hydroxyvitamin D in summer. J. Clin. Endocrinol. Metab. 53: 139-142.
Chiesura, P. (1970) Escrezione urinaria di cataboliti del piombotetraetile nell'uomo. [Urinary
excretion of tetraethyl lead catabolites in man.] Med. Lav. 61: 437-441.
Chisolm, J. J., Jr. (1981) Dose-effect relationships for lead in young children: evidence in
children for interactions among lead, zinc and iron. In: Lynam, D. R.; Piantanida, L. G;
Cole, J. R., eds. Environmental lead. New York, NY: Academic Press; pp. 1-7.
Chisolm, J. J., Jr.; Barltrop, D. (1979) Recognition and management of children with increased
lead absorption. Arch. Dis. Child. 54: 249-262.
Chisolm, J. J., Jr.; Harrison, H. E. (1956) The exposure of children to lead. Pediatrics 18:
943-958.
I11REF/A 11-165 7/29/83
-------
PRELIMINARY DRAFT
Chisolm, J. J. , Jr.; Mellits, E. D. ; Barrett, M. B. (1976) Interrelationships among blood lead
concentration, quantitative daily ALA-U and urinary lead output following calcium EDTA.
In: Nordberg, G. F. ; ed. Proceedings of third meeting of the subcommittee on the toxico-
logy of metals under the Permanent Commission and International Association on Occupa-
tional Health; November 1974; Tokyo, Japan. Amsterdam, The Netherlands: Elsevier Pub-
lishing Co.; pp. 416-433.
Clarkson, T. W.; Kench, J. E. (1958) Uptake of lead by human erythrocytes in vitro. Biochem. J.
69: 432-439.
Conrad, M. E. ; Barton, J. C. (1978) Factors affecting the absorption and excretion of lead in
the rat. Gastroenterology 74: 731-740.
Cramer, K. ; Goyer, R. A.; Jagenburg, R. ; Wilson, M. H. (1974) Renal ultrastructure, renal
function, and parameters of lead toxicity in workers with different periods of lead ex-
posure. Br. J. Ind. Med. 31: 113-127.
Cremer, J. E. (1959) Biochemical studies on the toxicity of tetraethyl lead and other organo-
lead compounds. Br. J. Ind. Med. 16: 191-199.
Cremer, J. E. ; Callaway, S. (1961) Further studies on the toxicity of some tetra and trialkyl
lead compounds. Br. J. Ind. Med. 18: 277-282.
Dacre, J. C.; Ter Haar, G. L. (1977) Lead levels in tissues from rats fed soils containing
lead. Arch. Environ. Contain. Toxicol. 6: 111-119.
Danscher, G. ; Hall, E. ; Fredens, K. ; Fjerdingstad, E.; Fjerdingstad, E. J. (1975) Heavy metals
in the amygdala of the rat: zinc, lead and copper. Brain Res. 94: 167-172.
Davis, R. K.; Horton, A. W.; Larson, E. E.; Stemmer, K. L. (1963) Inhalation of tetramethyllead
and tetraethyllead: a comparison of the affects in rats and dogs. Arch. Environ. Health 6:
473-479.
Day, J. P.; Fergusson, J. E.; Chee, T. M. (1979) Solubility and potential toxicity of lead in
urban street dust. Bull. Environ. Contam. Toxicol. 23: 497-502.
Day, J. P.; Hart, M. ; Robinson, M. S. (1975) Lead in urban street dusts. Nature (London) 235:
343-345.
Delves, H. T. ; Clayton, B. E.; Carmichael, A.; Bubear, M.; Smith M. (1982) An appraisal of the
analytical significance of tooth-lead measurements as possible indices of environmental
exposure of children to lead. Ann. Clin. Biochem. 19: 329-337.
Der, R.; Fahim, Z.; Hilderbrand, D. ; Fahim, M. (1974) Combined effect of lead and low protein
diet on growth, sexual development, and metabolism in female rats. Res. Commun. Chem.
Pathol. Pharmacol. 9: 723-738.
DeSilva, P. E. (1981) Determination of lead in plasma and studies on its relationship to lead
in erythrocytes. Br. J. Ind. Med. 38: 209-217.
Dobbins, A.; Johnson, D. R.; Nathan, P. (1978) Effect of exposure to lead on maturation of
intestinal iron absorption of rats. J. Toxicol. Environ. Health 4: 541-550.
I11REF/A 11-166 7/29/83
-------
PRELIMINARY DRAFT
Drill, S. ; Konz, J.; Mahar, H.; Morse, M. (1979) The environmental lead problem: an assessment
of lead in drinking water from a multi-media perspective. Washington, DC: U.S. Environ-
mental Protection Agency; EPA report no. EPA-570/9-79-003. Available from: NTIS, Spring-
field, VA; PB 296556.
Duggan, M. J. ; Williams, S. (1977) Lead-in-dust in city streets. Sci. Total Environ. 7: 91-97.
El-Gazzar, R. M.; Finelli, V. N.; Boiano, J.; Petering, H. G. (1978) Influence of dietary zinc
on lead toxicity in rats. Toxicol. Lett. 1: 227-234.
Everson, J.; Patterson, C. C. (1980) "Ultra-clean" isotope dilution/mass spectrometric analyses
for lead in human blood plasma indicate that most reported values are artificially high.
Clin. Chem. (Winston Salem N.C.) 26: 1603-1607.
Facchetti, S. (1979) Isotope study of lead in petrol. In: International conference: management
and control of heavy metals in the environment; September; London, United Kingdom. Edin-
burgh, United Kingdom: CEP Consultants, Ltd.; pp. 95-102.
Facchetti, S.; Geiss, F. (1982) Isotopic lead experiment: status report. Luxembourg: Commission
of the European Communities; Publication no. EUR 8352 EN.
Fairey, f. S.; Gray, J. W., III (1970) Soil lead and pediatric lead poisoning in Charleston,
S.C. J. S.C. Med. Assoc. 66: 79-82.
Fergusson, J. E. ; Hibbard, K. A.; Ting, R. L. H. (1981) Lead in human hair: general survey -
battery factory employees and their families. Environ. Pollut. Ser. B 2: 235-248.
Filippini, L.; Simmler, F. (1980) Blei-Intoxikation durch Schnupftabak. [Lead intoxication due
to taking snuff.] Dtsch. Med. Wochenschr. 105: 1504-1506.
Fischbein, A.; Cohn, J.; Ackerman, G. (1980a) Asbestos, lead, and the family: household risks.
J. Fam. Pract. 10: 989-992.
Fischbein, A.; Nicholson, W. J.; Weisman, I. (1980b) Comparative, lead emissions from conven-
tional and jacketed ammunition. Am. Ind. Hyg. Assoc. J. 41: 525-527.
Fischbein, A.; Rice, C. ; Sarkozi, L.; Kon, S. H. ; Petrocci, M. ; Selikoff, I. J. (1979) Exposure
to lead in firing ranges. J. Am. Med. Assoc. 241: 1141-1144.
Flindt, M. L. H.; King, E.; Walsh, D. B. (1976) Blood lead and erythrocyte 6-aminolevulinic
acid dehydratase levels in Manchester taxi drivers. Br. J. Ind. Med. 33: 79-84.
Forthofer, R. N. (1983) Investigation of nonresponse bias in NHANES II. Am. J. Epidemiol. 117:
IN PRESS.
Fosse, G.; Wesenberg, G. B. R. (1981) Lead, cadmium, zinc and copper in deciduous teeth of
Norwegian children in the pre-industrial age. Int. J. Environ. Stud. 16: 163-170.
Foster, J. D.; Louria, D. B; Stinson, L. (1979) Influence of documented lead poisoning on en-
vironmental modification programs in Newark, New Jersey. Arch. Environ. Health 34: 368-
371.
I11REF/A 11-167 7/29/83
-------
PRELIMINARY DRAFT
Fugas, M. (1977) Biological significance of some metals as air pollutants. Part I: Lead.
Research Triangle Park, NC: U.S. Environmental Protection Agency, Health Effects Research
Laboratory; EPA report no. EPA-600/1-77-041. Available from: NTIS, Springfield, VA; PB
274055.
Fugas, M. ; Wilder, 8.; Paukovic, R. ; Hrsak, J. ; Steiner-Skreb, 0. (1973) Concentration levels
and particle size distribution of lead in the air of an urban and an industrial area as a
basis for the calculation of population exposure. In: Barth, D. ; Berlin, A.; Engel, R.;
Recht, P.; Smeets, J. , eds. Environmental health aspects of lead: proceedings, inter-
national symposium; October 1972; Amsterdam, The Netherlands. Luxembourg: Commission of
the European Communities, Centre for Information
Galke, W. A.; Hammer, D. I.; Keil, J. E.; Lawrence, S. W. (1975) Environmental determinants of
lead burdens in children: international conference on heavy metals in the environment;
October; Toronto, ON, Canada. Toronto: Institute for Environmental Studies; pp. 53-74. See
also: Washington, DC: U.S Environmental Protection Agency; EPA report no. EPA-600/J-78-022
1975. Available from: NTIS, Springfield, VA; PB 283567.
Gallant, A. R. (1975) Testing a subset of the parameters of a nonlinear regression model. J.
Am. Stat. Assoc. 70: 927-932.
Garibaldi, P.; Facchetti, S. ; Quagliardi, A.; Vanini, G. ; Gaddo, P. P.; De Bortoli, M. ;
Gaglione, P. (1975) Petrols additivated with isotopically differentiated lead proposal of
an experiment to estimate the incidence of traffic on the environment pollution by lead -
first experimental results. In: Proceedings, international symposium on recent advances in
the assessment of the health effects of environmental pollution. Vol. 3; June 1974; Paris,
France. Luxembourg: Commission
Gartside, P. S. ; Buncher, C. R. ; Lerner, S. (1982) Relationship of air lead and blood lead for
workers in an automobile factory. Int. Arch. Occup. Environ. Health 50: 1-10.
Gause, D. ; Chase, W. ; Foster, J. ; Louria, D. B. (1977) Reduction in lead levels among children
in Newark. J. Med. Soc. N.J. 74: 958-960.
Gilbert, C.; Tuthill, R. W. ; Calabrese, E. J. ; Peters, H. A. (1979) A comparison of lead
hazards in the housing environment of lead poisoned children versus nonpoisoned controls.
J. Environ. Sci. Health A14: 145-168.
Goldsmith, J. R. (1974) Food chain and health implications of airborne lead. Sacramento, CA:
State of California, Air Resources Board; report no. ARB-R-102-74-36. Available from:
NTIS, Springfield, VA; PB 248745.
Goldsmith, J. R.; Hexter, A. C. (1967) Respiratory exposure to lead: epidemiological and exper-
imental dose-response relationships. Science (Washington D.C.) 158: 132-134.
Grandjean, P.; Nielsen, 0. V.; Shapiro, I. M. (1978) Lead retention in ancient Nubian and con-
temporary populations. J. Environ. Pathol. Toxicol. 2: 781-787.
Graovac-Leposavic, L.; Djuric, D.; Valjarevic V.; Senicar, H.; Senicar, L.; Milic", S.; Delid,
V. (1973) Environmental lead contamination of Meza Valley study on lead exposure of
population. In: Barth, D.; Berlin, A.; Engel, R. ; Recht, P.; Smeets, J., eds. Environ-
mental health aspects of lead: proceedings, international symposium; October 1972; Amster-
dam, The Netherlands. Luxembourg; Commission of the European Communities, Centre for In-
formation and Documentation; pp. 685-705.
I11REF/A 11-168 7/29/83
-------
PRELIMINARY DRAFT
Griffin, T. B.; Coulston, F.; Wills, H.; Russell, J. C. ; Knelson, 0. H. (1975) Clinical studies
of men continuously exposed to airborne particulate lead. In: Griffin, T. B.; Knelson, J.
H., eds. Lead. New York, NY: Academic Press; pp. 221-240. (Coulston, F.; Korte, F., eds.
Environmental quality and safety: supplement v. 2).
Gross, S. B. (1979) Oral and inhalation lead exposures in human subjects (Kehoe balance experi-
ments). New York, NY: Lead Industries Association.
Gross, S. B. (1981) Human oral and inhalation exposures to lead: summary of Kehoe balance ex-
periments. J. Toxicol. Environ. Health 8: 333-377.
Gross, S. B. ; Pfitzer, E. A.; Yeager, 0. W.; Kehoe, R. A. (1975) Lead in human tissues.
Toxicol. Appl. Pharmacol. 32: 638-651.
Hall, W. (1974) Survey plans and data collection and analysis methodologies: results of a pre-
survey for the magnitude and extent of the lead based paint hazard in housing. Washington,
DC: U.S. Department of Commerce, National Bureau of Standards; report no. NBSIR 74-426.
Available from: NTIS, Springfield, VA; COM-74-11074.
Hammer, D. I.; Finklea, J. F.; Hendricks, R. H.; Hinners, T. A.; Riggan, W. B. ; Shy, C. M.
(1972) Trace metals in human hair as a simple epidemiologic monitor of environmental ex-
posure. In: Hemphill, D. D., ed. Trace substances in environmental health - V: [pro-
ceedings of University of Missouri's 5th annual conference on trace substances in environ-
mental health]; June; Columbia, MO. Columbia, MO: University of Missouri-Columbia; pp.
25-38.
Hammond, P. B.; Lerner, S. I.; Hong, C. G. (1980) New perspectives on lead. Am. J. Ind. Med.
1: 401-404.
Hammond, .P. B.; O'Flaherty, E. J.; Gartside, P. S. (1981) The impact of air-lead on blood-lead
in man - a critique of the recent literature. Food Cosmet. Toxicol. 19: 631-638.
Hansen, K. S.; Sharp, F. R. (1978) Gasoline sniffing, lead poisoning, and myoclonus. J. Am.
Med. Assoc. 240: 1375-1376.
Hasselblad, V.; Stead, A. G.; Galke, W. (1980) Analysis of coarsely grouped data from the log-
normal distribution. J. Am. Stat. Assoc. 75: 771-778.
Heyworth, F.; Spickett, J.; Dick, M.; Margetts, B.; Armstrong, B. (1981) Tailings from a lead
mine and lead levels in schoolchildren: a preliminary report. Med. J. Aust. 2: 232-234.
Hopkins, D. R.; Houk, V. N. (1976) Federally-assisted screening projects for childhood lead
poisoning control: the first three years (July 1972-June 1975). Am. J. Public Health 66:
485-486.
Hopper, J. L.; Mathews, J. D. (1983) Extensions to multivariate normal models for pedigree
analysis. II: Modeling for the effect of shared environment in the analysis of variation
in blood lead levels. Am. J. Epidemiol. 117: 344-355.
Horn* J. D. (1976) Significant overestimation of blood lead levels in practice: a consequence
of using the micro method of blood lead analysis. ON, Canada: University of Western
Ontario, (Prepublication draft).
jllREF/A 11-169 7/29/83
-------
PRELIMINARY DRAFT
Hubermont, G.; Buchet, J-P. ; Roels, H.; Lauwerys, R. (1978) Placenta! transfer of lead, mercury
and cadmium in women living in a rural area: importance of drinking water in lead ex-
posure. Int. Arch. Occup. Environ. Health 41: 117-124.
Jacquez, J. A. (1972) Compartmental analysis in biology and medicine: kinetics of distribution
of tracer-labeled materials. New York, NY: Elsevier Publishing Co.
Johanson, W. C., Jr.; Luby, J. P. (1972) A report on a study to determine the blood lead levels
in Dallas children. Dallas, TX: Dallas Health Department.
Johnson, D. E. ; Prevost, R. J. ; Tillery, J. B. ; Thomas, R. E. (1978) The distribution of
cadmium and other metals in human tissue. Research Triangle Park, NC: U.S. Environmental
Protection Agency; EPA report no. EPA-600/1-78-035. Available from: NTIS, Springfield, VA;
PB 285200.
Johnson, D. E.; Prevost, R. J.; Tillery, J. B.; Camann, D. E.; Hosenfeld, J. M. (1976) Baseline
levels of platinum and palladium in human tissue. Research Triangle Park, NC: U.S. En-
vironmental Protection Agency, Health Effects Research Laboratory; EPA report no. EPA-
600/1-76-019. Available from: NTIS, Springfield, VA; PB 251885.
Johnson, D. E.; Tillery, J. B.; Hosenfeld, J. M.; Register, J. W. (1974) Development of analyt-
ical techniques to measure human exposure to fuel additives. Research Triangle Park, NC:
U.S. Environmental Protection Agency, National Environmental Research Center; EPA report
no. EPA-650/1-74-003. Available from: NTIS, Springfield, VA; PB 232124.
Johnson, D. E. ; Tillery, J. B.; Prevost, R. J. (1975) Levels of platinum, palladium, and lead
in populations of southern California. Environ. Health Perspect. 12: 27-33.
Jones, R. D. ; Commins, B. T. ; Cernik, A. A. (1972) Blood lead and carboxyhaemoglobin levels in
London taxi drivers. Lancet 2(7772): 302-303.
Rang, H. K. ; Infante, P. F. ; Carra, J. S. (1983) Determination of blood-lead elimination pat-
terns of primary lead smelter workers. J. Toxicol. Environ. Health, 11: 199-210.
Kaufman, A.; Wiese, W. (1978) Gasoline sniffing leading to increased lead absorption in chil-
dren. Clin. Pediatr. (Philadelphia) 17: 475-477.
Kehoe, R. A. (1961a) The metabolism of lead in man in health and disease: the normal metabolism
of lead. (The Harben lectures, 1960). J. R. Inst. Public Health Hyg. 24: 81-97.
Kehoe, R. A. (1961b) The metabolism of lead in man in health and disease: the metabolism of
lead under abnormal conditions. (The Harben lectures, 1960). J. R. Inst. Public Health
Hyg. 24: 129-143.
Kehoe, R. A. (1961c) The metabolism of lead in man in health and disease: present hygienic
problems relating to the absorption of lead. (The Harben lectures, 1960). J. R. Inst.
Public Health Hyg. 24: 177-203.
King, E. ; Conchie, A.; Hiett, D. ; Milligan, B. (1979) Industrial lead absorption. Ann. Occup.
Hyg. 22: 213-239.
I11REF/A 11-170 7/29/83
-------
PRELIMINARY DRAFT
Knelson, J. H.; Johnson, R. J.; Coulston, F.; Golberg, L. ; Griffin, T. (1973) Kinetics of re-
spiratory lead uptake in humans. In: Barth, D.; Berlin, A.; Engel, R.; Recht, P.; Smeets,
J., eds. Environmental health aspects of lead: proceedings, international symposium;
October 1972; Amsterdam, The Netherlands. Luxembourg; Commission of the European Com-
munities, Centre for Information and Documentation; pp. 391-399.
Landrigan, P. J. ; Baker, E. L. , Jr.; Feldman, R. G.; Cox, D. H.; Eden, K. V.; Orenstein, W. A.;
Mather, J. A.; Yankel, A. J. ; von Lindern, I. H. (1976) Increased lead absorption with
anemia and slowed nerve conduction in children near a lead smelter. J. Pediatr. (St.
Louis) 89: 904-910.
Landrigan, P. J.; Baker, E. L., Jr.; Himmelstein, J. S.; Stein, G. F.; Wedding, J. P.; Straub,
W. E. (1982) Exposure to lead from the Mystic River Bridge: the dilemma of deleading. N.
Engl. J. Med. 306: 673-676.
Landrigan, P. J.; Gehlbach, S. H.; Rosenblum, B. F.; Shoults, J. M.; Candelaria, R. M.;
Barthel, W. F.; Liddle, J. A.; Smrek, A. L.; Staehling, N. W.; Sanders, J. F. (1975)
Epidemic lead absorption near an ore smelter: the role of particulate lead. N. Engl. J.
Med. 292: 123-129.
Lead Development Association. (1973) Medical aspects of lead absorption in industrial pro-
cesses. London, United Kingdom.
Lepow, M. L.; Bruckman, L.; Gillette, M.; Markowitz, S.; Robino, R.; Kapish, J. (1975) Investi-
gations into the source of lead in the environment of urban children. Environ. Res. 10:
415-426.
Lin-Fu, J. S. (1973) Vulnerability of children to lead exposure and toxicity: parts one and
two. N. Engl. J. Med. 289: 1229-1233, 1289-1293.
Lucas, J. M. (1981) Effect of analytical variability on measurements of population blood lead
levels. Am. Ind. Hyg. Assoc. J. 42: 88-96.
Mahaffey, K. R.; Annest, J. L.; Barbano, H. E.; Murphy, R. S. (1979) Preliminary analysis of
blood lead concentrations for children and adults: HANES II, 1976-1978. In: Hemphill, D.
D., ed. Trace substances in environmental health—XIII: [proceedings of University of
Missouri's 13th annual conference on trace substances in environmental health]; June;
Columbia, MO. Columbia, MO: University of Missouri-Columbia; pp. 37-51.
Mahaffey, K. R.; Annest, J. L.; Roberts, J.; Murphy, R. S. (1982) National estimates of blood
lead levels: United States, 1976-1980: association with selected demographic and socio-
economic factors. N. Engl. J. Med. 307: 573-579.
Manton, W. I. (1977) Sources of lead in blood: identification by stable isotopes. Arch.
Environ. Health 32: 149-159.
Martin, A. E.; Fairweather, F. A.; Buxton, R. St. J.; Roots, L. M. (1975) Recent epidemio-
logical studies on environmental lead of industrial origin. In: Proceedings, international
symposium: recent advances in the assessment of the health effects of environmental pol-
lution, vol. 2; June 1974; Paris, France. Luxembourg: Commission of the European Com-
munities; pp. 1113-1122.
I11REF/A 11-171 7/29/83
-------
PRELIMINARY DRAFT
McDowell, A.; Engel, A.; Massey, J. T.; Maurer (1981) Plan and operation of the second national
health and nutrition examination survey, 1976-80: Programs and
collection procedures. DHHS Publication no. PHS 81-1217. National Center for
Health Statistics. (Vital and Health Statistics: Series 1, no. 15).
Mclntire, M. S.; Angle, C. R. (1973) Air lead/blood lead in G-6-PD deficient black school chil-
dren. In: Barth, D.; Berlin, A.; Engel, R.; Recht, P.; Smeets, J. ,
eds. Environmental health aspects of lead: proceedings, international symposium;
October 1972; Amsterdam, The Netherlands. Luxembourg: Commission of the European
Communities, Centre for Information and Documentation; pp. 749-761.
Milic, S. ; Stankovic, M.; Delid', V.; Djordjevic, V. (1973) Biochemical parameters in evaluation
of environmental lead exposure. In: Barth, D. ; Berlin, A.;
Engel, R. ; Recht, P.; Smeets, J., eds. Environmental health aspects of lead:
proceedings, international symposium; October 1972; Amsterdam, The Netherlands.
Luxembourg; Commission of the European Communities, Centre for Information and
Documentation; pp. 561-568.
Moore, M. R. (1977) Lead in drinking water in soft water areas—health hazards. Sci. Total
Environ. 7: 109-115.
Moore, M. R.; Goldberg, A.; Fyfe, W. M.; Low, R. A.; Richards, W. N. (1981a) Lead in water in
Glasgow - a story of success. Scott. Med. J. 26: 354-355.
Moore, M. R.; Goldberg, A.; Fyfe, W. M. ; Richards, W. N. (1981b) Maternal lead levels after
alterations to water supply [Letter]. Lancet 2(8239): 203-204.
Moore, M. R. ; Goldberg, A.; Meredith, P. A.; Lees, R. ; Low, R. A.; Pocock, S. J. (1979) The
contribution of drinking water lead to maternal blood lead
concentrations. Clin. Chim. Acta 95: 129-133.
Morse, D. L. ; Watson, W. N. ; Housworth, J.; Witherell, L. E.; Landrigan, P. J. (1979) Exposure
of children to lead in drinking water. Am. J. Public Health
69: 711-712.
Morton, D. E.; Saah, A. J.; Silberg, S. L.; Owens, W. L.; Roberts, M. A.; Saah, M. D. (1982)
Lead absorption in children of employees in a lead-related industry.
Am. J. Epidemiol. 115: 549-555.
\
Nathanson, B.; Nudelman, H. (1980) Ambient lead concentrations in New York City and their
health implications. Bull. N.Y. Acad. Med. 56: 866-875.
National Academy of Sciences; National Research Council. (1976) Recommendations for the pre-
vention of lead poisoning in children. Washington, DC: National
Academy of Sciences. Available from: NTIS, Springfield, VA; PB 257645.
National Bureau of Standards. (1973) Lead paint survey sampling plan and preliminary screening.
NBS report 10958. Available from: National Bureau of
Standards, U.S. Department of Commerce, Gaithersburg, MD.
Neri, L. C.; Johansen, H. L.; Schmitt, N.; Pagan, R. T.; Hewitt, D. (1978) Blood lead levels in
children in two British Columbia communities. In: Hemphill, 0. 0.,
ed. Trace substances in environmental health-XII: [proceedings of University of
Missouri's 12th annual conference on trace substances in environmental health];
June; Columbia, MO. Columbia, MO: University of Missouri-Columbia; pp. 403-410.
I11REF/A 11-172 7/29/83
-------
PRELIMINARY DRAFT
Nordman, C. H. (1975) Environmental lead exposure in Finland: a study on selected population
groups. Helsinki, Finland: University of Helsinki.
Nordman, C. H.; Hernberg, S. ; Nikkanen, J.; Ryhanen, A. (1973) Blood lead levels and erythro-
cyte 6-aminolevulinic acid dehydratase activity in people living around a secondary lead
smelter. Work Environ. Health 10: 19-25.
Nutrition Foundation, Inc. (1982) Assessment of the safety of lead and lead salts in food: a
report of the Nutrition Foundation's Expert Advisory Committee. Washington, DC. The
Nutrition Foundation.
O'Flaherty, E. J.; Hammond, P. B.; Lerner, S. I. (1982) Dependence of apparent blood lead
half-life on the length of previous lead exposure in humans. Fundam. Appl. Toxicol. 2:
49-54.
Ontario Ministry of the Environment. (1975) Report of the Lead Data Analysis Task Force.
Toronto, ON, Canada: Ontario Ministry of the Environment.
Oxley, G. R. (1982) Blood lead concentrations: apparent reduction over approximately one de-
cade. Int. Arch. Occup. Environ. Health 49: 341-343.
Piomelli, S.; Corash, L.; Corash, M. B.; Seaman, C.; Mushak, P.; Glover, B.; Padgett, R. (1980)
Blood lead concentrations in a remote Himalayan population. Science (Washington D.C.) 210:
1135-1137.
Pocock, S. J.; Shaper, A. G.; Walker, M. ; Wale, C. J.; Clayton, B.; Delves, T.; Lacey, R. F.;
Packham, R. F.; Powell, P. (1983) The effects of tap water lead, water hardness, alcohol,
and cigarettes on blood lead concentrations. J. Epidemic!. Comra. Health 37: 1-7.
Poole, C.; Smythe, L. E. (1980) Blood lead levels in Papua, New Guinea children living in a re-
mote area. Sci. Total Environ. 15: 17-24.
Popovac, D.; Graziano, J.; Seaman, C.; Kaul, B.; Colakovic, B.; Popovac, R.; Osmani, I.;
Haxhiu, M.; Begraca, M.; Bozovic, Z.; Mikic, M. (1982) Elevated blood lead in a population
near a lead smelter in Kosovo, Yugoslavia. Arch. Environ. Health 37: 19-23.
Rabinowitz, M. B.; Kopple, J. D.; Wetherill, G. W. (1980) Effect of food intake and fasting on
gastrointestinal lead absorption in humans. Am. J. Clin. Nutr. 33: 1784-1788.
Rabinowitz, M. B.; Needleman, H. L. (1982) Temporal trends in the lead concentrations of
umbilical cord blood. Science (Washington D.C.) 216: 1429-1432.
Rabinowitz, M. B.; Wetherill, G. W.; Kopple, J. D. (1977) Magnitude of lead intake from respir-
ation by normal man. J. Lab. Clin. Med. 90: 238-248.
Rabinowitz, M. B.; Wetherill, G. W.; Kopple, J. D. (1976) Kinetic analysis of lead metabolism
in healthy humans. J. Clin. Invest. 58: 260-270.
Rabinowitz, M.; Wetherill, G. W.; Kopple, J. 0. (1974) Studies of human lead metabolism by
using stable isotope tracers. Environ. Health Perspect. 7: 145-152.
Reece, R. M.; Reed, A. J.; Clark, C. S.; Angoff, R.; Casey, K. R.; Challop, R. S.; McCabe, E.
of wall paint by X-ray fluorescence. Am. J. Dis. Child. 124: 6500-6502.
I11REF/A 11-173 7/29/83
-------
PRELIMINARY DRAFT
Rice, C.; Fischbein, A.; Lilis, R.; Sarkozi, L.; Kon, S.; Selikoff, I. J. (1978) Lead contami-
nation in the homes of employees of secondary lead smelters. Environ. Res. 15: 375-380.
Roberts, T. M.; Hutchinson, T. C. ; Paciga, J.; Chattopadhyay, A.; Jervis, R. E.; Van Loon, 0.;
Parkinson, D. K. (1974) Lead contamination around secondary smelters: estimation of
dispersal and accumulation by humans. Science (Washington D.C.) 186: 1120-1123.
Roels, H. A.; Buchet, J-P.; Lauwerys, R. R. ; Bruaux, P.; Claeys-Thoreau, F.; Lafontaine, A.;
Verduyn, G. (1980) Exposure to lead by the oral and the pulmonary routes of children
living in the vicinity of a primary lead smelter. Environ. Res. 22: 81-94.
Roels, H. A.; Buchet, J-P.; Lauwerys, R. ; Bruaux, P.; Claeys-Thoreau, F.; Lafontaine, A.;
van Overschelde, J. ; Verduyn, G. (1978) Lead and cadmium absorption among children near a
nonferrous metal plant: a follow-up study of a test case. Environ. Res. 15: 290-308.
Roels, H.; Buchet, J-P.; Lauwerys, R.; Hubermont, G.; Bruaux, P.; Claeys-Thoreau, F. ;
Lafontaine, A.; Van Overschelde, J. (1976) Impact of air pollution by lead on the heme
biosynthetic pathway in school-age children. Arch. Environ. Health 31: 310-316.
Ryu, J. E. ; Ziegler, E. E. ; Nelson, S. E.; Fomon, S. J. (1983) Dietary intake of lead and
blood lead in early infancy. Pediatr. Res: (IN PRESS)
Saltzman, B. E. ; Yeager, D. W. ; Meiners, B. G. (1983) Reproducibility and quality control in
the analysis of biological samples for lead and mercury. Am. Ind. Hyg. Assoc. J. 44:
263-267.
Sayre, J. W.; Charney, E.; Vostal, J.; Pless, I. B. (1974) House and hand dust as a potential
source of childhood lead exposure. Am. J. Dis. Child. 127: 167-170.
Schlegel, H.; Kufner, G. (1979) Long-term observation of biochemical effects of lead in human
experiments. J. Clin. Chem. Clin. Biochem. 17: 225-233.
Schubert, J.; Brodsky, A.; Tyler, S. (1967) The log-normal function as a stochastic model of
the distribution of strontium-90 and other fission products in humans. Health Phys. 13:
1187-1204.
Searle, C. E.; Harnden, D. G. (1979) Lead in hair-dye preparations [Letter]. Lancet 2(8151):
1070.
Shaper, A. G. ; Pocock, S. J. ; Walker, M.; Wale, C. J. ; Clayton, B.; Delves, H. T. ; Hinks, L.
(1982) Effects of alcohol and smoking on blood lead in middle-aged British men. Br. Med.
J. 284(6312): 299-302.
Shapiro, I. M. •, Mitchell, G.; Davidson, I.; Katz, S. H. (1975) The lead content of teeth: evi-
dence establishing new minimal levels of exposure in a living preindustrialized human
population. Arch. Environ. Health 30: 483-486.
Shellshear, I. D. (1973) Lead poisoning in childhood: a case report with environmental implica-
tions. N.Z. Med. J. 78: 251-254.
Shellshear, I. D.; Jordan, L. D. ; Hogan, D. J.; Shannon, F. T. (1975) Environmental lead ex-
posure in Christchurch children: soil lead a potential hazard. N.Z. Med. J. 81: 382-386.
I11REF/A 11-174 7/29/83
-------
PRELIMINARY DRAFT
Sherlock, J. ; Smart, G.; Forbes, G. I.; Moore, M. R.; Patterson, W. J.; Richards, W. N.;
Wilson, T. S. (1982) Assessment of lead intakes and dose-response for a population in Ayr
exposed to a plumbsolvent water supply. Hum. Toxicol. 1: 115-122.
Shier, D. R.; Hall, W. G. (1977) Analysis of housing data collected in a lead-based paint
survey in Pittsburgh, Pennsylvania: part I. Washington, DC: National Bureau of Standards.
Report no. NBSIR 77-1250. Available from: National Bureau of Standards, Washington, DC.
^. w /iqan^ jjber den Zusammenhang von Luftbleikonzentration und Bleigehalt des Blutes von
'AnttohWnV'uird Berufstatigen im Kerngebiet einer Grossstadt (Blutbleistudie Frankfurt). I:
Versuchsanlage und differenzprllfung. [On the correlation between air lead concentration
and blood lead level of employees in the inner city (Frankfurt blood lead study). I: Study
design and discrepancy trial.] Int. Arch. Occup. Environ. Health 47: 93-118.
Sinn, W. (1981) Relationship between lead concentration in the air and blood lead levels of
people living and working in the center of a city (Frankfurt blood lead study). II: Cor-
relations and conclusions. Int. Arch. Occup. Environ. Health 48: 1-23.
Snee, R. D. (1981) Evaluation of studies of the relationship between blood lead and air lead.
Int. Arch. Occup. Environ. Health 48: 219-242.
Snee, R. D. (1982a) Development of an air quality standard for lead from community studies.
Environ. Sci. Techno!. 16: 241-246.
Snee, R. D. (1982b) Models for the relationship between blood lead and air lead. Int. Arch.
Occup. Environ. Health 50: 303-319.
Snee, R. D. (1982c) Silver Valley lead study: further analysis of the relationship between
blood lead and air lead. J. Air Pollut. Control Assoc. 32: 170-175.
Snee, R. D.; Pfeifer, C. (1983) (submission to Docket, Jan. 16) Available for inspection at:
U.S. Environmental Protection Agency, Environmental Criteria and Assessment Office,
Research Triangle Park, NC.
Stark, A. D.; Quah, R. F.; Meigs, J. W.; DeLouise, E. R. (1982) The relationship of environ-
mental lead to blood-lead levels in children. Environ. Res. 27: 372-383.
Steenhout, A. (1982) Kinetics of lead storage on teeth and bones: an epidemiologic approach.
Arch. Environ. Health 37: 224-231.
Stephens, R. (1981) Human exposure to lead from motor vehicle emissions. Int. J. Environ. Stud.
17: 73-83.
Stuik, E. J. (1974) Biological response of male and female volunteers to inorganic lead. Int.
Arch. Arbeitsmed. 33: 83-97.
Tepper, L. B.; Levin, L. S. (1975) A survey of air and population lead levels in selected
American communities. In: Griffin, T. B. ; Knelson, J. H. , eds. Lead. New York, NY:
Academic Press; pp. 152-195. (Coulston, F.; Korte, F. , eds. Environmental quality and
safety: suppl. v. 2).
Ter Haar, G. (1981) Lead debate [Letter]. Environ. Sci. Technol. 15: 722. Ter Haar, G.; Aronow,
R. (1974) New information on lead in dirt and dust as related to the childhood lead
problem. Environ. Health Perspect. 7: 83-89.
I11REF/A 11-175 7/29/83
-------
PRELIMINARY DRAFT
Ter Haar, G.; Chadzynski, L. (1979) An investigation of elevated blood lead levels in Detroit
children. Arch. Environ. Health 34: 145-150.
Thomas, H. F. (1980) Domestic water usage and blood lead levels. Public Health London 94:
294-295.
Thomas, H. F.; Elwood, P. C.; Welsby, E.; St. Leger, A. S. (1979) Relationship of blood lead in
women and children to domestic water lead. Nature (London) 282: 712-713.
Thomas, H. V.; Milmore, B. K.; Heidbreder, G. A.; Kogan, B. A. (1967) Blood lead of persons
living near freeways. Arch. Environ. Health 15: 695-702.
Tsuchiya, K. ; Sugita, M. ; Seki, Y. ; Kobayashi, Y. ; Hori, M. ; Park, Ch. B. (1975) Study of lead
concentration in atmosphere and population in Japan. In: Griffin, T. B.; Knelson, J. H. ,
eds. Lead. New York, NY: Academic Press; pp. 95-147. (Coulston, F. ; Korte, F. , eds. En-
vironmental quality and safety: suppl. v. 2).
Tyler, R. L. (1970) Philadelphia combats "silent epidemic" in the "ghetto" lead poisoning
control. J. Environ. Health 33: 64-71.
Tyrer, F. H. (1977) A cautionary tale of a lead hazard in recycling of waste. J. Soc. Occup.
Med. 27: 26-30.
U.S. Centers for Disease Control. (1973) Human lead absorption - Texas. Morbidity and Mortality
Weekly Report 22: 405-407.
U.S. Centers for Disease Control. (1977a) Surveillance of childhood lead poisoning - United
States. Morbidity and Mortality Weekly Report 26: 49.
U.S. Centers for Disease Control. (1977b) Increased lead absorption in children of lead workers
- Vermont. Morbidity and Mortality Weekly Report 26: 61-62.
U.S. Centers for Disease Control. (1979) Surveillance of childhood lead poisoning - United
States. Morbidity and Mortality Weekly Report 28: 117-118.
U.S. Centers for Disease Control. (1980) Surveillance of childhood lead poisoning - United
States. Morbidity and Mortality Weekly Report 29: 170-171.
U.S. Centers for Disease Control. (1982a) Surveillance of childhood lead poisoning - United
States. Morbidity and Mortality Weekly Report 31: 118-119.
U.S. Centers for Disease Control. (1982b) Blood-lead levels in U.S. population. Morbidity and
Mortality Weekly Report 31: 132-134.
U.S. Centers for Disease Control. (1982c) Surveillance of childhood lead poisoning - United
States. Morbidity and Mortality Weekly Report 31: 118-119.
U.S. Consumer Product Safety Commission. (1974) A report to Congress in compliance with the
lead based paint poisoning prevention act, as amended (P.L. 93-151). Washington, DC: U.S.
Consumer Product Safety Commission. Available from: NTIS, Springfield, VA; PB 245225.
U.S. Environmental Protection Agency. (1972) Helena Valley, Montana, area environmental pol-
lution study. Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of
Air Program; Office of Air Programs publication no. AP-91. Available from: NTIS, Spring-
field, VA; PB 207126.
I11REF/A 11-176 7/29/83
-------
PRELIMINARY DRAFT
U.S. Environmental Protection Agency. (1980) Ambient water quality criteria for lead.
Washington, DC: U.S. Environmental Protection Agency; EPA report no. EPA 440/5-80-057.
Available from: NTIS, Springfield, VA; PB81-117681.
United Kingdom Central Directorate on Environmental Pollution. (1982) The Glasgow duplicate
diet study (1979/1980): a joint survey for the Department of the Environment and the
Ministry of Agriculture Fisheries and Food. London, United Kingdom: Her Majesty's
Stationery Office; pollution report no. 11.
Urban, W. D. (1976) Analysis of blood lead levels of children surveyed in Pittsburgh, Pennsy-
lvania: analytical methodologies and summary results. Washington, DC: U.S. Department of
Commerce, National Bureau of Standards; report no. NBSIR 76-1024. Available from: NTIS,
Springfield, VA; PB 255876.
von Lindern, I.; Yankel, A. 0. (1976) Presentation to the Shoshone heavy metals project com-
mittee (Seattle, WA; September 1974). In: Shoshone lead health project: work summary.
Boise, ID: Idaho Department of Health & Welfare; pp. 73-89.
Vostal, J. J. ; Taves, E. ; Sayre, J. W.; Charney, E. (1974) Lead analysis of the house dust: a
method for the detection of another source of lead exposure in inner city children.
Environ. Health Perspect. 7: 91-97.
Waldron, H. A. (1975) Lead levels in blood of residents near the M6-A38(M) interchange,
Birmingham. Nature (London) 253: 345-346.
Walter, S. D.; Yankel, A. J.; von Lindern, I. H. (1980) Age-specific risk factors for lead
absorption in children. Arch. Environ. Health 35: 53-58.
Watson, W. N.; Witherell, L. E.; Giguere, G. C. (1978) Increased lead absorption in children of
workers in a lead storage battery plant. J. Occup. Med. 20: 759-761.
Wedeen, R. P.; Mallik, D. K.; Batuman, V.; Bogden, J. D. (1978) Geophagic lead nephropathy:
case report. Environ. Res. 17: 409-415.
Wigle, D. T.; Charlebois, E. J. (1978) Electric kettles as a source of human lead exposure.
Arch. Environ. Health 33: 72-78.
Williams, M. K.; King, E.; Walford, J. (1969) An investigation of lead absorption in an
electric accumulator factory with the use of personal samplers. Br. J. Ind. Med. 26: 202-
216.
Worth D.; Matranga, A.; Lieberman, M.; DeVos, E.; Karelekas, P.; Ryan, C.; Craun, G. (1981)
Lead in drinking water: the contribution of household tap water to blood lead levels.
In: Lynam, D. R.; Piantanida, L. G.; Cole, J. F., eds. Environmental lead: proceedings
of the second international symposium on environmental lead research; December 1978; Cin-
cinnati, OH. New York, NY: Academic Press; pp. 199-225.
Yankel, A. J.; von Lindern, I. H.; Walter, S. D. (1977) The Silver Valley lead study: the re-
lationship of childhood lead poisoning and environmental exposure. J. Air Pollut. Control
Assoc. 27: 763-767.
Zielhuis, R. L.; del Castilho, P.; Herber, R. F. M. ; Wibowo, A. A. E.; Salle, H. J. A. (1979)
Concentrations of lead and other metals in blood of two and three year-old children living
near a secondary smelter. Int. Arch. Occup. Environ. Health 42: 231-239.
I11REF/A 11-177 7/29/83
-------
PRELIMINARY DRAFT
APPENDIX 11A
COMPARTMENTAL ANALYSIS
Many authors have noted that under conditions of constant lead exposure, blood lead con-
centrations change from one level to another apparent equilibrium level over a period of
several months. A mathematical model is helpful in estimating the new apparent equilibrium
level even when the duration of the experiment is not sufficiently long for this equilibrium
level to have been achieved. The model assumes that lead in the body ,is held in some number
of homogeneous and well-mixed pools or compartments. The compartments have similar kinetic
properties and may or may not correspond to identifiable organ systems. In a linear kinetic
model it is assumed that the rate of change of the mass of lead in compartment i at time t,
denoted X.(t), is a linear function of the mass of lead 1n each compartment. Denote the frac-
tional rate of transfer of lead into compartment i from compartment j by K. . (fraction per
day), and let I.(t) be the total external lead input into compartment i at time t in units
such as ug/day. The elimination rate from compartment i is denoted K... The compartmental
model is:
dX.(t)/dt = I.
for each of the n compartments. If the inputs are all constant, then each X.(t) is the sum of
(at most) n exponential functions of time (see for example, Jacquez, 1972).
For the one-compartment model:
dXi(t)/dt = I1 - KQ1 Xx(t)
with an initial lead burden X^O) at time 0,
Xx(t) = X1(0) exp(-KQ1t) * Id1/K01) (l-exp(-K01t)]
The mass of lead at equilibrium is I-I/KQ-I M9- We may think of this pool as "blood lead". If
the pool has volume V, then the equilibrium concentration is Ii/Kni V, (jg/dl. Intake from
several pathways will have the form:
I1 = AI (Pb-Air) + A2 (Pb-Diet)+ ' ' '
so that the long term concentration is
VKoi vi= (YKoiV Pb'Air
OUPll/C 11A-1 7/29/83
-------
PRELIMINARY DRAFT
The inhalation coefficient is p = A,/!^.^. The blood lead half-life is 0.693/KQ1.
Models with two or more compartments will still have equilibrium concentrations in blood
and other compartments that are proportional to the total lead intake, and thus increase
linearly with increasing concentrations in air, dust, and diet. The relationship between the
exponential parameters and the fractional transfer coefficients will be much more complicated,
however.
Models with two or three pools have been fitted by Rabinowitz et a1-. (1976, 1977) and by
Batschelet et al. (1979). The pools are tentatively identified as mainly blood, soft tissue
and bone. But as noted in Section 11.4.1.1, the "blood" pool is much larger than the volume
of blood itself, and so it is convenient to think of this as the effective volume of distri-
bution for pool 1. A five-pool model has been proposed by Bernard (1977), whose pools are
mainly blood, liver, kidney, soft bones and hard bone.
The major conclusion of this Appendix is that linear kinetic mechanisms imply linear
relationships between blood lead and lead concentrations in environmental media. Any extended
discussion of nonlinear kinetic mechanisms is premature at this point, but it is of some
interest that even simple nonlinear kinetic models produce plausible nonlinear blood lead vs.
concentration relationships. For example, if the rate of blood lead excretion into urine or
storage "permanently" in bone increases linearly with blood lead, then at high blood lead
levels, blood increases only as the square root of lead intake. Let M denote the mass of lead
in pool 1 at which excretion rate doubles. Then:
dX1(t)/dt = Ij_ - KQ1(1 - X1(t)/M1)X1(t)
has an equilibrium level:
Xx - Mj(V 1 + 4VK01M1 " 1)/2
This is approximately linear in intake I when I- is small, but a square root function of in-
take when it is large. Other plausible models can be constructed.
DUP11/C 11A-2 7/29/83
-------
PRELIMINARY DRAFT
APPENDIX 11B
FITTING CURVES TO BLOOD LEAD DATA
The relationship between blood lead and the concentrations of lead in various environ-
mental media is a principal concern of this chapter. It is generally accepted that the geo-
metric mean blood lead is some function, f, of the concentration of air lead and of lead in
diet, dust, soil and other media. It has been observed that blood lead levels have a highly
skewed distribution even for populations with relatively homogeneous exposure, and that the
variability in blood lead is roughly proportional to the geometric mean blood lead or to the
arithmetic mean (constant coefficient of variation). Thus, instead of the usual model in
which random variations are normally distributed, a model is assumed here in which the random
deviations are multiplicative and lognormally distributed with geometric mean 1 and geometric
standard deviation (GSD) e°. The model is written
Pb-Blood = f (Pb-Air, etc.) eoz
where z is a random variable with mean 0 and standard deviation 1. It has a Gaussian or
normal distribution. The model is fitted to data in logarithmic form
In(Pb-Blood) = In (f)
even when f is assumed to be a linear function, e.g.,
f = P Pb-Air + BQ + p1 Pb-Dust + ...
The nonlinear function, fitted by most authors (e.g., Snee, 1982b), is a power function with
shape parameter X,
f = (p Pb-Air + P0 + PJ Pb"Dust + •••)*'
These functions can all be fitted to data using nonlinear regression techniques. Even when
the nonlinear shape parameter X has a small statistical uncertainty or standard error as-
sociated with it, a highly variable data set may not clearly distinguish the linear function
(X = 1) from a nonlinear function (X 1- 1). In particular, for the Azar data set, the residual
sum of squares is shown as a function of the shape parameter \, in Figure 11B-1. When only a
PB11D/D 11B-1 7/29/83
-------
PRELIMINARY DRAFT
9.3
[2 9.1
DC
<
o
<» 9.0
0 8.9
*
Q
55 flfl
LU O.O
CC
8.7
8.6
I I
MINIMUM SIGNIFICANT
DIFFERENCE FOR 1 DF
I I I I
I I
SSE FOR In (Pb-Blood) = A In (p Pb-Air + I/). C.)
A = 0.26
MINIMUM SIGNIFICANT
DIFFERENCE FOR 5 DF
SSE FOR In (Pb-Blood) = A In (/? Pb-Air+ 1/3. C.+Z/J'. C. Age)
1 J J 1 J J
i -
I I I
I I I I
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
POWER EXPONENT. A
Figure 11 B-1. Residual sum of squares for nonlinear regression models for Azar
data (N = 149).
PB11D/D
11B-2
7/29/83
-------
PRELIMINARY DRAFT
separate intercept (background) is assumed for each subpopulation, the best choice is A. =
0 26; but when age is also used as a covariate for each subpopulation, then the linear model
is better. However, the approximate size of the difference, in residual sum of squares
required to decide at the 5 percent significance level that a nonlinear model is better (or
worse) than a linear model, is larger than the observed difference in sum of squares for any
A>0.2 (Gallant, 1975). Therefore a linear model is used unless evidence of nonlinearity is
very strong, as with some of Kehoe's studies and the Silver Valley/Kellogg study. Non-
linearity is detectable only when blood lead is high (much above 35 or 40 ug/dl), and intake
is high, e.g., air lead much above 10 ug/m3. Additional research is needed on the relation-
ship between lead levels and lead intake from all environmental pathways.
The "background" or intercept term p,, in most models requires some comment. As the
Wanton and Italian lead isotope studies show, lead added to a regional environment by combus-
tion of gasoline accumulates a large non-inhalation component even after only 2 years (see
Figure 11-26). .The non-inhalation contribution in the Turin region was nearly independent of
location (air lead). It is not possible to assign causes, e.g., ingestion of food, dust, or
water by adults, so no direct extrapolation to U.S. populations is possible at this time due
to unknown differences in non-air exposures between the U.S. and Italy. It is probable that
the non-inhalation contribution to blood lead increases with time as lead accumulates in the
environment. After many years, one might obtain a figure like Figure 11B-2. Another concept
is that such a curve should predict zero blood lead increase at zero air lead. If the blood
lead curve is forced to pass through 0 when air lead = 0, a nonlinear curve is required. It
has been concluded that a positive intercept term is needed to account for intake from
accumulated lead in the environment, which precludes fully logarithmic models such as
In (Pb-Blood) = In (£Q) + p In (Pb-Air) + PX In (Pb-Dust) + ...
It must be acknowledged that such models may provide useful interpolations over a range of air
lead levels; e.g., the Goldsmith-Hexter equation predicts blood lead 3.4 ug/dl at an air lead
<0.004 ug/m3 in the Nepalese subjects in Piomelli et al. (1980).
The final concern is that the intercept term may represent indirect sources of lead expo-
sure that include previous air lead exposures. To the extent that present and previous air
lead exposures are correlated, the intercept or background term may introduce apparent curvi-
linearities in the population studies of inhalation. The magnitude of this effect is unknown.
PB11D/D 11B-3 7/29/83
-------
PRELIMINARY DRAFT
O
I
O
O
O
TOTAL CONTRIBUTION Of AIR LEAD AFTER
LONG INTERVAL OF EXPOSURE AND DEPOSITION
NON-INHALATION
BACKGROUND
CONCENTRATION
AFTER LONG
INTERVAL
OF AIR LEAD
EXPOSURE AND
DEPOSITION ^
TOTAL CONTRIBUTION OF AIR
LEAD AFTER SHORT INTERVAL
OF EXPOSURE AND DEPOSITION
NON-INHALATION
BACKGROUND
CONCENTRATION
AFTER SHORT
INTERVAL OF
AIR LEAD
EXPOSURE AND
DEPOSITION *>
DIRECT INHALATION
OF AIR LEAD FROM
CURRENT EXPOSURE
AIR LEAD CONCENTRATION
Rgure 11 B-2. Hypothetical relationship between blood lead and air lead by
inhalation and non-inhalation.
PB11D/D
11B-4
7/29/83
-------
PRELIMINARY DRAFT
APPENDIX 11C
ESTIMATION OF GASOLINE LEAD CONTRIBUTIONS TO ADULT
BLOOD LEAD BURDENS BASED ON ILE STUDY RESULTS
As discussed in Chapter 11 (pp. 11-118 to 11-123) the results of the Isotopic Lead Ex-
periment (ILE) carried out in Northern Italy provide one basis by which to estimate contribu-
tions of lead in gasoline to blood lead burdens of populations exposed in the ILE study area.
Figures 1 to 5 of this appendix, reprinted from the ILE Status Report (1982) illustrate
changes in isotopic lead (206/207) ratios for 35 adult subjects, for whom repeated measure-
ments were obtained over time during the ILE study. The percent of total blood lead in those
subjects contributed by Australian lead-labelled gasoline (petrol) used in automotive vehicles
in the ILE study area was estimated by the approach reprinted below verbatim from Appendix 17
of the ILE Status Report (1982):
The main purpose of the ILE project was the determination of the contribution of petrol
lead to total lead in blood. A rough value for the fraction of petrol lead in blood can be
derived from the following equations:
R! X + f (1-X) = R1 (1)
R2 X + f (1-X) = R" (11)
each of them referring to a given time at which equilibrium conditions hold.
R1 and R" represent the blood lead isotopic ratios measured at each of the two times; if
R- and R2 represent the local petrol lead isotopic ratios measured at the same times, X is the
fraction of local petrol lead in blood due to petrols affected by the change in the lead iso-
topic ratio, irrespective of its pathway to the blood i.e. by inhalation and ingestion (e.g.
from petrol lead fallout). The term (1-X) represents the fraction of the sum of all other
external sources of lead in the blood (any «other» petrol lead included), factor f being the
unknown isotopic ratio of the mixture of these sources. It is assumed that X and f remained
constant over the period of the experiment, which implies a reasonable constancy of both the
lead contributing sources in the test areas and the living habits which, in practice, might
not be entirely the case.
Data from individuals sampled at the initial and final equilibrium phases of the ILE
study together with petrol lead isotopic ratios measured at the same times, would ideally pro-
vide a means to estimate X for Turin and countryside adults. However, for practical reasons,
calculations were based on the initial and final data of the subjects whose first sampling was
DUP11/B 11C-1 7/29/83
-------
PRELIMINARY DRAFT
done not later than 1975 and the final one during phase 2. Their complete follow-up data are
shown in Table 27. For R.^ and R2 the values measured in the phases 0 and 2 of ILE were used
(R^ = 1.186, R2 = 1.060). Hence, as averages of the individual X and f results, we obtain:
Turin
countryside
<25 km
countryside
>25 km
X = 0.237 ± 0.054
fj = 1.1560 ± 0.0033
X? = 0.125 ± 0.071
f2 = 1.1542 ± 0.0036
X. = 0.110 ± 0.058
f = 1.1576 ± 0.0019
i.e 24%
i.e. 12%
i.e 11%
Pb206/Pb207
116-
115-
114-
113-
112-
PhaseO
Phase I
Phase 2
74 I 75 I 76 ' 77 ; 78 ' 79 I 80
Fig. 1. Individual values of blood Pb-206/Pb-207 ratio for subjects follow-up in Turin (12 subjects)
DUP11/B
11C-2
7/29/83
-------
PRELIMINARY DRAFT
Pb206/Pb207
116-
115-
114-
113-
1.12-
PhaseO
Phase 1
Phase 2
I I
74
75
76
77 ' 78
79
80
Fig. 2. Individual values in blood Pb-206/Pb-207 ratio for subjects follow-up in Castagneto (4 subjects)
DUP11/B
Pb206/Pb207 2
1.16-
1.15-
114-
1.13
1.12
--0--DRUENTO
—•— FIANO
Phase 0
.0 J
•o z
Phase 1
Phase 2
74
75
76
77
78
79
80
Fig. 3. Individual values of blood Pb-206/Pb-207 ratio fot subjects follow-up in Dtuento and Fiano (6 subjecu)
11C-3 7/29/83
-------
PRELIMINARY DRAFT
I
Pb206/Pb207
116
115-
114-
1.13-
112-
—•— SANTENA
—o~ NOLE
Phase 0
Phase 1
Phut 2
74 75 ' 76 77 78 ! 79 ! 80
Fig. 4. Individual values of blood Pb-206/Pb-207 ratio for subject! follow-up in Note and Santena (9 subjects)
Pb206/Pb207
115-
1.14
1.13-
112
Phase 0
Pkasel
Pkase2
• 2
• I ' ' ' '
76
11 •
77
I
74 I 75 I 76 I 77 ' 78 ' 79
Fig. 5. Individual values of blood Pb-206/Pb-207 ratio for subjects follow-up in Viu (4 subjects)
DUP11/B
11C-4
7/29/83
-------
APPENDIX 11-D
REPORT
OF THE
NHANES II TIME TREND ANALYSIS REVIEW GROUP
June 15, 1983
SRD/NHANES (llD-l) 6/22/83
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
Environmental Criteria and Assessment Office (MD-52)
Research Triangle Park, North Carolina 27711
The materials contained in this report were generated as the result of critical
evaluations and deliberations by members (listed below) of the NHANES II Time Trend
Analysis Review Group. All members of this Review Group unanimously concur with
and endorse the findings and recommendations contained in the present report as
representing the collective sense of the Review Group.
Dr. Joan Rosenblatt (Chairman)
Deputy Director
Center for Applied Mathematics
National Bureau of Standards
Washington, D. C. 20234
Dr. Harry Smith, Professor
Chairman, Department of
Biomathematical Science
Mt. Sinai School of Medicine
New York, New York 10029
Dr. Richard Royall, Professor
Department of Biostatistics
Johns Hopkins University
615 North Wolfe Street
Baltimore, Maryland 21205
Dr. J. Richard Landis, Professor
Department of Biostatistics
School of Public Health II
University of Michigan
Ann Arbor, Michigan 18109
Dr. Roderick Little
American Statical Assoc. Fellow
Bureau of Census
Department of Commerce
Washington, D. C.
(11D-2)
-------
Table of Contents
Summary i i
Introduction 1
Time Trends in Blood-Lead Values 2
Measurement Quality Control 2
Nonresponse 3
Survey Design 3
Sample Weights 5
Estimated Time Trends 6
Summary 6
Correlation Between Blood-Lead and Gasoline-Lead Levels 7
Preliminary Remarks 7
Variables Used in the Analyses 8
Statistical Techniques Used in the Analyses 11
Models Used in the Analyses 11
Gasoline Lead as a Causal Agent for the Decline
in Blood-Lead Levels 12
Use of NHANES II Data for Forecasting Results of
Alternative Regulatory Policies 13
Summary 13
References 14
Appendix 01 - Questions for the Review Group 15
Appendix D2 - Documents Considered by the Review Group 16
Appendix D3 - List of Attendees at Review Group Meetings 19
(nd-3)
-------
Summary
The Review Group finds strong evidence that there was a substantial decline in
the average level of blood lead in the U.S. population during the NHANES II
survey period. After adjustment for relevant demographic covariables, the
magnitude of the change can be estimated for the total U.S. population and for
some major subgroups, provided careful attention is given to underlying model
assumptions.
The Review Group also finds a strong correlation between gasoline-lead usage
and blood-lead levels. In the absence of scientifically plausible alternative
explanations, the hypothesis that gasoline lead is an important causal factor
for blood-lead levels must receive serious consideration. Nevertheless,
despite the strong association between the decline in gasoline-lead usage and
the decline in blood-lead levels, the survey results and statistical analyses
do not confirm the causal hypothesis. Rather, this finding is based on the
qualitatively consistent results of extensive analyses done in different but
complementary ways.
The gasoline lead coefficient in regressions of blood-lead levels on that
variable, adjusted for observed covariates, has been used to quantify the
causal effect of gasoline lead on blood-lead levels. The Review Group
considers that such inferences require strong assumptions about the absence of
effects from other unmeasured lead sources, the adequacy of national gasoline
lead usage as a proxy for local exposure, and the adequacy of a sample design
which does not measure changes in blood-lead levels for individuals in the
sample. The validity of these assumptions could not be determined froir. the
NHANES II data or from other data supplied to the Review Group. Furthermore,
the Review Group cautions against extrapolation of the observed relationship
beyond the limits of the four year period.
(11D-4)
-------
Introduction
This Review Group was appointed in February, 1983 by the Director of the
Environmental Criteria and Assessment Office, U.S. Environmental Protection
Agency (EPA), to consider a series of questions about the interpretation of
data from the second National Health and Nutrition Examination Survey (NHANES
II) to evaluate relationships over time between blood-lead levels and gasoline
lead usage. The questions addressed to the Review Group are listed in full in
Appendix Dl.
Documents describing NHANES II, analyses of the survey data, and analyses of
the relationships between blood-lead values and gasoline lead usage were
furnished for review. In two meetings, on March 10-11 and March 30-31, 1983,
the Review Group discussed these materials with officials of the EPA, and with
specialists from the several institutions that had conducted these studies.
The documents provided for review are listed in Appendix D2. The individuals
who attended the two meetings are listed in Appendix D3.
The panel members of the Review Group are statisticians with experience in
applications of statistics in the physical, biomedical, and social sciences,
but had no previous involvement in analyses of data about blood lead or
gasoline lead. The affiliations of the panel members are listed in Appendix
D3 for identification; views expressed by the panel in this report are their
own and not those of the institutions.
Agencies involved in the conduct of the NHANES II were the National Center for
Health Statistics (NCHS), the Centers for Disease Control (CDC) where the
chemical analyses were done, and the Food and Drug Administration (FDA).
Contributors to the analysis of the association between blood lead and
gasoline lead usage, in addition to NCHS and CDC, are E. I. DuPont de Nemours
& Co. (DuPont), The Ethyl Corporation (Ethyl), and the EPA Office of Policy
Analysis working in collaboration with ICF Incorporated (ICF) and Energy and
Resource Consultants, Inc. (ERC).
This report contains two major sections. The first, on time trends in
blood-lead levels, addresses a set of questions about the use of NHANES II
data to estimate changes over time. The second addresses statistical aspects
of evaluating the relationship of changes in blood-lead levels to gasoline
lead usage.
-1- 7/29/83
(11D-5)
-------
Time Trends In Blood-lead Values
At its first meeting on March 10-11, 1983, the Review Group considered only
the first of the set of questions presented to it (see Appendix 01), namely
questions about the extent to which the NHANES II data could be used to
"determine time trends for changes in nationally representative blood-lead
values for the years of the study (1976-1980)."
The phrases "define time trends" and "determine time trends ... (1976-1980)"
are interpreted throughout this report to mean "estimate changes in blood-lead
values during the survey period." In particular, such changes are not to be
interpreted as trends that might be extrapolated.
The Group recognized that the survey was designed as a cross-sectional survey,
and specifically inquired into three general kinds of possible sources of
time-related bias:
- the measurement quality control,
- the nonresponse experience, and
- the survey design.
As would be expected, only incomplete evidence could be made available in each
of these areas. The following assessment of this evidence indicates where it
depends on the expert opinion of others.
Measurement Quality Control
In order to analyze the time trends in NHANES II data, one must assume that
the procedures for collecting, handling, and analyzing blood specimens did not
change during the survey years. The Review Group is aware that contamination
can produce spuriously high values in determination of trace elements, and
sought evidence that quality control procedures were equally stringent at all
times.
Although no quality control specimens were prepared at the medical examination
sites, the Review Group has been assured that training, periodic retraining,
materials, equipment, and procedures were designed to prevent contamination,
and not changed. There was some turnover of personnel.
The CDC laboratory established and documented the results of extensive quality
control sampling (App. 02, item 14). The data on lead levels in the "blind"
samples, from two pools of bovine blood, exhibit essentially constant means
and standard deviations. The coefficient of variation for measurement error
was found to be about 17 percent for blood-lead levels near 13 ug/dL; it was
smaller, about 13 percent, for higher blood-lead levels near 25 M9/dl.
Additional evidence of the constancy of quality control is that data from
other analyses of the blood specimens (zinc, for example) exhibit little or no
change over time.
The Review Group finds no evidence that field and laboratory quality control
changes could account for the observed change in blood-lead levels.
-2- 7/29/83
(11D-6)
-------
Nonresponse
Nonresponse is an important potential source of bias in sample surveys. It is
of particular concern in the blood-lead analysis of the NHANES II since the
nonresponse rate is high--39.3 percent of sampled persons had missing lead
values due to nonresponse at various stages of participation in the survey
(App. D2, item 14, p.9). The NCHS attempted to adjust for nonresponse by
weighting responding individuals by estimates of the probability of response,
calculated within subclasses of the population formed by joint levels of age,
income, SMSA/non-SMSA, and region.
This is a standard adjustment method for unit nonresponse in surveys. The
method adjusts for differential nonresponse across the subclasses used to
calculate the weight, but does not account for residual association between
nonresponse and time and blood-lead level, which are the variables of primary
interest in the analysis under consideration. Thus there is the possibility
that nonresponse bias is a contributory factor to the trend in blood-lead
levels across time.
In order for nonresponse to have this effect it is necessary that, after
adjusting for the socioeconomic variables used to define the weights,
nonresponse be related to blood-lead level, and further that this relationship
change over time, so that a differential bias in the mean blood-levels of
respondents exists across time. Clearly this question cannot be addressed
directly, since the blood-lead levels of nonrespondents are not measured.
However, the Review Group considered such an interaction to be highly
unlikely, for the following reasons:
0 Nonresponse rates did not vary in a consistent way across
time. Examination of changes in response rates does not
indicate any relationship of importance (App. D2, item 18).
0 There does not appear to be evidence that the conditions of
the survey changed significantly across time, so that any bias
introduced by an association between nonresponse and
blood-lead level is unlikely to change across time.
Accordingly, the Review Group rejected nonresponse as a likely explanation for
the trend observed in the data.
Survey Design
The NHANES II was designed to provide U.S. national prevalence rates for a
wide range of characteristics and health conditions. Due to financial and
logistical constraints, the survey design required a four-year data collection
period. Consequently, the sample quantities, such as the blood-lead levels,
necessarily will provide period prevalence estimators, rather than point
prevalence estimators of the underlying population parameters. In general
practice, a fundamental assumption underlying the use of period data to
generate prevalence estimators is that the condition under investigation
remains relatively constant throughout the survey period.
-3- 7/29/83
(11D-7)
-------
Even though the NHANES II was not designed to detect and estimate changes in
prevalence throughout the survey period, one must consider the possibility
that the level of a particular target characteristic, such as blood lead,
actually may be changing over time. Consequently, one cannot ignore evidence
suggesting that the level of lead in blood in the U.S. population was
decreasing during the data collection period simply because the survey design
was cross-sectional, rather than longitudinal. Rather, the difficult question
is to what extent, if any, can these NHANES II data be used to determine time
trends.
Although a cross-sectional design such as the one utilized in the NHANES II
certainly is not optimal for investigating time trends, one can consider
making adjustments within the sample for the effects of relevant covariables
such as age, sex, race, residence, and income, if the distributions of these
covariables are not highly confounded with time. An additional requirement
for making adjustments is that there be reasonably large numbers of sample
persons for different covariable levels at various times. These internal
adjustments permit one to examine whether the decline in blood-lead levels can
be accounted for by differing proportions of individuals from subgroups
determined by relevant covariables. The extent of this type of selection bias
over time relative to primary demographic characteristics can be summarized
(App. D2, item 20, Tables M7, M8 for whites, and M13, M14 for blacks).
The Review Group considered carefully the potential bias due to changing
composition of the sample over time, especially since this had been emphasized
by Ethyl (App. D2, items 25, 26). The most striking problem occurs with urban
vs. rural groups. The fractions of blood samples obtained from white urban
residents are shown as follows:
% urban bloods Sample size
Jan - Jun 1976 64.2 795
Jul - Dec 1976 36.9 1255
Jan - Jun 1977 44.6 935
Jul - Dec 1977 57.3 1010
Jan - Jun 1978 46.3 1056
Jul - Dec 1978 40.6 981
Jan - Jun 1979 31.6 1228
Jul - Dec 1979 20.7 842
Jan 1980 0.0 267
Thus, there has been a striking decrease in the number of bloods taken from
white urbanites across the four years. If one assumes that exposure to lead
from gasoline is more prevalent in urban areas, then (without adjustment) the
observed mean blood levels across the four years would be biased because of
the NHANES II schedule.
Further examination of the CDC tabulation (App. 02, item 20) indicates sparse
information on blacks. The numbers are so small that time trend inferences
for blacks can be estimated with confidence only for overall mean blood-lead
level results without regard to sex, place of residence, and age.
(lfB-8)
7/29/83
-------
The Review Group finds that despite obvious trends over time for such
characteristics as degree of urbanization and the proportion of children aged
0.5 to 5 years, the sample size is distributed across the grid of covariable
levels sufficiently to permit reasonable adjustments. In support of this
finding, the Review Group notes that similar trends appeared whenever
demographic subgroups were examined separately. These subgroups included
white males, white females, white children, white teenagers, white adults, and
blacks, as well as breakdowns by income and urban-rural status.
Sample Weights
Another possibility is that the sample mean blood-lead level changes resulted
from trends in more subtle statistical characteristics of the sample over
time, such as characteristics related to the way sample weights are used to
calculate averages. But this explanation appears to be inconsistent with the
fact that analyses of the unweighted NHANES II data lead to essentially the
same results as the weighted data and analysis.
In response to questions raised by both industry representatives and other
observers, the Review Group explored the effects of the complex weighting
scheme inherent in all the CDC and EPA/ICF analyses. Each sample observation
has both a basic weight (related to the probability of selection), a final
weight (reflecting additional adjustments to the basic weight accounting for
nonresponse patterns of selected demographic subgroups), and a final examined
lead subsample weight (corresponding to the entire set of adjustments due to
the probability of selection, nonresponse, and post-stratification, and the
subsampling of individuals selected for the measurement of blood lead). All
the weighted analyses in the CDC and EPA/ICF reports were conducted relative
to the final examined lead subsample weight.
One potential problem associated with this final lead subsample weight is the
possibility that differential nonresponse patterns for various demographic
subgroups may lead to marked differences between the basic weight (without
nonresponse adjustments) and this final weight. For that reason, the Review
Group requested a data display of the total nonresponse rate and the average
blood-lead levels by the 64 separate stands using three different weighting
schemes in computing the averages:
i) unweighted;
ii) basic weights;
iii) final lead subsampling weights.
As shown in Table 1, item 18 of App. D2, the average blood-lead levels are
quite consistent under each weighting scheme for each of the 64 stands.
Furthermore, there is no apparent trend in the nonresponse rate across time.
Consequently, one would expect that an analysis of these data under the basic
weights also would parallel the results obtained in the CDC and the ICF
reports.
These findings, in conjunction with the similarities between the weighted and
unweighted analyses, lend additional support to the overall consensus among
panel members that these data analyses are not dependent on the particular
choice of weights, including the intermediate basic weights.
-5- 7/29/83
(11D-9)
-------
Estimated Time Trends
There seems to be no doubt that, qualitatively, a downward trend of blood-lead
levels has been observed during the NHANES II survey.
The data appear to support reasonably precise estimates of the magnitude of
lliL Jiange for a few major subgroups of the population. In particular, the
change in mean blood-lead levels during the survey period can be estimated for
the population as a whole and for population sectors grouped by age, sex,
race, urban/rural, and income, if each of these demographic categories is
considered separately.
For estimating changes in mean blood-lead levels for combinations of
demographic factors, sufficient data appeared to be available for white-by-sex
and white-by-age breakdowns. These estimated changes, and others that might
be considered, can be made on the basis of a linear model that provides
adjustments for demographic and socioeconomic covariables that are known or
believed to be associated with blood-lead levels.
For finer subdivisions, estimates of change are subject to large sampling
error and are sensitive to correct specification of the regression model.
Hence, caution must be exercised in their interpretation. It is not possible
to show time changes in mean blood levels for specific cities, towns, or
locales using the NHANES II data, since no city or locale was sampled more
than once. No data which would allow estimates of time trends in mean
IjlcoJ-lead levels for different occupational categories were shown to the
Review Group. The only socioeconomic variable considered was income.
Estimates of change, e.g., those reported by CDC (App. D2, item 14, Table 6,
page 44), should be accompanied by standard errors. There should be
discussions of the use of regression diagnostics to evaluate the adequacy of
the model, and the possibility that a few observations exert an excessive
influence on the result. The calculation of standard errors should use
procedures that take into account the stratification and clustering properties
of the survy design. In response to the Review Group's questions, CDC
provided a document presenting standard errors and the methodology used to
estimate them (App. D2, item 38). The size of these standard errors suggests
that there are only weak indications of differences between subgroups with
respect to the percent drop in the average blood-lead level.
Summary
Although the survey was not specifically designed to measure trends, data from
the NHANES II can be used to estimate changes in blood-lead levels during the
four-year period, 1976-1980, of the survey. Changes can be estimated for the
U.S. population and for major population subgroups, as specified in the
previous subsection. Because of sampling error, laboratory measurement error,
a high nonresponse rate, and the need to adjust for time-related imbalance in
the survey design, such estimated changes should be interpreted with caution.
-6- 7/29/83
(11D-10)
-------
Correlation Between Blood-Lead and Gasoline-Lead Changes
At its second meeting on March 30-31, 1983, the Review Group considered three
sets of studies that examine the association between changes in blood-lead
levels estimated from the NHANES II data and changes in the use of leaded
gasoline:
- the Ethyl Corp. analysis (App. 02, items 25, 26)
- the ICF/EPA analysis (App. D2, items 11, 22, 23, 24), and
- the CDC/NCHS analysis (App. 02, item 14 and appendices).
The following discussions summarize the Review Group's assessment of the
strengths and weaknesses of the analyses.
Preliminary Remarks
The analyses propose and evaluate models for the relationship between
blood-lead levels and gasoline-lead usage. All of these analyses rely on
multiple linear regression methods, whose limitations with respect to
establishing causal relations are well known (See, e.g., reference 1). The
statistician-reviewer may adopt one or the other of two approaches in
considering the strengths and weaknesses of the several analyses:
(1) Assume (on external authority) the existence of a causal
relationship between gasoline lead usage and blood lead levels. Consider the
variables and models used to analyze the strength of the association and to
estimate the effect of gasoline-lead changes on blood-lead changes. In this
approach, the possible effects of other changes over time that affect
blood-lead levels are treated as second-order effects. CDC urges this
approach.
(2) Adopt a neutral position as to the causal relationships, and examine
the associations among the variables studied. In this approach, "time" serves
as a proxy for the combined effect of whatever changes affected blood-lead
levels and it is left to the interpreter of the analyses to assign relative
importance among suggested explanations for changes over time. DuPont and
Ethyl suggest this approach.
The ICF and CDC analyses both found a clear relationship between gasoline lead
and blood lead. The Ethyl analysis found no evidence of association between
these variables. The purpose of this commentary is to discuss the important
differences between the analyses and to assess their utility in establishing
or contradicting the hypothesized relationship between the decline in
blood-lead levels and the decline in gasoline lead emissions over the period
of the NHANES II Survey.
Table I (next page) classifies the three analyses by six factors which capture
the main differences between them, namely: 1) the choice of measure of
gasoline lead, 2) the scale of blood lead variable, raw or logarithm, 3) the
unit of analysis, 4) control variables in the regression, and in particular
-7- 7/29/83
(11D-11)
-------
the inclusion or omission of a time variable, 5) the weighting used in the
regressions, and 6) the method used to calculate standard errors. The panel
concludes that of these factors only (1) and (4) had a substantial impact on
the final results.
Table 1
CDC ICF Ethyl
1) measure of gasoline quarterly monthly sales pop. density
lead x lead cone. local lead usage
2) scale of dependent log raw raw
variable
3) unit of analysis individual individual individual stage 1
locality stage 2
4) control variables no time, season, time
include time lagged gas
5) weighting by both yes no
selection probs.
6) design based yes yes no
standard errors
The first three factors are discussed under the heading "Variables Used in the
Analyses". Factors (4), (5), and (6) are discussed under "Statistical
Techniques Used in the Analyses". Factor (4) is considered further in the
assessment of "Models Used in the Analyses".
Variables Used in the Analyses
Demographic and socioeconomic covariables were used as defined for the NHANES
II Survey. Differences between the analyses occurred in the choice of
specific representations for blood-lead levels and gasoline lead usage.
Blood Lead. All the studies used blood-lead values for individuals from the
NHANES II Public Use Data Tape, with associated demographic, economic, time,
and sampling-weights data.
Ethyl calculated adjusted blood-lead values for its principal analysis by
fitting a linear model to adjust for age, sex, race, and income to obtain the
residuals from this analysis. Ethyl did not adjust the individual data for
the effect of the degree of urbanization, a factor recognized to be related to
blood-lead levels. Averages of the adjusted values for 55 of the 64
examination sites were used in the principal (second-stage) analysis.
-8- 7/29/83
(11D-12)
-------
ICF used the NHANES II blood leads without adjustment or transformation.
Adjustment for socio-demographic variables was achieved by including these
variables as covariates in regression models for individual blood leads.
CDC adopted a similar approach, but used the natural logarithms of the NHANES
II blood leads, on the basis of an analysis showing that the distribution of
the values themselves was skewed and that the transformation successfully
corrected for the skewness.
The scale of the dependent variable (raw or logarithm) does not appear to have
a great influence on the fi.nal results. With the exception of race, the
blood-lead/gasoline-lead slope in the CDC and ICF analyses appeared stable
across demographic factors, whether the raw or log scale was used for the
dependent variable. The logarithm scale has the advantage of being more
likely to yield normal residuals.
The unit of analysis (factor 3) received a considerable amount of discussion
by reviewers. In particular, the Ethyl two-stage analysis was subjected to
some criticism. At the first stage, the blood lead variable was adjusted for
differences in the distributions of demographic variables by an indiyiudal
level regression on NHANES II data. At the second stage, the adjusted
locality mean blood-lead values were regressed on proxies for gasoline lead
which had not themselves been adjusted for the demographic variables. This
two-step regression procedure leads to bias (see reference 2), but the bias
does not appear important, as Ethyl later corrected the analysis with no
substantial change in the results.
Gasoline Lead Usage/Exposure. There were several different approaches to
defining variables that could be interpreted as indexes of the amount of lead
present in the environment at the time when blood samples were taken, as well
as during the antecedent months. Clearly, no index number or set of index
numbers can serve as an ideal surrogate for a measurement of the exposure
experiences of sampled persons. The Review Group recognizes the complexity of
the mixture of lead sources and uptake pathways.
The large differences between the results of the ICF/CDC analyses and the
Ethyl analysis are caused by different measures of gasoline lead exposure.
ICF and CDC used national period measures-quarterly EPA lead additive data for
CDC and adjusted monthly gasoline sales data for ICF, whereas Ethyl used two
proxy measures for lead exposure at each locality-population density and lead
use per unit area.
A fundamental assumption underlying the creation of a local estimate of
gasoline lead exposure is the notion that the volume of leaded gasoline
consumed locally, with the resulting "fallout", is the primary source of lead
in human blood. Although this determination requires substantive expertise
beyond that on our Review Group, the choice of a local vs. a global measure of
exposure is a pivotal one in all these analyses. If, in fact, lead enters the
human blood system via imported fallout through the food chain (and other
sources), as well as the inhalation of local "fallout", then ideally one would
require a summary measure of exposure which captures both of these sources.
-9- 7/29/83
(11D-13
-------
CDC used data from the quarterly EPA Lead Additive Reports (App. D2, item 14,
pages 37-40 and Appendix H). These are national values of the total amount
(by weight) of lead used in gasoline production. The series exhibits seasonal
fluctuations in gasoline production in addition to a general downward trend.
ICF developed a monthly series of national values of the average amount (by
weight) per day of lead used in gasoline, as follows: Monthly average
gasoline use (liquid volume per day) was obtained from the DOE Monthly Energy
Review. Quarterly values of the concentration of lead in gasoline (grams per
gallon, based on refiner reports) were obtained from EPA (App. D2, item 11).
The product of these produced a monthly series. This series, if aggregated to
a quarterly series, would be closely related to the series used by CDC.
The measures of lead use used by CDC and ICF capture the downward trend in
gasoline lead over time, but they suffer from specification error in that they
are national rather than localized measures of gasoline lead exposure. The
defect has two consequences:
(a.) The gasoline lead use variable does not capture variation in gasoline
lead exposure between localities.
(b.) The lead use variable can be only partially adjusted for correlations
with the demographic covariates.
The CDC analysis partially corrects for (a) by aggregating the gasoline lead
exposure over all sampled localites in a six month period of sampling. The
second problem remains, however. The panel does not believe that these
deficiencies invalidate the qualitative findings of a relationship between
lead usage and blood lead. However, the impact on the coefficient of lead
usage in the CDC analysis is not clear.
Ethyl adopted a different approach, seeking to represent gasoline-lead usage
at the survey locations and also to consider separately the effects of lead in
air and lead fallout. The variables used to represent the two kinds of lead
exposure were, respectively, population density and gasoline lead usage per
square mile for the sampled localities.
The Review Group applauded the intention of the Ethyl effort, but the
variables selected appear to be inappropriate. In the Ethyl discussion (App.
D2, item 26, Appendix page A-3) it is pointed out that population density is
strongly related to degree of urbanization, a factor for which adjustment is
made in the CDC and ICF analyses, but not in the Ethyl analysis. Furthermore,
Ethyl calculated population density by interpolation between censuses and it
is doubtful that it would reflect changes (if any) in the concentration of
lead in air within the four-year survey period.
Ethyl represented lead usage per unit area by annual values by state.
Department of Transportation reports of annual gasoline sales (by state) and
annual Ethyl estimates of the amount of lead in gasoline being sold (by state)
produced state estimates of annual totals of lead used. These were then
divided by the area of the state. Examination of the resulting values (App.
D2, item 26, Table 6, page 23) reveals anomalies. For example, the 1979 lead
usage value for Washington, DC, is 5 times larger than that for any other
-10- 7/29/83
(11D-14)
-------
location. The second-largest value is the one for New Jersey in 1977, used
for locations adjacent to New York City; it is more than 4 times the 1977
value used for both New York City and its Westchester County suburbs. As
another example, the computed exposure for Houston, TX (ID no. 28) is 101,
compared to 7174 for Washington, DC (ID no. 33). The naive implication of
these two data points is that persons living in Washington, DC received a
71-fold (7174/101) increase in dosage of air-lead (or food chain lead)
compared to persons living in Houston, TX. Whether we view this dosage as
exposure through air or food, this extreme differential is highly unlikely.
This variable appears to represent chiefly the statewide average population
density. The Review Group cannot accept it as an indicator of gasoline lead
usage at the sample locations.
Statistical Techniques Used in the Analyses
All final models reported by EPA/ICF and CDC were fitted to the NHANES II data
using the SURREGR procedure available in SAS. This computing software permits
sample weights and cluster design effects to be incorporated into the
variance-covariance estimators of the model parameters. Although unweighted
and weighted ordinary least squares model fitting provided the same
conclusions, SURREGR provides better estimates of standard errors for these
complex survey data. This estimation and hypothesis testing strategy is the
most conservative approach, since it will produce larger standard errors for
the parameter estimates due to the clustering in the data. Extensive
empirical investigations of the role of weights and design effects in the
NHANES I survey demonstrated that test statistics are decreased when including
weights, and decreased even further when adjusting for design effects (see
reference 3).
The two-stage procedure adopted by Ethyl was described in the preceding
subsection.
Models Used in the Analyses
There is no unique correct approach to analyzing the relationships within the
NHANES II data or between the NHANES II and other data sets. For this reason,
it has been useful to compare and contrast a variety of approaches and models.
All of the models have the general character that a measure of blood lead is
expressed as a linear combination of a measure (or measure) of exposure to
gasoline lead with various demographic and socioeconomic covariables and
(sometimes) time.
The primary difficulty with the Ethyl analyses (App. 02, item 26) lies in the
choice of constructed gasoline-lead variables. Neither the population density
variable (C19) nor the lead usage variable (C16) is an acceptable measure of
gasoline lead exposure.
The Ethyl report concludes with the observation
In summary, our analysis of the NHANES II data has shown that time
(T) is the major contributor to differences in blood lead between
-11- 7/29/83
(11D-15)
-------
1976 and 1980 ... The major contribution of time to the decrease in
blood lead indicates that other factors that vary with time are the
major causes of the 1976 to 1980 decrease in blood lead and not
gasoline lead usage.
Ironically, national gasoline lead usage (as defined in the CDC or ICF
analysis) is such a variable that varies with time and is known to be
causative of some portion of the lead in blood. The constructed variable
(C16) does not display a similar relationship with time.
The CDC and ICF/EPA analyses are similar in their general approach. In each
case, a variety of models was considered (adding and deleting various subsets
of the covariables and interaction terms). These variations had only minor
impact on the value of the coefficient for the lead usage variable.
Although both the CDC and EPA/ICF analyses used national data on leaded
gasoline sales, the EPA/ICF models utilized a gasoline lead use variable which
was estimated at each month of the survey (App. D2, item 11, Table 1, pp.
13-14). Consequently, since the data collection period for most of the 64
stands in the NHANES II survey spanned across two months, the gasoline lead
use variable could, and in some cases did, assume two different values for the
same site, according to the month of examination. Investigations of the
relationships between time and blood-lead levels involved comparisons within
sites (due to spanning two months), as well as among sites. Thus, even though
there is a high degree of correlation between time and gasoline lead usage,
these two variables are not completely confounded with the 64 different sites.
It is, nevertheless, a significant question whether the time variable is
included in the model as a covariate. The ICF analysis included a linear time
covariable and seasonal effects in the model, "to give the models the ability
to attribute temporal variations in blood lead to effects other than gasoline
lead" (App. D2, item 11, p. 8). Variables for time and gasoline lead were not
included simultaneously in the CDC analysis.
The intent of the ICF procedure is reasonable, but the confounding between
time and gasoline lead in the data make the simultaneous inclusion of these
variables in the model questionable. The data do not allow the relationship
between gasoline lead and blood lead to be estimated at any particular time
point. Thus the attempt to adjust for time is highly dependent on the
specification of the time effects in the model. Despite these problems, two
aspects of the ICF analysis yielded some circumstantial evidence that gasoline
lead is an important agent of the trend in blood lead. The gasoline lead
variable accounted for seasonal variation in blood lead, and the lagged
gasoline lead variables provided a plausible lag structure: the one-month
lagged variable had the strongest association with blood lead.
Gasoline Lead as a Causal Agent for the Decline in Blood-Lead Levels
The CDC and ICF analyses provide strong evidence that gasoline lead is a major
contributor to the decline in blood lead over the period of the NHANES study.
DuPont stressed the limitations of statistical theory and methods as tools for
assessing causal relationships.
-12- 7/29/83
(110-16)
-------
Analysis of the NHANES II data cannot prove whether changes in the use of
leaded gasoline caused a change in average blood-lead levels. Variables X and
Y can be correlated because changes in X cause changes in Y, or vice versa, or
because some third factor, Z, affects both X and Y. There are many other
possibilities as well, but these are enough for this discussion. If X stands
for some measure of average blood lead concentration and Y stands for the
amount of lead in gasoline, we can dismiss the first possibility as absurd.
But the relative plausibility of the other two is a matter for expert
scientific judgement. To date, no hypothesis of the third form which could
explain the NHANES II data has been presented to the panel. One hypothesis of
this form has been discussed. This hypothesis has Z representing regulatory
changes and publicity aimed at reducing lead exposure generally. This could
result in reductions in gas lead, lead in food, lead in paint, etc., and it
could be that the gas lead change had little effect on blood-lead levels --
the blood-lead changes might have been caused by the other factors (food,
paint, etc.). Although this hypothesis cannot be disregarded entirely, it
does not seem to explain the blood-lead drop adequately. We have seen little
evidence that food lead has dropped by a factor large enough to explain a
sizable part of the drop in blood lead. In fact, the FDA diet lead values
shown in the ICF Report (App. D2, item 11, Table 2) were increasing during the
study period. That changes in exposure to leaded paint caused the decrease in
blood-lead observed over all age and sex groups seems highly unlikely. The
existence of influences (other than gasoline lead usage) that are not included
in the models must be recognized as a limiting factor in the evaluation of all
of the analyses.
Use of NHANES II Data for Forecasting Results of Alternative Regulatory Policies
Regression models have been used in all three analyses to see if the NHANES II
time trend in average blood-lead levels can be explained in terms of changes
1n demographic variables or in terms of changes in gas and lead usage.
Extension of the use of these and other statistical techniques "to estimate
the distribution of blood-lead levels of whites, blacks, and black children
and to forecast the results of alternative regulations," as in Section III of
the ICF Report of December, 1982 (App. 02, item 11), raises questions and
involves assumptions that go much further than those the Review Group was able
to consider. In general, the Review Group would warn that the weaknesses that
have been discussed in the context of analyzing relationships within the
four-year survey period become enormously greater in any attempt to
extrapolate beyond that period. For example, the cautions mentioned in the
ERC review (App. D2, item 22, p. 6) of the ICF analysis probably do not go far
enough.
Summary
In general, there is a significant correlation between gasoline-lead levels
and blood-lead levels in persons examined in the NHANES II Survey. Major
obstacles interfere with the use of the available data to describe the
relationship. They are: the need to perform model-based adjustments to
compensate for imbalance in the design of the NHANES II, the possibility of
specification error in the regression models, and the lack of a satisfactory
measure of individual or local exposure to gasoline lead, in addition to
sampling error, laboratory measurement error, and the high nonresponse rate.
-13- 7/29/83
(11D-17)
-------
The Review Group finds that the Ethyl analyses contribute little to
understanding the association between blood lead and gasoline lead because the
variables adopted to represent lead exposure are deemed inappropriate.
The CDC and ICF/EPA analyses relating the NHANES II blood-lead data to a
national measure of the amount of lead used in gasoline indicate that the drop
in average blood-lead levels can be explained, in large part, by the
concurrent drop in gasoline lead. This by no means confirms the hypothesis
that the blood lead decrease was caused by the decrease in gasoline lead but,
in the absence of scientifically plausible alternative explanations, that
hypothesis must receive serious consideration.
References
Literature cited in this report, in addition to the documents furnished by the
EPA which are listed in Appendix D2.
(1) Ling, R. F. (1982). A review of Correlation and Causation by David A. Kenny,
John Wiley & Sons. J. Am. Statis. Assoc. 77, 490-491.
(2) Goldberger, A. S. (1961). Step wise Least Squares: Residual Analysis and
Specification Error. J. Am. Statis. Assoc. 56, 998-1000.
(3) Landis, J. R., Lepkowski, J. M., Eklund, S. A. and Stehouwer, S. A. (1982).
A General Methodolody for the Analysis of Data from the NHANES I Survey.
Vital and Health Statistics. NCHS Series 2- No. 92. DHHS Publ No. (PHS)
82-1366. Washington. U.S. Government P7TntTng~0~ffice.
-14- 7/29/83
(11D-18)
-------
Appendix Dl
Questions for the Review Group
The following questions were stated in letters to members of the Review Group
from Dr. Lester D. Grant, Director of the EPA Environmental Criteria and
Assessment Office, February 17, 1983.
1. To what extent is it valid to use the NHANE5 II data to determine time
trends for changes in nationally representative blood-lead values for the
years of the study (1976-1980)? More specifically, to what extent can the
NHANES II data appropriately be used to define time trends for blood-lead
levels (aggregated on an annual, semiannual, or any other time-related basis)
for the total NHANES II sample (all ages, sexes, races, etc.) or for
subsamples defined by the following demographic variables: (1) age (e.g.,
children <6 years old, children 6-12 years old, adults by 10- or 20- year age
groups); (2) sex; (3) race; (4) geographic location (e.g., urban vs. rural
residence; Northeast vs. Southeast, Midwest, or other large regional areas of
the U.S.; residence in specific cities, towns, or rural locales); (5)
socioeconomic status; (6) occupation of respondents or their parents/head of
household at main residence; or (7) any combination of such demographic
variables (e.g., black children <6 years or white children <6 years old living
in urban or rural areas, etc.).
2. If it is indeed possible to derive such time trends from the NHANES II
data, to what extent can the changes in NHANES II blood-lead levels over time
be correlated credibly with changes in the usage of leaded gasoline over the
same time period (i.e., the years 1976-1980)? Several analyses of this type
have already been conducted and submitted to us, and we would appreciate your
evaluation of those analyses.
3. Are there any other appropriate credible statistical approaches or
analyses, besides those alluded to as already having been done, that might be
carried out with the NHANES II data to evaluate relationships over time
between blood-lead levels and gasoline lead usage?
-15- 7/29/83
(11D-19)
-------
Appendix D2
Documents Considered by
NHANES II TIME TREND ANALYSIS REVIEW GROUP
1. Plan and Operation of the Second National Health and Nutrition Examina-
tion Survey. (1976-1980) National Center for Health Statistics, Series 1,
No. 15. July, 1981.
2. Public Use Data Tape Documentation. Hematology and Biochemistry, catalog
number 5411. NHANES II Survey, 1976-1980, NCHS. July, 1982.
3. NHANES II Weight Deck (one record for each SP). Deck #502. Attachment
I, NCHS.
4. NHANES II Sampling Areas. Document furnished by NCHS during site visit,
March 10, 1983.
5. Steps in Selection of PSU's for the NHANES II Survey. Document furnished
by NCHS during site visit, March 10, 1983.
6. Location of Primary Sampling Units (PSU) chronologically by pair of cara-
vans: NHANES II Survey, 1976-80. Document furnished by NCHS during site
visit, March 10, 1983.
7. Annest, J. L. et al. (1982) Blood lead levels for person 6 months - 74
years of age: United States, 1976-1980. NCHS ADVANCEDATA, No. 79, May 12,
1982.
8. Mahaffey, K. R. et al. (1982) National estimates of blood lead levels:
United States, 1976-1980. Association with selected demographic and socio-
economic factors. New England Journal of Medicine 307: 573-579.
9. Average Blood Lead Levels for White Persons, 6 months - 74 years strat-
ified chronologically by PSU's: NHANES II, 1976-80 by caravan. "Graph"
furnished by NCHS, March 17, 1983.
10. Schwartz, J. The use of NHANES II to investigate the relationship between
gasoline lead and blood lead. Memo to David Weil (ECAO) (March 3, 1983).
11. ICF Report: The Relationship between Gasoline Lead Usage and Blood Lead
Levels in Americans: A Statistical Analysis of the NHANES II Data..
December 1982.
12. Annest, J. L. et al. (1983) The NHANES II study. Analytic error and its
effect on national estimates of blood lead levels.
13. Pirkle, J. L. Comments on the Ethyl Corp. analysis of the NHANES II data
submitted to EPA October 8, 1982 (Feb. 26, 1983).
14. Pirkle, J. L. Chronological trend in blood lead levels of the second
NHANES, Feb. 1976-Feb. 1980 (Feb. 26, 1983).
-16- 7/29/83
(11D-20)
-------
15. Lynam, D. R. Letter to David Weil dated October 15, 1982 containing ad-
ditional comments on NHANES II data.
16. E. I. DuPont de Nemours & Co., Inc. Supplementary statement presented to
EPA in the matter of regulation of fuel and fuel additives - lead phase-
down regulations proposed rulemaking (Oct. 8, 1982).
17. Pirkle, J. L. An expanded regression model of the NHANES II blood lead
data including more than 100 variables to explain the downward trend from
Feb., 1976-Feb., 1980 (Dec. 23, 1982).
18. Annest, J. L. et al. Table 1. Average blood lead levels and total non-
response rates for persons ages 6 months - 74 years stratified chrono-
logically by primary sampling unit (PSU): NHANES II, 1976-1980 (Corrected
version; April 8, 1983).
19. Pirkle, J. L. (1983). Duplicate measurements differing by more than 7
mg/dl in the lead measurements done in NHANES II Survey. Document fur-
nished by CDC at Panels request, March 18, 1983.
20. Pirkle, J. L. Appendix M: Tabulation by demographic variables (March 18,
1983).
21. Pirkle, J. L. Appendix N: Regression analysis of urban and rural popu-
lation subgroups (March 18, 1983).
22. Miller, C. and Violette, D. Comments on studies using the NHANES II data
to relate human blood lead levels to lead use as a gasoline additive
(March, 1983).
23. Miller, C. and Violette, D. (March 4, 1983). The Usefulness of the
NHANES II Data for Discerning the Relationship between Gasoline Lead
Levels and Blood Lead Levels in Americans and a Review of ICF's Analysis
using the NHANES II Data. Energy and Resource Consultants, Inc. ;
Boulder, Colorado.
24. Schwartz, J. Analysis of NHANES II data to determine the relationship be-
tween gasoline lead and blood lead. Memo to David Weil (ECAO). (March
18, 1983).
25. Excerpt - (Section I. C. - "Discussion of NHANES II Blood Lead Data")
from the Ethyl submission to the EPA's docket on the Lead Phasedown dated
May 14, 1982.
26. Excerpt - (Section III. A. - entitled "Correlation of Blood Lead to Gaso-
line Lead" and Appendix "Discrete Linear Regression Study") from the
Ethyl submission to EPA's docket on the Lead Phasedown. (October 8, 1982)
27. Ethyl Analyses of the NHANES II Data. This item was distributed at the
Criteria Document meeting held on January 18-20, 1983.
28. Comments by Dr. Norman R. Draper on Ethyl Corporation's comments and ICF,
Inc.'s comments.
-17- 7/29/83
(11D-21)
-------
23. Comments by Dr. Ralph A. Bradley entitled "A Discussion of Issues and
Conclusions on Gasoline Lead Use and Human Blood Lead Levels".
30. Comments by Dr. Ralph A. Bradley in a letter to B. F. Fort. (Ethyl Corp.)
31. Ethyl Corp. NHANES II - blood lead data correlation with air lead concen-
tration data.
32. Ethyl Corp. Summary of analyses of the NHANES II blood lead data (Janu-
ary, 1983).
33. E. I. DuPont de Nemours & Co. Comments submitted March 21, 1983.
34. E. I. DuPont de Nemours & Co. Comments by R. Snee and C. Pfieffer on
paper by Annest et al. on analytic error (see item #5).
35. Pirkle, J. L. The relationship between EPA air lead levels and population
density. (March, 1983).
36. Pirkle, J. Consecutive numbering of points on plots of 6-month average
NHANES II blood lead levels versus 6-month total lead used in gasoline
(April 11, 1983).
37. Pirkle, J. L. Distribution of the NHANES II lead subsample "weight" vari-
able (April 11, 1983).
38. Pirkle, J. L. Appendix 0: Propagation of error in calculating the percent
decrease in blood lead levels over the NHANES II survey period (April 11,
1983).
39. Pirkle, J. L. Appendix P: Regressing In (blood lead) on the demographic
covariates and then regressing the residuals on GASQ compared to regres-
sing In (blood lead) simultaneously on the demographic covanates + GASQ
(April 11, 1983).
40. Pirkle, J. L. Appendix Q: Regression of In (blood lead) on the demo-
graphic covariates only and subsequently adding GASQ: F statistics, R
square and Mallows C (p) (April 11, 1983).
-18- 7/29/83
(110-22)
-------
Appendix 03
List of Attendees at March 10-11 and March 30-31, 1983
meeting of
NHANES II TIME TREND ANALYSIS REVIEW GROUP
Panel Members
Joan Rosenblatt (Chairman)
National Bureau of Standards
J. Richard Landis
University of Michigan
Roderick Little
Bureau of the Census
Richard Royal 1
Johns Hopkins University
Harry Smith, Jr.
Mt. Sinai School of Medicine
David Weil (Co-chairman)
U.S. EPA
Observers
Dennis Kotchmar*
U.S. EPA
Vic Hasselblad
U.S. EPA
Allen Marcus
U.S. EPA
George Provenzano
U.S. EPA
Joel Schwartz
U.S. EPA
Earl Bryant*
NCHS
Trena Ezzote*
NCHS
J. Lee Annest
NCHS
Mary Kovar*
NCHS
Bob Casady*
NCHS
Jean Roberts*
NCHS
*attended March 10-11 meeting only.
tattended March 30-31 meeting only.
Robert Murphy
NCHS
Vernon Houkt
Centers for
James Pirkle
Centers for
Disease Control
Disease Control
Don Lynam
Ethyl Corporation
Ben Forte
Ethyl Corporation
Jack Pierrard*
DuPont
Chuck Pfieffer
DuPont
Ron Snee
DuPont
Asa Janney
ICF
Kathryn Mahaffey*
FDA
KIM - 6tl-f»4/100l
-19-
(110-23)
7/29/83
------- |